University of Ghana http://ugspace.ug.edu.gh ENHANCING GENETIC VARIABILITY FOR PHOSPHORUS-USE EFFICIENCY IN SORGHUM (Sorghum bicolor L. Moench) USING CHEMICAL MUTAGENESIS BY AZU, ELAINE (10235622) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF DOCTOR OF PHILOSOPHY DEGREE IN PLANT BREEDING WEST AFRICA CENTRE FOR CROP IMPROVEMENT COLLEGE OF BASIC AND APPLIED SCIENCES UNIVERSITY OF GHANA LEGON December, 2017 i University of Ghana http://ugspace.ug.edu.gh 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. ……………………………….. AZU, ELAINE (Student) ....………………………… PROF HARRY M. AMOATEY (Supervisor) ………………………… PROF ERIC Y. DANQUAH (Supervisor) ………………………… DR. ERIC NARTEY (Supervisor) ii University of Ghana http://ugspace.ug.edu.gh ABSTRACT In Ghana, sorghum (Sorghum bicolor L. Moench) ranks third after maize and rice, mainly as a dietary staple and is crucial for ensuring food security in the three northern regions. However, yields in farmers’ fields are extremely low (less than 2.0 tonnes ha-1) compared to the 4.5 to 5.0 tonnes ha-1 achievable in developed countries. A series of investigations were therefore carried out to identify farmers’ production constraints to facilitate development of strategies towards finding sustainable, long-term solutions within the context of plant breeding. A participatory rural appraisal was therefore conducted among smallholder sorghum farmers in four communities in the Talensi-Nabdam and Binduri districts of the Upper East Region of Ghana. The survey also sought to determine farmers’ perceptions of soil fertility as well as their current soil management strategies. Data were collected through focus group discussions, questionnaires and direct field observations. Within the communities, drought, high cost of farm input (labour and ploughing services) and declining soil fertility were the most important constraints to sorghum production. In spite of widespread low soil fertility, soil management methods such as fertiliser application, cover cropping and retention of crop residues were seldom practiced within the communities. Drought tolerance, high grain yield, earliness, grain quality (suitability for local foods and beverages) as well as low fertiliser requirement were the most preferred sorghum traits. Considering resources available to farmers and researchers in Ghana, mutation induction was considered a workable strategy in a breeding programme to expand the genetic diversity in the crop towards selection of phosphorus (P) use efficient genotypes. Mutation induction was achieved through chemical treatment of an inbred sorghum line BTx623 using ethyl methanesulphonate. A subset of 547 mutagenised lines at M3 generation were subjected to phenotypic assessment for expression of key morpho-agronomic traits of sorghum, both quantitative and qualitative. iii University of Ghana http://ugspace.ug.edu.gh Significant differences (p < 0.05) were observed among the M3 families for all the traits studied as well as between the families and BTx623. High genotypic coefficients of variation (GCV) established the existence of considerable variability. Broad sense heritability was also high ranging from 0.68 for stem girth to 0.99 for number of leaves. Positive and significant (p < 0.05) correlations were detected between grain yield and number of leaves, panicle width and stem girth. Principal component analysis revealed three principal components that accounted for 64% of the total variability. Grain yield (0.364), panicle width (0.436) and panicle length (0.411) were the most contributing traits. Hierarchical agglomerative clustering based on Unweighted Pair Group Method with Arithmetic mean (UPGMA), resolved the population into three divergent clusters at 52% level of dissimilarity. Two hundred and fifty three M3 lines were grouped into clusters separate from the wildtype BTx623, which indicates considerable variability between the mutagenised lines and the control. Close inspection for novel or distinct phenotypes revealed putative mutants for brown midrib, erect leaf and bloomless traits. Considerable genetic variability exists within the population which can be exploited for further breeding work. Further, genetic variation for P-use efficiency among the mutagenised lines was evaluated alongside selected cultivated varieties. Two hundred lines, comprising 170 mutagenised lines (agronomically- superior) and 30 cultivated varieties were evaluated under screen-house and field conditions and in two contrasting soil-P environments using a randomized complete block design and an augmented (incomplete block) design respectively. Measured traits included seedling vigour, days to anthesis, yield (grain and stover) as well as P concentration in the grain and stover. Analysis of variance for each P environment revealed significant (p <0.01) differences among the genotypes for all measured traits. Combined analysis of variance also showed significant (p <0.01) effects between genotype and P environment for all traits except plant height and stem girth, indicating iv University of Ghana http://ugspace.ug.edu.gh differences in genotypic response under contrasting P environments. Overall, grain and stover yield decreased by 49 and 27% respectively with the omission of P. Based on Pearson’s correlation analysis, grain yield was positively (p < 0.05) correlated with stover yield and P utilisation for grain production, but negatively correlated with P utilisation for stover production in both P environments, respectively. On the other hand, the relationship between stover yield and harvest index was significant (p < 0.05) in low-P soils but not significant in optimum-P soils. The contribution of each trait to observed genetic variation was determined by principal component analysis. The first three principal components (PC1, PC2 and PC3) explained 77% and 79% of the total variation in the low-P and optimum-P environments respectively. Traits that contributed the most variability in both P-environments were grain and stover yield. Subsequently, cluster analysis using UPGMA categorised the 200 genotypes into three divergent clusters at 49% dissimilarity. Cluster I, II and III contained 65, 106 and 29 genotypes respectively. Cluster I was characterised by reduced height and relatively late anthesis, Cluster II comprised the genotypes such as Grinkan, MR732, KuyumaWSV387 and CE151262A1 which produced the highest grain (mean = 56 g/plant) and stover yield (mean = 164 g/plant) whereas the genotypes in Cluster III were the tallest (mean height = 106 cm) in the population. The lines were then classified into four categories as P- uptake efficient, P-utilisation efficient, low-P stress tolerant or low-P stress sensitive, based on relative performance of the genotypes in both P environments. Kadaga West and Mut3412 were the most tolerant to low-P stress. Grinkan and ICSV1049 were the most P-uptake efficient whereas TxARG1 and Mut3708 were the most P-utilisation efficient. A comparison between the top-ranked mutant lines and cultivated genotypes indicated that the cultivated genotypes were more efficient at P-uptake whereas the putative mutants were more efficient at P-utilisation. There is potential to exploit the genotypic variability observed within the population for further breeding work. The v University of Ghana http://ugspace.ug.edu.gh identified P-use efficient mutagenised lines may either be released directly as mutant varieties following multi-locational trials or used as parents in future recombinant breeding programmes. vi University of Ghana http://ugspace.ug.edu.gh DEDICATION This thesis is dedicated to Adwoa, Ama and Aseda Elegba. Be encouraged. vii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGMENTS I am grateful to the West African Centre for Crop Improvement (WACCI) and the Alliance for a Green Revolution in Africa (AGRA) for providing the scholarship and logistics that were necessary to undertake my studies. I thank the Bill and Melinda Gates Foundation, which through Purdue University, USA provided funds for aspects of my work and gave me the opportunity to visit the Department of Agronomy, Whistler Centre for Carbohydrate Research, West Lafayette. I am particularly grateful to Dr. Mitch Tuinstra and Dr. Cliff Weil for the wealth of knowledge and experience I gained during my time at Purdue University. Thank you for sharing your germplasm with me. I am exceeding grateful to Prof. Eric Yirenkyi Danquah for the opportunity to work with the team at Purdue University. I am deeply grateful to my supervisors, Prof. Harry M. Amoatey, Prof. Eric Yirenkyi Danquah and Dr. Eric Nartey for their support and guidance throughout the entire study and to Prof. Kwadwo Ofori for his guidance and encouragement. My sincerest thanks also goes to all staff of WACCI and my colleagues, both past and present. You all made my time at WACCI memorable. I am grateful to Ghana Atomic Energy Commission for granting me the study leave required to complete this degree. I thank the Director of BNARI, Prof Kenneth Ellis Danso for the support of the institute, both infrastructural and technical. To all staff at the Biotechnology Centre of BNARI, particularly Mrs. Abigail Asare, Ms. Vida Adu Afrakoma, Ms. Patience Asare, Mr. Clement Annor, and Mr. Samuel Azure who in diverse ways helped to make this work successful, I am grateful. Special thanks go to staff of University of Ghana research farm particularly Mr. Agyekum and Mr. Stephen Atsu. Your unwavering support is fully-recognised and deeply-appreciated. To Mr. viii University of Ghana http://ugspace.ug.edu.gh Precious Blege of the University of Development Studies and the late Mr. Mohammed of the MoFA office, Talensi-Nabdam, I am grateful for the insight and guidance during the participatory rural appraisal in the Upper East Region of Ghana. Mr. James Azu and Ms. Irene Garbrah, my parents as well as Samuel Azu, my brother, I deeply appreciate your sacrifice and encouragement. Wilfred, you encouraged me like no other. I am eternally grateful. Adwoa, Ama and Aseda, you bore the brunt of this challenging period. Thank you for bearing with me. Finally, to you heavenly Father, only You are worthy. Thank you for grace. Indeed, you are the lifter up of my head. ix University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION........................................................................................................................... ii ABSTRACT .................................................................................................................................. iii DEDICATION............................................................................................................................. vii ACKNOWLEDGMENTS ......................................................................................................... viii LIST OF ABBREVIATIONS ................................................................................................... xxi CHAPTER ONE ........................................................................................................................... 1 1.0 GENERAL INTRODUCTION .............................................................................................. 1 CHAPTER TWO .......................................................................................................................... 7 2.0 LITERATURE REVIEW ...................................................................................................... 7 2.1 Origin domestication and distribution of sorghum ............................................................. 7 2.2 Classification of sorghum ....................................................................................................... 8 2.3 Sorghum production ............................................................................................................. 10 2.4 Uses of sorghum .................................................................................................................... 11 2.5 Edaphic and nutrient requirements of sorghum ............................................................... 12 2.6. Phosphorus in soils ............................................................................................................... 15 2.6.1 Forms of phosphorus in soils ............................................................................................ 15 2.6.2 Phosphorus availability and its acquisition by plants .................................................... 16 2.6.3 Phosphorus status of West African soils .......................................................................... 18 2. 7 Effect of phosphorus deficiency on sorghum productivity in West Africa..................... 19 x University of Ghana http://ugspace.ug.edu.gh 2.8 Crop adaptation to soil P deficiency.................................................................................... 20 2.8 Strategies for improving sorghum productivity in low P soils ......................................... 22 2.8.1 Traditional management strategies .................................................................................. 22 2.8.2 Breeding strategies ............................................................................................................. 24 2.8.2.1 Conventional approaches to breeding for phosphorus-use efficiency ....................... 24 2.8.2.2 Molecular approaches to breeding for phosphorus-use efficiency ............................. 27 2.8.2.3 Induced mutagenesis as an alternative breeding approach ........................................ 28 CHAPTER THREE....……………………………………………………………………..… 31 3.0 FARMERS’ PRODUCTION CONSTRAINTS, PERCEPTIONS OF SOIL FERTILITY AND PREFERENCE FOR SORGHUM TRAITS .................................................................. 31 3.1 Introduction ........................................................................................................................... 31 3.2 Materials and methods ......................................................................................................... 34 3.2.1 Description of study sites ................................................................................................... 34 3.2.2 Sampling procedure and data collection techniques ...................................................... 36 3.2.3 Data analysis ....................................................................................................................... 37 3.3 Results .................................................................................................................................... 37 3.3.1 Demographic characteristics of households .................................................................... 37 3.3.2 Socio-economic characteristics of households ................................................................. 39 3.3.3. Farm size, area allocated to sorghum and estimated grain yield ................................. 41 3.3.4 Sources of seed and information ....................................................................................... 43 xi University of Ghana http://ugspace.ug.edu.gh 3.3.5 General constraints to sorghum production.................................................................... 44 3.3.6 Farmers’ description and perception of soil fertility ...................................................... 46 3.3.7 Farmers’ cropping systems and soil management practices .......................................... 49 3.3.8 Cultivated sorghum varieties and farmers’ preference for traits ................................. 51 3.4 Discussion............................................................................................................................... 54 3.5 Conclusions ............................................................................................................................ 60 CHAPTER FOUR ....................................................................................................................... 61 4.0 PHENOTYPIC ANALYSES FOR IDENTIFICATION OF ARTIFICIALLY- INDUCED VARIABILITY IN SORGHUM ............................................................................ 61 4.1 Introduction ........................................................................................................................... 61 4.2 Materials and methods ......................................................................................................... 63 4.2.1 Description of study site .................................................................................................... 63 4.2.2 Generation of mutagenised population ............................................................................ 64 4.2.4 Experimental design and layout ....................................................................................... 65 4.2.5 Data collection .................................................................................................................... 66 4.2.6 Data analysis ....................................................................................................................... 67 4.3. Results ................................................................................................................................... 68 4.3.1 Variation within the mutagenised population for quantitative traits ........................... 68 4.3.2. Genetic and phenotypic variances and heritability within the mutagenised population ....................................................................................................................................................... 73 xii University of Ghana http://ugspace.ug.edu.gh 4.3.3 Correlation between the studied traits ............................................................................. 74 4.3.5 Divergence between the genotypes based on cluster analysis ........................................ 79 4.3.6 Variation within the mutagenised population for qualitative traits ............................. 83 4.5 Discussion............................................................................................................................... 94 4.6 Conclusion ........................................................................................................................ 98 CHAPTER FIVE ........................................................................................................................ 99 5.0 GENETIC VARIATION FOR PHOSPHORUS-USE EFFICIENCY AMONG MUTAGENISED SORGHUM LINES ..................................................................................... 99 5.1 Introduction ........................................................................................................................... 99 5.2 Materials and methods ....................................................................................................... 101 5.2.1 Evaluated genotypes ........................................................................................................ 101 5.2.2 Description of study sites ................................................................................................. 103 5.2.3 Soil sampling and analyses .............................................................................................. 103 5.2.4 Experimental design and layout ..................................................................................... 104 5.2.5 Data collection .................................................................................................................. 105 5.2.6 Data analysis ..................................................................................................................... 107 5.2.6.1 Data analysis for pot experiments ............................................................................... 107 5.2.6.2 Data analysis for field experiment ............................................................................... 109 5.3 Results .................................................................................................................................. 112 5.3.1. Pot experiment ................................................................................................................ 112 xiii University of Ghana http://ugspace.ug.edu.gh 5.3.1.1. Characteristics of soils in pot experiment .................................................................. 112 5.3.1.2. Genetic variation among the genotypes under contrasting P environments .......... 112 5.3.1.3. Correlation between morpho-agronomic traits and P-tolerance ratios .................. 117 5.3.1.4. Principal component analyses under contrasting P environments ......................... 119 5.3.1.5. Variation within the population based on cluster analysis. ..................................... 123 5.3.1.6. Relative performance of the genotypes in contrasting P environments.................. 126 5.3.1.7. Selection of P-use efficient genotypes ......................................................................... 126 5.3.2. Field experiment.............................................................................................................. 128 5.3.2.1. Characteristics of soils in field experiments .............................................................. 128 5.3.2.2. Genotypic variation for morpho-agronomic traits in contrasting P environments128 5.3.2.3. Genotypic variation for P-use efficiency indicators in contrasting P environments ..................................................................................................................................................... 133 5.3.2.4. Heritability and genotypic coefficient of variation within the population.............. 138 5.3.2.5. Correlation between morpho-agronomic traits and P-use efficiency indicators in contrasting soil-P environments .............................................................................................. 139 5.3.2.6. Differences between mutants and cultivated varieties for morpho-agronomic traits ..................................................................................................................................................... 142 5.3.2.7. Selection of low soil P-stress tolerant and sensitive genotypes ................................ 145 5.4 Discussion............................................................................................................................. 148 5.6 Conclusions .......................................................................................................................... 151 xiv University of Ghana http://ugspace.ug.edu.gh CHAPTER SIX ......................................................................................................................... 153 6.0 GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS ............ 153 REFERENCES .......................................................................................................................... 156 APPENDIX ................................................................................................................................ 187 Appendix 1: PRA questionnaire .............................................................................................. 187 Appendix 2: p- values for Pearson’s correlation coefficient of quantitative traits among 550 genotypes .................................................................................................................................... 193 Appendix 3: Members of individual clusters in Figure 4.5 ................................................... 194 Appendix 4: Procedures for soil characterization and analysis ........................................... 197 Appendix 5: Determination of phosphorus in leaf and grain samples ................................. 200 Appendix 6: Names of putative mutants (codes) used in PCA biplot of 200 genotypes as shown in Figure 5.3 ................................................................................................................... 201 Appendix 7: Distribution of 200 genotypes in three clusters – Figure 5.4 ........................... 203 xv University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 3.1. Geographical representation of Talensi-Nabdam and Binduri districts. .................... 35 Figure 3.2. Sources of seed and (A) and information (B) across the communities ...................... 51 Figure 3.3. Diversity of sorghum varieties cultivated in the communities ................................... 51 Figure 4.1. Box plots showing the distribution of studied traits in the mutagenised population.. 71 Figure 4.2. Loading plot of PC1 and PC2 showing contribution of quantitative traits to variability ....................................................................................................................................................... 78 Figure 4.3. Distribution of sorghum genotypes for first two principal components..................... 79 Figure 4.4. Dendrogram of 550 sorghum genotypes derived by UPGMA method (Pearson’s dissimilarity matrix) ...................................................................................................................... 81 Figure 4.5. Boxplots showing the distribution of traits in each cluster. ....................................... 82 Figure 4.6a: Variations in leaf midrib and blade colour ............................................................... 88 Figure 4.6b: Variations in leaf colour ........................................................................................... 89 Figure 4.6c: Variations in leaf size ............................................................................................... 90 Figure 4.7d: Variations in plant architecture ................................................................................ 91 Figure 4.6e: Variations in tillering and panicles ........................................................................... 92 Figure 4.6f: Variations in bloom: Senescence and wilting ........................................................... 93 Figure 5.1. Distribution of traits in low and optimum P-environments. ..................................... 114 Figure 5.2. Loading plot of PC1 and PC2 showing contribution of morpho-agronomic traits to variability .................................................................................................................................... 121 Figure 5.3. Distribution of the 200 genotypes for PC1 and PC2 ................................................ 122 Figure 5.4. Dendrogram of 200 sorghum genotypes derived by UPGMA method (Pearson’s dissimilarity matrix) .................................................................................................................... 124 xvi University of Ghana http://ugspace.ug.edu.gh Figure 5.5. Boxplots showing the distribution of traits in each cluster ...................................... 125 xvii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 3.1. Number of respondents across the study sites ............................................................. 36 Table 3.2. Demographic characteristics of households ................................................................ 38 Table 3.3. Socio-economic and farm characteristics of households ............................................. 40 Table 3.4. Farm size and area allocated to sorghum within the communities .............................. 42 Table 3.5. Ranking of sorghum production constraints across the four communities .................. 45 Table 3.6. Farmers’ description of soils ........................................................................................ 47 Table 3.7. Farmers’ perceptions of soil fertility and its effect on sorghum .................................. 48 Table 3.8 Farmers’ cropping systems and soil management practices ......................................... 50 Table 3.9. Ranking of farmers’ preference for sorghum traits ..................................................... 53 Table 4.1. Meteorological data for the sites during the study period ........................................... 64 Table 4.2. Descriptive statistics of nine evaluated traits ............................................................... 70 Table 4.3. Analysis of variance within the mutagenised population for quantitative traits .......... 72 Table 4.4 Summary of estimated variances and heritability for quantitative traits ...................... 73 Table 4.5. Pearson’s correlation coefficients of quantitative traits among 550 genotypes........... 75 Table 4.6. Eigenvalues of first three principal components .......................................................... 77 Table 4.7. Percentage contribution (%) of the traits PC1, PC2 and PC3 ...................................... 77 Table 4.8. Mutant phenotypes segregating in M3 families ........................................................... 86 Table 5.1. Summary of the cultivated genotypes and their important features .......................... 102 Table 5.2. Meteorological data for the experimental areas during the study period .................. 103 Table 5.3 Descriptions of evaluated traits ................................................................................... 106 Table 5.4. Format of ANOVA for two-way factorial experiment (RCBD) ................................ 107 Table 5.5. Estimations of P-uptake and utilisation parameters ................................................... 111 xviii University of Ghana http://ugspace.ug.edu.gh Table 5.6.1. Analysis of variance under contrasting P environments. ........................................ 115 Table 5.6.2. Comparison between means of traits in contrasting P environments ..................... 115 Table 5.6.3. Heritability, genetic and phenotypic variability of the traits under low-P stress .... 116 Table 5.6.4. Correlation coefficients between morpho-agronomic traits (under low P stress) .. 118 Table 5.6.5. Principal components of morpho-agronomic traits in the contrasting P environments ..................................................................................................................................................... 120 Table 5.6.6. Top-ranked (10%) genotypes for grain yield ratio, P-uptake and P-utilisation efficiencies .................................................................................................................................. 127 Table 5.6.7. Analysis of variance for morpho-agronomic traits in low-P stress ........................ 130 Table 5.6.8. Analysis of variance for morpho-agronomic traits under optimum soil P ............. 131 Table 5.6.9. Comparison between means of traits in low P and optimum P environments ....... 132 Table 5.7.1. Analysis of variance for P-uptake and use efficiency indicators in low P stress... 134 Table 5.7.2. Analysis of variance for P- uptake and use efficiency indicators under optimum soil P conditions ................................................................................................................................. 135 Table 5.7.3. Comparison between means of traits in low P and optimum P environments ....... 136 Table 5.7.4. Combined analysis of variance for morpho-agronomic traits ............................... 137 Table 5.7.5. Combined analysis of variance for P- uptake and use efficiency traits .................. 137 Table 5.7.6. Estimates of genetic variance, heritability and genotypic coefficient of variation in low and optimum P trials ............................................................................................................ 138 Table 5.7.7. Correlation between morpho-agronomic traits and P-use efficiency indicators under low soil P conditions ................................................................................................................... 140 Table 5.7.8. Relationship between yield components and P-use efficiency indicators under optimum soil P conditions .......................................................................................................... 141 xix University of Ghana http://ugspace.ug.edu.gh Table 5.7.9. Highest-ranking mutants and cultivated genotypes for yield and PUE indicators . 144 Table 5.8.1. Top-ranked genotypes (grain yield) in low-P soil and their performance in optimum- P soil ……………………………………………………………………………………….. 146 Table 5.8.2. Low-P stress tolerant and sensitive genotypes within the population .................... 147 xx University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS AGRA Alliance for Green Revolution in Africa ANOVA Analysis of variance ATP Adenosine Triphosphate BM Biomass BNARI Biotechnology and Nuclear Agriculture Research Institute CEC Cation exchange capacity CV Coefficient of variation DAP Di-ammonium phosphate DFR Days to flowering ratio DNA Deoxyribonucleic Acid DTF Days to flowering ECV Environment coefficient of variation EMS Ethyl methanesulphonate EPA Environmental Protection Agency FAO Food and Agriculture Organization FAOSTAT Food and Agriculture Organization Statistical Databases GCV Genotypic coefficient of variation GSS Ghana Statistical Service GY Grain yield GYP Grain yield per plant GYR Grain yield ratio H2 Broad sense heritability xxi University of Ghana http://ugspace.ug.edu.gh HSW Hundred seed weight IAEA International Atomic Energy Agency IBM International Business Machines ICRISAT International Crops Research Institute for Semi-Arid Tropics IFAD International fund for agricultural development IFDC International Fertiliser Development Centre K Potassium LMWOA Low Molecular Weight Organic Acids MAS Marker Assisted Selection MoFA Ministry of Agriculture, Ghana Mut Mutant N Nitrogen NL Number of leaves MNU Methyl nitrosourea NGOs Non-Governmental Organisations P Phosphorus PAE Phosphorus Acquisition efficiency PBM Phosphorus content in biomass PC Principal Component PCA Principal Component analysis PCG Phosphorus concentration in grain PCS Phosphorus concentration in stover PCV Phenotypic coefficient of variation xxii University of Ghana http://ugspace.ug.edu.gh PE Peduncle exsertion PG Phosphorus content in grain PH Plant height PHI Phosphorus harvest index PHR Plant height ratio PL Panicle length PRA Participatory Rural Appraisal PS Phosphorus content in stover PSTOL-1 Phosphorus Starvation Tolerance 1 PUE Phosphorus Use Efficiency Pup Phosphorus Uptake Efficiency PUP-1 Phosphorus Uptake 1 PUtE Phosphorus Utilisation Efficiency PUTIL-BM Phosphorus utilisation for biomass production PUTIL-G Phosphorus utilisation for grain production PUTIL-S Phosphorus utilisation for stover production PW Panicle width QTL Quantitative Trait Loci RILs Recombinant inbred lines REML Restriction maximum likelihood RNA Ribonucleic acid RUDEMA Rural Development Management SARI Savannah Agriculture Research Institute xxiii University of Ghana http://ugspace.ug.edu.gh SG Stem girth SNP Single-Nucleotide Polymorphisms SPSS Statistical Package for the Social Sciences SSP Single super phosphate SVR Seedling vigour ratio TSBF Tropical soil biology and fertility. TSP Triple super phosphate UPGMA Unweighted pair group method with arithmetic mean USA United State of America USDA United State Department Agency VAM Vesicular-Arbuscular Mycorrhiza xxiv University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE 1.0 GENERAL INTRODUCTION Sorghum (Sorghum bicolor L. Moench) is the fifth most important cereal in the world after wheat, maize, rice and barley (Dahlberg et al., 2011). In Africa, it is the second major cereal after maize and ranks third in Ghana after maize and rice (FAOSTAT, 2012). It is a highly versatile crop and is valued for both its grain and stover. In developed countries, its primary use is as feed for livestock and poultry as well as the production of ethanol (Rajasekher et al., 2005). On the other hand, in Africa, sorghum is a dietary staple and is crucial to the food security of over 500 million people (Dahlberg et al., 2011). Due to its ability to perform favourably under harsh and unpredictable climatic conditions, the crop provides a less risky alternative to sole maize cropping and is a reliable source of income for many smallholder farmers in the Sahelian and Savannah zones (Buerkert et al., 2001; Sultan et al., 2013). Although sorghum is adapted to the harsh climatic and edaphic conditions of West Africa, yield in this region is low (Wang et al., 2010). Yield potential in West Africa is estimated at 2.0 to 3.0 tonnes/ha. However, in several countries, sorghum productivity is approximately 1.8 tonnes ha-1 (Wang et al., 2010). This is woeful compared to the 4.5 to 5.0 tonnes ha-1 produced by the USA and Europe (FAO, 2012). While sorghum cultivation in developed regions is commercial, high- input and highly productive, its production in West Africa is subsistent, low-input and characterised by the cultivation of inherently low-yielding varieties (FAO, 2010; Vom Brocke et al., 2010). Additionally, sorghum production in West Africa is constrained by drought, diseases such as downy mildew (Peronosclerospora sorghi [Weston and Uppal; (Shaw)]) and anthracnose (Colletotrichum graminicola) and pests such as midge (Stenodiplosis sorghicola Coquillett) and 1 University of Ghana http://ugspace.ug.edu.gh shoot fly [Atherigona soccata (Rondani)] (Dhillon et al., 2005; Riyazaddin et al., 2016 ). Another key contributor to low sorghum yields in Africa is soil infertility (Leiser et al., 2014). Soil infertility is a widespread problem in Africa; over 121 million hectares of the continent’s total land is deficient in essential plant nutrients (Bationo et al., 2006). This is as a result of decades of unsustainable cropping systems leading to nutrient mining and erosion (Bationo et al., 2012). In northern Ghana for instance, major causes of soil infertility include nutrient mining and erosion worsened by seasonal bush fires, deforestation and overgrazing (Yahaya and Amoah, 2013). In general, the main soil nutritional constraint in small-scale farming systems in Africa is nitrogen (N) deficiency (Nziguheba, 2007). However, African soils are not only deficient in N but are extensively deficient in phosphorus (P) (Sweeney et al., 2011). In West African soils, P is one of the main limiting macronutrients with over 50% of arable land lacking sufficient quantities for maximum crop yield (Nandwa et al., 2011; Leiser et al., 2014). Smallholder fields rarely exceed the critical level of 15 mg P kg-1 and are often characterised by available P content of 7 to 10 mg P kg-1 (Doumbia et al., 1993; Buerkert et al., 2001). P levels as low as 2 to 5 mg P kg-1 have been reported in Mali and eastern Kenya (Obura, 2008; Leiser et al., 2012). According to Owusu- Bennoah and Acquaye (1996), available P in soils in northern Ghana varied from 1.7 to 8.7 mg P kg-1. However, Nyanteng and Asuming-Brempong (2003) reported slightly higher values of between 4.38 and 9.43 mg P kg-1. The low P status of West African soils is a major constraint to sorghum production (Verde and Matusso, 2014). P deficiency results in poor seedling emergence and establishment and an overall reduction in crop growth leading to low yields (Buerkert et al., 2001). Poor sorghum yields coupled with increasing food demand as a result of rapid growth in human population threatens the food 2 University of Ghana http://ugspace.ug.edu.gh security and livelihood of millions of people in West Africa. It has therefore become increasingly important to improve sorghum productivity in P-stressed soils. Several strategies have been proposed to increase sorghum productivity on marginal soils in general. These include traditional practices such as shifting cultivation or land fallowing, crop rotation, cover cropping and intercropping (Samake et al., 2005). Unfortunately, in recent times, the use of these methods is limited by dwindling land availability, increasing population densities and other socio-economic pressures (Gregory and Bumb, 2006). The key strategy for maintaining crop productivity on poor soils has been fertiliser application (AGRA, 2012). As with other crops, the application of fertilisers to sorghum fields has been shown to result in significant increases in grain yield (Kaizzi et al., 2012). However, in spite of demonstrated gains, sorghum cultivation in West Africa is characterised by little or no added nutrient input (Buah et al, 2012). Average fertiliser use in sub-Saharan Africa is estimated as 16.25 kg/ha, a sixth of the world’s consumption of 98.20 kg/ha (Hernandez and Torero, 2011; Sommer et al., 2013). For instance fertiliser application rates in Ethiopia, Burkina Faso and Cameroon are estimated at approximately 18, 9 and 7 kg/ha compared to 125, 167 and 238 kg/ha in Brazil, India and the Netherlands respectively (FAO, 2009). Although recent reports indicate an increase in the application of N-based fertiliser, the use of P-based fertilisers has remained minimal at 2.2 kg/ha compared to 7.9 kg/ha in Latin America (Kelly, 2006; MacDonald et al., 2011). This imbalance in N and P levels has been reported as a major contributor to yield reductions (van der Velde et al., 2014). The main constraints to fertiliser-use in West Africa are cost of the commodity and its unavailability (van der Velde et al., 2014). Unfortunately, the impact fertiliser subsidy policies is diminished by huge fiscal and administrative costs, late delivery of subsidised fertilisers as well as smuggling (Druilhe and Barreiro-Hurle, 2012). 3 University of Ghana http://ugspace.ug.edu.gh While increasing fertiliser use on smallholder sorghum fields in West Africa will undoubtedly improve yields, for mineral P fertilisers, this strategy is unsustainable. Mineral P is a finite resource. Rock phosphate reserves which account for 90% of inorganic P fertilisers are expected to be exhausted in 50 to 100 years (Cordell et al., 2009; Sattari et al., 2012). Thus, availability of inorganic P fertilisers is likely to be a major challenge in a few decades (Van Vuuren et al., 2010). Its unavailability is also likely to increase cost of the commodity, making it even more inaccessible to smallholder farmers in West Africa. Besides, increased P fertilisation has been implicated in environmental hazards such as the eutrophication of water supplies in developed countries (Hash et al., 2002). Perhaps, a cheaper and more sustainable option for mitigating the effects of low soil P on sorghum yield is the application of organic fertilisers such as manure, compost and plant residues. While these have been shown to increase P levels in soils, the amount of P in organic fertilisers is highly variable and often insufficient to meet crop requirements (Risse et al., 2006). Moreover, the use of organic fertilisers on small-scale farms is constrained by unavailability of manure (where livestock is not a major component of the farming system) and the removal of plant residue at harvest for use as dry season fodder and building material (Verde and Matusso, 2014). Widespread soil-P deficiency, dwindling rock phosphate reserves, high cost, unavailability and inefficient use of fertilisers strongly suggest that a more integrated approach is necessary to improve sorghum yield in West Africa. A more sustainable approach is the development of P-use efficient varieties. Until recently, research on improving sorghum adaptation to P deficiency has been minimal (Leiser et al., 2012). One of the main obstacles to breeding for P-use efficiency in sorghum was the limited degree of diversity for the trait (Ribaut and Poland, 2000). Recently, Leiser et al. (2014) reported sufficient genetic variability for P use-efficiency among West African 4 University of Ghana http://ugspace.ug.edu.gh inbred lines and landraces. However, given the alarming decline in soil fertility and its associated low sorghum productivity in West Africa as well as the extremely diverse levels of P deficiency in sorghum-growing regions, there is a need to broaden the gene pool for phosphorus-use efficiency (PUE) in sorghum and breed for higher levels of the trait in the crop (Chianu et al., 2012; Gemenet et al., 2016). The challenge for breeders is to increase the genetic variability for PUE in sorghum, followed by selection of desirable phenotypes in low-P soils and the subsequent introgression of these cultivars into elite Africa lines. While methods such as hybridisation, genetic engineering and somaclonal variation are available for increasing genetic diversity in crops, artificial mutagenesis offers an efficient and rapid means of generating new genotypes with desirable traits (Xin et al., 2008). Such genotypes may then be released either as direct mutants or used as parents in recombinant breeding programmes (Ahloowalia et al., 2004). According to the International Atomic Energy Agency, artificial mutagenesis has been used for the development of over 2000 new crop varieties with improved tolerance to several biotic and abiotic stresses (Ahloowalia et al., 2004). The main strategy in this case is the mutagenic treatment of elite genotypes followed by selection for morphological and/or physiological characters that influence the trait of interest (Parentoni and Souza, 2008; Sweeney et al., 2011). Thus, this project seeks to identify and develop new P use-efficient sorghum genotypes through induced mutagenesis of the in-bred line BTx623. The availability and cultivation of such genotypes is likely to contribute to improving yields in the P-stressed and low-input cropping environments in West Africa. In turn, food security and livelihood of smallholder sorghum farmers may be improved. The cultivation of P use-efficient sorghum cultivars may reduce excessive fertiliser 5 University of Ghana http://ugspace.ug.edu.gh application in high-input cropping systems. Thus, the goal of this project was to improve genetic variation for phosphorus-use efficiency in sorghum via induced mutagenesis. The specific objectives were to; 1. determine farmers’ production constraints, perceptions of soil fertility and preference for sorghum traits in the Upper East Region of Ghana, 2. assess the genetic variability within a chemically-mutagenised sorghum population at M3 and 3. identify phosphorus-use efficient mutants within the population (at M6). 6 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 Origin domestication and distribution of sorghum Sorghum is an ancient crop; it is thought to have diverged from a common ancestor as maize (Smith and Frederiksen, 2000; Paterson et al., 2004). According to Singh and Jauhar (2006), it was consumed by hunters and gatherers as early as 8000 B.C. Although some reports suggest diverse origin, archaeological evidence indicates that sorghum originated in north-eastern Africa (de Wet et al., 1967). Cultivated sorghum originated from wild races of sorghum bicolor ssp. verticilliflorum in the Ethiopian highlands as recently as 1000 BC (Kimber, 2000; Harlan, 1972). This is supported by the fact that the largest diversity of cultivated sorghum and several closely- related wild species can be found in this region presently (Bhattacharya et al., 2011). Thus, Ethiopia is generally accepted as the primary centre of origin (Vavilov, 1992). It is likely that wild sorghum was domesticated through natural and deliberate selection for specific traits (Doggett, 1970; Kimber, 2000). Domestication may also have resulted from crosses between wild and cultivated types (House, 1985). This process resulted in gradual changes in several plant characteristics including seed size as well as the transformation of open inflorescence into compact types (House, 1985). These new types may have spread to eastern and southern Africa through the migration of the Bantu people (Westengen et al., 2014). Subsequently, the crop became adapted to a wide range of environments on the continent. Smith and Frederiksen (2000) suggest that this movement, genetic adaptation and inter-crossing in the isolated ecosystems of Africa gave rise to the five basic races of cultivated sorghum. From Africa, sorghum spread to the Middle East, the Far East and India; considerable evidence exists of sorghum cultivation in India more than 4,500 years ago (Vavilov, 1992). Thus today, 7 University of Ghana http://ugspace.ug.edu.gh India is regarded as the secondary centre of origin of the crop. Sorghum is relatively new to South America and Australia and was introduced to the USA in the middle of the 19th century (1857) through slave trade (Doggett, 1965b). Today, a diverse range of genotypes are cultivated in tropical and subtropical regions of the world (Legwaila et al., 2003). This includes several countries in the Americas, Asia, Oceania and Africa. Its cultivation in Europe however is limited to a few areas in Italy, France and Spain (Wang et al., 2014). 2.2 Classification of sorghum Sorghum was first described by Linnaeus in 1753 under the name Holcus. He described three species of cultivated sorghum as Holcus sorghum, Holcus saccaratus and Holcus bicolor (House, 1985). In 1794, the genus Sorghum was separated from Holcus by Moench (House, 1978). Over time, the crop has been re-described by many taxonomists. However, the most definitive description was provided by Snowden, who classified sorghum into 52 species composed of 31 cultivated, 17 wild and 4 weedy species (Snowden, 1935). These species were later combined into a single species on the basis of the absence of genetic barriers (de Wet and Huckabay, 1967). The species now includes all cultivated sorghums as well as related semi-domesticated plants (de Wet, 1978). The Sorghum genus is further divided into five taxonomic sub-generic sections: Eu- Sorghum, Chaetosorghum, Heterosorghum, Para-Sorghum and Stiposorghum (Garber, 1950). All cultivated varieties and landraces belong to the sub-generic section Eu-Sorghum (Lazarides et al., 1991). Presently, the genus consists of 25 species (USDA-ARS, 2012). The scientific name for all cultivated sorghums is Sorghum bicolor (L.) Moench; Clayton and Renvoize, (1986) classified sorghum as follows; Kingdom – Plantae 8 University of Ghana http://ugspace.ug.edu.gh Sub-kingdom – Tracheobionta Super-division – Spermatophyta Division – Magnoliophyta Class – Liliopsida Sub-class – Commelinidae Order – Cyperales Family – Poaceae Tribe – Andropogonae Sub-tribe – Sorghinae Genus – Sorghum Species – Sorghum bicolor The species is divided into 5 races and 10 intermediate races under S. bicolor spp. bicolor (de Wet and Harlan, 1970). These races include bicolor, caudatum, kafir, durra and guinea which are distinguishable by panicle morphology (Harlan and de Wet, 1972). The predominantly West African Guinea, is characterised by glabrous panicles and open sessile spikelets (Harlan, 1972). The kafir, found primarily in East and southern Africa is characterised by semi-compact panicles with hairy, sessile spikelets. Panicles of the Caudatum race (found in east Nigeria and eastern Sudan) are dense or slightly open with obovate to elliptical sessile spikelets while the Durra race (dominant in Ethiopia and Central Africa) is characterised by obovate panicles. The most diverse and primitive race, bicolor is characterised by open panicles and pedicellate, persistent spikelets (Smith and Frederiksen, 2000). Like maize, sorghum has a haploid chromosome number of 10 (2n = 2x = 20) (Fedorov, 1969). The sorghum genome is small [approximately 730 Mbp] (Paterson, 2009). Its genome is approximately 60% larger than that of rice, 25% the size of the maize genome and 20 times smaller than that of wheat (Arumuganathan and Earle, 1991; Peterson et al., 2002). This makes it an ideal model for studying the structure and function of other cereal genomes (Paterson et al., 2004). The 9 University of Ghana http://ugspace.ug.edu.gh ancestral linkage between sorghum and other cereals also provides insights into genome evolution (Bowers et al., 2005). 2.3 Sorghum production Presently, total area under sorghum cultivation is estimated at 42 million hectares (FAO, 2014). Approximately 80% of this is located in developing countries in Asia and Africa (Christiaensen et al., 2010). In Africa, cultivation of the crop extends from the Atlantic coast to the borders of the Sahara and more arid parts of eastern and southern Africa (Wang et al., 2014). Sorghum acreage within sub-Saharan Africa is an estimated 24.8 million hectares, corresponding to 53% of the world’s production area (FAO, 2012). The largest areas under cultivation in Africa are located in Nigeria, Sudan and Ethiopia (FAO, 2013). Other major sorghum-cultivating countries in Africa are Burkina Faso, Senegal, Tanzania, Mali, Cameroun and Ghana. In Ghana, it is cultivated on approximately 243, 000 ha of land (Addai and Alimiyawo, 2015). The world produces a total of 59.3 million metric tonnes annually (FAO, 2014). The USA is currently the world’s largest producer with an estimated 15 million tonnes per annum (FAOSTAT, 2015). In general, sorghum productivity in Africa has increased steadily over the decades from 11.6 million tonnes in 1976 to 20.9 million tonnes in 2009 (Atokple, 2010). However, this improvement in productivity has been attributed to increases in acreage and not in average yields on farmers’ fields (FAO, 2010). Currently, Africa produces about a third of the world crop, an annual average of 26 million tonnes (FAOSTAT, 2015). Nigeria, Africa’s largest producer, contributes an average of 6.1 million tonnes yearly (FAOSTAT, 2015). Ghana ranks 22nd in the world in terms of sorghum production (USDA, 2012). Most of country’s sorghum is cultivated in the three northern regions mainly as a subsistence crop by smallholder farmers with land holdings 10 University of Ghana http://ugspace.ug.edu.gh of less than 2 hectares (Kudadjie et al., 2004; Atokple, 2012). In Ghana, as is the case in several parts of West Africa, the crop is usually planted as sole stands or under irregular cropping systems (Ncube et al., 2007). Average yields in the United States and Europe are estimated at 4.5 and 5 tonnes/ha respectively (Dahlberg et al., 2011). In Africa, average yields vary considerably at regional and national levels. For instance, productivity in West and Central Africa is an estimated 1.8 tonnes/ha compared to 2.5 tonnes/ha in East and Southern Africa (Wang et al., 2014). Yield in farmers’ fields may be as low as 1 tonne/ha in Mali or as high as 2.8 tonnes/ha in South Africa (Leiser et al., 2014). Although Ghana has a yield potential of 2 to 3 tonnes /ha, average yield usually ranges from 700 to 900 kg/ha in the Upper East and Upper West Regions to higher values of 1200 kg/ha in Northern Region (Atokple, 2010). 2.4 Uses of sorghum In developed regions of the world, sorghum is used primarily as feed for beef and dairy cattle, poultry and pigs (Berenji and Dahlberg, 2004). Sweet sorghums, due to their high sugar and biomass yield are used extensively for biofuel production in countries such as the USA and India (Shoemaker and Bransby, 2010). Other industrial uses of sweet sorghums include the production of sugar and syrup from the juice in stalks (Srinivasa Rao et al., 2012). Sorghum bagasse may also be used as feedstock for generation of electricity (Berenji and Dahlberg, 2004). In Africa, sorghum is a very important source of food (Kudadjie et al., 2004). Traditionally, it is processed into a wide variety of foods such as thin and stiff porridges, bread and couscous (Kudadjie-Freeman and Dankyi-Boateng, 2012). Sorghum grains are nutritious and constitute a principal source of energy for many of Africa’s rural poor (Thompson, 2000; Zohary and Hopf, 11 University of Ghana http://ugspace.ug.edu.gh 2000). Red-grain varieties are an excellent source of antioxidants (Klopfenstein and Hoseney, 1995; Chung et al., 1998). More recently, sorghum grain has been recommended for diabetics due to its slowly-digested starch and has gained attention as a substitute for wheat in the production of gluten-free food products (Rai et al., 2014; Dahlberg et al., 2011). Sorghum plays an important role in the socio-economic development of many African rural folk (Kudadjie et al., 2004). For instance, it is the grain of choice for brewing traditional beers - an important income-earning activity for many rural women, as is the case in northern Ghana (Kudadjie et al., 2004). In Nigeria, sorghum grain is used as a substitute for barley in the industrial brewing of lager beer thereby offering income for growers and providing additional jobs in the value chain (Goode and Arendt, 2003). Although stover is typically used as dry season fodder, fencing material and thatch, it may also be used in small-scale production of brooms, mats and baskets (ICRISAT, 2006). In most parts of Africa, its use as biofuel crop is still emerging (Deenanath et al., 2012). However, in countries such as South Africa, the sorghum bioethanol industry is quite advanced and provides income for farmers (Mokhema, 2013). Due to its untapped potential as an industrial crop, the World Bank has tagged sorghum cultivation as one of the strategies for improving the lives of vulnerable people in Africa (Angelucci, 2013). 2.5 Edaphic and nutrient requirements of sorghum Sorghum tolerates a diverse range of soils from heavy clays to medium loams (El-Bassam, 2010). Although it can be productive on shallow sandy soils, its yield is optimised on deep, well-drained loamy soils with high organic matter content; poorly-drained and compact soils may restrict root growth and overall performance (Verheye, 2010). Its growth and development is also influenced by soil pH (Butchee et al., 2012). Crop performance is best in soils with a pH between 6.0 and 6.5. 12 University of Ghana http://ugspace.ug.edu.gh (Doggett, 1970). However, sorghum appears to be more tolerant to soil acidity and alkalinity compared to maize and other grain crops (Undersander, 2003). Nevertheless, in acid soils with pH below 4.7, root growth and overall plant growth may be severely constrained (Kidd and Proctor, 2001; Menezes et al., 2014). Under such conditions, liming is recommended to improve nutrient solubility and availability, while the application of acidifying fertilisers may improve growth in alkaline soils (Tan et al., 1991). Sorghum is moderately-tolerant to salinity; it is more tolerant than maize, wheat and rice but less tolerant than barley (Prasad and Staggenborg, 2009). Thus, it could be considered as an alternative crop in salt-stressed soils (de Lacerda et al., 2005). However, a high accumulation of salts such as sodium calcium and magnesium particularly during germination and early vegetative stage may severely limit germination and primary root growth (Oliveira et al., 2011). An adequate supply of nutrients is necessary for plant establishment and growth. Nutrient application rates for sorghum differ according to climate, cultivar, soil type, amount of residual nutrients in the soil, target yield and cropping sequence (Daba and Zewedie, 2001; Kumar et al., 2011). As is the case with most plants, nitrogen is required in large quantities and is essential for vigorous growth and maximum yield. It may be applied to the crop in organic forms (manures and compost) or in inorganic forms such as urea, di-ammonium phosphate, ammonium nitrate, calcium nitrate or sodium nitrate (Tucker, 1999). Its availability is most critical from the 5-leaf to flowering stage (Rose et al., 2013). An inadequate supply of nitrogen may reduce leaf size and quality, delay crop maturity and affect seed development (Roy et al., 2005). Potassium (K) is critical for optimum photosynthesis and the activation of enzymes (Roy et al., 2005). In sorghum, maximum uptake of K occurs during the early growth stages; thus, application at sowing is recommended particularly for fields with less than 150 kg K2O per hectare and in 13 University of Ghana http://ugspace.ug.edu.gh cropping systems where all plant residue is removed at harvest (Curt et al., 1995). A top-dressing may be applied 20 to 30 days after sowing under severe conditions of K deficiency (Agropedia, 2009). An optimum supply of potassium improves disease and pest resistance, water balance and stalk strength in the crop (Smith and Frederiksen, 2000). Critical soil P values for sorghum are slightly less than maize and half that of wheat, indicating the relatively high uptake-efficiency of the crop (Gemenet et al., 2016). However, significant amounts of P are required throughout growth; from the seed germination and establishment to grain filling (Young and Long, 2000; Gelli et al., 2014). P application is most critical in soils with less than 5 mg P/kg available P and may be necessary in alfisols than in vertisols (El-Swaify et al., 1985; Sahrawat et al., 1996). Phosphorus should be applied at sowing typically by furrow placement or side dressing (Agropedia, 2009). An adequate supply of P promotes seedling establishment, root formation, flowering and maturation of grain which translates into higher biomass and grain yield (Bolland and Baker, 1988). The most important climatic factors for optimum sorghum growth includes rainfall and temperature (Konate, 1984). In West Africa, sorghum is cultivated in the humid/sub-humid Guinea-savannah, semi-arid Sudan-savannah and arid Sudano-Sahelian zones which vary widely in rainfall distribution (Leiser et al., 2012). While acceptable yields can be produced with limited moisture, maximum yield requires a moisture distribution of about 600 to 700 mm particularly during the reproductive stage (Prasad and Staggenborg, 2009). Interestingly, the crop is also more tolerant to periodic flooding compared to maize (Borell et al., 2000). Warm temperatures such as between 21oC and 35oC are required for seedling germination and establishment whereas temperatures between 26oC to 34oC are necessary for vegetative growth 14 University of Ghana http://ugspace.ug.edu.gh and reproductive development (Tiryaki and Andrews, 2001; Alagarswamy and Ritchie 1991; Prassad et al., 2006a). 2.6. Phosphorus in soils 2.6.1 Forms of phosphorus in soils Soil phosphorus exists in two main forms; inorganic and organic (Hansen et al., 2004). Organic soil P occurs in plant residues, manures and microbial organisms (Kwabiah et al., 2003). Its amount in soils can vary widely from 20 to 80% of total soil P (Holford, 1997). It may be present as stabilised forms such as inositol phosphates and phosphonates or active forms such as orthophosphate diesters and organic polyphosphates (Turner et al., 2002). Organic P is released to plants via mineralisation processes mediated by soil organisms and plant roots (Buresh and Smithson, 1997). To a large extent, availability of organic P to plants is affected by the rate of microbial conversion to available forms (orthophosphates) rather than its total amount (Richardson, 1994). Mineralisation processes in turn are highly influenced by soil moisture, temperature and pH (Turner et al., 2007). Inorganic forms of P in soil consist of hydrous sesquioxides, aluminium and iron compounds in acidic soils and calcium compounds in alkaline, calcareous soils (Gahoonia et al., 1992). Inorganic pools are derived mainly through the dissolution of apatite and other primary P minerals and the mineralisation of soil organic matter (Buresh and Smithson, 1997). The form of inorganic P in solution is influenced largely by soil pH (Furihata et al., 1992). The primary orthophosphate ion (H PO -2 4 ) is more predominant in acid soils (below pH 6) whereas the secondary orthophosphate (HPO 2-4 ) is predominant in alkaline soils (Condron et al., 2005). However, both forms are present in approximately equal proportion at neutral pH. The various forms of P have notable differences 15 University of Ghana http://ugspace.ug.edu.gh in chemical behaviour, mobility and availability in soils (Reddy and DeLaune, 2008). Transformations between inorganic and organic P forms occur continuously to maintain equilibrium conditions (Tiecher et al., 2012). 2.6.2 Phosphorus availability and its acquisition by plants Only 1 to 5% of total soil P occurs in plant-available forms (Molla et al., 1984). Thus, although total P in fertile soils may exceed plant requirements, its availability to plants may be severely restricted (Shen et al., 2011). P availability in soils is extremely complex and involves several physical, chemical and biological processes (Fink et al., 2016). It may be described in two broad contexts; spatial availability and bioavailability (Shen et al.., 2011). Bioavailability of P is linked to the concentration of soluble or dissolved forms of P present in soil for plant uptake (Hinsinger, 2001). It is influenced by several factors including mineralogy, composition and rate of hydrolysis of organic matter (Marschner, 1995). To a large extent, it is also influenced by the rate of adsorption and desorption of phosphates which in turn is highly affected by soil pH, composition of soil solution, the degree of weathering, drainage potential of the soil and clay fraction (Schaefer et al., 2008). For instance, P adsorption is significantly higher in clay loam soils than in sandy soils (Hinsinger, 1998). Similarly, P adsorption is less at slightly acidic to neutral pH than in extremely acidic and alkaline soils (Reddy and DeLaune, 2008). In acid soils, P is severely adsorbed by oxides and hydroxides of aluminium and iron, forming insoluble complexes such as strengite [FePO4.2H2O] and variscite [AlPO4.2H2O]) (Fairhurst et al., 1999). In neutral and alkaline soils, P is also likely to form stable complexes with calcium or magnesium (Devau et al., 2010). In parts of West Africa, P sorption rates can be as high as 405 mg P/kg leading to low 16 University of Ghana http://ugspace.ug.edu.gh bioavailability of P (Bationo et al., 2003). According to Leiser et al., (2012), available P may be as low as 2 mg P/kg in some sorghum-growing areas. Spatial availability of P on the other hand refers to the presence of sufficient amounts of soluble P within the rhizosphere for plant acquisition (Hinsinger, 1998). Due to its low mobility in soil, P stocks within the rhizosphere may be depleted (Marschner, 1995). This may be worsened by low soil moisture content, high temperature, aeration and compaction (Hash et al., 2002). Spatial availability of P may be improved by the placement of fertilisers within the effective rhizosphere or its slow diffusion from bulk soil to the rhizosphere (Lynch, 1995). P is absorbed by plants as inorganic phosphate ions H2PO - 4 or HPO 2- 4 (Hinsinger, 1998). Due to the steep concentration gradient (created by the low concentrations of P ions in soil solution and its high concentrations in root cells), uptake across the plasma membrane of root epidermal and cortical cells requires high-affinity active transport systems mediated by H+ co-transporters (Shin et al., 2004). Upon acquisition, P is loaded into the xylem and translocated upward into shoots where it plays structural, biochemical and physiological roles in plant growth and development (Ai et al., 2009). This includes its involvement in metabolic processes such as photosynthesis, respiration, glycolysis, enzyme activation, membrane synthesis and stability (Schachtman et al., 1998). Within plants, P occurs as sugar phosphates in the cytoplasm, phospholipids in chloroplasts, chlorophyll and plant hormones, inorganic phosphates in the vacuole and as phytin in seeds (Tucker, 1999). The major function of phosphorus is in energy storage and transfer as it is an essential component of the energy molecule adenosine triphosphate (ATP). It is also an integral component of nucleic acids; phosphate groups link DNA or RNA molecules, thus form the backbone of nucleic acids (Shen et al., 2011). 17 University of Ghana http://ugspace.ug.edu.gh 2.6.3 Phosphorus status of West African soils The available P content of cultivated soils in developed countries exceeds critical amounts required for plant growth (Melese et al., 2015). On the other hand, West African soils are typically vertisols, entisols, alfisols or inceptisols which are low in available P, highly-weathered, light-textured and highly-saturated with cations (Kang, 1985; Bationo and Mokwuenye, 1991; Gemenet et al., 2016). The low P status of West African soils can be attributed to two main factors. First is the geologic origin or the parent material from which these soils are derived (Fink et al., 2016). Due to a low occurrence of phosphate-bearing rocks such as apatite, strengite and variscite in the region, soils are derived mainly from rocks of low P content (van Kauwenbergh, 2006). Thus, West African soils are inherently P deficient. For instance, approximately 80% of soils in northern Ghana are formed from acidic parent material with low P availability of 8 mg kg-1 (Braimoh and Vlek, 2004). The second cause of West Africa’s low P status is nutrient depletion (Shepherd and Soule, 1998). In most cropping systems, nutrient output generally exceeds input resulting in a negative nutrient balance (Bationo et al., 2012). Soil nutrients are lost through erosion, leaching, burning of cover vegetation and the removal of crop residues at harvest (Kisinyo et al., 2012). In the last 30 years, an estimated 15 million tonnes of P has been lost from cultivated soils in Africa (Smaling, 1993). Bationo et al., (2012) reports that an average of 75 kg P ha-1 is lost annually from soils in several African countries. This situation is not limited to West Africa as annual depletion rates exceed 6.6 kg P ha-1 in Eastern African countries such as Kenya, Ethiopia and Rwanda (Smaling et al., 1997). Unlike nitrogen, P replenishment is mainly fertiliser dependent (Nziguheba, 2007). Unfortunately, such soil enhancement interventions are scarce in smallholder systems and nutrient replenishment depends to a large extent on the slow mineralization of rocks and minimum recycling of organic matter (Verde and Matusso, 2014). 18 University of Ghana http://ugspace.ug.edu.gh 2.7 Effect of phosphorus deficiency on sorghum productivity in West Africa Soil P deficiency has been established as one of the main constraints to sorghum production (Gemenet et al., 2016). Its negative effect on seedling vigour, plant height, grain and stover yield has been documented in Burkina Faso, Mali and Ghana (Bationo et al., 2012, Leiser et al., 2015). When deficiency is mild, clearly recognisable symptoms may be absent. However, plants often appear stunted and exhibit a general lack of vigour (Sanginga and Woomer, 2009). On the other hand, severe P deficiency results in delayed maturity, marked reductions in overall growth and leaf size, with plants appearing spindly with dark green leaves (Katyal and Das, 1993). In certain varieties, leaves turn brown resulting in poor radiation interception and consequently sub-optimal rates of photosynthesis and carbon assimilation (Smalberger et al., 2006). Perhaps, the most- recognisable symptom of P deficiency in sorghum is the development of reddish-purple colouring on leaf sheaths as a result of the accumulation of anthocyanin (Grundon et al., 1986). The degree of purpling may vary from cultivar to cultivar but typically, red overtones appear first on leaf tips and margins of older leaves and progress towards the midrib and the full length of the blade (Sanginga and Woomer, 2009). Low soil P reduces grain filling, seed size and tillering and has also been shown to increase disease pressure on sorghum fields (Wall et al., 1994; Jones et al., 2003). For instance, the production and exudation of strigolactones for the parasitic weed Striga is enhanced in low P soils resulting in serious yield losses (Yoneyama et al., 2007). P deficiency symptoms are exacerbated by the marked reduction in crop response to N (Zhao et al., 2007). When P and N deficiency occur simultaneously, chlorophyll content is reduced and leaves appear pale green. 19 University of Ghana http://ugspace.ug.edu.gh 2.8 Crop adaptation to soil P deficiency Plant adaptation mechanisms to soil P deficiency are based on acquisition, translocation, storage and utilisation of P (Ciarelli et al., 1998). The two main adaptation strategies are those that enhance P acquisition from surrounding soil (P acquisition efficiency) and those that improve physiological use of internal P [P utilisation efficiency] (Vance, 2001). Phosphorus Acquisition efficiency (PAE) is defined as the total P in the above-ground tissue at maturity per unit area while P utilisation efficiency (PutE) is the grain yield per unit of acquired P (Wang et al., 2010). Mechanisms of improved uptake are probably more beneficial in soils rich in total P than in deficient soils (Rose et al., 2013). Such mechanisms are aimed at maximizing exploration of surrounding soil for continuous uptake (Lynch and Ho, 2005). These include alterations in root architecture, morphology and distribution patterns (Liao et al., 2001; Shen et al., 2011). For instance, under P-deficient conditions, some plant species may exhibit increases in root/shoot ratio, root hair formation, root branching and elongation (Shen et al., 2011). For example, white lupin (Lupinus albus) and Arabidopsis (Arabidopsis thaliana) plants may develop a denser cluster of roots under conditions of low P stress thereby facilitating effective exploration of soil (Vance, 2001). Similar increases in root/shoot ratio are observed in the common bean (Phaseolus vulgaris) under low P stress (Lynch and Brown, 2008). An increase in the number and length of lateral roots and root hairs have been observed in wheat, barley, maize and Arabidopsis (Zhu and Lynch, 2004). Due to the relative immobility of P in soils and its abundance in top layers, modifications in branching such as a more horizontal growth angle may also enhance topsoil foraging of P (Hodge et al., 2009). This phenomenon has been observed in P-efficient genotypes of common bean and maize (Lynch and Brown, 2008). 20 University of Ghana http://ugspace.ug.edu.gh Another common adaptation of plants under soil P starvation is mycorrhizal symbioses (Brundrett, 2009). Such associations can improve P uptake by extending the nutrient absorptive area via the formation of hyphae or mycelia (Harrison, 2005). Differences in responsiveness to mycorrhizal inoculation under low P conditions can distinguish between P-use efficient genotypes and non- efficient ones (Feddermann et al., 2010). Raju et al. (1990) reports that in sorghum, symbiosis with vesicular-arbuscular mycorrhiza (VAM) resulted in increases in root length and a corresponding increase in P uptake. This is in contrast to reports by Leiser et al. (2016) which suggest that sorghum showed no positive relationship between VAM colonisation and grain yield. Kaeppler et al. (2000) suggest that symbiotic P acquisition may be more beneficial under severely low P conditions and less in fertile soils. Apart from alterations in morphology and architecture, P uptake may be improved through root- induced biochemical changes (Shen et al., 2011). The synthesis and exudation of compounds such as low molecular weight organic acids (LMWOA), carboxylates, phosphatases and phytases in the rhizosphere play an essential role in enhancing the availability of soil P and hence its uptake (Marschner, 1995; Zhang et al., 2010). For instance, LMWOA reduce the pH of alkaline soils thereby enhancing the dissolution of soluble P minerals (Marschner, 1995). Carboxylates such as citrate, malate and oxalate increase P availability by mobilising inorganic P from precipitated forms through ligand exchange and chelation of Al, Fe and Ca while phosphatases enhance greater mobilisation of organic P through enzyme-catalysed hydrolysis (Vance, 2001). Such changes are reported in rice and sorghum variety SC 283 in which roots produce citrate that complexes with Al and enhance the dissolution of P fixed by Al in acid soils (Magalhaes, 2007). While improved P acquisition in low P soils is undoubtedly beneficial in increasing yields, some have argued that such mechanisms are likely to cause further P depletion in cultivated soils (Lynch, 21 University of Ghana http://ugspace.ug.edu.gh 2007; Veneklaas et al., 2012). This argument is based on the existing imbalance between P fertilisation and its uptake/removal in small-scale farming systems (Rose et al., 2013). Thus, some authors suggest that a more efficient utilisation of internal P will ensure that limited P resources are channeled into producing higher yields (Veneklaas et al., 2012; Gemenet et al., 2016). It has also been suggested that breeding for internal P use-efficiency would be more relevant to the low input systems of West Africa. Arguably, due to declining root growth during reproductive stages, an efficient utilisation of internal P for grain development may be more valuable than the amount of P available for uptake (Rose et al., 2008). Efficient utilisation of internal P includes the production of the maximum amount of grain with the smallest accumulation of the nutrient (Moll et al., 1982). Efficient internal utilisation of P may also be defined as increased growth per unit of acquired P (Schachtman et al., 1998). The activation of these mechanisms in low P soils may differentiate between P-efficient and inefficient genotypes (Lynch and Brown, 2006). Further research is required to explain the physiological factors that contribute to improved utilisation of internal P in low P soils. 2.8 Strategies for improving sorghum productivity in low P soils 2.8.1 Traditional management strategies Several measures have been adopted to improve agricultural productivity on low P soils (Kaizzi et al., 2007). Key among these is the application of organic and inorganic fertilisers (Maerere et al., 2001; Verde and Matusso, 2014). Inorganic P fertilisers available in West Africa (though on a limited scale) include single super phosphate (SSP), triple super phosphate (TSP), diammonium phosphate (DAP) and rock phosphate (Nekesa et al., 2005). The advantage of these inorganic sources of P in improving crop yield is their immediate availability for plant uptake and use (Verde 22 University of Ghana http://ugspace.ug.edu.gh and Matusso, 2014). Nevertheless, organic fertilisers such as crop residues, green manures, compost and animal manures are also useful in enhancing crop performance in low P soils as have been demonstrated in many smallholder farms in Nigeria, Burkina Faso and Ghana (Mwangi et al., 2002; Odedina et al., 2011). Unfortunately, the use of fertilisers as a strategy for improving sorghum production in low P soils in West Africa is constrained by several factors. Perhaps, the most important factor is the high cost of fertilisers, farmers’ inability to afford the commodity and lack of access to credit (Leiser et al., 2015). Also, in West Africa, mineral fertilisers may be unavailable in rural areas due to lack of infrastructure, inefficient distribution systems and lack of fertiliser management policies (Chianu et al., 2012). Given that affordability and access to mineral fertiliser remain a challenge for smallholder farmers, many policies encourage the use of organic nutrient sources (Chianu et al., 2012). However, as is the case of mineral fertilisers, manure and compost are often unavailable in situations where livestock production is not intensive or a main component of the farming system (Parentoni et al., 2008). In arid parts of West Africa for instance, plant residues are more likely to be removed at harvest for use as dry season fodder than as a soil enhancement strategy (Lynch, 2007). Even when available, the P content of organic matter is highly variable and often insufficient to meet crop requirements (Risse et al., 2006). Furthermore, the impact of fertilisers on crop productivity in smallholder farms is diminished by farmers’ low literacy levels and lack of information leading to improper application rates and placement methods (Buresh and Smithson, 1997). Thus, a large proportion of added P is lost through volatilisation and run-off leading to expensive annual applications (Driessen et al., 2000). Perhaps the greatest concern associated with P fertilisation is the rapid depletion of rock phosphate reserves and the potential phosphate crisis (Vance, 2001). Prices of P fertilisers are expected to increase with depleting 23 University of Ghana http://ugspace.ug.edu.gh reserves and increasing demand; this is expected to make the commodity even more inaccessible to smallholder farmers in West Africa (Cordell et al., 2009). Lastly, fertiliser is seldom applied to traditional crops such as sorghum as is the case in many parts of West Africa; the introduction of fertiliser subsidies has had little effect on the fertilisation of the crop (Chianu et al., 2012). In order to improve sorghum productivity on marginal soils in West Africa, fertiliser-use by small- scale, resource-poor farmers must be complemented with other approaches. Proposed strategies include improved crop-livestock integration for manure production, crop-tree-livestock integration to reduce erosion and provide green manure and the extraction of P from urban waste water (Zomer et al., 2009; Bayala et al., 2013). Another proven, environmentally-friendly and relatively inexpensive approach is the development of P use-efficient varieties (Lynch, 2007). This strategy is beneficial in two ways; it is likely to increase productivity in the low input systems in West Africa while encouraging a reduction in fertiliser use in high input systems without compromises in yield (Weaver and Wong, 2011). 2.8.2 Breeding strategies 2.8.2.1 Conventional approaches to breeding for phosphorus-use efficiency Phosphorus use efficiency (PUE) is defined as the amount of biomass (stover and grain) produced per unit of available P in the soil (Moll et al., 1982). Rose et al. (2013) defined PUE as the grain yield per unit of P fertiliser applied or the grain yield per unit of P in the aboveground biomass. The concept of PUE has been reported in rice, maize, wheat and sorghum (Holford et al., 1997; Duan et al., 2007). In Africa, breeding for PUE in sorghum, although limited, has been approached mainly through the use of conventional methods such as pedigree selection, backcrossing and population 24 University of Ghana http://ugspace.ug.edu.gh improvement (Maqbool et al., 2001). Usually, such studies aim at identifying genotypes that produce acceptable yields under limited P conditions followed by crossing selected genotypes with sensitive ones (Gemenet et al., 2016). Conventional breeding employs two main strategies; Selection based on overall performance (grain yield and/or biomass) in low P soils or selection for phenotypic traits associated with P-use efficiency (Rose et al., 2013). Performance-based selections involve rigorous screening of genotypes on P-deficient and/or high P soils to reveal genetic variations in crop response (grain yield and biomass) (Quijano-Guerta et al., 2002). This has been suggested as one of the best predictors of genotypic variation for PUE and is based on the premise that certain genotypes produce similar yield under both low and high P conditions (Rao, 2001; Schaffert et al., 2001). In Mali, this method was used by Leiser et al. (2015) to identify P use-efficient sorghum genotypes among West African sorghum. The challenge of such phenotypic selections is for breeders to determine whether selections should be conducted under the conditions of low P (direct selection) or high P (indirect selection) (Gemenet et al., 2016). There have been contrasting reports on the efficacy of direct selection methods as opposed to indirect methods (Gemenet et al., 2016). While some researchers suggest that indirect selection in high P soils are likely to provide higher heritability estimates, others suggest that evaluations under low P stress are more likely to reveal P use-efficient genotypes (Banzinger and Cooper, 2001; Leiser et al., 2012). However, Banziger et al. (1997) reports that genotypes selected under low P stress may not be responsive to P fertilisation. Thus, Gallais and Coque (2006) recommended selection under P conditions that do not reduce yield beyond 40%. Alternatively, breeding for PUE may be approached indirectly through the assessment morphological characters associated with the trait. These include plant height, leaf area, stem thickness, percent dry matter, amount of P partitioned to stover and grain and the degree of 25 University of Ghana http://ugspace.ug.edu.gh exudation of P- mobilizing compounds (Sinclair, 1998; Broadley et al., 2007). The most-evaluated morphological traits are those associated with root architecture and morphology such as the ability to produce denser or more branched roots in P-depleted soils (Marschner, 1995; Kamara et al., 2003). Extensive genetic variation has been observed in sorghum root systems under low P stress, suggesting the critical role of roots in sorghum adaptation to low P soils (Jordan et al., 1979). However, the main challenge in evaluating root architecture and morphology in breeding work is the requirement for detailed analyses of root systems and rhizosphere interactions (Kondo et al., 2003). This usually involves destructive sampling or removal of the entire intact root system from the soil which may be impractical under field conditions. Also, many root traits have low heritability and are affected by environmental factors such as soil type and water availability, both of which are difficult to control in the field (Hash et al., 2002). Thus, although root evaluations may suit the low-cost nature of African breeding programmes, this form of selection must be used with caution (Hash et al., 2002). Some conventional breeding programmes for PUE in sorghum have attempted selections based on genotypic variations in the level of VAM association (Hash et al., 2002). However, according to Hash et al., 2002), breeding for improved VAM association in sorghum was not effective in enhancing its adaptation to low P soils as variations in VAM colonisation did not significantly improve P uptake and yield in low P soils. Besides, such experiments are limited by variations in quantities of the fungi in soil and difficulties in inoculating plant roots (Vance, 2001). It is also difficult to evaluate the contribution of the fungi to P uptake compared to uptake by the plant’s own root system (Hash et al., 2002). While these forms of selection are straightforward, they require the development of near-isogenic lines which may take several years of backcrossing (Guo et al., 2009). Another major disadvantage 26 University of Ghana http://ugspace.ug.edu.gh of these forms of selection is the tedious testing in multi-environments and the irregular distribution of P across field plots (Leiser et al., 2015). The heterogeneous availability of P in field plots may result in low heritability or the masking of genetic effects by experimental effects (Wissuwa et al., 2001). Thus, Kaeppler et al., (2000) suggest the use of controlled conditions such as pots, lysimeters and hydroponic systems as substitutes for field conditions. However, these require huge investments in equipment and screening procedures may not closely mimic or represent normal field conditions (Hash et al., 2002). According to Leiser et al. (2015), soil heterogeneity in field trials may be accounted for by spatial adjustments and the use of mixed models in data analysis to increase precision and reveal the extent of genotype x phosphorus interaction (Leiser et al., 2015). 2.8.2.2 Molecular approaches to breeding for phosphorus-use efficiency Due to the limitations of sole phenotypic selections, genomic tools such as marker-assisted selection (MAS) have been proposed for breeding for PUE. MAS can increase the precision of selection under field conditions and hasten the breeding process by facilitating the choice of parental lines and reducing the need for costly multi-locational screening (Hash et al., 2002; Wissuwa et al., 2016). However, the use of MAS in breeding for improved PUE in sorghum depends on the identification and validation of genes or quantitative trait loci (QTL) associated with the trait (Gelli et al., 2014; Gemenet et al., 2016). Although the genetic basis of PUE traits is complex, the completion of the sorghum genome has provided new avenues for elucidating gene function in the crop (Paterson et al., 2009). Upon identification, these genes or genomic regions may be isolated and their expression manipulated to improve P-use efficiency in the crop (Crawford and Forde, 2002). Attempts to identify genes or QTLs associated with PUE in sorghum include work by Koegel et al. (2013) who identified two genes (SbAMT3; 1 and SbAMT4) coding 27 University of Ghana http://ugspace.ug.edu.gh for ammonium transporters that were found to positively influence P uptake. The most important QTL for P- acquisition efficiency however, ‘Phosphorus Uptake 1’ (PUP-1) was mapped on chromosome 12 in the rice variety Kasalath (Wissuwa et al., 1998; 2002). The major gene Phosphorus Starvation Tolerance 1 - PSTOL1 which promotes early root development, thus improving uptake and grain yield in low P soils was identified in the PUP 1 (Gamuyao et al., 2012). Subsequently, other QTLS associated with tolerance to low P have reported in pearl millet and maize (Gemenet et al., 2016). QTL have also been identified for other PUE traits such as dry weight of shoots, tiller number (in wheat), leaf area and chlorophyll content in grains (in maize) (Su et al., 2006; Cai et al., 2012). The identification of these QTL and genes in sorghum followed by their introgression into elite African lines holds huge promise for developing P use-efficient African sorghum varieties. 2.8.2.3 Induced mutagenesis as an alternative breeding approach Both conventional and molecular approaches in plant breeding require the availability of sufficient levels of genetic variation for the desired trait (Gemenet et al., 2016). A large variation for PUE exists among crop species; for instance, wheat, white lupin, maize, rice and barley differ widely in their levels of PUE (Hammond et al., 2009). More recently, significant levels of genetic variation for PUE has been reported among West African sorghums (Leiser et al., 2015; Gemenet et al., 2016). Their work showed Guinea-race genotypes to be more uptake-efficient whereas Caudatum- based ones were more utilisation-efficient. It has been suggested, that a combination of these two sources of germplasm may be necessary to develop a superior genotype that exhibits both uptake and utilisation efficiencies in low P environments (Gemenet et al., 2016). Whilst this is entirely possible, it may require several years of backcrossing and the development of recombinant inbred lines. Hybridisation may also be hampered by undesirable linkage drag and in some cases 28 University of Ghana http://ugspace.ug.edu.gh incompatibility of the selected parents (Acquaah, 2015). Additionally, given that a substantial proportion of applied fertiliser is not acquired by plants, there still exists a significant margin for improving P acquisition in the crop (Hash et al., 2002). In West Africa, where fertilisation of traditional crops such as sorghum is almost non-existent, the availability of a more diverse genetic base for PUE in the crop is likely to be beneficial (Doumbia et al., 1993). In addition to hybridisation, several methods are available for increasing genetic diversity in crops. These include genetic engineering and induced mutagenesis. Artificial mutagenesis offers a highly useful method for expanding or broadening the gene pool of any cultivated species (Dillon et al., 2007). Its primary role in plant breeding is the generation of hitherto unavailable alleles (Sikora et al., 2011). One of its key advantages over hybridisation is its ability to introduce random changes throughout the genome, resulting in a wide variety of mutations in target genes whereas hybridisation exploits existing variability within crop species for the development of new individuals by creating new combinations of existing traits (Sikora et al., 2011). Compared to transgenic methods, artificial mutagenesis can be readily exploited without regulatory restrictions (Parry et al., 2009). Artificial mutagenesis may be achieved through the use of physical mutagens such as gamma rays, x-rays and UV light or chemical mutagens such as sodium azide, methylnitrosourea (MNU) and ethyl methanesulphonate (EMS) (Kodym and Afza, 2003). The main advantage of chemical mutagens is that they often result in single-nucleotide polymorphisms (SNPs) rather than deletions and translocations that are typical of radiological methods (Sikora et al., 2011). EMS is widely- used for seed-propagated crops due to its potential for producing a relatively high density of irreversible mutations (nucleotide substitution) (Talebi et al., 2012). This mutagen typically results in random point mutations (GC to AT base-pair transitions) by the alkylation of guanine bases 29 University of Ghana http://ugspace.ug.edu.gh which subsequently causes the replacement of cytosine with thymine by DNA-polymerase (Till et al., 2007). It has been shown to be highly mutagenic in sorghum and optimized protocols are readily available in published reports (Xin et al., 2008). The use of artificial mutagenesis for the improvement of tolerance to abiotic stresses such as drought and salinity have been reported in cereals (Brar and Jain, 1998). For example, increased tolerance to aluminium and low P stress via mutagenic treatments have been reported in Brazilian and Chilean wheat respectively (Camargo et al., 1995). Some 15 induced sorghum mutant accessions with tolerance to various biotic and abiotic stresses have been registered according to the International Atomic Energy Agency database (http://www-mvd.iaea.org/MVD/default.htm). Given that the availability of a genotype which possesses both uptake and utilisation efficiencies is beneficial for West Africa, this work is an attempt to identify new sources of both traits in sorghum (Gemenet et al., 2016). If successful, those genotypes may be used either directly as new varieties or as parents in recombinant breeding programmes. The approach is to increase the genetic variability for PUE in the crop through the use of artificial mutagenesis followed by the selection of desirable mutants in multi-locational trials. 30 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE 3.0 FARMERS’ PRODUCTION CONSTRAINTS, PERCEPTIONS OF SOIL FERTILITY AND PREFERENCE FOR SORGHUM TRAITS 3.1 Introduction Ninety percent of Ghana’s sorghum is cultivated in the three northern regions (Upper East, Upper West and Northern Regions) mainly as a rain-fed crop by subsistence farmers (RUDEMA, 2006). Its importance in these regions cannot be over-emphasized. Sorghum grains are an important food staple for many rural folk and are used in the preparation of several local beverages (Kudadjie et al., 2006). More recently, sorghum grains have been used as a substitute for barley in industrial breweries, allowing local farmers to earn additional income and breweries to save foreign exchange (Angelucci, 2013). Stover may also be used as fodder, household fuel, and as raw material for roofing traditional homesteads (Kudadjie et al., 2004). Weaving baskets and mats from sorghum stover is also an important cottage industry for rural women (Kudadjie-Freeman and Dankyi-Boateng, 2012). In spite of its importance, sorghum productivity in northern Ghana is extremely low; rarely exceeding 2 t/ha (Kombiok et al., 2012). For instance in the Upper East Region, annual yields are estimated at 700 kg/ha (Al-Hassan and Jatoe, 2002). According to FAO (2013), sorghum production in northern Ghana dropped from 287,069 tonnes in 2011 to 256,736 tonnes in 2013. Such low yields threaten the food security and livelihood of many of the regions’ rural poor (Quaye, 2008). Low sorghum yields can be attributed to factors such as drought, high incidence of diseases and pests, cultivation of inherently low-yielding varieties and low plant densities on farms (SARI, 1995; Braimoh and Vlek, 2006; Atokple, 2010). Another key constraint to sorghum production in 31 University of Ghana http://ugspace.ug.edu.gh northern Ghana is widespread soil infertility (Kombiok et al., 2012). In view of this, several strategies have been proposed to increase crop productivity on marginal soils, key of which is the promotion of fertiliser use among smallholder farmers (SARI, 1995; Kombiok et al., 2012). However, in spite of persistent efforts, reports indicate weak demand and low fertiliser use among sorghum farmers in Ghana and Africa as a whole (IFDC, 2006; Bumb et al., 2011). This has largely been attributed to high cost of fertilisers and its unavailability on rural markets (Laube, 2012). However, even when fertilisers and subsidies are available, there have been difficulties in convincing smallholder farmers to apply fertiliser to the crop (MoFA, 2009; Rware et al., 2016). This suggests that other factors in addition to unaffordability and unavailability of fertilisers might influence farmers’ attitudes towards its use. Farmers’ perceptions have been shown to influence their adoption of agricultural practices and technologies (Marenya et al., 2008; Fosu-Mensah et al., 2012; Ndamani and Watanabe, 2015). Studies in Kenya, Ethiopia and Nigeria for instance revealed that farmers’ perceptions of soil fertility influenced their attitudes towards fertiliser use and other soil management strategies (Mairura et al., 2007; Moges and Holden, 2007; Baba, 2017). In Ghana, given persistent reports of low fertiliser use among sorghum farmers, there is a need to seek more information on farmers’ perceptions of soil infertility as well as their preferred soil management strategies (MoFA, 2009). This information may provide more insight on farmers’ attitudes towards sorghum fertilisation which in turn may guide stakeholders on the need or otherwise to prescribe alternative approaches to increasing sorghum productivity in low-input systems. Although similar studies have been conducted across West Africa, perceptions can be location-specific and may also depend on socio- economic status and cultural considerations of the target group (Mairura et al., 2007; Ndamani and 32 University of Ghana http://ugspace.ug.edu.gh Watanabe, 2015; Yeshaneh, 2015). Thus, it is important to focus on assessing the perceptions of these farmers at the community level. Additionally, a better understanding of the constraints to sorghum production from the farmers’ perspective will enable stakeholders effectively prioritise and address constraints in the target areas. Also, an assessment of farmers’ preference for traits will help scientists set relevant breeding goals and enhance the willingness of farmers to accept and use prescribed measures. The role of participatory rural appraisal (PRA) is to facilitate a greater interaction between farmers and researchers and provide a platform for assessing farmers’ indigenous knowledge and perceptions through semi-structured interviews and questionnaires (Theis and Grady, 1991; Chambers, 1992c). The specific objectives of this study were to; 1. identify production constraints of smallholder sorghum farmers in the Upper East Region of Ghana, 2. determine farmers’ perceptions of soil fertility, 3. identify farmers’ soil management methods and 4. identify farmers’ preferences for sorghum traits. 33 University of Ghana http://ugspace.ug.edu.gh 3.2 Materials and methods 3.2.1 Description of study sites The study was conducted in four villages; Gbane and Yameriga in the Talensi-Nabdam district and Gumyoko and Tempelim in the Binduri district. Both districts are located in the Upper East region of Ghana (Lat. 10o15' and 10o10'N, Long. 0oand 1o4'W). The Talensi-Nabdam district lies between Latitude 10.15o and 10.60o north of the Equator and Longitude 0.31o and 10.5o (GSS, 2013). The topography is predominantly undulating, thus erosion-prone (Kpongor, 2007). Rainfall pattern is unimodal and unpredictable and is characterised by a short rainy season from May to October (annual mean of 950mm) followed by a long dry season of six to seven months. Temperatures are generally high (between 26oC and 45oC). Natural vegetation is guinea savannah woodland characterised by short scattered trees. Cultivated crops include sorghum, millet, rice and a range of legumes and vegetables (Kudajie-Freeman and Boateng, 2012). Soils here are generally alfisols which vary from light-textured sandy loams to heavier-textured soils (Kpongor, 2007). Soil pH ranges from 5.1 to 8.7, total nitrogen from 0.61 to 11.6 mg/kg soil while available phosphorus ranges from 1.75 to 14.75 mg/kg soil (Braimoh and Vlek, 2004). Binduri is located between Latitudes 11o 111 and 10o 401 N and Longitudes 0o 181 W and 0o 61 E (GSS, 2013). The climate of Binduri is similar to that of Talensi-Nabdam, however, mean annual rainfall is lower (800 mm). Natural vegetation in Binduri district is mainly the Sahel Savannah type with scattered trees (GSS, 2013). Similar to Talensi-Nabdam, its topography is undulating. Thus, low-lying areas are prone to erosion. Soils here are generally of the savannah ochrosol type, ranging from sandy loams of good water retention capacities to clay loams (Kpongor, 2007). In both districts, sorghum is a major crop and it is grown mainly by small-scale and/or resource-poor 34 University of Ghana http://ugspace.ug.edu.gh farmers. These districts were selected based on the importance of the crop as well as yield and socio-economic status. A geographical representation of the two districts is shown in Figure 3.1. Figure 3.1. Geographical representation of Talensi-Nabdam and Binduri districts. 35 University of Ghana http://ugspace.ug.edu.gh 3.2.2 Sampling procedure and data collection techniques Focused group discussions were organised with stakeholders (farmers, agricultural extension officers and researcher) at each village to obtain information on general constraints. One group was organised per village, each comprising a maximum of seven farmers including male and female representation. Subsequently, heads of households or available family members involved in farming activities were interviewed individually using semi-structured questionnaires (Appendix 1). The questionnaire comprised five sections with questions related to demographics, constraints to sorghum productivity, perceptions and assessment of soil fertility, fertiliser use and other soil management strategies and lastly, preferences for traits. Interviews were followed by visits to some sorghum fields during which direct observations were made. A total of 122 sorghum farmers were sampled in the two districts. The number of respondents that were sampled in each village is given in Table 3.1. Table 3.1. Number of respondents across the study sites Districts Villages Number of respondents Gbane 28 Talensi-Nabdam Yameriga 32 Gumyoko 32 Binduri Templenim 30 Total 122 36 University of Ghana http://ugspace.ug.edu.gh 3.2.3 Data analysis Data were summarized into averages and frequencies using XLSTAT 2012. Descriptive statistics, t-tests and chi-square tests were done using IBM Statistical Package for Social Sciences (SPSS) version 23. 3.3 Results 3.3.1 Demographic characteristics of households Out of 122 farmers, 73 were male, representing approximately 60%. A male majority was the trend in all four communities. The highest percentage of female farmers (47%) was observed in Yameriga, compared to Gumyoko (44%), Tempelim (37%) and Gbane (32%) (Table 3.2.). In terms of number, there was no significant variation in gender across the communities. Farmers’ age ranged from 25 to 75 years. Respondents belonged to three main religions; Christianity (41%), Islam (37%) and Traditional (22%). Overall, majority of respondents (57%) had no formal education, whilst the proportion of farmers with primary and secondary education was 32% and 8% respectively (Table-2). Only 3% of farmers had been educated to the tertiary level. The highest literacy levels were recorded in Gbane where 57.1% of respondents had at least basic education. This was followed by Yameriga (50%), Gumyoko (34.4%) and Tempelim (33.3%). 37 University of Ghana http://ugspace.ug.edu.gh Table 3.2. Demographic characteristics of households Communities Variables Characteristics Gbane Gumyoko Tempelim Yameriga Total Percentage p - values Gender Male 19 18 19 17 73 59.8 0.643 Female 9 14 11 15 49 40.2 Total 28 32 30 32 122 100.0 Age <30 1 5 2 10 18 14.8 0.094 (years) 31-45 15 13 14 12 54 44.3 46-60 10 11 11 5 37 30.3 >60 2 3 3 5 13 10.7 Total 28 32 30 32 122 100.0 Religion Christians 13 14 11 12 50 41.0 0.346 Muslims 9 10 16 10 45 36.9 Traditionalists 6 8 3 10 27 22.1 Total 28 32 30 32 122 100.0 Marital status Married 25 31 26 20 102 83.6 0.002 Single 2 1 4 12 19 15.6 Divorced 1 0 0 0 1 0.80 Total 28 32 30 32 122 100.0 Number of No child 2 1 3 3 9 7.4 0.098 children 1 to 5 14 23 8 19 64 52.5 6 to 10 7 6 14 6 33 27.0 11 to 15 5 1 4 4 14 11.5 >15 0 1 1 0 2 1.6 Total 28 32 30 32 122 100.0 Literacy level None 12 21 20 16 69 56.6 0.373 Basic education 12 9 7 11 39 32.0 Secondary 2 1 2 5 10 8.2 Tertiary 2 1 1 0 4 3.3 Total 28 32 30 32 122 100.0 38 University of Ghana http://ugspace.ug.edu.gh 3.3.2 Socio-economic characteristics of households The longest farming experience observed among the farmers was 36 years (Table 3.3). Thirty percent had been involved in farming for over 15 years. More than half (51%) had at least 11 years of farming experience. The highest percentage of experienced farmers (> 10 years) was recorded in Gbane (60.7%) and Tempelim (56.7%). Only 14% of the respondents had less than 5 years’ experience in farming. Fifty five percent of farmers cultivated sorghum solely for household use (food and feed) whereas for 44% of farmers, the crop was cultivated as a source of food, feed and income. The farmers made an average income of $50 (224 cedis) from farming in the previous season. 39 University of Ghana http://ugspace.ug.edu.gh Table 3.3. Socio-economic and farm characteristics of households Communities Variables Characteristics Gbane Gumyoko Tempelim Yameriga Total Percentage p -values Farming as Yes 28 32 20 32 112 91.8 0.000 main occupation No 0 0 10 0 10 8.2 Total 28 32 30 32 122 100.0 Farming experience ≤5 0 8 3 6 17 13.9 0.248 (years) 6 to 10 11 8 10 13 42 34.4 11 to 15 8 6 8 4 26 21.3 >15 9 10 9 9 37 30.3 Total 28 32 30 32 122 100.0 Purpose for Household use only 8 23 11 25 67 54.9 0.000 sorghum cultivation Income only 0 1 0 0 1 0.8 Both 20 8 19 7 54 44.3 Total 28 32 30 32 122 100.0 Number of One 6 13 9 8 36 29.5 0.059 varieties grown Two 12 15 19 16 62 50.8 Three 10 4 1 6 21 17.2 Four 0 0 1 2 3 2.5 Total 28 32 30 32 122 100.0 Total farm size <1 ha 17 14 5 9 45 36.9 0.006 1 to 2 ha 9 17 22 23 71 58.2 3 to 4 ha 2 1 3 0 6 4.9 Total 28 32 30 32 122 100.0 Area allocated < 1 ha 26 29 18 30 103 84.4 0.000 to sorghum 1 to 2 ha 2 3 12 2 19 15.6 Total 28 32 30 32 122 100.0 40 University of Ghana http://ugspace.ug.edu.gh 3.3.3. Farm size, area allocated to sorghum and estimated grain yield A large majority (93%) of farmers owned their farm lands. A total of 163.7 ha of farmland was available for farming in the four communities; 78.89 ha of total land was allocated to sorghum production, representing 48.2% (Table 3.4). Overall, farm sizes varied from 0.2 ha to 4.86 ha, with an average of 1.34 ha. Farm sizes in Gbane ranged from 0.2 to 4.45 ha. In Yameriga farm lands ranged from 0.2 to 2.83 ha, whereas in Gumyoko and Tempelim, farm sizes ranged from 0.2 to 3.24 and 0.4 to 4.86 respectively. Details of farm sizes in the various communities are shown in Table 3.4. Almost 50% of total farm land was allocated to sorghum production in all the communities with the exception of Gbane where the proportion was 58%. By farmers’ estimations, on the average, 471.9kg of grain was produced per hectare in the last growing season. Approximately 39% harvested less than 250 kg of grain, 48% harvested between 250 and 500 kg whereas 13% had a yield of between 550 and 700 kg. Sorghum productivity was highest in Gbane (606 kg/ha) and lowest in Tempelim (364 kg/ha). Gumyoko and Yameriga had average yields of 525 kg/ha and 450 kg/ha respectively. 41 University of Ghana http://ugspace.ug.edu.gh Table 3.4. Farm size and area allocated to sorghum within the communities District Village Number of Total Average Land allocated Average land Proportion of land respondents farm size farm size to sorghum (ha) allocated to allocated to sorghum (ha) (ha) sorghum (ha) (%) Talensi-Nabdam Gbane 28 27.52 0.98 14.93 0.53 54.25 Yameriga 32 40.87 1.28 19.27 0.60 47.14 Binduri Gumyoko 32 40.27 1.26 20.09 0.63 49.88 Tempelim 30 55.04 1.83 24.6 0.82 44.69 Total 122 163.7 1.34 78.89 0.65 48.19 42 University of Ghana http://ugspace.ug.edu.gh 3.3.4 Sources of seed and information A large majority of farmers (72%) saved seeds from previous cropping seasons (Figure 3.2A). Some farmers (15%) purchased seed from the local market than from the MOFA office (11%) whereas a small minority (4%) obtained their seeds from relatives or colleagues. In general, information on improved cultivars, agronomic practices and technology was from more than one source. Seventy seven percent cited farmer associations and colleagues as major sources of information (Figure 3.2B). The Ministry of Agriculture through extension officers was also a major source of information (71%). 4% Research institutions 9% Family/Colleagues 15% Radio NGOs 72% Agric extension officers Saved seeds Local market 0 20 40 60 80 100 A MOFA Farmer colleagues B Percentage of farmers Figure 3.2. (A) Sources of seed (B) sources of information across the communities 43 Sources of information University of Ghana http://ugspace.ug.edu.gh 3.3.5 General constraints to sorghum production Eight major constraints to sorghum production were cited among the farmers in the two districts. They included drought, high cost of farming inputs (agro-chemicals, hired labour, tractors or oxen and seeds), declining soil fertility, Striga, diseases, post-harvest losses, land unavailability and pests (bird damage) [Table 3.5]. Overall, drought was the most important constraint whereas declining soil fertility was ranked third (Table 3.5). The trend for most important constraint (drought) remained the same within all the communities except Yameriga where high cost of farm input was the most important constraint. Declining soil fertility was ranked second in Gumyoko, fourth in Tempelim and Yameriga and fifth in Gbane (Table 3.5). 44 University of Ghana http://ugspace.ug.edu.gh Table 3.5. Ranking of sorghum production constraints across the four communities Communities Gbane Gumyoko Tempelim Yameriga Overall Std. Overall Major constraints Mean Rank Mean Rank Mean Rank Mean Rank Means Dev Rank p - values Drought 1.6 1 2.2 1 1.8 1 2.6 2 1.59 0.44 1 0.003 Cost of farm input 3.2 2 2.8 3 2.6 2 2.4 1 2.58 0.35 2 0.001 Declining soil fertility 4.7 5 2.3 2 3.9 4 4.4 4 3.76 1.09 3 0.006 Striga infestation 4.3 4 4.7 4 5.6 5 3.7 3 4.45 0.80 4 0.001 Diseases 4.0 3 5.2 5 5.8 6 4.7 5 4.82 0.78 5 0.001 Post-harvest losses 4.8 6 6.8 8 3.2 3 5.4 6 5.31 1.48 6 0.006 Land unavailability 7.1 8 5.5 6 6.3 7 6.6 8 6.65 0.66 7 0.000 Pests (bird damage) 6.3 7 6.4 7 6.7 8 6.2 7 6.79 0.21 8 0.000 ** The highest score is 1 and the lowest is 8. 45 University of Ghana http://ugspace.ug.edu.gh 3.3.6 Farmers’ description and perception of soil fertility More than half of farmers (51.6%) described their soils as moderately fertile (Table 3.6). Almost 42% described their soils as low in fertility whilst 6.6% described their soils as highly fertile. In general, farmers described fertile soils as darker, more compact and moist whereas infertile soils were described as stony, reddish or ‘lighter’ in colour and texture. The main criteria for assessment of soil fertility (88.5%) was colour (Table3.6). This was followed closely by crop performance (85%). Soil texture and high moisture retention were also cited as good indicators of soil health as indicated by 66% and 46% of farmers respectively. A small minority (13%) mentioned the presence of weeds and other plants as indicators of soil fertility. A large majority (95%) perceived symptoms of nutrient-stress in sorghum as easily-recognisable. In this regard, the most-mentioned symptoms were poor yields (75%) and stunted growth (43%). A much lesser proportion mentioned delayed maturity (22%), poor seedling establishment (9.8%) and leaf discolorations (0.8%) as symptoms of nutrient-stress on sorghum. Sixty seven percent of farmers indicated that sorghum yield could be improved by fertiliser application (Table 3.7). However, only 38% of farmers applied fertiliser to the crop (Table 3.8). The number of farmers who applied fertiliser to sorghum was highest in Gumyoko (63%) and lowest in Tempelim (20%). More farmers (43.5%) fertilised the crop with manure and/or household waste than with mineral fertilisers (23.9%). Almost 33% fertilised the crop with a combination of inorganic and organic fertilisers. The most frequently cited constraint (82%) to fertiliser use was farmers’ inability to afford the commodity. A lesser proportion (48%) mentioned unavailability of fertiliser as a constraint. 46 University of Ghana http://ugspace.ug.edu.gh Table 3.6. Farmers’ description of soils Variables Characteristics Gbane Gumyoko Tempelim Yameriga Total Percentage p - values Description of soil Highly fertile 4 0 1 3 8 6.6 0.008 Moderately fertile 17 10 17 19 63 51.6 Poor 7 22 12 10 51 41.8 Total 28 32 30 32 122 100.0 Crop performance Yes 25 25 27 27 104 85.2 0.53 No 3 7 3 5 18 14.8 Total 28 32 30 32 122 100.0 Soil texture Yes 20 26 16 19 81 66.4 0.09 No 8 6 14 13 41 33.6 Total 28 32 30 32 122 100.0 Soil moisture Yes 15 16 14 11 56 45.9 0.458 No 13 16 16 21 66 54.1 Total 28 32 30 32 122 100.0 Soil colour Yes 26 27 26 29 108 88.5 0.729 No 2 5 4 3 14 11.5 Total 28 32 30 32 122 100.0 Soil cover Yes 8 4 1 3 16 13.1 0.032 No 20 28 29 29 106 86.9 Total 28 32 30 32 122 100.0 47 University of Ghana http://ugspace.ug.edu.gh Table 3.7. Farmers’ perceptions of soil fertility and its effect on sorghum Communities Variables Characteristics Gbane Gumyoko Tempelim Yameriga Total Percentage p-values Is sorghum yield Yes 23 21 18 20 82 67.2 0.273 improved by fertiliser No 5 11 12 12 40 32.8 Total 28 32 30 32 122 100.0 Is nutrient-stress Yes 26 31 27 32 116 95.1 0.281 recognisable on sorghum No 2 1 3 0 6 4.9 Total 28 32 30 32 122 100.0 Stunted growth Yes 22 29 15 25 91 74.6 0.002 No 6 3 15 7 31 25.4 Total 28 32 30 32 122 100.0 Reduced yield Yes 14 8 14 17 53 43.4 0.099 No 14 24 16 15 69 56.6 Total 28 32 30 32 122 100.0 Delayed maturity Yes 10 6 6 5 27 22.1 0.253 No 18 26 24 27 95 77.9 Total 28 32 30 32 122 100.0 Poor seedling establishment Yes 3 3 2 4 12 9.8 0.89 No 25 29 28 28 110 90.2 Total 28 32 30 32 122 100.0 Leaf discolorations Yes 1 0 0 0 1 0.8 0.336 No 27 32 30 32 121 99.2 Total 28 32 30 32 122 100.0 48 University of Ghana http://ugspace.ug.edu.gh 3.3.7 Farmers’ cropping systems and soil management practices Mono-cropping of sorghum was practiced by a 4% of farmers (Table 3.8). Seventy three percent of farmers typically cultivated the crop in variable mixed cropping systems with crops such as millet, soybean, groundnut, cowpea, bambara bean, okro and other vegetables. Twenty three percent indicated inter-cropping sorghum with millet and legumes as their main cropping system. Mixed cropping was highest in Yameriga (91%) and lowest in Gbane (54%). On the other hand, intercropping was highest in Gbane (36%) and lowest in Gumyoko (9%). No defined crop rotational system was mentioned by the farmers. Shifting cultivation was not a common practice. Seventy five percent cultivated the crop on the same piece of land each season. With regard to other soil management practices, less than half (46%) retained some crop residue on harvested fields. This form of mulching was highest in Gbane (50%) and lowest in Tempelim (40%). At least one erosion control method was practiced by 60% of farmers. Stone bunding, earth bunding or grass strips were the main erosion control methods. 49 University of Ghana http://ugspace.ug.edu.gh Table 3.8 Farmers’ cropping systems and soil management practices Variable Characteristics Gbane Gumyoko Tempelim Yameriga Total Percentage p-values Cropping Mono-cropping 3 2 0 0 5 4.1 0.028 system Mixed cropping 15 21 24 29 89 73.0 Intercropping 10 9 6 3 28 23.0 Total 28 32 30 32 122 100.0 Retention of Yes 14 15 12 15 56 45.9 0.888 crop residues No 14 17 18 17 66 54.1 Total 28 32 30 32 122 100.0 Shifting Yes 10 15 4 2 31 25.4 0.000 cultivation No 18 17 26 30 91 74.6 Total 28 32 30 32 122 100.0 Erosion Yes 24 18 12 19 73 59.8 0.005 control No 4 14 18 13 49 40.2 Total 28 32 30 32 122 100.0 Application of Yes 13 20 6 7 46 37.7 0.001 fertilisers No 15 12 24 25 76 62.3 Total 28 32 30 32 122 100.0 Type of Organic 6 6 5 3 20 43.5 0.000 fertiliser Inorganic 4 5 0 2 11 23.9 Both 3 9 1 2 15 32.6 Total 13 20 6 7 46 100.0 Cost of fertiliser Yes 22 28 26 24 100 82.0 0.497 as a constraint No 6 4 4 8 22 18.0 Total 28 32 30 32 122 100.0 Unavailability of Yes 16 18 13 11 58 47.5 0.217 fertiliser as a constraint No 12 14 17 21 64 52.5 Total 28 32 30 32 122 100.0 50 University of Ghana http://ugspace.ug.edu.gh 3.3.8 Cultivated sorghum varieties and farmers’ preference for traits Farmers cultivated between two and ten different crops. Apart from sorghum, other cultivated crops included millet, rice, maize, groundnuts, soybean, cowpea, tomatoes, garden eggs and okro. Seven sorghum varieties, Naga Red, Belko-peleg, Belko-zia, Naga white, Kapaala, Dorado and Kowerig were recorded across the four communities. The highest number of sorghum varieties grown by a farmer was four (Table 3.3). Overall, the most-cultivated variety was Naga-red; as indicated by 66% of farmers (Figure 3.3). This was followed closely by Belko-peleg (64.8%). Belko zia, Naga white, Kapaala, Dorado and Kowerig varieties were cultivated by 53.3%, 31.2%, 28.7%, 13.9% and 9.8% respectively. In the Talensi-Nabdam district, the most-cultivated variety was Naga Red (73%). Belko-peleg was the most-cultivated variety in Binduri (66%). Kowerig Kapaala Dorado Belko-zia Belko-peleg Naga White Naga Red 0 20 40 60 80 Percentage of farmers (%) Figure 3.3. Diversity of sorghum varieties cultivated in the communities Overall, drought tolerance was the most-preferred trait, followed by high grain yield and earliness (Table 3.9). Grain quality was the fourth most-preferred trait. These included all characteristics directly or indirectly linked to the grain such as taste, storage quality, suitability for food and beverage and threshability. 51 Cultivated varieties University of Ghana http://ugspace.ug.edu.gh Crop adaptation to poor soils ranked fifth, ahead of disease tolerance and resistance to bird damage. At the community level, high grain yield was the most-preferred trait in Gbane and Gumyoko, whereas drought tolerance and earliness were the most-preferred traits in Tempelim and Yameriga respectively. Varieties with low fertiliser requirement ranked third in Gumyoko but fifth in all other communities. 52 University of Ghana http://ugspace.ug.edu.gh Table 3.9. Ranking of farmers’ preference for sorghum traits Communities Gbane Gumyoko Tempelim Yameriga Overall Std. Preference criteria Mean Rank Mean Rank Mean Rank Mean Rank Mean Deviation Rank p-values Drought tolerance 2.3 2 2.1 2 1.8 1 3.1 4 2.3 0.6 1 0.004 High grain yield 2.0 1 1.8 1 3.3 4 3.0 3 2.5 0.7 2 0.006 Early-maturing 3.1 3 3.9 5 2.8 2 2.0 1 3.0 0.8 3 0.005 Good grain quality 3.3 4 3.8 4 3.2 3 2.4 2 3.2 0.6 4 0.001 Low fertiliser requirement 5.7 5 3.7 3 4.2 5 5.4 5 4.8 1.0 5 0.002 Resistance to Striga 6.3 6 6.5 6 6.8 7 6.4 6 6.5 0.2 6 0.000 Disease tolerance 6.5 7 7.1 8 6.6 6 6.9 8 6.8 0.3 7 0.000 Tolerance to pests 6.9 8 6.7 7 7.1 8 6.7 7 6.9 0.2 8 0.000 ** The highest score is 1 and the lowest is 8. 53 University of Ghana http://ugspace.ug.edu.gh 3.4 Discussion In the surveyed communities, sorghum was cultivated more as a means of providing food for the household than as a source of income. Farmers indicated that income generation from sorghum was limited by poor yields and their inability to meet household food demand. This observation is corroborated by other surveys in northern Ghana (IFAD 2010; Eguavoen (2013). On the other hand, farmers derive income from sorghum either directly from the sale of grains or indirectly by brewing local beer. The farmers also rely on off-farm economic activities for additional income. For women, these include the sale of firewood and cooked food, gathering of fruit, cooked food and shea butter preparation whereas artisanship (auto-mechanic and blacksmithing) and employment as casual labourers are the main activities for men. For smallholders, off-farm activities are important strategies for minimizing food insecurity and improving livelihood (MoFA, 2015; Osarfo et al., 2016). Farm sizes were generally below two hectares. However, almost half of cropped lands was allocated to sorghum cultivation, an indication of its importance in the communities (Kudadjie et al., 2004). The small farm sizes may be attributed to scarcity of arable land and/or high cost of renting additional land (Ohene-Yankyera, 2004). Also in the Upper East Region, as is the case in many parts of Ghana, farming is extremely labour-intensive with limited mechanisation (Diao et al., 2014). Thus, larger farms are likely to require the hiring of expensive casual labour, tractors or oxen, an investment many smallholder farmers may be unable to make (Whitehead, 2003). Literacy levels were low among the farmers. According to Enu and Attah-Obeng (2013), lack of formal education is a hindrance to agricultural productivity in northern Ghana. Other reports also suggest that farmers with higher education often possess greater capabilities for assessing their environments and implementing technologies (Akinbile, 2003; Uematsu and Mishra, 2010). Thus, 54 University of Ghana http://ugspace.ug.edu.gh it might be expected that the general lack of formal education among the farmers would be a barrier to effective assessment of soil fertility. On the contrary, the high experience levels among the farmers seemed to compensate for low literacy levels. For instance, most farmers described their soils with easily-observable indicators such as soil colour, texture and crop performance. They also recognized stunted growth and reduced yield as symptoms of nutrient stress in sorghum. In a similar study in Ethiopia, farmers also cited crop performance as one of the main criteria for assessing soil fertility (Yeshaneh, 2015). Although leaf discolorations and poor stand establishment were also cited as symptoms of nutrient-stress in sorghum, farmers associated these symptoms more with drought than with poor soils. This observation might be due to the fact that, both abiotic factors present similar symptoms (Chen et al., 2015). Moreover, nutrient unavailability is interrelated with drought as nutrient adsorption by plants is influenced by soil moisture content (Schoonover and Crim, 2015). Seven sorghum varieties were identified in the communities. Due to differences in ethnicity and language, identical sorghum varieties were referred to by different vernacular names. For example, red-grained varieties were interchangeably referred to as Kezie, Keto or Kimolga, an observation that has also been reported in Ghana by Kudadjie et al., (2004) and in Benin by Missihoun et al., (2012). Preference for the Naga Red variety was clear with farmers citing its suitability for brewing local beer as a major incentive. Farmers also prefer Naga Red due to its early-maturing characteristic as it provides a less risky option for farmers in case of insufficient rain (Kudadjie- Freeman and Dankyi-Boateng (2012). The popularity of the late-maturing, low-yielding Belko- peleg within the communities is likely due to its suitability for preparing tuo zaafi and other local foods (Kudadjie et al., 2004). According to Buah et al., (2012), Naga white and Kapaala are 55 University of Ghana http://ugspace.ug.edu.gh cultivated on a lesser scale due to farmers’ perceptions of their higher fertiliser requirements, low grain quality and susceptibility to bird damage. The most important sources of information for the farmers were family and friends. Some information was also received through extension officers, research institution and NGOs. Access to information on agricultural innovations has been shown to increase farm output in many rural parts of Africa (Kimaru-Muchai et al., 2012). However, informal channels of information such as observed among the farmers may expose them to the risks of inadequate or inaccurate information which is likely to hinder productivity (Benard et al., 2014). The impact of agricultural extension services was particularly obvious in Gbane where estimated sorghum productivity was highest. Electronic media such as radio and television were not important sources of information due to limited or no access to electricity. Majority of the farmers save seeds for subsequent planting seasons, a frequent practice in northern Ghana and other parts of West Africa (Kudadjie-Freeman and Boateng, 2012; Dossou-Aminon et al., 2015). Although such seed systems have been practiced for decades, reports suggest that poor post-harvest handling of seeds as well as damage by pests may lead to poor plant establishment, a factor which may contribute to low yields among these farmers (Ochieng et al., 2011). Sorghum yield among the farmers was low as evidenced by productivity estimates of 364 to 606 kg/ha). In the Upper East Region sorghum yield is estimated at 700 kg/ha (Alhassan and Jatoe, 2009). This observation suggests that food insecurity and low income are critical issues for many farm households in the surveyed districts. The poor yields observed in this study may be attributed to factors such as low plant densities on farmers’ fields and the cultivation of inherently low- yielding varieties (Atokple, 2010). However, from the farmers’ perspective, the major constraint 56 University of Ghana http://ugspace.ug.edu.gh to sorghum productivity was drought. Drought has been reported to be one of the main constraints to crop productivity in northern Ghana (Ndamani and Watanabe, 2015). Given that crop is cultivated mainly under rain-fed conditions, the low and erratic rainfall patterns characteristic of this agro-ecological zone often exposes the crop to moisture stress leading to poor yields (Zougmore, 2003). The second most important constraint was high cost of farm inputs. Farmers described this mainly as cost of hiring tractors, oxen or labour for land preparation and maintenance. Farmers reiterated that limited access to ploughing services results in delays in sowing and limits farm sizes. Millar et al., (2007) report that although the use of tractors and bullocks for land preparation has increased in northern Ghana, the patronage of such services is limited by high cost. Similar observations have been made by Nin-Pratt and McBride (2014). The third major constraint was declining soil fertility. Soil nutrient levels in smallholder farms in West Africa are often extremely low as a result of deforestation, overgrazing and burning of cover vegetation (Mapfumo and Giller, 2001; Twomlow and Ncube, 2001; Henao and Baanante, 2006; Omotayo and Chukwuka, 2009). Farmers’ approach to tackling the problem of declining soil fertility was mainly erosion control through methods such as stone bunding, earth bunding and vetiver grass strips. Soil erosion is severe in many parts of northern Ghana and farmers were quite knowledgeable of its effects and the benefits associated with its control (Veihe, 2002; EPA, 2003). This is most likely due to extension services by MOFA officers and other stakeholders. On the other hand, shifting cultivation and mulching were uncommon in the communities. Farmers cited increasing economic pressure as the main driving force. In recent times, land fallow as a conservation method is limited in many parts of West Africa, an observation that has been attributed to due to increasing population densities and escalating land pressure (Laube, 2007; Kanchebe, 2010). Also, the retention of plant residue at harvest (mulching) as a soil enhancement 57 University of Ghana http://ugspace.ug.edu.gh method was low due to removal of stover for use as feed and household fuel. This practice is common in northern Ghana, particularly in the Upper East Region due to scarcity of grazing land and high population density (Kombiok et al., 2012). The main cropping systems were mixed cropping and intercropping, cropping patterns which have been practiced for decades in northern Ghana (Diehl, 1992). Among smallholders, mixed cropping is usually an insurance against crop failure or as a strategy for providing a variety of food for the household (Eguavoen, 2013). In general, farmers were aware of the role of fertilisers in enhancing sorghum yield. This is likely due to personal experiences, on-farm trials and/or demonstrations. However, fertiliser application was almost non-existent. The amount of applied fertiliser was not well-defined due difficulties in estimating quantities. However, the trend among smallholder sorghum farmers is the application of quantities far below recommended rates (Kombiok et al., 2012). Often, there is no deliberate fertilisation of sorghum and the crop most likely benefits from fertiliser applied to other crops in the mixed cropping system. According to the farmers, two main factors influence their attitude towards sorghum fertilisation. The most important was farmers’ inability to afford fertiliser. This observation has been widely reported in Ghana and has been attributed to extreme poverty rates among farmers in the region (Yilma, 2006; Laube, 2007; Quaye, 2008; Abane, 2015). Access to government-subsidised fertiliser, which was the case for some of the farmers in this study had little impact on fertilisation of the crop. Unavailability of mineral fertilisers was a lesser problem with many reporting of sufficient quantities on local markets. On the other hand, farmers described unavailability in terms of late arrival of government-subsidised fertiliser. Application of manure and/or household waste to the crop was an alternative for some farmers. This form of soil fertilisation is common in northern Ghana (Gyasi and Uitto 1997; Issaka et al., 2016). Unfortunately, its use is constrained by insufficient quantities, laborious gathering from free-range 58 University of Ghana http://ugspace.ug.edu.gh animals as well as its transportation to farmlands (Tossah, 2000). Thus, a combination of both organic and inorganic fertilisers was a more popular choice among the farmers. Although high cost of fertilisers and limited availability were the most cited constraints in this survey, other reviews suggest that farmers’ perceptions of the crops’ low response to applied fertiliser is a key impediment to its fertilisation (Kpongor, 2009; Kihara et al., 2016). In this study, 33% of farmers were of the perception that sorghum yield does not improve significantly with fertiliser application. Kombiok et al., (2012) suggests that such perceptions are due to the minimal yield increases farmers may observe when fertiliser is applied to local varieties. This perceived unprofitability is likely to influence farmers’ attitude towards fertilising the crop (Nkegbe and Shankar 2014). Tolerance to drought was identified as the most-preferred trait. In their survey of parts of northern Ghana, Ndamani and Watanabe (2015) suggest that although farmers recognize the effect of drought on crop production they often lack the capacity to implement adaptation practices. This may account for farmers’ preference for drought-tolerant types. Preference for earliness in sorghum may be linked to lower risks of crop failure in case of insufficient rain and timely provision of food after the long dry season (Kombiok et al., 2012). Qualities of the grain such as taste, storage quality, suitability for food and beverage, threshability and market value have been reported to be of importance to farmers (Kudadjie et al., 2004). Sorghum is widely consumed in northern Ghana MacCarthy and Vlek, 2012). The fifth most-preferred trait was low fertiliser requirement. Low adoption of some improved sorghum varieties in northern Ghana has been linked to associated high fertiliser requirements (Buah et al., 2012). Thus, farmers are likely to recognize the need for this trait particularly due to their inability to afford mineral fertilisers. 59 University of Ghana http://ugspace.ug.edu.gh 3.5 Conclusions Sorghum production in the sampled communities is faced with several constraints, the most important of which were drought, limited access to ploughing services and declining soil fertility. Farmers were cognizant of the effects of declining soil fertility on sorghum yield. However, application of fertiliser to the crop was low. High cost of mineral fertilisers, untimely access to the commodity and limited availability of organic nutrient sources were the most important constraints to fertiliser use. Farmers employed other soil management strategies such as erosion control through the use of stone and earth bunding. Mixed cropping of the sorghum with legumes and other food crops was the cropping system of choice. The most-preferred sorghum traits were drought tolerance, high grain yield, earliness, suitability of grain for food and beverages as well as varieties with low fertiliser requirements. 60 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR PHENOTYPIC ANALYSES FOR IDENTIFICATION OF ARTIFICIALLY-INDUCED VARIABILITY IN SORGHUM 4.1 Introduction Genetic diversity is a key component of plant breeding; to a large extent, the success of any breeding programme depends on the availability of sufficient levels of genetic variation for the trait of interest (Govindaraj et al., 2015; Gemenet et al., 2016). Over decades, plant breeders have relied on the natural or existing genetic variation within species for crop improvement (Wooten, 2001). Sorghum improvement programmes for instance, heavily exploit the existing variation within cultivated genotypes and/or their wild relatives (Chantereau et al., 2001). The main approach in these programmes has been hybridisation (Xu, 2010). Hybridisation, particularly of diverse parents, is highly-useful for expanding the genetic diversity within crop species, in that, several possible genotypes may be obtained (Breseghello, 2013; Schnell, 2015). However, modern hybridisation programmes usually focus on a relatively narrow gene pool resulting in a continuous recycling of genes within the species (Reddy et al., 2006). This, coupled with artificial selection during the breeding process has led to significant losses in diversity in sorghum (van Harten, 1998; Rakshit et al., 2012). The problem is worsened by the gradual replacement of landraces with high- yielding cultivars on farmers’ fields (Breseghello, 2013; Dossou-Aminon et al., 2015). Losses in genetic diversity may affect the crop’s adaptation to environmental stresses (Phillips and Wolfe, 2005). This is of concern particularly in West Africa where farming environments are highly variable and often plagued by several biotic and abiotic stresses (Kong et al., 2000). More importantly, given the critical role that sorghum breeding plays in meeting food demand in West 61 University of Ghana http://ugspace.ug.edu.gh Africa, any reduction in the genetic variability available to breeders threatens the food security of the region (Beresghello, 2013). Besides hybridisation, genetic diversity of crops may be broadened through tissue culture (somaclonal variation), genetic engineering and mutagenesis (Larkin and Scowcroft, 1981, Ye et al., 2000; Xin et al., 2008). Unfortunately, somaclonal variation has not been routinely applied to sorghum and has also been reported to add little value for creating genetic variability in the crop (Omondi et al., 2013). Transgenic methods, although highly effective for introducing hitherto unavailable genes within species, are not readily-exploited in West Africa due to tedious biosafety regulatory restrictions and associated high cost (Henikoff and Comai, 2003). On the other hand, artificial mutagenesis is a readily-available and effective tool for generating germplasm with diverse characteristics (Dillon et al., 2007; Liang, 2009). The key advantage of mutagenesis over hybridisation is its ability to introduce random changes throughout the plant genome resulting in a wide variety of mutations in target genes whereas hybridisation is only effective in creating new combinations of existing alleles (House, 1995; Zhang et al., 2007). Mutagenesis is particularly beneficial when the genetic variability for any trait of interest is limited or non-existent in cultivated or wild genotypes (Rakszegi et al., 2010). The technique has been successfully applied to cereals including rice, wheat, barley and maize (Ahloowalia et al., 2004; Caldwell et al., 2004; Till et al., 2004b; Suzuki et al., 2008). These mutant varieties exhibit a wide array of improved traits such as resistance to blast, yellow mottle virus and Striga asiatica as well as tolerance to acidity, drought and salinity (Jenks et al., 1994; Kiruki et al., 2006; Peters et al., 2009). In sorghum, the method was used to increase genetic variability for lysine content as well as develop multi-seeded and nuclear male sterile genotypes (Axtell et al., 1979; Xin et al., 2017). 62 University of Ghana http://ugspace.ug.edu.gh Subsequent to mutagenic treatment, assessment of the mutagenised population for any induced variability is necessary to confirm its usefulness or otherwise as a germplasm resource (Xin et al., 2008). This may be achieved through morphological characterisation which is a relatively simple, inexpensive and efficient method (Rakshit et al., 2012; Elangovan et al., 2013). Morphological characterisation provides an opportunity for studying the performance of the mutagenised lines under actual growing conditions which in turn facilitates the identification of desirable genotypes (van Beuningen and Busch, 1997; Elangovan et al., 2014). For mutant populations, characterisation over several generations also helps validate the stability of any altered traits (Luz et al., 2016). Subsequently, desirable genotypes be released either directly as new mutant varieties or used as parents in recombinant breeding programmes (Ahloowalia, 2004; Xin et al., 2008). Thus, the goal of this study was to enhance genetic variability for agro-morphological traits in sorghum through artificial mutagenesis. The specific objectives were to; 1. assess a chemically-mutagenised population for any induced genetic variability and 2. identify within this population putative mutants with desirable morpho-agronomic traits. 4.2 Materials and methods 4.2.1 Description of study site The experiment was conducted during the 2013 cropping season at the research station of the University of Ghana, Legon (Latitude 5o36’N, Longitude 0o10W). The area lies within the Coastal Savannah agro-ecological zone. The natural vegetation is mainly grassland (short and medium grasses) interspersed with dense thickets and a few trees (EPA, 2003). Rainfall pattern in this area is bi-modal. There is a major rainy season from March to July and a minor rainy season from September to October. Annual average rainfall is approximately 800 mm with a range of 600 to 63 University of Ghana http://ugspace.ug.edu.gh 1200 mm (AQUATSAT, 2005). The average minimum and maximum temperatures are 22oC and 35oC respectively. The predominant soil type is well-drained Savannah Ochrosol (Ferric Acrisol) derived from quartzite schist, classified as the Haatso series (FAO, 1994).These are sandy loams, light in texture and inherently poor in fertility (Brammer, 1962). Rainfall and temperature conditions of the study site during the experimental period are shown in Table 4.1. Table 4.1. Meteorological data for the sites during the study period Mean monthly temperature Month Total monthly rainfall (mm) Maximum (oC) Minimum (oC) May 130 31 23 June 210 29 23 July 50 28 22 August 30 28 22 Source: Ghana meteorological agency (2013) 4.2.2 Generation of mutagenised population The mutagenised population was developed at and by the Whistler Agronomy Laboratory, Purdue University, USA. The mutagenised population was derived from mutagenic treatment of BTx623. Btx623 is a white-seeded inbred line developed and released in Texas in 1976 (Miller 1976). Its pedigree is BTx3197 x SC170. It is a widely-used maintainer line (B line) classified as a Kafir/Zera-Zera derivative (Menz et al., 2002). It is also the genotype selected for genome sequencing of the crop (Paterson et al., 2009). Homogeneity of M0 seeds was ensured by repeated selfing over six generations and selection by single seed descent as recommended by Xin et al., (2008). At each generation, seeds were selected from one plant that presented the most typical traits of BTx623. The mutagen used was ethyl methane sulphonate (EMS) [Sigma Aldrich, St Louis, MO]. Initial experiments were carried out to determine optimum concentration of EMS for mutagenesis. Subsequently, batches of 100g of dry seed (approximately 3300 seeds) were soaked in 200 ml of tap water containing EMS at a 64 University of Ghana http://ugspace.ug.edu.gh concentration of 0.25% (v/v). The soaked seeds were maintained under continuous agitation for 16 hours at 50 revolutions per minute on rotary shaker. The treated seeds were then washed thoroughly in 400 ml of tap water for five hours at room temperature, changing the water every 30 minutes. They were then air-dried on tissue paper for 24 hours at room temperature. Air dried seeds were planted in the field at high densities (120,000 seeds/hectare) to produce M1 plants. Each fertile panicle was harvested manually and threshed individually for use as M2 seeds which were then planted in head rows. Head rows were inspected closely for aberrant phenotypes in comparison to the wild type BTx623. M2 head rows which showed two or more plants with altered and/or distinct phenotypes were tagged as putative mutants for further evaluation. Panicles from these tagged lines were harvested individually, threshed and advanced as M3 seed. At each generation, panicles were bagged before anthesis with rain-proof paper pollination bags (Lawson Bags, Northfield, IL) to prevent cross pollination. M3 seeds were stored in paper envelopes which were in turn stored in plastic containers at - 20 oC. The germplasm used in this study comprised a population of 547 of these mutagenised M3 families. Three additional genotypes were used as controls; wildtype BTx623, Tx631 and Grinkan. Tx631 is a common grain sorghum line released in 1985 as a selection from the cross (BTx378 x SC110- 9) x BTx615. Grinkan is a popular short-day and dwarf guinea-caudatum open-pollinated variety with intermediate maturity (Abdulai, 2012). 4.2.4 Experimental design and layout The M3 families were planted in an augmented design. Twenty nine blocks were planted in total. Each block comprised twenty two genotypes; nineteen putative mutant families and three control varieties planted randomly. Each genotype was planted in a single row plot of approximately 3 65 University of Ghana http://ugspace.ug.edu.gh metres, using a planting distance of 0.3 metres within rows and 0.8 metres between rows. Ten hills were sown per family at 3 seeds per hill. Test plots were later thinned to one plant per hill. Plants were maintained mainly under rain-fed conditions. However, irrigation was applied when necessary. Fertilizer was applied at a rate of 60 kg N ha-1, 30 kg/ha of phosphorus (P2O5) and 30 kg/ha potassium (K2O) in two split doses; at three and six weeks after planting. Standard agronomic and plant protection (weed and insect control) measures were applied during the cropping season. 4.2.5 Data collection Five plants were randomly-selected in each row and tagged. Data on nine quantitative traits were recorded at different growth stages. This included days to flowering (DTF), plant height (PH), number of leaves per plant (NL) and stem girth (SG) at anthesis. At maturity, panicles were harvested, air-dried for 2 weeks at ambient temperature and threshed. Data were then collected on panicle length (PL), peduncle exsertion (PE), panicle width (PW), hundred-seed weight (HSW) and grain yield per plant (GYP). For qualitative traits, throughout the season, all plants in each family were closely-inspected for novel or distinct phenotypes in comparison to wild-type BTx623. Qualitative traits that were considered included leaf characteristics (size, arrangement, midrib colour, lesions and necrotic areas), plant size, tendency to tiller, tendency to wilt or senesce, the presence or absence of epicuticular wax on stems and threshability (freely threshable, partly threshable and difficult to thresh). All the mutagenised families were then categorised as described by Xin et al. (2008) and USDA mutant descriptors for sorghum (www.lbk.ars.usda.gov/psgd/sorghum/till/index.aspx). 66 University of Ghana http://ugspace.ug.edu.gh 4.2.6 Data analysis Descriptive statistics were obtained using summary statistics procedure in XLSTAT 2017 (Addinsoft, Paris, France). A one-way analysis of variance (ANOVA) for each quantitative trait was performed. Where means were statistically significant (p<0.05), ANOVA was followed by mean separation using Dunnett’s test (p<0.05) to determine which putative mutant families were significantly different from the control (BTx623). Genetic parameters were then calculated to predict the extent of induced genetic variability. Genotypic and phenotypic variances (σ2g and σ2p respectively), genotypic coefficient of variation (GCV) and phenotypic coefficients of variation (PCV) were computed according to the formula proposed by Johnson et al. (1955). The coefficients were categorised as proposed by Khorgade et al. (1985) where values of less than 10%, 10 to 20% and greater than 20% were considered as low, intermediate and high respectively (Khorgade et al., 1985). The heritability was computed based on the methods given by Falconer (1960a). All computations were done using Microsoft Excel 2013. Pearson’s correlation coefficients (r) were calculated between the phenotypic traits to assess trait associations. The percentage contribution of each trait to total genetic variation was estimated by Principal Component Analysis (PCA) using the mean values of the nine quantitative traits. Cluster analysis was performed using Unweighted Pair Group Method with Arithmetic mean (UPGMA) was employed to determine the extent of heterogeneity induced within the population. These analyses were done using XLSTAT 2017 software (Addinsoft, Paris, France). 67 University of Ghana http://ugspace.ug.edu.gh 4.3. Results 4.3.1 Variation within the mutagenised population for quantitative traits Highly significant differences (p≤ 0.01) were observed within the mutagenised population for all the studied quantitative traits (Table 4.3). Highly significant differences were also observed between several mutagenised families and the wildtype BTx623. Descriptive statistics, results of ANOVA and mean separation for each trait are elaborated below. Number of days to anthesis ranged from 51 and 70 days. The earliest anthesis was observed in Mut2619 whereas the latest flowering (70 days) was observed in Mut0878. Seventy five M3 families were significantly different (p<0.01) from BTx623. Of these, fifteen families were earlier than the wildtype whereas sixty reached anthesis later than the control variety. Plant height ranged from 23 to 142 cm. The shortest families were Mut0992 and Mut2094#1, both of which recorded heights of less than 50 cm whereas Mut3053 presented the tallest plants. Compared to an average of 109.8 cm for the wild-type BTx623, 386 of the putative mutant families were significantly (p < 0.01) shorter than the control whereas ten M3 families were significantly taller than the control. Number of leaves per plant was least (7 ) in Mut3057 and Mut1668 and highest (12) in twelve M3 families (Mut1504, Mut1505, Mut1457#1, Mut0755, Mut3809, Mut2716, Mut2811, Mut2807, Mut0931, Mut1884, Mut0561, Mut2039 and Mut2038#1). The wildtype BTx623 also presented an average of 12 leaves per plant. Compared to this, the number of leaves per plant was significantly reduced in twenty eight families. For all other families, the trait was not significantly different (p<0.01) from the control. 68 University of Ghana http://ugspace.ug.edu.gh Stem girth was thinnest in Mut1668 and Mut3515-2 (7.3 and 12.6 mm respectively) and thickest (25.2 cm) in Mut2807. An average of 24.5 cm was observed by the wildtype. Compared to the average stem girth observed for the wildtype (24.5 cm), 359 families were significantly thinner. There were no putative mutants with significantly larger stem diameters compared to BTx623. Panicle length ranged from 14.5 cm in Mut3515-2 to 48.5 cm in Mut2038#1. Three hundred and forty five M3 families were significantly different from BTx623, forty two of which were significantly longer. Panicle width was least (2.9 cm) in Mut4213 and widest (8.4 cm) in Mut2616. One hundred and fifty five M3 families had mean values significantly different from BTx623. Of these, 69 presented mean values significantly longer than BTx623 whilst 87 families had shorter panicles. Twelve families had no peduncle exsertion. These included Mut1204#2, Mut1041#1, Mut1199, Mut1156, Mut0682, Mut0456#1, Mut0672, Mut0056, Mut1113#1, Mut3567, Mut3218 and Mut3220. The longest peduncle exsertion (22.6cm) was observed in family Mut4602#1. One hundred and eighty seven families were significantly different from the control in terms of peduncle exsertion. Hundred-seed weight ranged from 1.94 g in Mut2899#1 to 4.05 g in Mut3008. Compared to BTx623, significant decreases in seed weight were observed in 44 families whereas higher values were observed in 12 families. The least grain yield (19.4 g) was recorded for Mut3515-2 whilst the highest (203.2 g) was recorded in Mut2157. BTx623 recorded an average grain yield of 67.5 g. Three hundred and thirty mutant families recorded higher grain yields than the control of which 293 were significantly higher (p < 0.01). 69 University of Ghana http://ugspace.ug.edu.gh Coefficient of variation (CV) ranged from 0.9% for number of leaves to 7.27% for grain yield. Descriptive statistics for the nine quantitative traits are given in Table 4.2 and Figure 4.1. Details of analysis of variance is shown in Table 4.3 Table 4.2. Descriptive statistics of nine evaluated traits DTF PH NL SG PL PE PW HSW GYP Minimum 51.10 23.20 7.30 9.69 14.54 0.00 2.85 1.94 19.43 Maximum 70.00 141.90 13.00 28.50 48.45 22.61 10.10 4.05 203.19 Mean of M3 families 61.53 92.27 10.02 18.48 29.65 6.71 5.77 2.93 73.36 Mean of BTx623 58.60 109.80 12.00 24.50 28.60 3.30 5.70 3.10 67.46 Grand mean 61.86 95.24 10.28 19.39 30.17 6.46 5.98 2.94 79.24 Standard error 0.14 0.61 0.04 0.13 0.16 0.16 0.05 0.01 1.21 Sample variance 13.01 233.90 1.12 10.24 17.14 15.82 1.45 0.11 923.01 DTF= Days to flowering; PH = Plant height; NL= Number of leaves; SG = Stem girth; PL = Panicle length; PE = Peduncle exsertion; PW = panicle width; HSW = Hundred seed weight; GYP; Grain yield per plant. 70 University of Ghana http://ugspace.ug.edu.gh Figure 4.1. Box plots showing the distribution of studied traits in the mutagenised population. The upper, median and lower quartiles represent the 75th, 50th and 25th percentiles of the clusters respectively. The vertical lines represent the variation within the clusters. Dots represent outliers. 71 University of Ghana http://ugspace.ug.edu.gh Table 4.3. Analysis of variance within the mutagenised population for quantitative traits Sources of variation Df PL PE PW PH DTF NL SG HSW GYP Genotypes 548 19.34** 18.23** 1.64** 268.97** 14.91** 1.29** 11.48** 0.13** 1059.3** Blocks 28 6.35** 0.52** 0.28** 8.72** 0.01** 0.01** 32.94** 0.02** 22.8** Error 84 2.91 0.24 0.17 7.78 0.75 0.01 1.90 0.01 44.70 CV (%) 4.35 8.00 6.99 2.59 1.44 0.92 6.56 4.00 7.27 ** Values significant at 0.01 probability levels. PL= Panicle length; PE=Peduncle exsertion; PW= Panicle width; PH= Plant height; DTF= Days to flowering; NL= Number of leaves per plant; SG= Stem girth; HSW= Hundred-seed weight; GYP = Grain yield/plant 72 University of Ghana http://ugspace.ug.edu.gh 4.3.2. Genetic and phenotypic variances and heritability within the mutagenised population The lowest GCV values were observed for days to flowering (2.99%) and plant height (3.97%). Peduncle exsertion and grain yield per plant had the highest GCV values of 31.31 and 19.63% respectively. PCV values were slightly higher than GCV values for all traits. Peduncle exsertion, hundred seed weight and panicle width showed the highest PCV values 33.15, 22.76 and 19.31% respectively. Heritability estimates varied from 0.60 for stem girth to 0.90 for number of leaves. Estimates for broad sense heritability, genetic and phenotypic coefficients of variation are presented in Table 4.4. Table 4.4 Summary of estimated variances and heritability for quantitative traits Traits σ2g σ2p GCV% PCV% H2 Panicle length 3.62 5.32 6.31 7.64 0.68 Peduncle exsertion 4.10 4.59 31.31 33.15 0.89 Panicle width 0.92 1.33 16.07 19.31 0.69 Plant height 13.61 16.42 3.97 4.25 0.83 Days to flowering 3.42 4.29 2.99 3.35 0.80 Number of leaves 0.87 0.96 9.08 9.55 0.90 Stem girth 2.00 3.35 7.29 9.44 0.60 Hundred seed weight 0.33 0.45 19.63 22.76 0.74 Grain yield per plant 24.34 31.05 6.23 7.03 0.78 σ2g = genetic variance; σ2p = phenotypic variance; GCV = Genotypic coefficient of variation, PCV = Phenotypic coefficient of variation; H2 = broad sense heritability. 73 University of Ghana http://ugspace.ug.edu.gh 4.3.3 Correlation between the studied traits Significant (p<0.05) associations were observed between several traits with majority (15 pairs) exhibiting correlations greater than 0.3. These included grain yield/number of leaves (r=0.447) and grain yield/hundred seed weight (r=0.482). Both grain yield and panicle width were also significantly (p < 0.05) and positively correlated with panicle length and stem girth. The highest correlation coefficient (r=0.716) was between grain yield / panicle width. Peduncle exsertion was negatively correlated with all traits except plant height and days to flowering. A few of the correlations between the traits were low and not significantly different (p<0.05). This was mostly observed between DTF and other traits. For example, DTF/PH (r=0.019), DTF/NL (r=-0.062) and DTF/GYP (r=0.053). The phenotypic correlation among all pairs of traits is shown in Table 4.5. 74 University of Ghana http://ugspace.ug.edu.gh Table 4.5. Pearson’s correlation coefficients of quantitative traits among 550 genotypes Traits PL PE PW PH DTF NL SG HSW GYP PL 1 PE -0.222 1 PW 0.610 -0.318 1 PH 0.453 0.058 0.349 1 DTF 0.003 0.019 -0.110 -0.044 1 NL 0.279 -0.259 0.378 0.313 -0.026 1 SG 0.368 -0.393 0.393 0.123 0.106 0.509 1 HSW 0.183 -0.051 0.213 0.212 -0.062 0.166 0.156 1 GYP 0.644 -0.285 0.716 0.409 0.053 0.447 0.482 0.249 1 Values in bold are different at 0.05 significance level PL= Panicle length; PE=Peduncle exsertion; PW= Panicle width; PH= plant height; DTF= Days to flowering; NL= number of leaves per plant); SG= Stem girth; HSW= Hundred-seed weight; GYP= Grain yield/plant. P-values can be found in Appendix 2. 75 University of Ghana http://ugspace.ug.edu.gh 4.3.4 Grouping of genotypes based on principal component analysis The first three principal components (PC1 to PC3) with eigenvalues greater than 1 were extracted. These accounted for 63.8% of the total variability observed among the 550 sorghum genotypes (Table 4.6). The first principal component (PC1) accounted for 38.4% of the total variance. Grain yield per plant (0.463), panicle width (0.436) and panicle length (0.411) were the most important traits contributing to PC1. On PC1, the highest negative loading was peduncle exsertion (-0.238). The second principal component (PC2) accounted for 13.9% of variability. It showed a strong positive loading with peduncle exsertion (0.556) and plant height (0.527). On the other hand, PC2 showed negative loading with days to flowering (-0.306). The third principal component (PC3) accounted for 11.5% of the total variation with a very high factor loading for days to flowering (0.888). The loading plot shows that peduncle exsertion, plant height, panicle weight, grain yield and stem girth contributed the highest towards total genetic variation. The first axis differentiated genotypes that had larger panicles, higher grain yield and larger stem girth. The second axis however differentiated genotypes with well-exserted panicles. Results of the principal component analysis showing eigenvalues, percentage variance and cumulative percentage variance are given in Table 4.6. The contribution of the variables to the three principal components are shown in Table 4.7. Loading plots of the traits in the first two factors is shown Figure 4.3. The biplot of the 550 genotypes based on the first two principal components is presented in Figure 4.2. 76 University of Ghana http://ugspace.ug.edu.gh Table 4.6. Eigenvalues of first three principal components Traits PC1 PC2 PC3 Panicle length 0.411 0.178 0.127 Panicle exsertion -0.238 0.556 0.330 Panicle width 0.436 0.076 -0.112 Plant height 0.288 0.527 0.208 Days to flowering -0.009 -0.306 0.888 Number of leaves 0.347 -0.155 -0.053 Stem girth 0.358 -0.417 0.036 Hundred seed weight 0.192 0.286 -0.119 Grain yield per plant 0.463 0.043 0.112 Eigenvalue 3.459 1.248 1.032 Variability (%) 38.429 13.871 11.466 Cumulative % 38.429 52.300 63.765 The highest loadings are shown in bold Table 4.7. Percentage contribution (%) of the traits PC1, PC2 and PC3 Traits PC1 PC2 PC3 Panicle length 16.93 3.16 1.62 Panicle exsertion 5.69 30.90 10.88 Panicle width 19.05 0.57 1.26 Plant height 8.30 27.80 4.34 Days to flowering 0.01 9.35 78.82 Number of leaves 12.05 2.40 0.28 Stem girth 12.85 17.43 0.13 Hundred seed weight 3.68 8.20 1.42 Grain yield per plant 21.46 0.19 1.25 The highest percentages are shown in bold 77 University of Ghana http://ugspace.ug.edu.gh Traits (axes PC1 and PC2: 52.30 %) 1 0.75 PE PH 0.5 HSW PL 0.25 PW GYP 0 -0.25 NL DTF -0.5 SG -0.75 -1 -1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1 PC2 (38.43 %) Figure 4.2. Loading plot of PC1 and PC2 showing contribution of quantitative traits to variability 78 PC1 (13.87 %) University of Ghana http://ugspace.ug.edu.gh PC1 and PC2: (52.30 %) 4 m295 3 m10 mm541203m46441 m381 m325 m512 m490 m m258079m26m043m6 508m210 m416 m13m9209 m15m7m13m45 m464 917m72177670 m141 mm425 m4m345 2 m75m378m45m7m 743277m11m351899 m315 284mm20m8334m832m98268 m3m30 m04m9m29201288 2m427m135 m32589 m 6 m 2100m m14m4 75082591m343m6mm093m351m m420m4829m8308 m57 m2mm2m4m4 m98 75961438513m33m942m5339m303mm0161m98 m243 1472m49257 m422946375m123 5m13974 m278 m131091 55 m296 m395 mm33m26933mm8268387 mmm35233122 m 16547816m5 2m32m5m3 2714491m859 m2m1 245 m346214 m50m6 4m002972 m52 m3 2 m103mm41276m3751258 12 250mm244m70m29302 1 mm29543m0m2m2563215m4m5m24326m8691m36m342m9167341m924m692m44m617m387 398 8205135m7 3m2 4m4115m1299m473 m2m483m22m2340m1m3512m0m91m4239 m587m5422 m221 41m9m334632 m333 m456m364m4022810m7m23m73 0m58m2024098m5418132m8942m69360 m323 2mm453548m52034117m1m952m293m2m4m1331722684 m263 mm445m13283m038243m261349m434 54251 m140 m351729 801 5385m194738944 m416171 841m2162 7 m444m3m574m7m1370438m73m9142582387255m6m6184 m367 m256 mmm 1310m19m4632m40349m1 3m1494m270m6447m46m5450m63m641 m 41664 54 m25063 2m072264m 2 2 mm3034170 m1 19 m 4 m5m89157040 8m282 m1m27m843925 3mm454m20m23164594m90 26 m51 0 4682 6 m431 04593381mm7m2454m91254m218925466mm314922267810 m414 m32m6531 m7043 m57mm9410 m m255522m0154m7 4 7m8453 8592m3mm5 m 11m24m128 137 m2 153203189826069m89754m24113861 14c3m49 1 m5348 mm216423 m37mm01m2m3 5639m934m0mm34405117525mm4241201182700m739748m14 1413 0m52m511m9141175m2510139m08635m504 50 m3m7298796 m2 m m18m82027243m24mm5 320143571mm8613 m368 m16 mm2m1m105192232 1 m5309m78 242290m2 31 32030 8m0737m31215 61m2 m643136m87514 m366 m18 m496m54356 m244m3m145m44mm1824638m206m4m28m8131m6m781191mmm111611 m64740m7501 c12m34m34m4231mm44m38865m0393m491m271 mmm1m6461265 424 7037395m52m 49 mm452m3m5 161 m3m77 03 49m2m1m71614m36137 25 00mm5311m69m0m118429m8318 25 90 c2 -1 m547 m m474 m6339 m235m236 m8m59197 48m1m954407m57m101m583739 461 m3m1934m266 mm2512m3637m34176 m526 3m3725 m4m3397 m350 m535 mm10 m108 5343m285 m1 4418 m470 m36 m385 m447 m4m5423m9m380m3 m8487m446 m3m1121m213 mm61 m316 -2 462 m69 m327 m73m373 m442 m154 m3m22424 m68 m536 m188 m11m5110 mm38293 m109 -3 m146 -4 -8 -6 -4 -2 0 2 4 6 8 PC1 (38.43 %) Figure 4.3. Distribution of sorghum genotypes for first two principal components 4.3.5 Divergence between the genotypes based on cluster analysis The 550 genotypes were resolved into three divergent clusters at 0.52 level of dissimilarity. Cluster I consisted of the largest number of genotypes (297) including the wild-type (BTx623) and the Tx631. Cluster II comprised 133 members whereas Cluster III comprised 120 genotypes including Grinkan. The highest inter-cluster distance (3.07) was observed between Clusters II and III whereas for intra-cluster distances, the highest (3.82) was observed between Clusters I and II. Means of the three clusters for the nine traits are presented in Figure 4.4. The highest grain yield (83.0 g/plant was observed for Cluster I whereas the lowest number of days to flowering (63) was 79 PC2 (13.87 %) University of Ghana http://ugspace.ug.edu.gh observed in Cluster II. The distribution of the genotypes in the different clusters is shown in Appendix 3. 80 University of Ghana http://ugspace.ug.edu.gh - 0.7 0.6 Cluster I 0.5 Cluster III 0.4 Cluster II 0.3 0.2 0.1 0 Figure 4.4. Dendrogram of 550 sorghum genotypes derived by UPGMA method (Pearson’s dissimilarity matrix) based on standardized data of all morpho-agronomic traits. Cluster I (Green); Cluster II (Blue); Cluster III (Red). Details of the members in each cluster can be found in Appendix 3. 81 Dissimilarity University of Ghana http://ugspace.ug.edu.gh Panicle length Panicle exsertion Panicle width 50 25 9 45 8 2040 7 35 15 630 5 25 10 20 4 5 15 3 10 0 2 Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Plant height Days to flowering Number of leaves 160 70 13 140 12 65 120 11 100 60 10 80 9 60 55 40 8 20 50 7 Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Stem girth Hundred seed weight Grain yield/plant 30 4.5 250 25 4 200 3.520 150 3 15 100 2.5 10 2 50 5 1.5 0 Cluster Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Cluster I Cluster II III Figure 4.5. Boxplots showing the distribution of traits in each cluster. The upper, median and lower quartiles represent the 75th, 50th and 25th percentiles of the clusters respectively. The vertical lines represent the variation within the clusters. Dots represent outliers. 82 University of Ghana http://ugspace.ug.edu.gh 4.3.6 Variation within the mutagenised population for qualitative traits Several distinct traits were observed in the M3 families with each family exhibiting variation for at least one distinctive phenotype. Segregation for more than one distinct phenotype was observed in 11 mutant families. For example, mutations in both leaf colour and plant height (dwarfing) was displayed in Mut0889. The most frequently-observed mutant phenotypes were leaf-related, most of which were obvious by the flowering stage. Albinism was observed in 16 families. Twenty families segregated for leaf abnormalities such as lesions or spots of various sizes ranging from tiny spots to large necrotic areas. Alterations in leaf pigmentation was also very evident within the population. These included green, yellow and bronze hues. For example, vertical green stripes were observed on leaves of families Mut1865 #1 and Mut0460, yellow leaves in Mut2858 and pale green leaves in families 2899, 1865, 0755, 1043, 1202 and 1302. Leaves with red/purple margins were observed in Mut3092. Leaf striations (narrow white vertical stripes) were observed in Mut2904 while zebra cross-bands were observed in Mut1027#2. Survival of these chlorophyll mutants ranged from slight delay in vigour and growth to complete lethality in albinos (white seedlings). Several variations were also observed in midrib colour (white, dull green or brown). The putative brown midrib (bmr) mutant phenotype was easily distinguishable in the field by the distinct brownish-coloured midrib of leaves. This trait was observed in five mutant families (Mut5095, Mut5181, Mut5199 #1, Mut5398 and Mut0008). In addition to leaf abnormalities and pigmentation mutations, several lines also showed alterations in leaf arrangement. Some plots segregated for multiple plants with erect leaf architecture. This was most distinct in two independent mutant families, Mut0393 and Mut0992 which appeared to segregate for the erect leaf (erl) phenotype. These families showed more upright leaves (reduced 83 University of Ghana http://ugspace.ug.edu.gh or acute leaf angle between the main shoot and the line from the leaf base to the leaf tip) as described by Xin et al., (2015). Other mutant phenotypes included altered agronomic traits such as number of tillers and early senescence. Tillering ranged from light to profuse across the population. A few mutants showed a tendency for earlier senescence or wilting compared to the wildtype. This was observed in lines Mut0222 #1, Mut0572, Mut2952 and Mut1195. Segregation in plant size was also observed among the mutant families. In comparison to the wild- type BTx623, mutant families were either tall, semi-dwarf (normal), dwarf or tiny. Plants scored as dwarf were those that presented reduced height but largely normal leaf size and stem thickness. Tiny plants were those that presented a reduction in size of all organs, such as small leaves and thin stems. Mutant family Mut3092 for instance presented extremely short plants with no distinct internodes (rosette). The tallest plants (more than 140 cm) were observed line Mut0120 which maintained erectness in spite of height. Multiple tillering, some of which were upper nodal, was displayed in mutant families 3042, 2927, 2518, 2570, 2571 2573, 2574 and 2575. The bloomless trait (reduction of epicuticular wax) described as well-defined shiny green stems was observed in the line Mut3663. Variations in seed colour were limited in the population as most lines maintained the white-seeded characteristic of BTx623. Also, most mutants presented normal, fully-developed panicles that were comparable to the wildtype in both shape and size. One mutant family Mut0589, appeared to have the multi-seeded (msd) mutant trait as indicated by the relatively large and bulky heads. On the other hand, some lines displayed some abnormalities in panicle development. This ranged from highly-underdeveloped panicles to compact ones that bore no pollen. All distinct and altered phenotypes that were observed within the mutant population are listed in Table 4.8. To illustrate 84 University of Ghana http://ugspace.ug.edu.gh the diversity in qualitative traits observed in this mutant sorghum population, a gallery of selected mutant phenotypes is presented in Figures 4.7a to 4.7f. 85 University of Ghana http://ugspace.ug.edu.gh Table 4.8. Mutant phenotypes segregating in M3 families Category Mutant name Abbreviation Phenotype description Leaf abnormalities Crinkly leaf 1 crl1 Leaves have a crinkled appearance-raise and lowered bumps Leaf rolling 1 rl1 Tips of leaves roll, otherwise plant is normal Leaf rolling 2 rl2 entire leaves roll Leaf colour Brown midrib bmr1 leaves have a brown midrib Albino plant 1 wht1 new seedlings are white or albino Chlorotic leaf chl1 leaves have an interveinal chlorotic appearance Green stripe gst1 leaves have vertical green stripes Light green leaf lgl1 leaves are lighter green than normal Red margins rmg1 leaf margins are red Virescent seedling 1 vrs1 seedling is light green Leaf striations 1 str1 leaves have many narrow, yellow vertical stripes Leaf striations 2 str2 leaves have many narrow, white vertical stripes Variegated leaf 1 var1 leaves have variable widths of white vertical stripes Yellow leaf 1 ylf1 leaves are yellow Yellow leaf splotch 1 ysp1 leaves have yellow patches Yellow-Green leaf 1 ygl1 leaves are light green, yellowing at the tips Yellow-Green leaf 2 ygl2 leaves are yellow-green, poor growth Zebra Cross-bands zb1 leaves mostly yellow-green, with green cross-bands Zebra Cross-bands zb2 leaves are green, with yellow-green cross-bands White margins 1 wmg1 leaf margins and tips are white Leaf size and shape Erect leaf 1 erl1 leaves erect with very narrow leaf angle Grass leaf 1 grl1 leaves very narrow with –grass-like appearance Grass leaf 2 grl2 leaves very narrow with grass-like appearance with many tillers Narrow leaf 1 nrl1 leaves narrow, normal otherwise Wide leaf 1 wdl1 leaves are wider than normal Pineapple leaf 2 pnl2 leaves stacked (short internodes), normal otherwise Stacked leaf 1 stl1 leaves stacked- short internodes *Phenotypes were named according to Xin et al., (2008) and USDA Mutant description for sorghum 86 University of Ghana http://ugspace.ug.edu.gh Table 4.8 (Continued). Mutant phenotypes segregating in M3 families Category Mutant name Abbreviati Phenotype description on Lesions and necrotic Brown lesions 1 blsn1 leaves have brown spots and splotches in some area areas Brown lesions 2 blsn2 leaves have vertical brown lesions Leaf bronzing 1 lbr1 some leaves turn bronze colour Leaf bronzing 2 lbr 2 all leaves turn bronze colour, sometimes plant dies Spot leaf lesion 1 sp1 leaves have tiny whitish spots Plant size Dwarf plant 1 dw1 dwarf with stunted growth Rosette plant 1 rst1 plant stays close to the ground and resembles a rosette-type plant Short plant 1 sht1 plants shorter than normal Tiny plant 1 tny1 plants are very small with narrow leaves Tiny plant 2 tny2 plants are very small with stunted growth Tillering Single-stalked 1 Tx1 plants have single-stem, no tillers Multiple tillers 1 mtl1 plants have multiple tillers, normal otherwise Multiple tillers 2 mtl2 plants have multiple tillers, shorter than normal Multiple tillers 3 mtl3 plants have multiple tillers that sprawl out Nodal tillers 1 ntl1 plants have upper nodal tillers Senescence and wilting Early senescence 1 esn1 leaves die earlier than normal Wilt 1 wlt1 leaves wilt and turn yellow Panicle abnormalities Underdeveloped panicle 1 udp1 head not well-formed Underdeveloped panicle 2 udp2 head looks compact Underdeveloped panicle 3 udp3 head formed but never produces pollen Other Bloomless 1 blm1 stems do not have epicuticular wax Bloom blm stems have epicuticular wax *Phenotypes were named according to Xin et al., (2008) and USDA Mutant description for sorghum 87 University of Ghana http://ugspace.ug.edu.gh A B C D E F G H I J Figure 4.6a: Variations in leaf midrib and blade colour (A): crinkly leaf (crl1); (B): leaf rolling 1 (rl1); (C): leaf rolling 2 (rl2) Leaf colour: (D): cream midrib (wildtype); (E): brown midrib (bmr1); (F): albino plant (wht1); (G): chlorotic leaf (chl1); (H): green stripe (gst1); (I): light green leaf (lgl1) (J): Red margins (rmg1) 88 Figure 4.7a; Variations in leaf midrib and blade colour: (A): crinkly leaf (crl1); (B): leaf rolling 1 (rl1); (C): leaf rolling 2 (rl2) Leaf colour: (D): cream midrib (wildtype); (E): brown midrib (bmr1); (F): albino plant (wht1); (G): chlorotic leaf (chl1); (H): green stripe (gst1); (I): light green leaf (lgl1) (J): Red margins (rmg1) University of Ghana http://ugspace.ug.edu.gh A B C D E A B C D E A B C D E A B C D E F G H I J Figure 4.6b: Va riations in leaf colour (A): Virescent seedl ing 1 (vrs1); (B): Leaf striations 1 (str1); (C ): Leaf striations 2 (str2); (D): Variegated leaf 1 (var1); (E): Yellow leaf 1 (ylf1); (F): Yellow leaf splotch 1 (ysp1); G H I J (G): Yellow-greeFn leaf 1 (ygl1); (H): Yellow-green leaf 2 (ygl2) Leaf colour: (I): Zebra cross-bands 1 (zb1); (J): White margiins 1 (wmg1) g G H I J u 89 r G H I J e 4 University of Ghana http://ugspace.ug.edu.gh A B C A B C A B C A B C D E F G Figure 4.6c: Variati ons in leaf size (A): Erect le af 1 (erl1); (B): Grass leaf 1 (grl1); (C): Grass leaf 2 (grl 2); (D): Narrow leaf 1 (nrl1); (E): Wide leaf 1 (wdlE1) ; (F): Pineapple leaf 1 (pnlF1 ); (G): Stacked leaf 1 (stl1G) F i arrangement: ( g E F G u 90 Figure 7c; Variartions in leaf size andE a rrangement: (A): ErecFt leaf 1 (erl1); (B): Grass lGea f 1 (grl1); (C): Grass leea f 2 (grl2); (D): Narrow leaf 1 (nrl1); (E): Wide leaf 1 (wdl1); (F): Pineapple leaf 1 (pnl1); (G): Stacked leaf 1 (stl1) 7 University of Ghana http://ugspace.ug.edu.gh A B C D E A B C D E A B C D E A B C D E F G H I J Figure 4.7d: Varia tions in plant archite cture (A): Brown lesio ns1 (blsn1); (B): Brow n lesions (blsn2); (C): Leaf bronzing 1 (lbr1); (D): Leaf bronzing 2 (lbr2); (E): Spot leaf lesion 1 (sp1) Plant size: (F): Dwarf plant 1 (dw1); (G): G H I J Rosette plant 1 (rstF1): (H): Short plant 1 (sht1); (I): Tiny plant 1 (tny1); (J): Tiny plant (tny2) i g G H I J u 91 r G H I J e 7 University of Ghana http://ugspace.ug.edu.gh A B C D E A B C D E A B C D E A B C D E F G H I J Figure 4.6e: Var iations in tillering an d panicles (A): Single -stalked 1 (Tx1): (B): Multiple tillers 1 (mtl1 ): (C): Multiple tillers 2 (mtl2); (D): Multiple tillers (mtl3); (E): Upper nodal tillers (ntl1) Panicle abnormalities: (F): normal panicle (G): G H I J underdeveloped pFanicle 1 (udp1) (H): underdeveloped panicle 2 (udp2); (I): underdeveloped panicle (udp3); (J): undeveloped panicle i g G H I J u r G H I J e 7 92 e ; V University of Ghana http://ugspace.ug.edu.gh A B C D Figure 4.6f: Variati ons in bloom: Senescence and wilting: (A): early sen escence 1 (esn1); (B): Wi lt 1 (wlt1); Bloom: (C): Bloomless (blm1); (D): Normal bloom F B C D i g B C D u r B C D e 7 f V a r i a 93 t i o University of Ghana http://ugspace.ug.edu.gh 4.5 Discussion In any crop species an expansion of genetic divergence within the species is very useful particularly for traits of limited variability (Jiao et al., 2016). Thus, in this study, an attempt was made to enhance genetic variability in sorghum via induced mutagenesis. Significant differences (p < 0.01) were observed among the mutagenised families as well as between the mutagenised families and the wildtype BTx623 for all assessed quantitative traits. For instance, wide variability was observed for grain yield with several putative mutant families producing significantly higher grain yield than Btx623. A similar observation was made by Burow et al. (2014) where phenotypic analyses of a mutagenised BTx623 population also revealed putative mutants which exhibited significantly higher grain yield than the original. Such mutants, tagged as multi-seeded (msd), typically exhibit large bulky panicles and may be exploited for increased grain yield in sorghum (Burow et al., 2014). Besides grain yield, variations in heading dates were also observed within the mutagenised population. Families which showed significant reductions in number of days to anthesis may be classified as early flowering types. Early flowering mutants have been shown to be beneficial in improving grain production whereas late flowering types have potential in improving biomass yield (Xin et al., 2009). Variations in plant height within mutant populations such as observed in this study have been reported by several authors. For instance, Xin et al. (2008) reported reduced plant height among sorghum mutants. According to Yamaguchi et al. (2016) such dwarf types show increased resistance to lodging and are advantageous for mechanical harvesting. Variations in panicle characteristics such as length and width are important determinants of grain yield in sorghum (Hart et al., 2001). As was the case for other assessed traits, panicle length, width and exsertion varied 94 University of Ghana http://ugspace.ug.edu.gh widely within the population. Phenotypic variability for these traits give scope for selection and advancement. This indicates an induction of appreciable levels of variability and suggests an expansion of sorghum genetic resources. Estimates of genotypic coefficients of variation for the traits confirmed the significant variation revealed by ANOVA. GCV was highest for peduncle exsertion and grain yield which suggests wide differences among the mutant lines for those traits. Thus, selections may be possible for these characters. Based on the low GCV values for days to anthesis and number of leaves, it is likely minimal variability was induced by the mutagenic treatment for these traits. The relatively high heritability values associated with number of leaves suggest that the effect of the environment on the expression of this character is minimal compared to the other measured traits. This implies that selection based on this trait has potential to improve the crop. Correlation analysis revealed high association between grain yield and panicle traits such as width and length. Based on this, it can be expected that wider and longer panicles may result in higher grain yields, an observation corroborated by Hart et al. (2001) and El-Din et al. (2012). Grain yield was also positively correlated with stem girth which suggests that thicker stems were influential in increasing grain yield in sorghum. A positive correlation between stem girth and days to flowering indicates that late-flowering types are more likely to have larger stems, an observation also made by Zhao et al. (2016). The negative correlation between peduncle exsertion and grain yield is similar to an observation made by Mofokeng et al. (2017). In their study of the genetic diversity among South African sorghum genotypes a negative association between peduncle exsertion and grain yield was also observed. Information on associations between these traits may allow breeders to simultaneously select desired traits. 95 University of Ghana http://ugspace.ug.edu.gh Cluster analysis helped define the relationship between the mutagenised families and the wildtype. The grouping of 233 M3 families in clusters independent of that of BTx623 suggests considerable genetic divergence between these M3 families and BTx623. Selections from the various clusters for further breeding is likely to maintain the divergence exhibited within the mutant population. Genetic variation within the mutagenised population was also assessed using qualitative traits. It has been suggested that the occurrence of morphological alterations or novel phenotypes is more likely in mutant populations (Xin et al., 2009). Thus, the genotypes were closely inspected for traits such as plant pigmentation, midrib colour and nodal tillering. A wide variety of distinct phenotypes was observed several of which were similar to those reported by Xin et al. (2008) and Jiao et al., (2016). The most visual phenotypes were modifications in leaf pigmentation. Virescent, pale green, albino, variegated and yellow leaves were observed in several M3 families. Tan et al. (2008) explains that such alterations in leaf colour may be caused by disruptions in genes essential for chlorophyll synthesis. The chlorophyll mutations attests to the efficiency of EMS in inducing high density mutations in sorghum (Wu et al., 2005). The mutant population also showed alterations for several relevant agronomic traits. For example, the brown midrib (bmr) trait was observed in several mutant families. Brown midrib mutants derived through chemical mutagenesis have been reported in sorghum (Saballos et al., 2008). The trait, associated with reduced lignin content results in increased conversion efficiency of biomass to ethanol and improved forage digestibility (Sattler et al., 2014). The putative bmr mutants observed in this study may be useful as forage for livestock or for biofuel production (Yan et al., 2012). Other observed phenotypes include the erect leaf architecture. This feature is distinct from the typical open canopy in wildtype BTx623 in that, leaves are shorter with more acute angle between 96 University of Ghana http://ugspace.ug.edu.gh the leaf and the stem (Xin et al., 2015). Erect leaf types have been shown to contribute significantly to yield increases through improved distribution of solar radiation to lower leaves which results in greater photosynthetic efficiency (Duvick and Cassman, 1999; Narayan et al., 2014). The trait has also been reported to improve drought tolerance and allow higher planting densities on limited land resources (Saballos et al., 2008; Sattler et al., 2012). Until recently, the trait has not been fully exploited in sorghum due to limited genetic variation (Assefa and Staggenborg, 2011). However, two erect leaf (erl) sorghum mutants have been isolated from an EMS-mutagenised BTx623 in a similar study by Xin et al., (2015). Putative erect leaf types isolated in this study may be useful in improving yield in sorghum. The mutants may be confirmed by further characteristion through the measurement of leaf angles, length and width of leaves as well as panicle length and exsertion as reported by Xin et al. (2015). A putative bloomless mutant was also observed as indicated by the absence of the white epicuticular wax or the glossy appearance of the stem (Maiti et al., 1984). In contrast, wild-type BTx623 typically shows profuse deposition of epicuticular wax on stems and leaves, a feature which plays an important role in drought tolerance as well as tolerance to fungal pathogens such as Exserohilum turcicum (Burow et al., 2008). However, some reports associate bloomless sorghum with improved resistance to sheath blight (Rhizoctonia solani Kuehn) and green bug [Schizaphis graminum Rondani] (Jenks et al., 1994; Mizonu et al., 2013). The putative bloomless mutant identified here might have application in areas where the green bug and sheath blight are important stresses after resistance to these stresses have been confirmed. Lastly, other putative mutants which have potential application are those which exhibited multiple tillering and early senescence. Multiple tiller mutants were distinguishable from wildtype BTx623 by the presence of four to ten tillers as opposed to a maximum of three stalks in the wildtype (Xin 97 University of Ghana http://ugspace.ug.edu.gh et al., 2009). Multiple tillering types may be useful for enhancing biomass yield under intense management whereas early senescing types may be useful for early dehydration of biomass for use as livestock feed (Jiao et al., 2016). 4.6 Conclusion A mutagenised sorghum population was generated using the chemical mutagen EMS. The population was phenotyped in the field for traits such as panicle length, width and exsertion, plant height, grain yield and days to flowering. Significant variation was observed among the putative mutants and between the putative mutants and the wildtype BTx623. Multivariate analysis confirmed the variability within the mutagenised population by the clustering of the M3 families into three divergent clusters. An assessment of qualitative traits also revealed several phenotypes distinct from the wild-type BTx623. Of these, several desirable mutants were identified which have potential use in plant breeding and/or crop production. These include putative mutants for; early flowering (Mut2619), late flowering (Mut0878), brown midrib (Mut5095, Mut5181, Mut5199#1, Mut5398 and Mut0008), multiple-tillering (Mut3042, Mut2927, Mut2518, Mut2570, Mut2571, Mut2573 and Mut2574), bloomless (Mut3663), erect leaf architecture (Mut0393 and Mut0992) and multi-seeded (Mut0589).The diversity observed within the mutagenised population suggests that it is a useful germplasm resource for both forward and reverse genetic studies. 98 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE GENETIC VARIATION FOR PHOSPHORUS-USE EFFICIENCY AMONG MUTAGENISED SORGHUM LINES 5.1 Introduction In West Africa, available phosphorus (P) levels in cultivated soils can be as low as 2 mg P/kg (Buekert et al., 2001). This is a major constraint particularly for traditional crops such as sorghum which are usually cultivated under low-input conditions (Leiser et al., 2012). Under such conditions, seed germination, seedling vigor and overall plant growth are adversely affected resulting in significant losses in grain yield (Doumbia et al., 1993; Leiser et al., 2012). Given its key role as a food staple in many West African rural communities, low sorghum yield coupled with rapid growth in human population threatens to worsen food insecurity in the region (Gemenet et al., 2016). One of the key strategies for mitigating the effects of soil infertility on sorghum is increased fertiliser use among smallholder farmers (Gemenet et al., 2016). Unfortunately, fertiliser use in West Africa is severely constrained by farmers’ inability to afford the commodity, its frequent unavailability on rural markets as well as farmers’ misgivings on achieving maximum net returns on fertiliser use (Kaizzi et al., 2012; Rware et al., 2014). Even when applied, incorrect timing rates and rates of application are factors that affect efficient fertiliser-use among smallholders (Rware et al., 2014). Moreover, increased use of P fertilisers is not sustainable due to depletion of world phosphorus reserves (Gemenet et al., 2016). The cost of P fertilisers are expected to increase, thus making the commodity even more inaccessible to smallholder farmers in resource-poor regions (Cordell and White, 2015). In view of these constraints, increased fertiliser-use among smallholders as a strategy to improve sorghum productivity on West Africa’s marginal soils, must be complemented with other approaches. 99 University of Ghana http://ugspace.ug.edu.gh One such approach is the development of P-use efficient varieties. Until recently, research targeted at improving P-use efficiency in sorghum has been limited in West Africa (Leiser et al., 2012). The most significant work is that reported by Leiser et al. (2015), in which selection strategies for P-use efficiency were researched in West African germplasm. They report useful genetic variation for P-uptake efficiency in Guinea landraces and utilisation efficiency in Caudatum-derived cultivars. However, both uptake and utilisation efficiency was found to be lacking in any single genotype (Leiser et al., 2015). Given the extremely low-P status of West African soils and its negative impact on sorghum yield, the development of varieties which show improved levels of both uptake and utilisation efficiencies is worthwhile. An expansion of the genetic variability for both traits will also be useful in future breeding programmes. These goals can be approached in several ways. The traditional approach is hybridisation, for instance of the two genetic pools identified by Leiser et al. (2015). However, this method usually requires the development of inbred lines, followed by several years of backcrossing. In complement to recombinant breeding, induced mutagenesis offers a rapid and efficient method for generating novel variability for important agronomic traits in crops (Bentley et al., 2000). For example, it was applied in the development of sorghum varieties with improved lysine content and early flowering (Sree Ramulu, 1970a; Quinby, 1975; Oria et al., 2000). The primary advantage of artificial mutagenesis is that desirable mutants may be released directly as varieties without the need for several years of backcrossing (Ahloowhalia, 2004). Thus, the goal of this study was to assess the genetic variability for P-uptake and utilisation efficiency among a population of mutagenised sorghum as a step towards identifying new sources of P-use efficiency in sorghum. The specific objectives were to; 100 University of Ghana http://ugspace.ug.edu.gh 1. assess the genetic variation for P-use efficiency among mutant and cultivated genotypes under screen-house conditions, 2. confirm any observed diversity for P-use efficiency under field conditions and 3. determine differences between mutant genotypes and cultivated genotypes. 5.2 Materials and methods 5.2.1 Evaluated genotypes Two hundred sorghum genotypes were evaluated comprising 170 mutagenised families and 30 cultivated genotypes. The mutagenised lines were selected from previous experiments in 2014 and had been advanced from M3 to M6 by controlled self-fertilisation. Included in the 30 cultivated genotypes were Grinkan, a P-use efficient genotype (Leiser et al., 2012) and Kadaga, a variety frequently grown in northern Ghana. These two genotypes served as control varieties. With the exception of Kadaga, all genotypes were sourced from the Whistler laboratory, Purdue University. A summary of the 31 cultivated lines included in this study and their important features are listed in Table 5.1. 101 University of Ghana http://ugspace.ug.edu.gh Table 5.1. Summary of the cultivated genotypes and their important features Genotypes Important feature Kuyuma (WSV387) Food-grade – Zambia Sepon82 Food-grade – Niger Ajabsido Post-flowering drought tolerance – Sudan CE-151-262-A1 Food-grade – Senegal CSM-63 Popular Guinea – Mali Koro Kollo Pre-flowering drought tolerance Sudan Feterita Gishesh Preflowering drought tolerance – Sudan Segeolane Preflowering drought tolerance – Botswana PI609567 - ErtHdDur D/D N. Mali Erect-head dhurra - N. Mali Wassa Food-grade guinea – Mali Seguetana Food-grade guinea – Mali El Mota - S241 Pre-flowering drought tolerance – Niger Honey Drip Sweet sorghum Theis Sweet sorghum Mace Da Kunya (MDK) Late-season dune sorghum - Niger MR732 Male-parent – Niger Tx430 Elite yellow pollinator BTx2752 Elite red seed parent TX 436 Elite food-grade male TX631 Elite food-grade seed parent Framida Striga resistant - Burkina Faso ICSV1049 Striga resistant - Burkina Faso Sariaso 14 Striga resistant - LGS - Burkina Faso Grinkan P-use efficient (Food Grade Guinea – Mali) TxARG1 High grain yield BN223 Male-sterile parent SC599 Stalk rot resistant and post-flowering drought tolerant Mota Maradi Improved variety - Niger Honeydrip Sweet sorghum SC35 Charcoal rot resistant Kadaga East Popular variety (northern Ghana) 102 University of Ghana http://ugspace.ug.edu.gh 5.2.2 Description of study sites Pot experiments were carried out at the Biotechnology and Nuclear Agricultural Research Institute (BNARI) of the Ghana Atomic Energy Commission (Latitude 5o40’N and Longitude 0o 13’W). Field experiments were conducted at the research farm of the University of Ghana, Legon, (Latitude 5o36’N, Longitude 0o10W), approximately 8.8 km from BNARI. A more-detailed description of the research farm is given in the previous chapter. Meteorological data for the experimental areas during the study period is given in Table 5.2. Table 5.2. Meteorological data for University of Ghana research farm during the study period Mean monthly temperature (oC) Year/ Total monthly rainfall Maximum Minimum month (mm) 2015 May 139.3 32.7 25.0 June 295.1 - - July 53.8 29.1 - August 2.0 29.9 23.1 2016 July 49.7 29.2 23.6 August 16.9 28.8 23.1 September - - - October - - - Source: Ghana meteorological agency (2015, 2016) 5.2.3 Soil sampling and analyses For the pot experiment, soils were sampled from several sites at Berekusu in the Eastern Region of Ghana. The selected sites had no recent history of crop cultivation and/or fertiliser application. Samples were collected from the plough layer (0 to 20 cm in depth) using the auger after which all debris were removed. The soil samples were then air-dried for seven days, homogenized and 103 University of Ghana http://ugspace.ug.edu.gh passed through a 2mm sieve to obtain fine earth. Subsequently, physical and chemical analyses were carried out at the University of Ghana Soil Science Laboratory. Analyses included determination of soil texture (Bouyoucos, 1951) and pH. Organic carbon content was determined using the wet combustion method of Walkley and Black (1934) whereas Total N was determined by the Kjeldahl digestion method. Available-P was extracted by the Bray-1 method and analysed using the molybdate blue procedure (Murphy and Riley, 1962). Cation exchange capacity (CEC) was determined by saturation with 1 N ammonium acetate (NH4OAc, pH 7.0) and extraction of ammonium with 2M potassium chloride (TSBF, 1993). Cations (Ca, K and Mg) were extracted by the Mehlich- 3 procedure (Mehlich, 1984) and determined by absorption spectrophotometry. Undisturbed soil samples were taken for bulk density determination. All analyses were carried out in triplicate and the results averaged. Details of procedures for soil characterisation and analyses is given in Appendix 4. Soils for all experiments were selected based on available P content. For low P soils, critical point for available-P was less than 10 mg P/kg (Doumbia et al., 1993). The physico-chemical properties of the soils in all the trials are shown in the results. Following soil analyses, top soil was harvested from suitable sites for the pot experiment. Harvested soil was passed through sieves to remove all debris and obtain fine earth. Large polythene bags were used as pots; these were filled with equal quantities (approximately 10 kg) of sieved soils after which the soils were allowed to equilibrate for two weeks before sowing. 5.2.4 Experimental design and layout The experimental design for the pot experiment was a two-factor (fertiliser level and genotype) factorial arranged in a randomized complete block design. Each genotype was allocated twelve pots in total. Thus, the total number of experimental pots was 2,400. The pots were grouped into 104 University of Ghana http://ugspace.ug.edu.gh four blocks; two blocks per P treatment. In each block, each genotype was represented by three pots making a total of 6 pots per P environment. Two P treatments were imposed at sowing; 0 mg P/kg (control) and 50 mg P/kg P2O5 (from triple super phosphate – TSP). Three seeds were sown per pot. Fourteen days after planting seedlings were thinned to two per pot. All pots received nitrogen from urea (46-0-0) at 80 mg N /kg and potassium from muriate of potash at 40 mg K / kg of soil. Standard agronomic and plant protection practices were followed during the cropping season. Field trials were conducted in the 2016 cropping season. The experiment consisted of two separate and contrasting P fields; one with P fertilisation and one without. In each trial, the 200 genotypes were sown in an incomplete block/augmented design. The number of incomplete blocks in each trial was 20. Each block comprised 12 row plots (10 test genotypes plus the two check varieties). The plots measured approximately three metres with 0.8 m between the plots. Ten hills were planted per row at three seeds per hill. Hills were later thinned to two plants. At sowing, the optimum P field was treated with 40 kg ha -1 of triple super-phosphate (TSP). Urea and potash were applied at 50 kg ha -1 40 kg ha -1 respectively. To obtain similar units of nitrogen and potassium in both fields, the low P field was treated with urea and potash at 60 kg ha -1 and 40 kg/ha respectively. Trials were mainly rain-fed. However, irrigation was provided when necessary. Standard agronomic practices of weed and insect control were used. Soil analyses were repeated six weeks after planting. 5.2.5 Data collection Data were collected on eight traits. For pot experiments, data were collected on all twelve plants that represented each genotype. With the exception of grain and stover yield, field data were 105 University of Ghana http://ugspace.ug.edu.gh collected on five tagged plants in the middle of each plot. A summary of the data collected in the study is given in Table 5.3. Table 5.3 Descriptions of evaluated traits Traits Descriptions Seedling vigour Visual score based on a 1 to 5 scale, 1=least vigorous and 5 = most vigorous. This was recorded 28 days after planting. Days to anthesis Number of days from sowing to when each tagged plant had flowered. Plant height Distance from the base of the plant to the flag leaf at anthesis. Stem girth Width of the base of the stem at anthesis. Stover yield Average of weight of stover of all plants in a plot Grain yield Average of weight of seeds harvested from whole plot. P concentration in leaves Leaves were sampled from 3 tagged plants at anthesis and measured as the amount of P (mg) in 1 g shoot dry matter. P concentration in grain Grain sampled from 3 tagged plants at physiological maturity and measured as the amount of P (mg) in 1 g dry grain matter. 106 University of Ghana http://ugspace.ug.edu.gh Seedling vigour scores were a relative evaluation based on the range of variation for seedling size (height, length and width of leaves) in the genotype under study (Cisse and Ejeta, 2003). Harvested panicles and stover were air-dried at ambient temperature for 10 days prior to weighing. For pot experiments, stover and grain yield were analysed in grams/plant whereas for field experiments the parameters were measured in kg, then converted into tonnes per hectare. Digestion and determination of P concentration in the plant samples were done using the Bhargava and Raghupathi (1984) and Murphy and Riley (1962) protocols respectively (Described in more detail in Appendix 5). 5.2.6 Data analysis 5.2.6.1 Data analysis for pot experiments Analysis of variance (ANOVA) was performed on all traits using the format presented in Table 5.4. Table 5.4. Format of ANOVA for two-way factorial experiment (RCBD) Source of variation Df Mean square (MS) Expected mean square (EMS) Genotypes (G) 199 MSg σ 2 e + RP σ 2g + R σ2gp P – fertilisation (P) 1 MSp σ 2 e + RG σ 2 + R σ2gp G*E MS σ2 + R σ2gp e gp Error (E) MSe σ 2 e σ2e = error term; σ 2g = genotypic variance; σ2gp = variance due to genotypes by phosphorus interaction; MSg = mean square for genotype; MSp = mean square for phosphorus level; MSgp= mean square for genotype by phosphorus interaction; MS4 = mean square error Genetic and phenotypic variances as well as genotypic and phenotypic coefficients of variation were calculated using the following formulae; 107 University of Ghana http://ugspace.ug.edu.gh Genotypic variance (σ2g) = MSg-MSe/r Error variance (σ2e) = MSe/r; where MSg and MSe are the mean squares of the genotypes and error respectively and r is the number of replications. Phenotypic variance (σ2 ) = (σ2p g) + (σ 2 e) Genotypic and phenotypic coefficients of variation were calculated according to the formulae proposed by Burton (1952). PCV and GCV values greater than 20% were regarded as high, values between 10 and 20% were intermediate, whereas values less than 10% were considered low (Khorgade et al., 1985). GCV = √σ2g x 100; PCV = √σ 2 p x 100 x x ECV (Environmental Coefficient of Variation) = √σ2e x 100/ x GECV (Genotype by Environment interaction coefficient of variation) = √σ2gp x 100 / x Where x = sample mean Broad sense heritability (H2) was estimated using the formula suggested by Falconer (1989) as follows; H2 = σ2 / σ2g g + σ 2 p; where σ 2 g = genotypic variance and σ 2 p = phenotypic variance. Pearson’s correlation coefficients were estimated using the mean values in each P environment. Hierarchical agglomerative clustering was done using Pearson’s dissimilarity matrix. A dendrogram was formed based on unweighted pair group method with arithmetic mean (UPGMA). All mean values were standardized prior to analyses. Correlation and cluster analyses were done using XLSTAT 2017. 108 University of Ghana http://ugspace.ug.edu.gh P-uptake [PUp] (mg/plant) was calculated according to Gunes et al. (2006) where; PUp = Stover yield (g) x P concentration in the stover (mg/g).1 P-utilisation efficiency (PUtE) was calculated according to Rose and Wissuwa (2012); where P- utilisation efficiency is the stover yield produced per unit P accumulated in the stover. PUtE = Stover yield (g) / P-uptake (PUp). P-tolerance ratios (relative performance) were calculated for each genotype. The mean value of each trait in the low P trial was divided by its corresponding mean value in the optimum P trial for each genotype (Leiser et al., 2015). The ratios were calculated for seedling-vigor (SVR), days to flowering (DTFR), plant height (PHR) and grain yield (GYR). Genotypes that scored higher SVR, PHR and GYR and lower DTFR were expected to be better adapted to low-P stress (Leiser et al., 2015). The genotypes were ranked for tolerance to low-P stress based on a selection differential where; Selection differential, S = GYR of the genotype – Grand mean (GYR) (Leiser et al., 2012). Genotypes were tolerant if the GYR of that genotype was higher than the grand mean whereas inefficient genotypes had GYR values lower than the grand mean (Ozturk et al., 2005;Gunes et al., 2009). The top 10% P-use efficient genotypes were those that scored the highest selection differentials. 5.2.6.2 Data analysis for field experiment Analysis of variance for field experiments was performed in two stages. Initially, each P trial (single-environment) was analysed separately to evaluate the differences between the genotypes within each environment. Each trait was analysed with a restricted maximum likelihood (REML) mixed model where genotypes were treated as fixed effects and incomplete blocks were fitted as random effects. Differences between genotypic means across the optimum P and low P trials were 109 University of Ghana http://ugspace.ug.edu.gh tested using at two-sided t test. Correlations between traits were estimated for each P environment. Subsequently, a combined analysis of variance across the two environments was done where P treatments were treated as fixed and genotypes as random to estimate the genotype-by-P interaction. Genetic and phenotypic variances, genotypic and phenotypic coefficients of variation (GCV and PCV respectively) as well as broad sense heritabilities were estimated using formulae as described for the pot experiment. ANOVA were carried out using SAS 9.2 software (SAS Institute Inc., Cary, NC) whereas correlation analysis and genetic parameters were estimated using XLSTAT 2017. Phosphorus-uptake and utilisation efficiency parameters were calculated as described by Leiser et al., (2012). The parameters are described in Table 5.5. 110 University of Ghana http://ugspace.ug.edu.gh Table 5.5. Estimations of P-uptake and utilisation parameters Trait Description Calculation Unit GY Grain yield t/ha SY Stover yield t/ha BM Total biomass yield GY + SY t/ha PCS P concentration in stover mg/g PCG P concentration in grain mg/g PS P content in stover PCS x SY kg/ha PG P content in grain PCG x GY kg/ha PBM P content of total biomass PS + PG kg/ha PHI P harvest index PG/PBM PUTIL-G P-utilisation for grain GY/PBM kg/g P production PUTIL-S P-utilisation for stover SY/PBM kg/g P production PUTIL-BM P-utilisation for biomass BM/PBM kg/g P production Based on these parameters, the genotypes were classified into four categories as P-uptake efficient, P-utilisation efficient, low-P stress tolerant and low-P stress sensitive. 111 University of Ghana http://ugspace.ug.edu.gh 5.3 Results 5.3.1. Pot experiment 5.3.1.1. Characteristics of soils in pot experiment Soil was characterised as sandy clay. The physical and chemical properties of the soil are as follows; Texture (68.4% sand, 7.2% silt and 24.4% clay), pH (5.6), organic carbon (8.2 g/kg), Total N (1.3 g/kg) and CEC (6.1 cmolc/kg). Plant - available P for the optimum P pots was an average of 18.7 mg P /kg of soil whilst that for the low P pots was 4.6 mg P /kg of soil. 5.3.1.2. Genetic variation among the genotypes under contrasting P environments Highly significant differences (p<0.001) were observed among the genotypes for all traits (Table 5.6.1). Interaction between P-environments and genotypes was also significant (p<0.001). Grain yield reduced significantly from an average of 81.7 g/plant under optimum P conditions to 48.8 g/plant under low P- stress (Table 5.6.2). The least grain yield in both low P and optimum P environments (22.1 and 38.2 g respectively) was produced by Feterita Gishesh. Grinkan (121.8 g) and MDK (139.7 g) were the highest producers of grain in the low and optimum P environments respectively. Stover yield also decreased from 237.6 to 149.8 g between the two environments as was the case for plant height which decreased from 103 cm to 81 cm. Seedling vigour scores varied between the two environments with a mean of 3.4 and 4.7 in low and optimum P conditions respectively. The trend was reversed for days to anthesis, as the trait increased from 59.8 days under optimum conditions to 62.5 days under stressed conditions. Distribution of all the traits in both P environments is shown in Figure 5.1 Genetic variance among the assessed traits was lowest (0.21) for seedling vigour and highest (17693.66) for grain yield (Table 5.6.3). Similarly, the highest GCV was observed for grain yield 112 University of Ghana http://ugspace.ug.edu.gh (66.09), plant height (59.34) and biomass (51.38) whilst seedling vigour and days to flowering exhibited the lowest GCV of 11.4 and 29.27 respectively. Broad sense heritability were intermediate (approximately 0.50) for all the traits except seedling vigour (0.36). 113 University of Ghana http://ugspace.ug.edu.gh Figure 5.1. Distribution of traits in low and optimum P-environments. 114 University of Ghana http://ugspace.ug.edu.gh Table 5.6.1. Analysis of variance under contrasting P environments. Source of variation df SV PH DTF SG GY SY BM Genotypes (G) 199 0.73** 6031.59** 213.72** 59.88** 3877.70** 19892.09** 36153.16** P environment (P) 1 942.51** 269985.27** 4400.33** 7738.93** 648554.94** 4625364.20** 8757254.74** G*P 199 0.44** 96.09** 1.36ns 4.28** 587.10** 3744.72** 5565.28** Grand mean 4.04 92.40 61.14 18.28 65.31 193.68 258.87 R-square 0.66 0.98 0.92 0.80 0.84 0.89 0.92 ** Significant at 0.01 probability level; SV = Seedling vigour; PH = Plant height; DTF = Days to flowering; SG= Stem girth; GY= Grain yield per plant; SY = Stover yield per plant; BM = Biomass per plant Table 5.6.2. Comparison between means of traits in contrasting P environments Seedling Days to vigour Plant height flowering Stem girth Grain yield Stover yield Biomass Optimum P 4.66a 103.01a 59.78a 20.07a 81.73a 237.55a 319.28a Low P 3.41b 81.80b 62.49b 16.48b 48.85b 149.75b 198.60b Means followed with different letters in a column are significantly different at p < 0.05 in a two-sampled t test. 115 University of Ghana http://ugspace.ug.edu.gh Table 5.6.3. Heritability, genetic and phenotypic variability of the traits under low-P stress Trait σ2g σ2p GCV PCV ECV GECV Heritability Seedling vigour 0.21 0.37 11.40 14.97 9.70 3.64 0.367 Days to flowering 105.87 106.86 16.83 16.91 1.63 0.53 0.497 Plant height 3006.57 3015.79 59.34 59.43 3.29 3.89 0.499 Stem girth 28.63 29.94 29.27 29.93 6.28 2.87 0.488 Grain yield/plant 1863.25 1938.85 66.09 67.42 13.31 13.05 0.490 Stover yield/plant 9657.59 9946.05 50.74 51.49 8.77 11.86 0.492 Biomass 17693.66 18076.58 51.38 51.94 7.56 10.93 0.494 Where σ2g = genetic variance; σ 2 p = phenotypic variance; GCV = genetic coefficient of variation; PCV = phenotypic coefficient of variation; ECV = Environmental coefficient of variation; GECV = Genotype by Environment interaction coefficient. 116 University of Ghana http://ugspace.ug.edu.gh 5.3.1.3. Correlation between morpho-agronomic traits and P-tolerance ratios Plant height was positively correlated with seedling vigour (r=0.179) and grain yield (r = 0.301) [Table 5.6.4]. Positive and significant (p<0.05) correlations were also observed between grain yield and stover yield (r= 0.602), days to flowering (r = 0.380) and stem girth (r = 0.234). Stover yield was also positively correlated to days to flowering was positively (r = 0.234) and stem girth (r = 0.391). No negative correlations were observed except between the morpho-agronomic traits except between seedling vigour and stem girth (r = -0.191). 117 University of Ghana http://ugspace.ug.edu.gh Table 5.6.4. Correlation coefficients between morpho-agronomic traits (under low P stress) Seedling Plant Days to Grain Stover Traits vigour height flowering Stem girth yield/plant yield/plant Biomass Seedling vigour 1 Plant height 0.179 1 Days to flowering -0.046 0.116 1 Stem girth -0.191 -0.125 0.352 1 Grain yield/plant -0.056 0.301 0.380 0.418 1 Stover yield/plant -0.085 0.256 0.234 0.391 0.602 1 Biomass -0.083 0.297 0.307 0.439 0.798 0.962 1 Values in bold are significantly different at p < 0.05. 118 University of Ghana http://ugspace.ug.edu.gh 5.3.1.4. Principal component analyses under contrasting P environments The contribution of each trait to genetic variation in each P environment was determined using principal component analysis. Under low-P stress conditions, the first three principal components (PC1, PC2 and PC3) explained 77 % of the total variation. PC1, PC2 and PC3 accounted for 45.4%, 18.9 and 12.7% of the total variance respectively (Table 5.6.5). Under optimum P conditions, the first three principal components explained 79% of the total variation and accounted for 47.2, 17.4 and 14.5% of the total variance respectively. In both P environments, yield components (biomass, grain yield and stover yield) were the most important traits for PC1 whereas for PC2, plant height, seedling vigour and stem girth were the most important (Figure 5.2). Days to flowering (0.749) and seedling vigour (0.872) were the highest contributors to the variability in PC3 for low P and optimum P respectively. Contribution of all the traits in PC1 and PC2 are shown in Figure 5.2. The bi-plots (Figure 5.3) clearly separated the mutant genotypes from the cultivated lines with respect to the assessed traits. In both P conditions, genotypes in quadrant I (such as MDK, Grinkan, MR732 and Mut3864) were characterised by high grain and stover yield. Conversely, genotypes in quadrant III were characterised by lower values in yield. On the other hand, genotypes in quadrant IV comprised early-flowering types such as Mut2902#1, SC35, Mut0393#1. 119 University of Ghana http://ugspace.ug.edu.gh Table 5.6.5. Principal components of morpho-agronomic traits in the contrasting P environments Low P Optimum P Traits PC1 PC2 PC3 PC1 PC2 PC3 Seedling vigour -0.070 0.589 0.503 -0.028 -0.323 0.872 Plant height 0.192 0.642 -0.076 0.160 0.723 0.362 Days to flowering 0.287 -0.133 0.749 0.282 -0.281 0.259 Stem girth 0.334 -0.456 0.249 0.361 -0.490 -0.167 Grain yield/plant 0.494 0.076 -0.023 0.494 -0.042 -0.108 Stover yield/plant 0.477 0.063 -0.289 0.484 0.206 0.046 Biomass 0.542 0.077 -0.182 0.534 0.100 -0.028 Eigenvalue 3.180 1.326 0.890 3.301 1.219 1.012 Variability (%) 45.432 18.945 12.711 47.152 17.408 14.455 Cumulative % 45.432 64.377 77.088 47.152 64.560 79.015 The highest factor loadings are shown in bold. 120 University of Ghana http://ugspace.ug.edu.gh (PC1 and PC2: 64.38 %) (PC1 and PC2: 64.56 %) 1 1 Plant height Seedling Plant height 0.75 vigour 0.75 0.5 0.5 Stover yield 0.25 per plantGraSitno yvieerl dy iper 0.25 Biomaeslds Biomass ppelar nptlant 0 0 Grain yield per Days to plant -0.25 flowering -0.25 Days to Seedling flowering vigour -0.5 -0.5 Stem girth Stem girth -0.75 -0.75 -1 -1 -1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1 -1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1 PC1 (45.43 %) PC1 (47.15 %) A B Figure 5.2. Loading plot of PC1 and PC2 showing contribution of morpho-agronomic traits to variability in (A) low P environment and (B) optimum P environment. 121 PC2 (18.94 %) PC2 (17.41 %) University of Ghana http://ugspace.ug.edu.gh II I II I IV III III IV A B Figure 5.3. Distribution of the 200 genotypes for PC1 and PC2 under (A) low-P and (B) optimum P environments. Names of 200 putative mutants in the bi-plots can be found in Appendix 6. 122 University of Ghana http://ugspace.ug.edu.gh 5.3.1.5. Variation within the population based on cluster analysis. Three main clusters (I, II and III) were formed at 49% level of dissimilarity (Figure 5.4). Cluster I contained 65 genotypes, 63 of which were mutant families grouped with two cultivated varieties (Tx631 and SC599). The genotypes in this cluster were the late flowering types (63 days) within the population as well as the shortest in height (71 cm). Average grain yield for this cluster was an intermediate 41 g/plant. Cluster II contained 106 genotypes including cultivated varieties such as Grinkan, MR732, KuyumaWSV387 and CE151262A1. These genotypes were the highest in terms of grain (56 g/plant) and stover yield (164 g/plant) with relatively thicker stems (17 cm). Cluster III contained 29 genotypes. Fourteen of the genotypes included in this cluster were the cultivated varieties such as CSM63, Korokollo, Ajabsido, Honeydrip, Sariaso and Kadaga West. The genotypes in this cluster were the tallest (106 cm) and thinnest (13.5 cm) genotypes in the population. The resulting dendrogram is shown in Figure 5.4. Box plots summarising the means of the various clusters are shown in Figure 5.5. 123 University of Ghana http://ugspace.ug.edu.gh 0.7 0.6 Cluster II Cluster III Cluster I 0.5 0.4 0.3 0.2 0.1 0 Figure 5.4. Dendrogram of 200 sorghum genotypes derived by UPGMA method (Pearson’s dissimilarity matrix) based on standardized data of all morpho-agronomic traits. Details of the members in each cluster can be found in Appendix 7. 124 Dissimilarity University of Ghana http://ugspace.ug.edu.gh Days to flowering Stem girth Plant height Seedling vigour 80 24 250 4.5 4.3 75 22 200 4.1 20 3.9 70 18 150 3.7 65 3.5 16 100 3.3 60 14 3.1 50 2.9 55 12 2.7 50 10 0 2.5 Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Biomass Stover yield/plant Grain yield/plant 400 300 140 350 120 250 300 100 200 250 80 150 200 60 100 150 40 100 Cluster 50 20 Cluster I Cluster II III Cluster I Cluster II Cluster III Cluster I Cluster II Cluster III Figure 5.5. Boxplots showing the distribution of traits in each cluster. The upper, median and lower quartiles represent the 75th, 50th and 25th percentiles of the clusters respectively. The vertical lines represent the variation within the clusters. Dots represent outliers. 125 University of Ghana http://ugspace.ug.edu.gh 5.3.1.6. Relative performance of the genotypes in contrasting P environments. Relative performance of the genotypes were used as measures or indicators of low-P stress tolerance. P-tolerance ratios were estimated as the genotypic mean of each trait under low P to its corresponding genotypic mean under optimum P conditions. Based on this, seedling vigour ratio (SVR) was lowest (0.55) in Mut1665 and Mut5115 and highest (0.94) in Feterita Gishesh. Average seedling vigour ratio (SVR) within the population was 0.73. The least plant height ratio [PHR] (0.65) was shown in Mut2716 whilst the highest (0.94) was observed in Seguetana. An average of 0.79 was recorded for this trait. The lowest (1.02) days to flowering ratio (DTFR) was observed in six genotypes. These included Mut2856#2, MR732, Mut0460#1, Grinkan, CSM63 and Mut0056. The highest (1.10) was observed in El-Mota. The highest grain yield ratio (0.83) was shown by Wassa whilst the highest stover yield ratio (0.86) was shown by Mut0056. The least P-uptake [PUp] (38.2 mg/plant) was observed in Mut2899#1. Framida had the highest SPU of 142.58 mg/plant. Estimated P-utilisation efficiency (PUtE) was least (1.39) in Korokollo and highest (2.86) in TxARG1. 5.3.1.7. Selection of P-use efficient genotypes P-use efficient genotypes were selected based on relative performance of the genotype in both P environments with regards to grain yield (grain yield ratio). All genotypes that showed grain yield ratios (GYR) greater than the grand mean (GYR of the population) were categorised as P-use efficient. Estimated grand mean of the grain yield ratios was 0.61. Based on this, 44.5% of the genotypes were P-use efficient whereas 52.5% were inefficient. Three percent showed GYR values equal to the grand mean and were classified as average. Subsequently, the top 10% P-use efficient 126 University of Ghana http://ugspace.ug.edu.gh genotypes were selected based on a selection differential; where the selection differential = GYR of the genotype – Grand mean (GYR) of the population. The genotypes with the highest selection differentials were selected. Wassa, Segeolane, Kadaga West, CE151262A1 and Mut4459#2 comprise the first five. The top 10% genotypes are summarised in Table 5.6.6. Table 5.6.6. Top-ranked (10%) genotypes for grain yield ratio, P-uptake and P-utilisation efficiencies Rank Genotypes GYR Genotypes PUp Genotypes PUtE 1 Wassa 0.83 Framida 142.58 TxARG1 2.86 2 Segeolane 0.80 ICSV1049 135.26 Mut2854#1 2.78 3 Kadaga West 0.79 Mut3434 128.61 SC35 2.70 4 CE151262A1 0.78 Grinkan 118.91 Mut3003#1 2.63 5 Mut4459#2 0.78 MDK 117.25 Mut0122 2.56 6 Ajabsido 0.78 BN223 115.46 Mut3708 2.56 7 Theis 0.78 Mut4106#1 110.97 Mut3217 2.56 8 Mota Maradi 0.78 Mut2570 106.67 Mut3412 2.56 9 Mut4511 0.76 Mut4155 106.00 Mut3709 2.56 10 SC35 0.75 PI609567 104.41 Tx430 2.56 11 KoroKollo 0.75 Mut0839#2 103.46 Mut3960#1 2.56 12 Grinkan 0.75 Tx631 102.40 Mut2091#1 2.50 13 Seguetana 0.74 Mut3809 102.05 Mut4260-2 2.50 14 Mut3707 0.74 Mut4112 97.61 Mut3808 2.50 15 Mut2854#1 0.74 Mut2359 96.55 Mut2952#1 2.50 16 Mut4156 0.74 Mut4264 95.44 Mut3368 2.50 17 Mut4060#1 0.74 Mut4313-2 94.82 CSM63 2.50 18 Sepon82 0.74 Mut3105 94.55 Mut2899#1 2.50 19 El-Mota 0.73 Mut3005 93.83 Mut2098#1 2.50 20 Mut3708 0.73 Mut0454#1 92.75 Mut3813 2.50 GYR= Grain yield ratio; PUp= P-uptake efficiency; PUtE =P utilisation efficiency 127 University of Ghana http://ugspace.ug.edu.gh 5.3.2. Field experiment 5.3.2.1. Characteristics of soils in field experiments Soils of field experiments were characterised as sandy loam belonging to the Toje Series, Rhodic Khandiustalf (Eze, 2008). The physical and chemical properties of the soils in the two P environments were as follows; For the low P field, soil texture (72.3% sand, 18.1% silt and 8.7% clay), pH (5.5), organic carbon (4.1 g/kg), Total N (0.3 g/kg) and CEC (2.9 cmolc/kg). Plant - available P was an average of 5.7 mg P /kg. For the optimum P field, texture (74.4% sand, 19.4% silt and 6.2% clay), pH (5.8), organic carbon (5.0 g/kg), Total N (0.4 g/kg), CEC (3.3 cmolc/kg). Plant - available P was an average of 19.5 mg/kg. 5.3.2.2. Genotypic variation for morpho-agronomic traits in contrasting P environments Highly significant differences (p<0.001) were observed among the genotypes for all morpho- agronomic traits in both P environments (Tables 5.6.7 and 5.6.8). A significant shift in mean values was also observed between the two P-environments. Under low P conditions, days to anthesis decreased from 62 days to 60 days under adequate conditions. For all other traits, mean values increased significantly under optimum conditions. Plant height increased from 80 to 101 cm whereas stover yield increased from 3.5 tonnes/ha in the low P environment to 4.7 tonnes/ha under optimum conditions. Mean grain yield varied almost two-fold between the two environments (0.86 tonnes/ha to 1.69 tonnes/ ha). Within the P-stressed environment, grain yield was least (0.3 tonnes/ha) in CSM63 and Feterita Gishesh and highest (2.2 tonnes/ ha) in Grinkan. Within the mutant population, grain yield under P stress was least (0.4 tonnes/ha) in Mut4313-2, Mut1455#1 and Mut1817-2 and highest (1.5 tonnes/ha) in Mut2572, Mut3108, Mut3419 and Mut2570. On the other hand, under optimum P conditions, Mut1864 produced the least grain yield (0.6 tonnes/ha) 128 University of Ghana http://ugspace.ug.edu.gh whereas the highest (3.9 tonnes/ha) was produced by Mut4366. The differences between the test genotypes and the control varieties were also significant for all traits except stem girth and biomass. Details of the analysis of variance for the morpho-agronomic traits in the two P environments are presented in Tables 5.6.7 and 5.6.8. Results of two-sampled t-test comparing the means in the two environments are summarised in Table 5.6.9. 129 University of Ghana http://ugspace.ug.edu.gh Table 5.6.7. Analysis of variance for morpho-agronomic traits in low-P stress Source of variation Df SV DTF PH SG GY SY BM HI Blocks 19 0.026ns 1.046ns 123.759ns 1.853ns 0.019ns 0.109ns 0.160ns 0.001ns Entries 199 0.286* 43.223* 1735.983* 7.501* 0.179* 1.339* 1.775* 0.007* Test genotypes (T) 197 0.207* 19.861* 273.133* 4.645* 0.068* 1.238* 1.553* 0.004* Controls (C) 1 9.801* 50.625* 82910.130* 572.292* 10.100* 13.225* 47.089* 0.081* T vs C 1 6.282* 4620.530* 209132.053* 5.370ns 11.961* 9.415* 0.181ns 0.604* Error 19 0.014 0.625 152.936 1.650 0.010 0.147 0.183 0.001 CV (%) 3.775 1.234 13.269 7.798 10.671 11.402 9.906 9.350 * = significant at p < 0.01; ns = non-significant; Df = degrees of freedom; SV= Seedling vigour; DTF = Days to flowering; PH = Plant height; SG = Stem girth; GY = Grain yield; SY = Stover yield; BM = Biomass; HI = Harvest index; CV = Coefficient of variation 130 University of Ghana http://ugspace.ug.edu.gh Table 5.6.8. Analysis of variance for morpho-agronomic traits under optimum soil P Source of variation Df SV DTF PH SG GY SY BM HI Blocks 19 0.005ns 0.374ns 156.155ns 2.731* 0.023ns 0.102ns 0.180ns 0.0003ns Entries 199 0.125* 38.081* 3040.360* 11.658* 0.523* 1.588* 3.350* 0.0031* Test genotypes (T) 197 0.106* 18.368* 602.941* 8.133* 0.362* 1.144* 2.257* 0.0028* Controls (C) 1 0.169* 84.100* 171858.990* 701.406* 28.224* 90.601* 218.089* 0.0176* T vs C 1 3.953* 3862.903* 313211.093* 16.280* 4.430* 0.020ns 3.953* 0.0426* Error 19 0.011 0.521 178.495 1.196 0.029 0.062 0.129 0.0003 CV (%) 2.292 1.170 11.440 5.740 9.710 5.284 5.540 6.1796 * = significant at p < 0.01; ns = non-significant; Df = degrees of freedom; SV= Seedling vigour; DTF = Days to flowering; PH = Plant height; SG = Stem girth; GY = Grain yield; SY = Stover yield; BM = Biomass; HI = Harvest index; CV =Coefficient of variation. 131 University of Ghana http://ugspace.ug.edu.gh Table 5.6.9. Comparison between means of traits in low P and optimum P environments Traits SV PH DTF SG GY SY BM HI Low P 3.02a 80.74a 62.18a 16.38a 0.86a 3.45a 4.31a 0.21a Optimum P 4.54b 101.31b 59.98b 18.91b 1.69b 4.71b 6.40b 0.26b LSD0.05 1.52 20.57 2.20 2.53 0.83 1.26 2.09 0.06 Means followed with different letters in a column are significantly different at p < 0.05. SV= Seedling vigour; PH= Plant height; DTF = Days to flowering; SG= Stem girth; GY= Grain yield; SY= Stover yield; BM= Biomass; HI = Harvest index; LSD= Least significant difference at p<0.05. 132 University of Ghana http://ugspace.ug.edu.gh 5.3.2.3. Genotypic variation for P-use efficiency indicators in contrasting P environments All P-use efficiency indicators varied significantly (p< 0.01) between the genotypes and also between the two P environments (Tables 5.7.1 and 5.7.2). For instance, P concentrations in the grain (PCG), in the stover (PCS), P harvest index (PHI) and P utilisation for grain production (PUTIL-G) were all significantly affected by genotype and the environment. Genotype TxARG1 demonstrated the highest PUTIL-G of 0.71 kg/g P whereas Mut4313-2 had the least (0.17 kg/g P). Stover yield, P harvest index, P content of biomass increased significantly in the adequate P environment in comparison to low P (Table 5.7.3). Lowest PCG values varied from 0.60 mg/g in Kadaga West to 0.81 mg/g in Mut001 in low and optimum P soils respectively. The highest PCG in both P environments was shown by Sariaso (1.29 mg/g and 1.57 mg/g). At low P, PHI ranged from 0.13 in Mut3612 to 0.53 in Mut3779 whereas in optimum P, the trait ranged from 0.19 in Mut4512 to 0.58 in Mut3218. Under low P stress, there were no significant differences between the test genotypes and the control. In a combined ANOVA for the two P environments, genotype x environment interactions were highly significant for seedling vigour, days to flowering, grain yield, stover yield and biomass (Table 5.7.4). However, genotype x environment interaction was not significant for plant height and stem girth in the combined ANOVA. 133 University of Ghana http://ugspace.ug.edu.gh Table 5.7.1. Analysis of variance for P-uptake and use efficiency indicators in low P stress Source of variation Df PCG PCS PG PS PBM PHI PUTIL-G PUTIL-S PUTIL-BM Blocks 19 0.002ns 0.002ns 0.030ns 0.075ns 0.122ns 0.002ns 0.003ns 0.008ns 0.014ns Entries 199 0.009* 0.005* 0.166* 0.406* 0.741* 0.010* 0.015* 0.065* 0.050* Test genotypes (T) 197 0.005* 0.005* 0.045* 0.380* 0.541* 0.006* 0.011* 0.039* 0.036* Controls (C) 1 0.778* 0.018* 14.340* 5.027* 36.214* 0.218* 0.000ns 1.576* 1.624* Tests vs Controls 1 0.000ns 0.023* 9.808* 0.8878* 4.770* 0.539* 0.858* 3.829* 1.056* Error 19 0.002 0.001 0.023 0.081 0.123 0.001 0.003 0.005 0.009 CV (%) 5.929 7.569 20.157 17.558 14.762 11.402 12.100 5.016 5.111 * = significant at p<0.01; ns= non-significant; Df = degrees of freedom; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; CV = Coefficient of variation. 134 University of Ghana http://ugspace.ug.edu.gh Table 5.7.2. Analysis of variance for P- uptake and use efficiency indicators under optimum soil P conditions Source of Df PCG PCS PG PS PBM PHI PUTIL-G PUTIL-S PUTIL-BM variation Blocks 19 0.002ns 0.001ns 0.028ns 0.123ns 0.174ns 0.001ns 0.001ns 0.001ns 0.002ns Entries 199 0.019* 0.010* 0.716* 1.028* 2.917* 0.005* 0.005* 0.027* 0.028* Test genotypes (T) 197 0.017* 0.009* 0.465* 0.698* 1.772* 0.005* 0.005* 0.023* 0.025* Controls (C) 1 0.339* 0.155* 45.518* 66.487* 221.841* 0.018* 0.004* 0.538* 0.635* T vs C 1 0.077* 0.004* 5.254* 0.573* 9.304* 0.031* 0.052* 0.144* 0.026* Error 19 0.002 0.001 0.049 0.072 0.173 0.001 0.004 0.002 0.002 CV (%) 4.460 4.151 11.410 8.643 8.268 5.650 6.018 4.550 3.366 * = significant at p<0.01; ns= non-significant; Df = degrees of freedom; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; CV = Coefficient of variation. 135 University of Ghana http://ugspace.ug.edu.gh Table 5.7.3. Comparison between means of traits in low P and optimum P environments PCG PCS PG PS PBM PHI PUTIL-G PUTIL-S PUTIL-BM Low P 0.78a 0.47a 0.67a 1.65a 2.31a 0.30a 0.39a 1.50a 1.89a Optimum P 1.11b 0.65b 1.87b 3.07b 4.94b 0.38b 0.34b 0.98b 1.32b LSD0.05 0.33 0.18 1.21 1.42 2.63 0.08 0.05 0.52 0.57 Means followed with different letters in a column are significantly different at p < 0.05. PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; LSD = Least significant difference at p<0.05. 136 University of Ghana http://ugspace.ug.edu.gh Table 5.7.4. Combined analysis of variance for morpho-agronomic traits Sources of variation Df SV PH DTF SG GY SY BM Genotype 199 0.34*** 4704.00*** 83.30*** 19.20*** 0.60*** 2.58*** 4.60*** Environment 1 269.25*** 66231.00*** 675.40*** 793.30*** 75.92** 216.88*** 54.84*** G*E 199 0.15*** 115.00ns 1.70*** 0.80ns 0.12*** 0.59*** 0.80*** Environment (Block) 1 0.11** 244.00ns 3.10* 1.90ns 0.14** 0.19ns 0.80* Error 75 0.01 152.00 0.60 1.90 0.02 0.10 0.20 ***, **, * = significant at p < 0.001; 0.01; 0.1 respectively; ns = non-significant; Df = degrees of freedom; SV= Seedling vigour; DTF = Days to flowering; PH = Plant height; SG = Stem girth; GY = Grain yield; SY = Stover yield; BM = Biomass; HI = Harvest index; CV =Coefficient of variation. Table 5.7.5. Combined analysis of variance for P- uptake and use efficiency traits * = significant at p<0.01; Df = degrees of freedom; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P Sources of variation Df HI PCG PCS PG PS PMB PHI PUTIL-G PUTIL-S PUTIL-BM Genotype 199 0.0082 0.0220 0.0130 0.7200 1.2500 3.2000 0.0122 0.0169 0.0720 0.0700 Environment 1 0.1311 12.4980 3.5970 166.7000 259.3700 841.8000 0.4564 0.4957 27.2690 35.2300 G*E 199 0.0039 0.0080 0.0030 0.1900 0.3000 0.7000 0.0042 0.0052 0.0260 0.0200 Environment (Block) 1 0.0002 0.0010 0.3930 0.0900 0.0400 0.2000 0.0015 0.0002 0.0180 0.0100 Error 75 0.0006 0.0020 0.0010 0.0300 0.0900 0.1000 0.0011 0.0017 0.0040 0.0100 content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; CV = Coefficient of variation. 137 University of Ghana http://ugspace.ug.edu.gh 5.3.2.4. Heritability and genotypic coefficient of variation within the population Under low P-stress, heritability ranged from 0.57 for P concentration in the stover to 0.97 for days to flowering (Table 5.7.6). Heritability estimates tended to be higher under optimum conditions where it ranged from 0.77 for P concentration in the grain to 0.97 for days to anthesis. No low GCV values (< 10) were observed for the traits in the low P environment. In this environment, GCV ranged from 10.2 for days to anthesis to 48.8 for P content of the grain. Under optimum conditions however, low GCV (7.4) was observed for seedling vigour. The highest GCV (45.8) was observed for plant height. Table 5.7.6. Estimates of broad-sense heritability and genotypic coefficient of variation in low P trials H2 GCV (%) Traits Low P Optimum P Low P Optimum P SV 0.91 0.84 16.86 7.35 DTF 0.97 0.97 10.19 9.94 PH 0.84 0.89 42.69 45.81 SG 0.64 0.81 14.69 16.98 GY 0.89 0.90 43.24 40.18 SY 0.80 0.92 32.45 26.21 BM 0.81 0.93 29.24 27.79 HI 0.87 0.82 35.69 19.84 PCG 0.63 0.77 10.73 11.53 PCS 0.57 0.87 12.35 14.78 PG 0.76 0.87 48.82 42.11 PS 0.67 0.87 35.21 31.58 PBM 0.72 0.89 33.13 32.91 PHI 0.76 0.82 28.76 18.01 PUTIL-G 0.72 0.85 27.38 19.11 PUTIL-S 0.85 0.87 16.96 16.32 PUTIL-BM 0.69 0.87 10.84 12.38 H2 = broad sense heritability; GCV (%) = genetic coefficient of variation 138 University of Ghana http://ugspace.ug.edu.gh 5.3.2.5. Correlation between morpho-agronomic traits and P-use efficiency indicators in contrasting soil-P environments In both P environments, grain yield was positively correlated with stover yield (r = 0.408; r = 0.562), P content in the grain (r= 0.954; r = 0.947) and P harvest index (r = 0.449; r = 0.608) [Tables 5.7.7 and 5.7.8] but exhibited a negative correlation with P utilisation for stover production (r = -0.394; r= -0.494). No significant correlation was observed between grain yield and P concentration in the grain. On the other hand, whereas the relationship between stover yield and harvest index was significant (r= -0.540) in low P soils, there was no significant relationship between the two traits in optimum P soils. This was also the case for the relationship between stover yield and P utilisation for grain yield as well as between harvest index and P concentration in the grain (Tables 5.7.7 and 5.7.8). 139 University of Ghana http://ugspace.ug.edu.gh Table 5.7.7. Correlation between morpho-agronomic traits and P-use efficiency indicators under low soil P conditions PUTIL PUTIL PUTIL Traits GY SY BM HI PCG PCS PG PS PBM PHI -G -S -BM GY 1 SY 0.408 1 BM 0.576 0.981 1 HI 0.507 -0.54 -0.378 1 PCG -0.025 0.173 0.149 -0.193 1 PCS 0.026 0.121 0.114 -0.131 0.149 1 PG 0.954 0.464 0.616 0.411 0.259 0.066 1 PS 0.379 0.908 0.893 -0.501 0.201 0.503 0.443 1 PBM 0.596 0.894 0.926 -0.297 0.244 0.439 0.663 0.965 1 PHI 0.449 -0.499 -0.352 0.911 0.024 -0.437 0.420 -0.587 -0.367 1 PUTIL-G 0.432 -0.534 -0.387 0.937 -0.307 -0.448 0.309 -0.619 -0.426 0.939 1 PUTIL-S -0.394 0.311 0.196 -0.639 -0.139 -0.638 -0.402 -0.006 -0.123 -0.397 -0.343 1 PUTIL-BM -0.169 0.026 -0.012 -0.142 -0.317 -0.917 -0.246 -0.352 -0.366 0.111 0.201 0.851 1 GY= Grain yield; SY= Stover yield; BM= Biomass; HI = Harvest index; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; Values in bold are significant at p < 0.05. 140 University of Ghana http://ugspace.ug.edu.gh Table 5.7.8. Relationship between yield components and P-use efficiency indicators under optimum soil P conditions PUTIL- PUTIL- PUTIL- Traits GY SY BM HI PCG PCS PG PS PMB PHI G S BM GY 1 SY 0.562 1 BM 0.803 0.944 1 HI 0.732 -0.119 0.206 1 PCG -0.038 -0.131 -0.110 0.075 1 PCS 0.059 0.029 0.045 0.023 0.442 1 PG 0.947 0.489 0.730 0.735 0.264 0.190 1 PS 0.498 0.839 0.803 -0.083 0.103 0.552 0.503 1 PMB 0.803 0.786 0.887 0.324 0.201 0.450 0.834 0.897 1 PHI 0.608 -0.165 0.123 0.903 0.215 -0.271 0.658 -0.282 0.156 1 PUTIL-G 0.595 -0.091 0.171 0.815 -0.353 -0.508 0.467 -0.335 0.025 0.830 1 PUTIL-S -0.494 0.089 -0.133 -0.676 -0.567 -0.685 -0.650 -0.287 -0.516 -0.494 -0.145 1 PUTIL-BM -0.211 0.045 -0.051 -0.287 -0.703 -0.886 -0.418 -0.426 -0.486 -0.106 0.306 0.898 1 GY= Grain yield; SY= Stover yield; BM= Biomass; HI = Harvest index; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; Values in bold are significant at p < 0.05. 141 University of Ghana http://ugspace.ug.edu.gh 5.3.2.6. Differences between mutants and cultivated varieties for morpho-agronomic traits A comparison between the top-ranked mutant lines and the top-ranked cultivated genotypes (in terms of P utilisation for grain production) revealed differences for several P-use efficiency traits. Stover yield was significantly (p < 0.0l) lower in the mutants compared to the cultivated types. On the other hand, in terms of P utilisation for grain production and P harvest index, the mutants exhibited significantly (p < 0.0l) higher mean values compared to the cultivated varieties. In general, higher values for P-uptake traits were observed among the cultivated genotypes whereas the mutants had higher values for utilisation traits. Within the mutant population, Mut2572 showed the highest stover yield and utilisation of P for grain yield production. In terms of P concentration in the grain, the top-ranked mutant was Mut1771-2. No significant differences were observed between the two groups for grain yield as well as P concentration in the grain and stover. The differences between the two groups for the respective traits are shown in Figure 5.7. The highest- ranking mutants and cultivated types for the various categories are summarized in Table 5.7.9. 142 University of Ghana http://ugspace.ug.edu.gh Figure 5.6. Comparison between highest ranking mutants and cultivated genotypes for P- use efficiency traits in low P environment. GY= Grain yield; SY= Stover yield; BM= Biomass; HI = Harvest index; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; Where indicated, the two groups are significantly different at at 0.5 (*) and 0.01 (**) probability levels. 143 University of Ghana http://ugspace.ug.edu.gh Table 5.7.9. Highest-ranking mutants and cultivated genotypes for yield and PUE indicators Traits Mutants Cultivated genotypes Overall Yield GY Mut2572 Grinkan Grinkan SY Mut0538 MDK MDK BM Mut0538 MDK MDK P-uptake PCG Mut1771-2 Sariaso Sariaso PCS Mut4264 Korokollo Korokollo PG Mut3108 Grinkan Grinkan PS Mut3434 ICSV1049 ICSV1049 PBM Mut3434 ICSV1049 ICSV1049 P-utilisation PHI Mut2572 TxARG1 Mut2572 PUTIL-G Mut2572 TxARG1 TxARG1 PUTIL-S Mut3708 SC35 Mut3708 PUTIL-BM Mut3708 SC35 Mut3708 P-stress tolerance SVR Mut0317 SC599 Mut0317 DFR Mut0056 Sariaso Mut0056 PHR Mut2856#2 SC35 Mut2856#2 SYR Mut4107#1 Framida Mut4107#1 GY= Grain yield; SY= Stover yield; BM= Biomass; HI = Harvest index; PCG= P concentration in the grain; PCS= P concentration in the stover; PG= P content of grain; PS= P content of stover; PBM = P content of biomass; PHI= P harvest index; PUTIL-G = P utilisation for grain production; PUTIL-S = P utilisation for stover production; PUTIL-BM = P utilisation for biomass production; SVR= Seedling vigour ratio; DFR= Days to flowering ratio; PHR = Plant height ratio; SYR= Stover yield ratio. 144 University of Ghana http://ugspace.ug.edu.gh 5.3.2.7. Selection of low soil P-stress tolerant and sensitive genotypes Under optimum P conditions, the genotypes with the highest grain yield were MDK, MR732, Mut1253 and KuyumaWSV387 with average grain yield of 3.7, 3.4, 3.5 and 2.8 tonnes/ha respectively. In the low-P environment, the genotypes with the highest grain yield were Grinkan, ICSV1049, MDK and Mut2572 with 1.9, 1.8, 1.6 and 1.5 tonnes/ha respectively (Table 5.8.1). However, identification of P-stress tolerant and sensitive genotypes were based on the relative performance of the genotypes in the contrasting P environments (grain yield ratios - GYR). Based on this, the five highest performing varieties were Theis, El-Mota, Grinkan, Mota Maradi and Framida whereas for the mutant population, Mut3412, Mut4017, Mut0460#1, Mut4156 and Mut4218 were the best-performing (Table 5.). In general the mutants showed higher tolerance to low P stress in terms of GYR; GYR for the best performing varieties ranged from 0.65 to 0.74 whereas among the mutants the ratio ranged from 0.82 to 0.93. Least performing varieties included Seguetana, Tx430, PI609567, CSM63 and Sepon82. Among the mutants, most-sensitive types werMut3003#1, Mut3008, Mut1455#1, Mut5115 and Mut4313-2. GYR for the most-sensitive varieties were higher (0.34 to 0.38) compared to GYR for most-sensitive mutants (0.24 to 0.28). 145 University of Ghana http://ugspace.ug.edu.gh Table 5.8.1. Top-ranked genotypes (grain yield) in low-P soil and their performance in optimum-P soil Genotype GY Rank GY Rank GYR Rank Low P Low P Optimum P Optimum P (PUE) (PUE) Grinkan 1.88 1 2.74 13 0.69 33 ICSV1049 1.75 2 3.92 1 0.45 144 MDK 1.57 3 3.70 2 0.42 153 Mut2572 1.52 4 2.53 19 0.60 70 Mut3108 1.47 5 2.30 35 0.64 50 Mut3419 1.46 6 2.06 52 0.71 27 Mut2570 1.45 7 2.59 18 0.56 81 MR732 1.42 8 3.40 3 0.42 155 Mut2616 1.40 9 1.80 77 0.78 12 Mut0988 1.40 10 2.40 24 0.58 74 Grand mean 0.86 1.69 GY = Grain yield; GYR = Grain yield ratio; PUE = P-use efficiency. 146 University of Ghana http://ugspace.ug.edu.gh Table 5.8.2. Low-P stress tolerant and sensitive genotypes within the population Tolerant genotypes Sensitive genotypes Selection Selection Rank Genotype GYR differential Genotype GYR differential 1 Mut3412 0.93 0.40 Mut3003#1 0.24 -0.29 2 Kadaga West 0.88 0.35 Mut3008 0.25 -0.28 3 Mut4017 0.85 0.32 Mut1455#1 0.26 -0.27 4 Mut0460#1 0.83 0.30 Mut5115 0.26 -0.27 5 Mut4156 0.83 0.30 Mut4313-2 0.28 -0.25 6 Mut4218 0.82 0.29 Mut2954#1 0.29 -0.24 7 Mut2850#1 0.81 0.28 Mut2857#1 0.29 -0.24 8 Mut0001 0.80 0.27 Mut1817-2 0.30 -0.23 9 Mut4512 0.79 0.26 Mut1864 0.30 -0.23 10 Mut3160 0.78 0.25 Mut3813 0.30 -0.23 11 Mut3663 0.78 0.25 Mut3760#2 0.32 -0.21 12 Mut2616 0.78 0.25 Mut2205#2 0.32 -0.21 13 Mut3864 0.78 0.25 Mut1769 0.33 -0.20 14 Mut4060#1 0.77 0.24 Mut3659 0.34 -0.19 15 Mut3057 0.76 0.23 Mut2950#1 0.34 -0.19 16 Theis 0.74 0.21 Seguetana 0.34 -0.19 17 Mut3112 0.74 0.21 Tx430 0.35 -0.18 18 Mut4511 0.74 0.21 Mut3217 0.35 -0.18 19 Mut4308 0.73 0.20 Mut1253 0.36 -0.17 20 Mut3105 0.73 0.20 PI609567 0.36 -0.17 GYR= Grain yield ratio. 147 University of Ghana http://ugspace.ug.edu.gh 5.4 Discussion Average available-P levels of the low-P environment in this study was 4.6 and 5.7 mg P /kg for the pot and field experiments respectively. According to Doumbia et al. (1993), plant-available P in sorghum-cultivating regions of West Africa is generally below 10 mg P /kg of soil. Thus, the P- deficient soils used in this study were fairly representative of the low-input systems of smallholder sorghum farmers in the sub-region. The available P content of the P-deficient soils were also below the critical P threshold for optimum sorghum growth (7 mg P/kg) whereas in the optimum P soils, available P content of 18.7 and 19.5 mg P/ kg indicate adequate supply of P (Manu et al., 1991; Doumbia et al., 2003. Leiser et al. (2012) report similar values in studies aimed at selecting sorghum genotypes in contrasting P environments. The large decreases in plant height, grain yield and stover yield in the low-P environments compared to optimum P conditions indicate that the two P environments used in the study were sufficiently different for levels of available P (Rose and Wissuwa, 2012). In sorghum, as well as in other crops, such decreases in plant height, grain and stover yield, delays in flowering and poor seedling establishment in P-deficient soils compared to optimum P soils have been reported (Chen et al., 2008; Parentoni et al., 2010; Leiser et al., 2012). According to Leiser et al. (2012), selection based on the differences in these traits in contrasting P environments is appropriate for identifying P-use efficient genotypes. The results also indicated considerable genotypic variation for P-use efficiency was established within the mutagenised population. The higher genotypic variation (GCV) values observed for P- uptake traits compared to P utilisation traits has also been reported by Wissuwa et al. (1998). In general, the genotypic variation for PUE was higher under low P conditions compared to optimum conditions. According to Wissuwa and Ae (2001), this enables better selections under low P 148 University of Ghana http://ugspace.ug.edu.gh conditions. Relative grain yield under the two P environments was quite high for several genotypes including Mut3412, Kadaga West and Mut4017 which suggests sufficiently good adaptation to soil P deficiency. The performance of Kadaga West might be attributed to years of cultivation and selection under low-input conditions (Buah et al., 2012). On the other hand, significant increases in grain and biomass yield between the two environments indicates that most of the genotypes were quite responsive to P fertilisation. For example, genotypes such as Grinkan, ICSV1049, Mut2572, and Mut3108, although not categorised among the most tolerant genotypes, were the highest-ranking in terms of grain yield under optimum P conditions. This reflects the impressive response of these genotypes to adequate nutrient supply. The observation confirms that adequate amounts of the nutrient is required for maximum growth and productivity of sorghum which in turn corroborates opinions that P-use efficient genotypes must be used to complement fertiliser application (Gemenet et al., 2016). On the other hand, several inefficient and sensitive genotypes were identified within the population which showed large losses in grain yield under no P fertilisation with yields ranging from 0.3 to 2.2 tonnes/ha. Such yield levels are within the range reported for smallholders in Ghana and West Africa as a whole and reflects the low grain yield recorded in farmers’ fields across Ghana and many parts of West Africa (Atokple, 2010; FA0, 2010). A comparison between the top-ranked mutant lines and the top-ranked cultivated genotypes revealed higher P-uptake efficiency among the cultivated genotypes whereas the mutagenised lines were generally better at utilising P for grain production. According to Manske et al. (2002) and Rose and Wissuwa (2012), P-utilisation efficiency is likely to increase with harvest index, an observation which has been reported in cereals such as wheat and rice. Thus, the higher harvest index of the mutant genotypes might account for increased P-utilisation efficiency within the 149 University of Ghana http://ugspace.ug.edu.gh mutagenised population. On the other hand, the more efficient P acquisition among the cultivated varieties might be due to decades of selections by farmers (Leiser et al., 2012). Higher P-uptake has been linked with improved seedling establishment in low P soils; this might also account for the slightly higher seedling vigour scores observed among the cultivated genotypes (White and Veneklaas, 2012). The use of relative performance as a strategy to select for tolerance to P-deficient conditions has been shown by Leiser et al. (2012b). This is based on the fact that P deficiency causes marked decreases and delays in sorghum growth and development (Sahrawat et al., 1995). For instance, P availability at sowing affects early plant development or seedling vigour (Grant et al., 2002). It is likely that a higher grain P content of the grain in the cultivated varieties compared to the mutant lines contributed to the high seedling vigour ratios of the cultivated types. Unfortunately, selecting for increased grain P content although beneficial for improved plant establishment has been implicated in mining large amounts of soil P (Rose and Wissuwa, 2012; White and Veneklaas, 2012). In this study, high grain yield ratios (GYR) were used as a selection criterion to identify genotypes with better tolerance to low P-stress. This form of selection can increase specific adaptation to soil P deficiency as only genotypes that perform relatively well in both environments are selected as tolerant (Lesier et al., 2015). However, Blum (2005) suggests that GYR as a selection criterion can lead to lower yield potential. Similar observations were made in this study. The cultivation of PUE genotypes may be less advantageous if the genotypes are inherently low yielding. Thus, it might be important for breeders to select for both high grain yield and PUE. In this case, Grinkan might be more beneficial on farmers’ fields whereas the P-use efficient Mut3412 and Kadaga might be more useful as PUE parents in breeding programmes. Selection for earlier days to 150 University of Ghana http://ugspace.ug.edu.gh flowering in low-input systems may be useful in production systems where drought and soil P deficiency are major constraints. Although delays in flowering time under inadequate nutrient conditions might be inevitable, genotypes that flower relatively earlier might reduce risks associated with early cessation of rain (Dingkuhn et al., 2006). The positive relationship between P content of the grain, P harvest index and P utilisation for grain production in the low P field suggest that simultaneous selections for these traits is possible (Leiser et al., 2014). A very strong correlation was also observed between harvest index and P utilisation for grain production, a phenomenon which has been reported by Rose and Wissuwa (2012). Negative correlations between P concentration in the stover (PCS) and grain yield have also been reported by Lesier et al., (2014) which suggest that high concentrations of P in the stover did not result in a corresponding gain in grain yield. 5.6 Conclusions Phosphorus-uptake and P-utilisation efficiencies was influenced by genotype, P environment and their interaction. Although inadequate supply of P significantly reduced grain and stover yields in all the genotypes, considerable variation for yield still exists among the 200 genotypes under low soil P conditions. Grinkan and MDK were the highest producers of grain and stover. On the whole, P-uptake efficiency was higher among the cultivated genotypes than in the mutant varieties. On the other hand, the mutagenised lines showed better utilisation of P. The top-ranked putative mutants with regards to P-uptake were Mut1771-2, Mut4264, Mut3108 and Mut3434 whereas among the cultivated varieties, Sariaso, Korokollo, Grinkan and ICSV1049 were the most efficient at P-acquisition. In general, Mut2572, TxARG1 and Mut3708 were the most-efficient genotypes 151 University of Ghana http://ugspace.ug.edu.gh in terms of utilisation. The mutagenised lines also showed higher tolerance to low-P stress (relative grain yield) compared to the cultivated types. Overall, the most tolerant genotypes were Mut3412, Kadaga West, Mut4017, and Mut0460#1. Mut3003#1, Mut3008, Mut1455 and Mut5115 were the most sensitive types. New sources of P-use efficiency were identified in the mutagenised population. There is potential to exploit the genotypic variability observed within the population for further breeding work either through advancement of the mutagenised lines or as parents in recombinant breeding programmemes. 152 University of Ghana http://ugspace.ug.edu.gh CHAPTER SIX 6.0 GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS Sorghum is an important crop both as a source of food and income among smallholder farmers in the Talensi-Nabdam and Binduri districts of Ghana. However, its production is faced with numerous constraints including drought, limited tractor services for land preparation and declining soil fertility. Strategies to improve productivity of the crop on marginal soils such as the application of fertilisers are largely unexploited due to high cost of mineral fertilisers on rural markets, untimely access to the commodity and limited availability of organic alternatives. On the other hand, farmers’ perceptions of the low fertiliser requirements of the crop as well as uncertainties of maximum returns on fertiliser application also influenced their attitudes towards fertiliser application. Besides the use of mineral fertilisers, farmers employed other soil management strategies such as erosion control and the retention of crop residues. Among the farmers, key indicators of soil fertility were soil colour and crop performance with most farmers describing their soils as moderately fertile. Farmers preferred several traits including drought tolerance, high grain yield, earliness as well as varieties with lower fertiliser requirements. Farmers’ constraints, perceptions and preferences may differ with location. Thus, a more extensive survey is required in other sorghum-growing regions in Ghana (Upper West and Northern Regions). This may reveal more insight into farmers’ constraints and preferences which in turn will inform stakeholders and help prioritise breeding goals. The question also remains on whether on a wider scale, the behaviour of sorghum farmers’ in northern Ghana, in terms of fertiliser input and cropping practices are influenced by their perceptions, adequate information on soil fertility or other processes. Studies to identify any linkage between farmers’ perceptions, management practices and actual soil fertility are necessary to address soil fertility problems. In the meantime, access to 153 University of Ghana http://ugspace.ug.edu.gh fertilisers and credit facilities must be improved to boost fertiliser use among smallholders. Better agricultural extension services are needed to educate farmers on the benefits of fertiliser-use and other soil fertility management methods. In this study, induced mutagenesis was used as an efficient strategy for expanding genetic resources in sorghum. Several potentially-useful mutants were derived from the mutagenic treatment of BTx623. These include putative mutants for early flowering, late flowering, early senescence, brown mid-rib, multiple tillering, erect leaf architecture and bloomless characteristics. Characterisation of the putative mutants at more advanced generations and in diverse environments is required to eliminate any environmentally-influenced phenotypes and confirm the stability of the phenotypes observed at M3. Upon confirmation, any desirable genotypes can be exploited for sorghum improvement either through selection or recombinant breeding. Further characterisation of the mutagenised population under different biotic and abiotic stress conditions may also be beneficial for identifying other useful traits which were not immediately visible at M3. Given that the crop has been completely sequenced, it is also possible to isolate or confirm genes controlling identified traits in further genomic studies. Further characterisation of the mutagenised population was carried out at M6 as a strategy to identify and subsequently select P-use efficient genotypes. Significant variability was observed within the population for all measured traits in both low-P deficient and optimum-P environments. This suggests a fairly wide scope for further selections for P-use efficiency within this population. The P-use efficient genotypes identified in this study may be used to improve farmers’ yields on marginal soils. However, multi-environment testing is necessary to fully confirm levels of tolerance to low-P stress as the data reported here (although from both pot and field environments) were obtained from single year trials. Following confirmation, any P-use efficient genotypes may 154 University of Ghana http://ugspace.ug.edu.gh be advanced and used directly as mutant varieties or as parents in recombinant breeding programmes. 155 University of Ghana http://ugspace.ug.edu.gh REFERENCES Abane, J. A. (2015). The challenges of Millennium Development Goal 1 in Bongo District of the Upper East Region of Ghana. Global Social Welfare, 2 (1): 332–340. Abdulai, A.L.I, Parzies, H., Kouressy, M., Vaksmann, M., Asch, F. and Brueck, H. (2012). Yield Stability of Photoperiod Sensitive Sorghum (Sorghum bicolor L. Moench) Accessions under Diverse Climatic Environments. International Journal of Agricultural Research, 7: 17-32. Acquaah G. (2015). Conventional Plant Breeding Principles and Techniques. In: Al-Khayri J., Jain S., Johnson D. (eds) Advances in Plant Breeding Strategies: Breeding, Biotechnology and Molecular Tools. Springer, Cham. Addai, I.K. and Alimiyawo, M. (2015). Graphical Determination of Leaf Area Index and its Relationship with Growth and Yield Parameters of Sorghum (Sorghum bicolor L. Moench) as Affected by Fertilizer Application. Journal of Agronomy, 14: 272-278. Agropedia (2009). Phosphorus nutrition of sorghum. http://agropedia.iitk.ac.in/content/phosphorus-nutrition-sorghum. Accessed 12/11/2017. Ahloowahlia, B.S., Maluszynski, M. and Nichterlein, K. (2004). Global impact of mutation- derived varieties. Euphytica, 135: 187-204. Ai. P. H., Sun, S. B., Zhao, J. N., Fan, X. R., Xin, W. J., Guo, Q., et al. (2009). Two rice phosphate transporters, OsPht1; 2 and OsPht1; 6, have different functions and kinetic properties in uptake and translocation. Plant Journal, 57: 798–809. Akinbile, L.A. (2003). Technology dissemination, agricultural productivity and poverty reduction in the rural sector of Nigeria. In: F Okunmadewa (Ed.): Poverty Reduction and the Nigeria Agricultural Sector, 58-78. Alagarswamy, G. and Ritchie J.T. (2001). Phasic development in CERES Sorghum model. T. Hodges (Ed.), Predicting Crop Phenology, CRC Press, Boca Raton, FL: 143-152 Al-Hassan, R. and Jatoe, J.B.D. (2002). “Adoption and Impact of Improved Cereal Varieties in Ghana”. Workshop on the Green Revolution in Asia and its Transferability to Africa. Organized by Foundation for Advanced Studies in International Development (FASID), from 8th – 10th December 2002. Tokyo, Japan Alliance for a Green Revolution in Africa [AGRA] (2012). Annual report of 2011. Nairobi, Kenya. Angelucci, F. (2013). Analysis of incentives and disincentives for sorghum in Ghana. Technical notes series, MAFAP, FAO, Rome. 156 University of Ghana http://ugspace.ug.edu.gh AQUASTAT (2005). Irrigation in Africa in figures, AQUASTAT 2005 survey. Available at: http://www.fao.org/nr/water/aquastat/countries_regions/gha/GHA-CP_eng.pdf. Arumuganathan, K., and Earle, E.D. (1991). Estimation of nuclear DNA content of plants by flow cytometry. Plant Molecular Biology Reporter, 9: 229–241. Assefa, Y. and Staggenborg, S. (2011) Phenotypic changes in grain sorghum over the last five decades. Journal of Agronomy and Crop Science, 197: 249–257 Atokple, I.D.K. (2010). Sorghum and Millet Breeding in West Africa in Practice. http://www.afripro.org.uk/papers/Paper14Atokple.pdf. Accessed on 30th November, 2017 Atokple, IDK. (2012). Hybrid sorghum production. Presentation at the Bioscience for farming in Africa, Journalist training course, Accra, Ghana, September 19 – 22, 2012. Axtell, J. D., Van Scoyoc, S. W., Christensen, P. J. and Ejeta, G. (1979). In: Seed protein improvement in cereals and grain legumes (Vienna: IAEA), 2: 35 Baba, S. U. (2017). Farmers’ perceptions of Soil Degradation in Rural Kano, Northern Nigeria, Resources and Environment, 7: 8-12. Bationo, A. and Mokwunye, A.U. (1991). Role of manures and crop residue in alleviating soil fertility constraints to crop production: With special reference to the Sahelian and Sudanian zones of West Africa. Fertilizer Research, 29: 117. Bationo, A., Mokwunye U., Vlek, P.L.G., Koala, S. and Shapiro, B.I. (2003). Soil fertility management for sustainable land use in the West African Sudano-Sahelian Zone. In: Soil fertility management in Africa: A regional perspective. Academic Science Publishers. African Academy of Sciences, Nairobi, Kenya. Bationo, A., Waswa, B., Kihara, J. and Kimetu, J. (2006). Advances in integrated soil fertility management in sub-Saharan Africa: Challenges and opportunities. Nutrient Cycling in Agroecosystems, 76. Bationo, A., Waswa, B., Abdou, A., Bado, B.V., Bonzi, M., Iwuafor, F., et al., (2012). Overview of Long term experiments in Africa. In: Bationo, Lessons Learned from Long Term Soil Fertility Management Experiments in Africa. Springer, Dordrecht, 1–26. Bationo, A., Hartemink, A., Lungu, O., Naimi, M., Okoth, P., Smaling, E. et al. (2012). Knowing the African soils to improve fertilizer recommendations. In: Kihara, J. et al. (eds.). Improving Soil Fertility Recommendations in Africa using the Decision Support System for Agrotechnology Transfer (DSSAT). Springer Dordrecht, 19-42 Bänziger, M., Bétran, F.J. and Lafitte, H.R. (1997). Efficiency of high-nitrogen environment for improving maize for low-nitrogen environment. Crop Science, 37:1103-1109 157 University of Ghana http://ugspace.ug.edu.gh Banziger, M. and Cooper, M. (2001). Breeding for low-input conditions and consequences for participatory plant breeding: examples from tropical maize and wheat. Euphytica, 122: 503-519 Bayala, J., Bazié, H.R., Sanou, J. (2013). Competition and facilitation-related factors impacts on crop performance in an agroforestry parkland systems in Burkina Faso. African Journal of Agricultural Research 8 (43): 5303-5310. Bentley, A., MacLennan, B., Calvo, J. and Dearolf, C.R. (2000). Targeted recovery of mutations in Drosophila. Genetics 156:1169–1173. Berenji, J. and Dahlberg, J. (2004). Perspectives of sorghum in Europe. Journal of agronomy and crop science, 190 (5): 332 -338. Benard, R., Dulle, F. and Ngalapa, H. (2014). Assessment of information needs of rice farmers in Tanzania; A case study of Kilombero District, Morogoro. Philosophy and Practice, 1071. http://unllib.unl.edu//LPP (Accessed 18 September 2015). Bhargava, B. S.and Raghupathi, H. B. (1984). Analysis of plant materials for macro and micro- nutrients. In: H.L.S. Tandon (ed.), Methods of Analysis of Soils, Plants, Waters and Fertilizers. Fertilizer Development and Consultation Organization, 49-82. Bhattacharya, A., Rice, N., Shapter, F. M., Norton, S. L. and Henry, R.J. (2011). Sorghum and its wild crop relatives. In Kole, C. (ed.) Wild Crop Relatives: Genomic and breeding. Resources Cereals, Springer, Heidelberg. 397-406. Blum, A. (2005). Drought resistance, water-use efficiency, and yield potential—are they compatible, dissonant, or mutually exclusive? Australian Journal of Agricultural Research, 56:1159–1168. Bolland, M. D. A. and Baker, M. J. (1988). High phosphorus concentrations in seed of wheat and annual medic are related to higher rates of dry matter production of seedlings and plants. Australian Journal of Experimental Agriculture, 28: 765–770 Borrell, A. K., Hammer, G. L.and Douglas, A. C. (2000). Does maintaining green leaf area in sorghum improve yield under drought? Leaf growth and senescence. Crop Science, 40:1026–1037. Bouyoucos, G.H. (1951) A Recalibration of the Hydrometer for Making Mechanical Analysis of Soils. Agronomy Journal, 43: 434 - 438. Bowers, J. E., Arias, M.A., Asher, R., Avise, J.A., Ball, R.T., Brewer, G.A. et al. (2005). Comparative physical mapping links conservation of microsynteny to chromosome structure and recombination in grasses. Proceedings of the National Academy of Sciences,USA 102(37):13206–13211. 158 University of Ghana http://ugspace.ug.edu.gh Braimoh, K. and Vlek, P.L.G. (2004). The impact of land-cover change on soil properties in northern Ghana. Land degradation and Development, 15 (1): 65-74. Brammer, H. (1962). Soils: In Agriculture and Land use in Ghana. J.B Wills (ed.), Oxford University Press London, Accra, New York. Pp 88-126. Brar, D. S. and Jain, S.M. (1998). Somaclonal variation, mechanisms and applications in crop improvement. In: Somaclonal variation and induced mutations in crop improvement. Kluwer, Dordrecht. pp 15-38. Bray, R. and Kurtz, L. T. (1945). Determination of total, organic, and available forms of phosphorus in soils. Soil Science, 59: 39–46. Breseghello, F. and Coelho A. S. G. (2013). Traditional and modern plant breeding methods with examples in rice (Oryza sativa L.). Journal of Agricultural and Food Chemistry, 61: 8277– 8286. Broadley, M.R., White, P.J., Hammond, J.P., Zelko, I., Lux, A. (2007). Zinc in plants. New Phytologist, 173: 677–702. Brundrett, M.C. (2009). Mycorrhizal associations and other means of nutrition of vascular plants: understanding the global diversity of host plants by resolving conflicting information and developing reliable means of diagnosis. Plant Soil, 320: 37–77. Buah, S.S.J., Kombiok, J.M. and Abatania, L.N. (2012). Grain sorghum response to NPK fertilizer in the Guinea savanna of Ghana. Journal of Crop Improvement, 26:101–115. Buerkert, A., Bationo, A. and Piepho, H. P. (2001). Efficient phosphorus application strategies for increased crop production in sub-Saharan West Africa. Field Crops Research, 72: 1–15. Bumb, B.L., Johnson, M.E., Fuentes, P.A. (2011). IFPRI; Washington DC. Policy Options for Improving Regional Fertilizer Markets in West Africa. IFPRI Discussion Paper 01084: 73. Buresh, R.J. and Smithson, P.C. (1997). Building soil phosphorus capital in Africa. In: Buresh at al. (Eds). Replenishing Soil Fertility in Africa. SSA Special Publication, 51. Madison, Wisconsin, USA. Burow, G., Franks, C.D., Acosta-Martinez, V. and Xin, Z. (2008). Molecular mapping and characterization of BLMC, a locus for profuse wax (bloom) and enhanced cuticular features of Sorghum (Sorghum bicolor (L.) Moench.). Theoretical and applied genetics. 118. 423-31. Burow, G., Xin, Z., Hayes, C., Burke, J. (2014). Characterization of a multi-seeded mutant of Sorghum for increasing grain yield. Crop Science 54: 2030–2037. 159 University of Ghana http://ugspace.ug.edu.gh Burton, G.W. (1952). Quantitative inheritance in grasses. Proceedings of the 6th International Grasslands Congress Journal, 1:227 – 283. Butchee, K., Arnall, D.B, Sutradhar, A., Godsey C., Zhang, H., and Penn, C. (2012). “Determining Critical Soil pH for Grain Sorghum Production,” International Journal of Agronomy, 6:1 -6. Cai, H. G., Mi, G. H., Chen, F. J., Zhang, X. Z., Gao, Q. (2010). Genotypic variation of leaf SPAD value, nitrogen and nitrate content in maize. Acta Metallurgica Sinica, 16 (4): 866 - 873. Caldwell, D.G., McCallum, N., Shaw, P., Muehlbauer, G.J., Marshall, D.F. and Waugh, R. A (2004). Structured mutant population for forward and reverse genetics in barley (Hordeum vulgare L.). The Plant Journal, 40:143-150. Camargo, C.E. De O., Felicio, J.C., Tulmann Neto, A. Ferreira Filho, A.W.P., Pettinelli Jr., A. and Castro, J.L. (1995). Wheat breeding: XXVIII. New genotypes obtained by selections in interspecific segregate population submitted to gamma irradiation. Bragantia, 54: 51- 65. Chambers, R. (1992c). Rural appraisal: Rapid, relaxed and participatory. IDS Discussion Paper 311. Brighton: IDS. Chanterau, J., Trouche G., Rami, J. F., Deu M., Barro, C. Grivet, L. (2001). RFLP mapping of QTLs for photoperiod response in tropical sorghum. Euphytica, 120:183–194. Chen, J., L. Xu, Y. Cai, and Xu. J. (2008). QTL mapping of phosphorus efficiency and relative biologic characteristics in maize (Zea mays L.) at two sites. Plant Soil, 313:251–266. Chen, D., Wang, S., Xiong, B., Cao, B., & Deng, X. (2015). Carbon/Nitrogen Imbalance Associated with Drought-Induced Leaf Senescence in Sorghum bicolor. PLoS One, 10 (8). Chianu, J.N., Chianu, J.N., Mairura, F. (2012). Mineral fertilizers in the farming systems of sub- Saharan Africa. A review. Agronomy for Sustainable Development, 32:545–566. Christiaensen, L., and Demery, L., and Kuhl, J. (2010). The (Evolving) Role of Agriculture in Poverty Reduction: An Empirical Perspective, WIDER Working Paper 2010/36. Chung, K.-T., Wong, T.-Y., Huang, Y.-W., Lin, Y. (1998). Tannins and human health: a review. Critical Review of Food Science and Nutrition, 38: 421–464. Ciarelli, D.M., Furlani, A.M.C., Dechen, A.R. and Lima, M. (1998). Genetic variation among maize genotypes for phosphorus-uptake and phosphorus-use efficiency in nutrient solution. Journal of Plant Nutrition, 21: 2219-2229. 160 University of Ghana http://ugspace.ug.edu.gh Cisse, N. and Ejeta, G. (2003). Genetic variation and relationship among seedling vigor traits in sorghum. Crop Science, 43: 824-828. Clayton, W.D., and Renvoize, S.A. (1986). Genera Graminum grasses of the world, Kew Bulletin Additional Series XIII, Royal Botanic. 338-345. Condron, L.M, Turner, B.L., Cade-Menun, B.J. (2005). Chemistry and dynamics of soil organic phosphorus. In: Sims, J.T., Sharpley, A.N. (Eds.), Phosphorus: agriculture and the environment. ASA/CSSA/SSSA, Madison, Wisconsin. United States of America, 87-121. Cordell, D., Drangert, J.O and White, S. (2009). The story of phosphorus: global food security and food for thought. Global Environmental Change, 19: 292-305 Cordell, D. and White, S. (2015). Tracking phosphorus security: indicators of phosphorus vulnerability in the global food system. Food Security, Springer, 7(2): 337-350. Crawford, N.M and Forde, B.G. (2002). Molecular and developmental biology of inorganic nitrogen nutrition. The Arabidopsis Book: American society of plant biologists, 1. Curt, M .D. Fernandez, J. and Martinez, M. (1995). Productivity and water use efficiency of sweet sorghum (Sorghum bicolor (L.) Moench) cv. "Keller"; in relation to water regime. Biomass and Bioenergy, 8 (6): 401- 409. Daba, S. and Zewedie, E. (2001). Evaluation of the residual value of phosphorus fertilizer for sorghum (Sorghum bicolor L.) grown on a vertisol. Bodenkultur, 52:175-181. Dahlberg, J., Berenji, J., Sikora, V. and Latkovic, D. (2011). Assessing sorghum (Sorghum bicolor (L) Moench) germplasm for new traits: food, fuels and unique uses. Maydica, 56 (1750): 85-92. Deenanath, E. D., Iyuke, S., and Rumbold, K. (2012). The Bioethanol Industry in Sub-Saharan Africa: History, Challenges, and Prospects. Journal of Biomedicine and Biotechnology, 416-491. De Lacerda, C.F., Cambraia, J., Oliva M.A., Ruiz, H.A. (2005). Changes in growth and in solute concentrations in sorghum leaves and roots during salt stress recovery. Environmental and Experimental Botany, 54: 69-76. Dhillon, M. K., Sharma, H. C., Ram, S., Naresh, J. S. (2005). Mechanisms of resistance to shoot fly, Atherigona soccata in sorghum. Euphytica 144: 301–312. Dillon, S.L., Frances, M., Shapter, R. J. H., Cordeiro, G., Izquierdo, L., Slade Lee, L. (2007). Domestication to Crop Improvement: Genetic Resources for Sorghum and Saccharum (Andropogoneae). Annals of Botany, 100, (5): 975–989. Diao, X., Cossar, F., Houssou, N., Kolavalli.S. (2014). Mechanization in Ghana: Emerging demand and the search for alternative supply models. Food Policy, 48: 168–181. 161 University of Ghana http://ugspace.ug.edu.gh de Wet, J.M.J. and Huckabay, J.P. (1967). Origin of Sorghum bicolor. II. Distribution and Domestication. Evolution, 21: 787–802. de Wet, J.M.J., (1978). Systematics and evolution of Sorghum sect. bicolor (Gramineae). American Journal of Botany, 65: 477–484. de Wet, J., and Harlan, J. (1970). Apomixis, Polyploidy, and Speciation in Dichanthium. Evolution, 24(2): 270-277. Diehl, L. (1992). Smallholder Farming Systems in the Northern Region. In: Acquaye DK (ed.). Improving Farming Systems in the Interior Savannah Zone of Ghana. Ghana: Nyankpala Agricultural Research Report (8) 2-13. Devau, N., Le Cadre, E., Hinsinger, P, and Gérard F. (2010). A mechanistic model for understanding root-induced chemical changes controlling phosphorus availability. Annals of Botany, 105:1183–1198. Dingkuhn, M., B. Singh, B. Clerget, J. Chantereau, and Sultan, B. (2006). Past, present and future criteria to breed crops for water-limited environments in West Africa. Agriculture Water Management, 80 (1–3), 241–261. Doggett, H, (1965b). Disruptive selection in crop development. Nature, 206: 279-280. Doggett, H. (1970). Recurrent selection in sorghum populations. Heredity, 28: 9-29 Doggett, H. (1988). Sorghum, 2nd Ed, Longman Scientific & Technical, Harlow, UK. Doggett, H. (1988), Sorghum. Tropical Agriculture Series. 2nd ed. CTA, Wageningen, The Netherlands. Dossou-Aminon, I., Yêyinou Loko, L., Adjatin, A., Ewedje, E.B.K., Dansi, A., Rakshit, S., et al. (2015). Genetic Divergence in Northern Benin Sorghum (Sorghum bicolor L. Moench) Landraces as Revealed by Agromorphological Traits and Selection of Candidate Genotypes. The Scientific World Journal, 6-10. Doumbia, M.D., Hossner, L.R. and Onken, A.B. (1993). Variable sorghum growth in acid soils of sub-humid West Africa. Arid Soil Research and Rehabilitation, 7:335–346. Doumbia, M.D., Sidibé, A., Bagayoko, A., Diarra, M.A., Bationo, A., Kablan, R.A., Yost, R.S., Hossner, L.R., Hons, F.M. (2003). Specific fertiliser recommendations: Calibration and validation of the phosphorus module of the NuMaSS model. African Crop Science Journal. 11: 17-26 162 University of Ghana http://ugspace.ug.edu.gh Driessen, P., Deckers, J., Spaargaren, O., Nachtergaele, F. (2001). Lecture notes on the major soils of the world. FAO World Soil Resources report 94: 334. Druilhe, Z. and Barreiro - Hurle, J. (2012). Fertilizer subsidies in sub-Saharan Africa. ESA Working paper 12, Rome, FAO. Duan, Y.H., Zhang, Y.L., Ye, L.T., Fan, X.R., Xu, G.H. and Shen, Q.R. (2007). Responses of rice cultivars with different nitrogen use efficiency to partial nitrate nutrition. Annals of Botany, 99:1153-1160. Duvick, D.N. and Cassman, K.G. (1999). Post-Green Revolution Trends in Yield Potential of Temperate Maize in the North-Central United States. Crop Science, 39:1622-1630. Eguavoen, I. (2013). Climate change and trajectories of blame in northern Ghana. Anthropological Notebooks, 19 (1): 5-24. Elangovan, M., Jain S. K., Patel N. V. (2013). Characterization of sorghum germplasm collected from gujarat. Indian Journal of Plant Genetic Resources, 26 (1):42–46. Elangovan, M., Kiran babu, P., Seetharama, N, Patil, J. V. (2014). Genetic Diversity and Heritability, characters Associated in Sweet Sorghum [Sorghum bicolor (L.) Moench]. Sugar Technology, 16:200–210. El-Bassam, N. (2010). Handbook of Bioenergy Crops. A Complete Reference to Species, Development and Applications. Part II, 319- 333. El-Din, A.A.T., Hessein, E.M., Ali, E.A. (2012). Path coefficient and correlation assessment of yield and yield associated traits in sorghum (Sorghum bicolor L. Moench). American- Eurasian Journal of Agricultural and Environmental Science, 12: 815–819. El-Swaify, S.A, Pathak, P., Rego, T.J., Singh, S. (1985). Soil management for optimized productivity under rainfed conditions in the semi-arid tropics. Advances in Soil Sciences, 1: 1-64 Enu, P. and Attah-Obeng, P. (2013). Which Macroeconomic Factors Influence Agricultural Production in Ghana? Academic Research International, 4 (5): 333 – 346. Environmental Protection Agency [EPA], (2003). Annual Report: Revised Draft, Environmental Protection Agency, Accra, Ghana. Fairhurst, T. R., Lefroy, R. Mutert, E. and Batijes, N. (1999). The importance, distribution and causes of phosphorus deficiency as a constraint to crop production in the tropic. Agroforestry Forum, 9: 2-8, 1999. 163 University of Ghana http://ugspace.ug.edu.gh Falconer, D. S., (1960a). Introduction to Quantitative Genetics. Oliver & Boyd, Edinburgh, London. Food and Agriculture Organization of the United Nations [FAO] (1994). The state of Food and Agriculture. FAO Agriculture series (27). Available at: http://www.fao.org/docrep/t4450e/T4450E00.htm. Accessed on 14th October 2013. Food and Agriculture Organization of the United Nations [FAO] (2009). Food security and agricultural mitigation in developing countries: Options for capturing synergies. Rome, Italy: Available at: http://www.fao.org/docrep/012/i1318e/i1318e00.pdf. Food and Agriculture Organization of the United Nations Statistics [FAOSTAT] (2012). http://faostat.fao.org/site/575/default. Accessed on 2nd May 2014. Food and Agriculture Organization of the United Nations [FAO] (2010). Sorghum. http://faostat.fao.org/site/567/default.aspx. Accessed on 10th December 2015. Food and Agriculture Organization of the United Nations [FAO] (2013b). FAOSTAT Database. Rome: FAO. Food and Agriculture Organization of the United Nations [FAO] (2014). The state of food and agriculture: Innovation in family farming. Rome: Food and Agriculture Organization of the United Nations FAOSTAT (2015). http://www.fao.org/faostat/en/#data/QC. Accessed on 12th November 2014. Feddermann, N., Finlay, R., Boller, T., Elfstrand, M. (2010). Functional diversity in arbuscular mycorrhizal—the role of gene expression, phosphorus nutrition and symbiotic efficiency. Fungal Ecology, 3: 1–8. Fedorov, A. A. (1969). Chromosome Numbers of Flowering Plants. Leningrad: Academy of Sciences of the USSR. Komarov Botanical Institute. Goldblatt, P. (Ed). Index to plant chromosome numbers 1975 – 1978. Fink, J. R., Inda, A. V., Bavaresco, J., Barron, V., Torrent, J. and Bayer, C. (2016). Phosphorus adsorption and desorption in undisturbed samples from subtropical soils under conventional tillage or no-tillage. Journal of plant nutrition and soil science 179 (2): 198- 205. Fosu-Mensah, B. Y., Vlek, P. L. G., and McCarthy, D. S. (2012). Farmers’ perception and adaptation to climate change: A case study of Sekyedumase district in Ghana. Environment, Development and Sustainability, 14: 495-505. Furihata, T., Suzuki, M., Sakurai, H. (1992) Kinetic characterization of two phosphate uptake systems with different affinities in suspension-cultured Catharanthus roseus protoplasts. Plant Cell Physiology, 33:1151–1157. 164 University of Ghana http://ugspace.ug.edu.gh Gahoonia, T.S, Claassen, N. and Jungk, A. (1992). Mobilization of phosphate in different soils by ryegrass supplied with ammonium or nitrate. Plant and Soil, 140:241-248. Gallais, A., Coque, M., Quilleré, I., Prioul, J.L. and Hirel, B. (2006). Modelling post-silking N- fluxes in maize using 15N-labelling field experiments. New Phytologist, 172:696-707. Gamuyao, R., Chin, J.H., Pariasca-Tanaka, J., Pesaresi, P., Catausan, S., Dalid, I. Slamet- Loedin, C. et al. (2012). The Protein Kinase PSTOL1 from Traditional Rice Confers Tolerance of Phosphorus Deficiency. Nature, 488 (7412): 535-539. Garber, E.D. (1950). Cytotaxonomic studies in the genus Sorghum. University of California Publications in Botany, 23: 283 – 362. Gelli, M., Duo, Y., Konda, A. J., Zhang, C., Holding, D. and Dweikat, I. (2014). Identification of differentially expressed genes between sorghum genotypes with contrasting nitrogen stress tolerance by genome-wide transcriptional profiling. BMC Genomics, 15:179. Gemenet, D. C, Leiser, W.L., Beggi, F., Herrmann, L. H., Vadez, V. Rattunde, H.F.W. et al. (2016). Overcoming phosphorus deficiency in West African Pearl Millet and Sorghum Production systems; Promising Options for crop improvement. Frontiers in plant science, 7:1389. Ghana Statistical Service [GSS] (2013). Millennium Development Goals Report 2013. Ghana Statistical Service. Accra. Govindaraj, M., Vetriventhan, M. & Srinivasan, M. (2015). Importance of Genetic Diversity Assessment in Crop Plants and Its Recent Advances: An Overview of Its Analytical Perspectives. Genetics Research International, 1-14. Goode, D.L., Halbert, C. and Arendt, E.K. (2003). Optimization of mashing conditions when mashing with unmalted sorghum and commercial enzymes. Journal of American Society of Brewing Chemist, 61: 69-78. Grant, C.A., D.N. Flaten, D.J. Tomasiewicz, and Sheppard. S.C. (2001). The importance of early season phosphorus nutrition. Canadian Journal of Plant Science, 81:211–224. Gregory, D.I. and Bumb, B. L. (2006). Factors affecting Supply of Fertilizer in Sub Saharan Africa. Agriculture and Rural Development Discussion Paper 24 World Bank, Washington DC. Grundon, N.l., Asher, C.l., and Edwards, D.G. (1986). Symptoms of nutritional deficiencies and toxicities in grain sorghum cv. Texas 61 OSR. ln: Foale, M.A. and HenzelL R.G. eds, Proceedings First Australian Sorghum Conference, February 1986, Gatton, Australia, 6.1- 6.10. Gunes, A., Inal, A., Alpaslan, M., and Cakmak, I. (2006). Genotypic variation in phosphorus efficiency between wheat cultivars grown under greenhouse and field conditions. Soil Science and Plant Nutrition, 52: 470-478 165 University of Ghana http://ugspace.ug.edu.gh Guo, P., Baum, M., Grando, S., Ceccarelli, S., Bai, G., Li, R. et al. (2009). Differentially expressed genes between drought-tolerant and drought-sensitive barley genotypes in response to drought stress during the reproductive stage. Journal of Experimental Botany, 60:3531 – 3544 Gyasi, E. and Uitto, J. (1997). Environment, Biodiversity and Agricultural Change in West Africa; Perspectives from Ghana. United Nations University Press. Available at: http://www.unu.edu/unupress/unupbooks/80964e/80964E00.htm Hammond, J.P., Broadley, M.R., White, P.J, King, G.J., Bowen, H.C., Hayden, R. et al. (2009). Shoot yield drives phosphorus use efficiency in Brassica oleracea and correlates with root architecture traits. Journal of Experimental Botany, 60: 1953–1968. Hansen, J.C., Cade-Menun, B.J., Strawn, D.G. (2004) Phosphorus speciation in manure-amended alkaline soils. Journal of Environmental Quality, 33: 1521–1527. Harlan, J. and De Wet, J. (1972). A simplified classification of cultivated sorghum. Crop Science, 12(2):172-176. Harlan, J.R. (1972). A new classification of cultivated sorghum. In: Sorghum in Seventies N.G.P. Rao and L.R. House (eds.). Oxford & IBH, New Delhi, India. 512-516. Harrison, M. (2005) Signaling on the arbuscular mycorrhizal symbiosis. Annual Reviews in Microbiology, 59:19–42. Hart, G.E., Schertz, K.F., Peng, Y. and Syed, N.H. (2001). Genetic mapping of Sorghum bicolor (L.) Moench QTLs that control variation in tillering and other morphological characters. Theoretical Applied Genetics, 103: 1232–1242. Hash, C. T., Schaffert, R. E. and Peacock, J. M. (2002). Prospects for using conventional techniques and molecular biological tools to enhance performance of ‘orphan’ crop plants on soils low in available phosphorus. Plant and soil, 245: 135-146. Henikoff, S. and Comai, L. (2003). Single-nucleotide mutations for plant functional genomics. Annual Review of Plant Biology, 54: 375-401. Hernandez, M.A. and Torero, M.T. (2011). Fertilizer Market Situation. Market Structure, consumption and trade patterns, and pricing behavior. IFPRI Discussion Paper 1058. International Food Policy Research Institute. Henao, J, and Baanante, C. (2006). Agricultural production and soil nutrient mining in Africa: Implication for resource conservation and policy development. IFDC Tech. Bull. International Fertilizer Development Center. Muscle Shoals, Al. USA. Hinsinger, P. (1998). How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Advance Agronomy, 64: 225–265. 166 University of Ghana http://ugspace.ug.edu.gh Hinsinger P. (2001). Bioavailability of soil Inorganic P in the rhizosphere as affected by root- induced chemical changes: a review. Plant Soil, 237: 173-195. Hodge, A., Berta, G., Doussan, C., Merchan, F. and Crespi, M. (2009). Plant Root Growth, Architecture and Function. Plant and Soil, 321: 153-187. Holford, I.C.R. (1997). Soil phosphorus: its measurements and its uptake by plants. Australian Journal of Soil Research, 35: 227-239. Holford, I. C. R.; Hird, C.; Lawrie, R. (1997). Effects of animal effluents on the phosphorus sorption characteristics of soils. Australian Journal of Soil Research, 35: 365-373. House, L.R., (1978). A guide to Sorghum breeding. International Crop Research Institute for Semi- Arid Tropics (ICRISAT), Patancheru, A.P., India. House L.R. (1985). A guide to sorghum breeding, Second edition. Patancheru, Andhra Pradesh, India: International Crops Research Institute for the Semi-Arid Tropics: 165. ICRISAT (2006). Nurturing the seeds of success in the semi-arid tropics. ICRISAT Annual Report 2006. International Crops Research Institute for the Semi-Arid Tropics, Patancheru, Hyderabad. International Fertilizer Development Center [IFDC] (2006). Africa Fertilizer Summit Proceedings, Abuja, Nigeria June 9–13; IFDC, Florence, Alabama. Issaka, Y. B., Antwi, M. and Tawia, G. (2016). Organic and non-organic farms in West Mamprusi district of Ghana. Agriculture, 6: 13. International Fund for Agricultural Development [IFAD] (2010). Rural Poverty Report 2010: New Realities, New Challenges: New Opportunities for Tomorrow’s Generation. Rome: Quintily. Jenks, M. A., Joly, R. J., Peters, P. J., Rich, P. J., Axtell, J. D., Ashworth, E. A. (1994a). Chemically-induced cuticle mutation affecting epidermal conductance to water vapor and disease susceptibility in Sorghum bicolor (L.) Moench. Plant Physiology, 1051(1):1239– 1245. Jiao, Y., Burke, J., Chopra, R., Burow, G., Chen, J., Wang, B., Hayes, C., Emendack, Y., Ware, D. and Xin, Z. (2016) A sorghum mutant resource as an efficient platform for gene discovery in grasses. Plant Cell, 28 (7):1551–1562. Jones, C. D., Cox, P.M., Essery, R.L.H., Roberts, D.L. and Woodage, M.J. (2003), Strong carbon cycle feedbacks in a climate model with interactive CO2 and sulphate aerosols. Geophysical Research Letter, 30 (9):1479. 167 University of Ghana http://ugspace.ug.edu.gh Jordan, W.R., McCrary, M. and Miller, F.R., (1979). Compensatory growth in the crown root system of sorghum. Agronomy Journal, 71: 803-806. Kaeppler, S.M., Kaepller, H.F., Rhee, Y., (2000). Epigenetic aspects of somaclonal variation in plants. Plant Molecular Biology, 43: 179–18. Kaizzi, C.K., Ssali, H., Nansamba, A., Vlek, P.L.G. (2007). The potential benefits of Azolla, velvet bean (Mucuna pruriens) and N fertilizers in rice production under contrasting systems of eastern Uganda. In: Bationo A, Kihara J, Kimetu J, Waswa B (eds) Advances in Integrated Soil Fertility Research in sub-Saharan Africa: Challenges and Opportunities. 423- 433. Kaizzi, K.C., Byalebeka, J., Semalulu, O., Alou, I., Zimwanguyizza, W., Nansamba, A., et al. (2012). Sorghum response to fertilizer and nitrogen use efficiency in Uganda. Agronomy Journal. 104:83-90. Kamara, A.Y., Kling, J.G., Menkir, A., Ibikunle, G. (2003). Agronomic performance of maize (Zea mays L.) breeding lines derived from a low nitrogen maize population. Journal of Agricultural Science, 141:221-230. Kanchebe, E. (2010). Local Knowledge and Livelihood Sustainability under Environmental Change in Northern Ghana- PhD thesis. Center for Development Research, University of Bonn. Available from: http://hss.ulb.uni-bonn.de:90/2010/2336/2336.htm Kang B. T. (1985). Cropping systems and soil fertility management in the humid and sub-humid tropics with special reference to West Africa. In: Management of Nitrogen and Phosphorus Fertilizers in Sub-Saharan Africa, Mokwunye U., Vlek P. L. G., (eds) 83–94. Katyal, J.C. and Das, S.K. (1993). Fertiliser management in grain sorghum. In: Tandon HLS (ed) Fertiliser Management in Food Crops: 61–78. Fertiliser Development and Consultation Organisation, New Delhi, India. Kelly, V. (2006). Factors Affecting Demand for Fertilizer in Sub-Saharan Africa. Discussion Paper 23, Agriculture and Rural Development Department, the World Bank, Washington D.C. Khorgade, P.W., Narkhede. M.N. and Raut, S.K. (1985). Genetic variability studies in chickpea (Cicer arietinum L.). International Chickpea Newsletter, 12-13. Kidd, P.S. and Proctor, J. (2001). Why plants grow poorly on very acid soils: are ecologists missing the obvious? Journal of Experimental Botany, 52 (357): 791–799. Kihara, J., Nziguheba, G., Zingore, S., Coulibaly, A., Esilaba, A., Kabambe, V., Huising, J. (2016). Understanding variability in crop response to fertilizer and amendments in sub-Saharan Africa. Agriculture, Ecosystems & Environment, 229: 1–12. Kimaru-Muchai, S., Mucheru-Muna, M., Mugwe, J., Mugendi, D., Mairura, F., Tsobeng, A and Tchoundjue, Z. (2013). Communication channels used in dissemination of soil fertility 168 University of Ghana http://ugspace.ug.edu.gh management practices in the central highlands of Kenya. In: Agro-Ecological Intensification of Agricultural Systems in the African Highlands. B. Vanlauwe, P. van Asten, G. Blomme (Eds.). 281-291. Kimber, C.T. (2000). Origins of domesticated sorghum and its early diffusion into India and China. In ‘Sorghum: Origin, History, Technology, and Production’, C. Wayne Smith and R.A. Frederiksen (eds). 3-98. Kiruki, S., Onek, L.A., Limo, M., (2006). Azide-based mutagenesis suppresses Striga hermonthica seed germination and parasitism on maize varieties. African Journal of Biotechnology, 5: 866-870. Kisinyo, P.O., Gudu, S.O., Othieno, C.O., Okalebo, J.R., Opala, P.A., Maghanga, J.K., Agala et al. (2012).Effects of lime, phosphorus and rhizobia on Sesbania sesban performance in a Western Kenyan acid soil. African Journal of Agricultural Research, 7(18): 2800 -2809. Klopfenstein, C.F and Hoseney, R.C. (1995). Nutritional properties of sorghum and the millets. In: D.A.V Dendy (Ed), sorghum and the millets: Chemistry and Technology”, American Association of Cereal chemists Inc. St. Paul, 125-168. Kodym, A. and Afza, R. (2003). Physical and chemical mutagenesis. In: Grotewold E (ed) Plant functional genomics: methods and protocols. Methods in Molecular Biology. 236: 189 – 203. Koegel, S., Ait Lahmidi, N., Arnould, C., Chatagnier, O., Walder, F., Ineichen, K., Boller, T., Wipf, D., Wiekem, A. and Courty, P. E. (2013). The family of ammonium transporters (AMT) in Sorghum bicolor: two AMT members are induced locally, but not systemically in roots colonized by arbuscular mycorrhizal fungi. New Phytologist, 198 (3): 853-865. Kombiok, J.M., Buah, J.S.S. and Sogbedji, J.M. (2012). Enhancing soil fertility for cereal crop production through biological practices and the integration of organic and in-organic fertilisers in northern savanna zone of Ghana. In: Soil Fertility. R. Issaka (ed). INTECH online publishers. 3-31. Konate, M. (1984). Climate of sorghum and millet cultivation zones of the semi-arid tropics regions of West Africa. In: Agrometeorology of Sand Millet in the Semi-arid Tropics. M V K Sivakumar and S M Virmani (Eds) Proceedings of International Symposium ICRISAT Center, 15–20. November, 101–114. ICRISAT, Hyderabad, India. Kondo, M., Pablico, P.P., Aragones, D.V., Agbisit, R., Morita, S., Courtois, B. (2003). Genotypic and environmental variations in root morphology in rice genotypes under upland field conditions. Plant and Soil, 255:189-200. 169 University of Ghana http://ugspace.ug.edu.gh Kong, L., Dong, J. and Hart, G. E. (2000). Characteristics, linkage-map positions, and allelic differentiation of Sorghum bicolor (L.) Moench DNA simple-sequence repeats (SSRs). Theoretical and Applied Genetics, 101, (3): 438–448. Kpongor, D. (2007). Spatially explicit modeling of sorghum (Sorghum bicolor L) production on complex terrain of a semi-arid region of Ghana using APSIM. PhD thesis. Ecology and Development Series 51.Center for Development Research (ZEF), University of Bonn. Bonn, Germany. 132. Kpongor, D. S., Tabo, R., Bationo, A., Fosu, M. and Vlek, P. (2009). Simulating Sorghum Yield Response to Mineral Fertilizer in Semi-arid Northern Ghana. Paper from GLOWA-Volta Project and the CGIAR Challenge Program on Water and Food [online]. Available: www.ifwf2.org/addons/download.presentation. Date accessed18th July 2016. Kudadjie, C.Y., Struik, P. C., Richards, P. and Offei, S. K. (2004). Assessing production constraints, management and use of sorghum diversity in northeast Ghana: a diagnostic study. Net Journal of Agricultural Science, 52(3 - 4): 371-391. Kudadjie, E. (2006). Integrating science with farmer knowledge; sorghum diversity management in northern-east Ghana. PhD. Thesis, Wageningen University. 220. Kudadjie-Freeman, C. and Dankyi-Boateng, S. (2012). Gender and cultural dimensions of sorghum seed management in northeast Ghana. Global Journal of Biology, agriculture and health sciences, 1(2): 4-9. Kumar, A.A., Reddy,V.S., Sharma, H.C., Hash, C.T., Srinivasa Rao, P., Ramaiah, B. and Sanjana, R.P. (2011). Recent advances in sorghum genetic enhancement research at ICRISAT. American Journal of Plant Sciences, 2:589–600. Kwabiah, A.B., Palm, C.A., Stoskopf, N.C. and Voroney, R.P. (2003). Response of soil microbial biomass dynamics to quality of plant materials with emphasis on P availability. Soil Biology and Biochemistry, 35: 207–216. Larkin, P.J. and Scowcroft, W.R. (1981). Somaclonal variation: a novel source of variability from cell cultures for plant improvement. Theoretical and Applied Genetics, 60: 197–214. Laube, W., Schraven, B. and M. Awo. M. (2012). Smallholder adaptation to climate change: dynamics and limits in Northern Ghana. Climate change, 111: 753-774. Lazarides, M., Hacker, J.B. and Andrew, M.H. (1991). Taxonomy, cytology and ecology of indigenous Australian sorghums (Sorghum Moench: Andropogoneae: Poaceae), Australian Systematic Botany, 4: 591-635 Legwaila, G. M., Balole, T.V. and S. K. Karikari, (2003). Review of sweet sorghum: a potential cash and forage crop in Botswana. UNISWA Journal of Agriculture, 12: 5–14. 170 University of Ghana http://ugspace.ug.edu.gh Leiser, W.L., Rattunde, H.F., Piepho, H.P. and Parzies, H.K. 2012a. Getting the most out of sorghum low-input field trials in West Africa using spatial adjustment. Journal of Agronomy and Crop Science, 198:349–359. Leiser, W. L., Rattunde, H. F.W. Weltzien, E., and Haussmann, B.I.G. (2014). Phosphorus uptake and use efficiency of diverse West and Central African sorghum genotypes under field conditions in Mali. Plant Soil, 377:383–394. Leiser, W.L., Rattunde, H.F.W., Piepho, H.P., Weltzien, E., Diallo, A., Tourè, A.O., Haussmann, B.I.G. (2015) Phosphorous efficiency and tolerance traits for selection of sorghum for performance in phosphorous-limited environments. Crop Science, 55:1152 – 1162. Leiser, W. L., Olatoye, M. O., Rattunde, H. F. W., Weltzien, E., Neumann, G., Haussmann, B. I. G. (2016). No need to breed for enhanced colonization by arbuscular mycorrhizal fungi to improve low-P adaptation of West African sorghums. Plant Soil, 401 51–64. 10. Liang, Q. (2009). “Preface,” in Induced Plant Mutations in Genomics Era. Food and Agriculture Organization of the United States.1. Liao, H., Rubio, G. and Yan, X., (2001) Effect of phosphorus availability on basal root shallowness in common bean. Plant and Soil, 232: 69–79. Luz, V.K., Silveira, S.F.S., Fonseca, G.M., Groli, E.L., Figueiredo, R.G. et al. (2016). Identification of variability for agronomically important traits in rice mutant families. Bragantia, 75(1): 41-50. Lynch, J. P. (1995). Root architecture and plant productivity. Plant Physiology. 109: 7–13. Lynch, J.P. and Ho, M.D. (2005). Rhizoeconomics: carbon costs of phosphorus acquisition. Plant and Soil, 269:45-56. Lynch, J.P. (2007) Roots of the second green revolution. Australian Journal of Botany, 55:1–20. Lynch, J.P. and Brown, K.M. (2008). Root strategies for phosphorus acquisition. In: The Ecophysiology of Plant-Phosphorus Interactions. P. J. White & J. P. Hammond (Eds). New York: Springer. MacCarthy, D. S., Vlek, P. L. G. (2012). Impact of climate change on sorghum production under different nutrient and crop residue management in semi-arid region of Ghana: a modeling perspective. African Crop Science Journal, 20 (2): 275-291 MacDonald, G.K., Bennett, E.M., Potter, P.A. and Ramankutty, N. (2011). Agronomic phosphorus imbalances across the world’s croplands. Proceedings of the National Academy of Sciences, USA, 108: 3086-3091. 171 University of Ghana http://ugspace.ug.edu.gh Maerere, A.P, Kimbi, G.G. and Nonga, D.L.M. (2001).Comparative effectiveness of animal manures on soil chemical properties, yield and root growth of amaranthus (Amaranthus cruentus L.). African Journal of Science and Technology, 1 (4): 14-21. Magalhaes, J.V., Liu, J., Guimarães, C.T., Lana, U.G.P., Alves, V.M.C. Wang, Y.H et al. (2007). A gene in the multidrug and toxic compound extrusion (MATE) family confers aluminum tolerance in sorghum. Nature Genetics, 39:1156–1161. Mairura, F.S., Mugendi, D.N., Mwanje, J.I., Ramisch, J.J., Mbugua, P.K., Chianu, J.N. (2007). Integrating scientific and farmers' evaluation of soil quality indicators in Central Kenya. Geoderma, 139:134–143. Maiti, R. K., Ramaiah, K. V., Bisen, S. S., Chidley, V. L. (1984). A comparative study of the haustorial development of Striga asiatica (L.) kuntze on sorghum cultivars. Annals of Botany, 54: 447–457. Manske, G.G.B., Ortiz-Monasterio, J.I., van Ginkel, R.M. et al. (2002). Phosphorus use efficiency in tall, semi-dwarf and dwarf near isogenic lines of spring wheat. Euphytica, 125:113–119. Manu, A., A. Bationo, and Geiger. S.C. (1991). Fertility status of selected millet producing soils of West Africa with emphasis on phosphorus. Soil Sciences, 152:315–320. Mapfumo, P. and Giller, K.E. (2001). Soil fertility management strategies and practices by smallholder farmers in semi-arid areas of Zimbabwe. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) with permission from the Food and Agricultural Organization (FAO), Bulawayo, Zimbabwe and Rome, Italy. Marenya, P., Barrett, C.B. and Gulick. T. (2008). “Farmers Perceptions of Soil Fertility and Fertilizer Yield Response in Kenya.” SSRN Scholarly Paper ID 1845546, Social Science Research Network, Rochester, NY. Marschner, H. (1995) Mineral Nutrition of Higher Plants, Second edition. London: Academic Press. Maqbool, S.B., Devi, P. and Sticklen, M.B. (2001). Biotechnology- Genetic Improvement of Sorghum (Sorghum bicolor (L.). In Vitro Cellular and Developmental Biology- Plant, 37:504-515. Melese A., Gebrekidan, H., Yli-Halla M. and Yitaferu, B. (2015). Phosphorus Status, Inorganic Phosphorus Forms, and Other Physicochemical Properties of Acid Soils of Farta District, Northwestern Highlands of Ethiopia. Applied and Environmental Soil Science, 1-11. Mehlich, A. (1984). Mehlich three soil test extractant: A modification of the Mehlich two extractant. Communications in Soil Science and Plant Analysis, 15:1409 -1416. 172 University of Ghana http://ugspace.ug.edu.gh Menezes, C.B., Carvalho, J., de Silva, G.A., Assis, L., Bernardino, K., Costa, S., Vander, F., Tardin, F.D. and Schaffert, R.E. (2014). Combining ability of grain sorghum lines selected for Aluminum tolerance. Crop Breeding and Applied Biotechnology, 14 (1): 42-48. Menz, M.A., Klein, R.R., Mullet, J.E., Obert, J.A., Unruh, N.C. and Klein, P.E. (2002). A high- density genetic map of Sorghum bicolor (L.) Moench based on 2926 AFLP, RFLP and SSR markers. Plant Molecular Biology, 48(5–6):483–499. Millar, D., Abazaami, J. and Bonye, S. (2007). Women, land and agricultural productivity: a study into community level constraints to land availability for women in Kalbeo and Gowrie Kunkwa. Centre for Cosmovision and Indigenous Knowledge (CECIK). Miller, F.R. (1976). Release of A/BT×623. TAES Form 97-72. Report of Technical Committee on Seed Release and Increase. Texas Agriculture Experimental Station, College Station, TX. Ministry of Food and Agriculture (MoFA). (2009). Food and Agriculture Sector Development Policy (FASDEP II). Accra. In: West African Agriculture and Climate Change: A Comprehensive Analysis. Jalloh, A., Nelson, G. C., Thomas, T. S., Zougmoré, R. and Roy-Macauley, H. (Eds), IFPRI. Washington, DC. Accesed from http://www.ifpri.org/sites/default/files/publications/rr178.pdf Ministry of Food and Agriculture [MOFA] Ghana (2015). Talensi-Nabdam District in the Upper East. http://MoFA.gov.gh/site/?page id=1658 Missihoun, A.A., Agbangla, C., Adoukonou-Sagbadja, H., Ahanhanzo, C., Vodouhè, R. (2012a). Gestion traditionnelle et statut des ressources génétiques du sorgho (Sorghum bicolor L. Moench) au Nord-Ouest du Bénin. International Journal of Biological and Chemical Science 6(3):1003-1018. Mizuno, H., H. Kawahigashi, J. Ogata, H. Minami, H. Kanamori, H. Nakagawa et al. (2013). Genomic inversion caused by gamma irradiation contributes to downregulation of a WBC11 homolog in bloomless sorghum. Theoretical and Applied Genetics, 126:1513– 1520. Mofokeng, A. M., Hussein, A. S. and Laing, M.D. (2017) Agro-morphological diversity of South African sorghum genotypes assessed through quantitative and qualitative phenotypic traits, South African Journal of Plant and Soil, 34: (5) 361-370. Moges, W. and Holden, N.M. (2007). Farmers’ perceptions of soil erosion and soil fertility loss in southern Ethiopia. Land Degradation and Development, 18(5):543-554. Mokhema, T. (2013). South Africa biofuels seen raising sorghum output fivefold. www.https://advancedbiofuelsusa.info/south-africa-biofuels-seen-raising-sorghum- output-fivefold/. Accessed on 18th September 2016. 173 University of Ghana http://ugspace.ug.edu.gh Moll, R.H, Kamprath, E.J and Jackson, W.A. (1982). Analysis and interpretation of factors which contribute to efficiency of nitrogen utilization. Agronomy Journal, 74 (3):562–564. Molla, M., Chowdhury, A., Islam, A., and Hoque, S. (1984). Microbial mineralization of organic phosphate in soil. Plant and Soil, 78(3):393-399. Murphy, J. and Riley, J.P. (1962). A Modified Single Solution Method for the Determination of Phosphate in Natural Waters. Analytica Chimica Acta, 27:31-36. Mwangi, T.J., Ngeny, J.M., Wekesa, F., Mulati, J. (2002). Acidic soil amendment for maize production in Uasin Gishu District, North Rift Kenya. In: Gachene et al. (Eds), Participatory technology development by small holders in Kenya. Proceedings of the 2nd scientific conference of the soil management and legume research network projects. KARI, Nairobi, Kenya. Nandwa, S. A, Bationo, A, Obanyi, S. N., Rao, I. M., Sanginga, N., Vanlauwe, B. (2011). Inter and intra-specific variation of legumes and mechanisms to access and adapt to less available soil phosphorus and rock phosphate. In: A. Bationo et al. (eds.). Fighting poverty in Sub-Saharan Africa: The Multiple Roles of Legumes in Integrated Soil Fertility Management,pp. 47-66. Narayanan, S., Aiken, R.M., Vara, P.V., Zhanguo Xin, P., Paul, G., and Yu, J. (2014). A Simple Quantitative Model to Predict Leaf Area Index in Sorghum. Agronomy Journal- Crop Ecology and Physiology, 6, (1): 219 -226. Ncube, B., Twomlow, S.J., Van Wijk, M.T., Dimes, J.P., Giller, K.E. (2007). Productivity and residual benefits of grain legumes to sorghum under semi-arid condition in south western Zimbabwe. Plant Soil, 229:1-15. Ndamani, F. and Watanabe, T. (2015). Farmers’ Perceptions about Adaptation Practices to Climate Change and Barriers to Adaptation: A Micro-Level Study in Ghana. Water, 7:4593-4604. Nekesa, A.O., Okalebo, J.R., Othieno, C.O., Thuita, M.N., Kipsat, M., Bationo, A., Sanginga, N., Kimettu, J., Vanlauwe, B. (2005). The potential of minjingu phosphate rock from Tanzania as a liming material: effect on maize and bean intercrop on acid soils of western Kenya. African Crop Science conference proceedings, 7: 1121-1128. Nin-Pratt, A., McBride, L. (2014). Green Revolution in West Africa: Is increasing population density triggering an Asian-style agricultural intensification in Ghana? Food Policy. Nkegbe, P.K. and Shankar, B. (2014). Adoption intensity of soil and water conservation practices by smallholders: evidence from northern Ghana. Bio-based and Applied Economics, 3 (2): 159-174. Nyanteng, V. K. and Asuming-Brempong, S. (2003). The role of Agriculture in Food security in Ghana. Roles of Agriculture Project International Conference. 20-22 October, Rome, Italy. 174 University of Ghana http://ugspace.ug.edu.gh Nziguheba, G. (2007). Overcoming phosphorus deficiency in soils of Eastern Africa: recent advances and challenges. In: Bationo A, Waswa B, Kihara J, Kimetu J (eds). Advances in integrated soil fertility management in Sub - Saharan Africa: Challenges and opportunities, Springer, 149-160. Obura. P.A. (2008). Effects of soil properties on bioavailability of aluminium and phosphorus in selected Kenyan and Brazilian soils. Ph. D. Thesis, Purdue University, USA. Ochieng, L., Mathenge, P. and Muasya, R. (2011). A survey of on-farm seed production practices of sorghum (Sorghum bicolor L. Moench) in Bomet District of Kenya. African Journal of Food, Agriculture, Nutrition and Development, 11(5): 270–276. Odedina, J.N., Odedina, S.A. and Ojeniyi, S.O. (2011). Effect of types of manure on growth and yield of cassava (Manihot esculenta, Crantz). Researcher, 3 (5): 1 –8. Ohene-Yankyera, K. (2004). Determinants of Farm Size in Land-Abundant Agrarian Communities of Northern Ghana. Journal of Science and Technology, 24 (2): 45-53. Oliveira, A.B., Alencar, N.L.M., Prisco, J.T., and Gomes-Filho, E. (2011). Accumulation of organic and inorganic solutes in NaCl-stressed sorghum seedlings from aged and primed seeds. Scientia Agricola, 68 (6): 632-637. Omondi, E. G. O., Makobe, M.N., Matasyoh, L.G. and Onyango, C.A. (2013). Genetic variability of tissue cultured Sorghum bicolor (L) Moench as revealed by morphological traits and simple sequence repeats (SSR) markers. African Journal of Biotechnology, 12(1): 1-7. Omotayo, O. E. and Chukwuka, K.S. (2009). Soil fertility restoration techniques in sub-Saharan Africa using organic resources. African Journal of Agricultural Research, 4 (3): 144-150. Oria, M.P., Hamaker,B.R., Axtell, J.D., Huang, C.P. (2000). A highly digestible sorghum mutant cultivar exhibits a unique folded structure of endosperm protein bodies. Proceedings of the National Academy of Sciences, USA, 97: 5065-5070. Osarfo, D., Senadza, B. and Nketiah-Amponsah, E. (2016).The Impact of Nonfarm Activities on Rural Farm Household Income and Food Security in the Upper East and Upper West Regions of Ghana . Theoretical Economics Letters, 6: 388-400. Ozturk, L., Eker, S., Torun, B. and Cakmak, I. (2005). Variation in phosphorus efficiency among 73 bread and durum wheat genotypes grown in a phosphorus-deficient calcareous soil. Plant Soil, 269: 69– 80. Owusu-Bennoah, E. and Acquaye, D.K. (1996). Greenhouse evaluation of agronomic potential of different sources of phosphate fertilizer in a typical concretionary soil of Northern Ghana. Fertility Research, 44:1091–106. Parentoni, S.N. and Souza, C.L. Jr (2008) Phosphorus acquisition and internal utilization efficiency in tropical maize genotypes. Pesquisa Agropecuária Brasileira, 43:893–901. 175 University of Ghana http://ugspace.ug.edu.gh Parentoni, S., C. de Souza, Jr., V. de Carvalho Alves, E. Gama, A. Coelho, A. de Oliveira, P. et al. (2010). Inheritance and breeding strategies for phosphorus efficiency in tropical maize (Zea mays L.). Maydica, 55:1–15. Parry, M. A. J., Madgwick, P. J., Bayon, C., Tearall, K., Hernandez-Lopez, A., Baudo, M. et al. (2009). Mutation discovery for crop improvement. Journal of Experimental Botany, 60(10): 2817- 2825. Paterson, A.H., Bowers, J.E., Chapman, B.A. (2004). Ancient polyploidization predating divergence of the cereals, and its consequences for comparative genomics. Proceeding of the National Academy of Science, USA, 101: 9903–9908. Paterson, A.H., Bowers, J.E. and Bruggmann, R. (2009). The Sorghum bicolor genome and the diversification of grasses. Nature, 457:551-556. Parentoni, S.N., and C. Lopes de Souza, Jr. (2008). Phosphorus acquisition and internal utilization efficiency in tropical maize genotypes. Pesquisa Agropecuária Brasileira, 43:893–901. Peters, P.J., Jenks, M.A., Rich,P.J. Axtell, J.D. and Ejeta. G. (2009). Mutagenesis, selection, and allelic analysis of epicuticular wax mutants in sorghum. Crop Science, 49:1250–1258. Peterson DG, Schulze SR, Sciara EB, Lee SA, Bowers JE, Nagel ANJ, Tibbitts DC et al. (2002) Integration of Cot analysis, DNA cloning, and high-throughput sequencing facilitates genome characterization and gene discovery. Genome Research, 12: 795–807. Phillips, S.L. and Wolfe, M.S. (2005). Centenary Review. Evolutionary plant breeding for low input systems. Journal of Agricultural Science, 143: 245-254. Prasad, P.V.V., K.J. Boote, and L.H. Allen, Jr. (2006a). Adverse high temperature effects on pollen viability, seed-set, seed yield, and harvest index of grain-sorghum [Sorghum bicolor (L.) Moench] are more severe at elevated carbon dioxide due to higher tissue temperatures. Agricultural and Forest Meteorology, 139:237–251. Prasad, P.V.V. and Staggenborg, S.A. (2009). Growth and production of sorghum and millets. In soils, plant growth and crop production. In: Encyclopedia of life Support Systems, Eolss Publishers, Oxford, UK. http:// www.eolss.net. Quaye, W. (2008). Food security situation in northern Ghana, coping strategies, and related constraints. African Journal of Agricultural Research, 3 (5): 334-342. Quijano-Guerta, C., Kirk, G.J.D., Portugal, A.M., Bartolome, V.I. and McLaren, G.C. (2002) Tolerance of rice germplasm to zinc deficiency. Field Crop Research, 76:123–130. Quinby, J.R. (1975). The genetics of sorghum improvement. Journal of Heredity, 66:56 – 62. 176 University of Ghana http://ugspace.ug.edu.gh Rakshit S., Gomashe S. S., Ganapathy K. N., et al. (2012). Morphological and molecular diversity reveal wide variability among sorghum Maldandi landraces from India. Journal of Plant Biochemistry and Biotechnology, 21(2):145–156. Rakszegi, M., Kisgyörgy, B. N., Tearall, K., Shewry, P. R., Láng, L., Phillips, A., et al. (2010). Diversity of agronomic and morphological traits in a mutant population of bread wheat studied in the health grain program. Euphytica, 174, 409–421. Rai, S., Kaur, A., and Singh, B. (2014). Quality characteristics of gluten free cookies prepared from different flour combinations. Journal of Food Science and Technology, 51(4):785– 789. Rajasekher, R.A., Ravinder R.V, Parthasarathy R.P., Gurava R.K., Belum V.S Reddy (2005). Performance of Layers on Sorghum-Based Poultry Feed Rations. International Sorghum and Millets Newsletter, ICRISAT, Patancheru, India, 46: 75-78. Raju, P. S., Clark, R. B., Ellis, J. R., Duncan, R. R. and Maranville, J. W. (1990). Benefit and Cost Analysis and Phosphorus in Mycorrhizal Fungi Colonizations with Sorghum (Sorghum bicolor) Genotypes Grown at Varied Phosphorus Levels. Plant Soil, 124: 199–204. Rao, I. M. (2001). Role of physiology in improving crop adaptation to abiotic stresses in the tropics: The case of common bean and tropical forages. In: Handbook of Plant and Crop Physiology. M. Pessarakli (Ed), 583–613. New York: Marcel Dekker. Reddy, B.B.S., Ramesh, S., Reddy, P.S. (2006). Sorghum genetic resources, cytogenetics and improvement. In: Singh RJ, Jauhar PP, editors. Cereals. Vol. 2. Boca Raton, FL: CRC Taylor and Francis, pp. 309–363. Genetic resources, chromosome engineering, and crop improvement. Reddy B.VS, Sharma, H.C., Thakur, R.P., Ramesh, S., Rattunde, F. and Mgonja, M. (2006) Sorghum hybrid parents research at ICRISAT: retrospect and prospects. SAT e-journal 2(1):1–24. Reddy, K.R. and DeLaune, R.D. (2008). Biogeochemistry of Wetlands: Science and Applications. CRC press. 335-345. Ribaut, J.M. and Poland, D. (2000). Molecular Approaches for the Genetic Improvement of Cereals for Stable Production in Water-limited Environments. A strategic planning workshop held at CIMMYT, El Batan, Mexico, 21–25 June 1999. International Maize and Wheat Improvement Center (CIMMYT).Mexico, DF, Mexico. Richardson, A. E. (1994). Soil microorganisms and phosphorus availability. In: Soil Biota Management in Sustainable Farming Systems, Pankhurst C. E., Doubeand B. M. Gupta V. V. S. (Eds) (Melbourne, VIC: CSIRO Publishing): 50–62. 177 University of Ghana http://ugspace.ug.edu.gh Risse, L.M., Cabrera, M.L., Franzluebbers, A.J., Gaskin, J.W., Gilley, J.E., Killorn, R., Radcliffe, D.E., Tollner, W. E., Zhang, H. (2006). Land application of manure for beneficial reuse. Biological Systems Engineering: Papers and Publications. Paper 65. Riyazaddin, M., Ashok Kumar, A., Rajendra Sudhakar, M., Ramaiah, B., Kavi Kishor, B. Polavarapu, H. C. S (2016). Inheritance of resistance to Sorghum Shoot Fly, Atherigona soccata in sorghum, Sorghum bicolor (L.) Moench. Frontiers in Plant Sciences, 7: 543. Rose T. J., Rengel Z., Ma Q., Bowden J. W. (2008). Post-flowering supply of P, but not K, is required for maximum canola seed yields. European Journal of Agronomy, 28: 371–379. Rose, T.J. and Wissuwa, M. (2012). Rethinking internal phosphorus utilization efficiency. Advances in agronomy, Elsevier: 85–217. Rose, T. J., Liu, L., & Wissuwa, M. (2013). Improving phosphorus efficiency in cereal crops: Is breeding for reduced grain phosphorus concentration part of the solution? Frontiers in Plant Science, 4; 444. Roy, P. K. Niyogi, K. Sengupta, D. N. Ghoosh, B. (2005). Spermidine treatment to rice seedlings recovers salinity stress-induced damage of plasma membrane and PM-bound H-ATPase in salt-tolerant and salt-sensitive rice cultivars. Plant Science, 168 (3): 583-591. RUDEMA (Rural Development Management (2006). National Strategic Plan for Sorghum Development in Ghana. Final Report. 3-16 Rware, H., Wairegi, L., Oduor, G., Macharia, M., Romney, D., Tarfa, B.D. et al. (2014). Assessing the Potential to Change Stakeholders Knowledge and Practices on Fertilizer Recommendations in Africa. Agricultural Science, 5(14):1384. Saballos, A., Vermerris, W., Rivera, L., Ejeta, G. (2008). Allelic association, chemical characterization and saccharification properties of brown midrib mutants of sorghum (Sorghum bicolor (L.) Moench). Bioenergy Research, 1: 193–204. Sahrawat, K.L., T.J. Rego, J.R. Burford, M.H. Rahman, J.K. Rao, and Adam, A. (1995). Response of sorghum to fertilizer phosphorus and its residual value in a Vertisol. Fertility Research, 41:41–47. Sahrawat, K.L, Mulbah, C.K., Diatta, S., DeLaune, R.D., Patrick Jr. W.H., Singh B.N., Jones M.P. (1996). The role of tolerant genotypes and plant nutrients in the management of iron toxicity in lowland rice. Journal of Agricultural Science, Cambridge, 126:143-149. Samake, O., Smaling, E.M.A., Kropff, M.J., Stomph, T. J. and Kodio, A. (2005). Effects of cultivation practices on spatial variation of soil fertility and millet yields in the Sahel of Mali. Agriculture, Ecosystems and Environment, 109 (4), 335- 345. 178 University of Ghana http://ugspace.ug.edu.gh Sanginga, N. and Woomer, P.L. (2009). Integrated Soil Fertility Management in Africa: Principles, Practices and Developmental Process. Tropical Soil Biology and Fertility Institute of the International Centre for Tropical Agriculture. Nairobi. 263. SARI (1995). Savanna Agricultural Research Institute (SARI) annual Report (1995), Nyakpala, Ghana. 133. Sattari, S.Z., Bouwman, A.F., Giller, K.E. and Van Ittersum, M.K. (2012). Residual soil phosphorus as the missing piece in the global phosphorus crisis puzzle. Proceedings of the National Academy of Sciences, USA, 109: 6348-6353. Sattler, S.E., Saballos, A., Xin, Z., Funnell-Harris, D.L., Vermerris, W., Pedersen, J.F. (2014). Characterization of novel Sorghum brown midrib mutants from an EMS-mutagenized population. G3 (Bethesda), 4: 2115–2124 Schachtman, D.P., Reid, R.J., Ayling, S.M. (1998). Phosphorus uptake by plants: from soil to cell. Plant Physiology, 116:447–453. Schaefer, C. E. G. R., Fabris, J. D. and Ker, J. C. (2008). Minerals in the clay fraction of Brazilian Latosols (Oxisols): a review. Clay Minerals, 43 (1): 137-154. Schaffert, R.E., Alves, V.M.C., Pitta, G.V.E., Bahia, A.F.C. and Santos, F.G. (2001). Mechanisms of phosphorus efficiency in maze. In: Plant nutrition, food security and sustainability of agroecosystems through basic and applied research. WJ. Horst et al. (Eds.) Plant Nutrition-Food Security and sustainability of Agro-ecosystems, 72-73. Schnell, J., Steele, M., Bean, J., Neuspiel, M., Girard, C. and Dormann, N. (2015). A comparative analysis of insertional effects in genetically engineered plants: Considerations for pre-market assessments. Transgenic Research, 24:1–17. Schoonover, J. E. and Crim, J. F. (2015). An Introduction to Soil Concepts and the Role of Soils in Watershed Management. Journal of Contemporary Water Research & Education, 154(1):21–47. Shen J. B., Yuan L. X., Zhang J. L., Li H. G., Bai Z. H., Chen X. P., et al. (2011). Phosphorus dynamics: from soil to plant. Plant Physiology, 156: 997–1005. Shepherd, K.D., and Soule M.J. (1998). Soil fertility management in West Kenya; a dynamic simulation of productivity, profitability and sustainability at different resource endowment levels. Agricultural Ecosystems and Environment 71, 131-145. Shin H, Shin HS, Dewbre GR, Harrison MJ. (2004) Phosphate transport in Arabidopsis: Pht1;1 and Pht1;4 play a major role in phosphate acquisition from both low- and high-phosphate environments. Plant Journal, 39: 629–642 179 University of Ghana http://ugspace.ug.edu.gh Shoemaker, C.E. and D.I. Bransby. (2010). The role of sorghum as a bioenergy feedstock. In: R. Braun, D. Karlen, and D. Johnson, editors, Sustainable alternative fuel feedstock opportunities, challenges and roadmaps for six U.S. regions, Proceedings of the Sustainable Feedstocks for Advance Biofuels Workshop, Atlanta, GA. 28–30 September. Soil and Water Conservation Society, Ankeny, 149–159. Sikora, P., Chawade, A., Larsson, M., Olsson, J. and Olsson, O. (2011). Mutagenesis as a Tool in Plant Genetics, Functional Genomics, and Breeding. International Journal of Plant Genomics, 1-13. Sinclair, T.R. (1998). Historical changes in harvest index crop nitrogen accumulation. Crop Science, 38:638-643. Singh, R. J. and Jauhar, P. (2006). Genetic Resources, Chromosome. Engineering and Crop Improvement, Volume 2 Cereals. CRC Press, New York. Smaling, E.M.A., (1993). An ecological framework for integrated nutrient management with special reference to Kenya. Ph.D. thesis, Wageningen Agricultural University, Wageningen. Smalberger, S. A., Singh, U., Chien, S. H., Henao, J., and Wilkens, P. W. (2006). Development and validation of a phosphate rock decision support system. Agronomy Journal, 98: 471– 483. Smaling, E.M.A, Nandwa, S.M. and Janssen, B.H. (1997). Soil fertility is at stake. In: Buresh RJ, Sanchez PA & Calhoun F (Eds). Replenishing soil fertility in Africa. Special publication No. 51 Madison, Wisconsin: American Society of Agronomy and Soil Science society of America. 47-61. Smith, C.W. and Frederiksen, R.A. (2000). Sorghum: origin, history, technology, and production, New York, NY John Wiley and Sons, 824. Snowden, J.D. (1935). A classification of cultivated sorghum. Kew Bulletin, 5:221 – 255. Sommer, R., Bossio, D., Desta, L., Dimes, J., Kihara, J., Koala, S., Winowiecki, L. (2013). Profitable and Sustainable Nutrient Management Systems for East and Southern African Smallholder Farming Systems-Challenges and Opportunities: A synthesis of the Eastern and Southern Africa situation in terms of past experiences, present and future opportunities in promoting nutrient use in Africa. Cali, Columbia, CIAT. Queensland, Australia: The University of Queensland, Queensland Alliance for Agriculture and Food Innovation (QAAFI), 91. Sree Ramulu, K. (1970a). Induced systematic mutations in sorghum. Mutation Research- Fundamental and Molecular Mechanism of Mutagen, 10:77–80. 180 University of Ghana http://ugspace.ug.edu.gh Srinivasa Rao, P., Kumar, C.G., Malapaka, J., Kamal, A. and Reddy, B. V. S. (2012). Feasibility of Sustaining Sugars in Sweet Sorghum Stalks during Post-Harvest Stage by Exploring Cultivars and Chemicals: A Desk Study. Sugar Tech, 14 (1): 21-25. Su J., Xiao Y., Li M., Liu Q., Li B., Tong Y., et al. (2006). Mapping QTLs for phosphorus- deficiency tolerance at wheat seedling stage. Plant Soil 281, 25–36. Su, Y J., Qin, Y. T., Zhang, S. L., Qin, G. W., Xu, G. J., Lu, R. Q., Mei, Z. J. (2012). Effects of planting density on growth and yield of summer maize Xundan 28. Acta Agriculture Jiangxi, 24 (6): 49- 53. Sultan, B., Roudier, P., Quirion, P., Alhassane, A., Muller, B., Dingkuhn, M. et al. (2013). Assessing climate change impacts on sorghum and millet yields in the Sudanian and Sahelian savannas of West Africa. Environmental Research Letters, 8, (1). Suzuki, T., Eiguchi, M., Kumamaru, T., Satoh, H., Matsusaka, H., Moriguchi, K. et al. (2008). MNU-induced mutant pools and high performance TILLING enable finding of any gene mutation in rice. Molecular Genetics and Genomics, 279:213-223. Sweeney, D. W., Moyer, J. L., Jardine, D. J. and Whitney, D. A. (2011). Nitrogen, phosphorus and potassium effects on grain sorghum production and stalk rot following alfalfa and birdsfoot trefoil, Journal of Plant Nutrition, 34, (9):1330-1340. Talebi, A. B., Talebi, A. B. and Shahrokhifar, B. (2012). Ethyl methane sulphonate (EMS) induced mutagenesis in malaysian rice (cv. MR219) for lethal dose determination. American Journal of Plant Sciences, 3:1661-1665. Tan K., Keltjens, W.G. and Findenegg, G.R. (1991). Role of magnesium in combination with liming in alleviating acid-soil stress in the aluminium-sensitive sorghum genotype CV323. Plant and Soil, 136: 65-71. Tan, W., Bogre, L. and Lopez-Juez, E. (2008). Light fluence rate and chloroplast are sources of signal controlling mesophyll cell morphogenesis and disvision. Cell Biology International, 32: 363-565 Theis, J. and Grady, H.M., (1991). Participatory Rapid Appraisal for Community Development: A Training Manual Based on Experience in the Middle East and North Africa. IIED / Save the Children Foundation, London. Thompson, T. (2000). Questionable foods and the gluten ‐ free diet: Survey of current recommendations. The Journal of the American Dietetic Association, 100: 463‐ 465. Tiecher, T.; Rheinheimer, D.S.; Kaminski, J. & Calegari, A. (2012). Forms of inorganic phosphorus in soil under different long term soil tillage systems and winter crops. Revista Brasileira de Ciência do Solo, 36:271-281. 181 University of Ghana http://ugspace.ug.edu.gh Till, B.J., Reynolds, S.H., Weil, C., et al. (2004b). Discovery of induced point mutations in maize genes by TILLING. BMC Plant Biology, 4:12. Till, B. J., Cooper, J. and Tai, T. H. et al., (2007). Discovery of chemically induced mutations in rice by TILLING. BMC Plant Biology, 7:19. Tiryaki, I. and Andrews, D. (2001). Germination and seedling cold tolerance in sorghum; Crop Science, 42:1391–1397. Tossah, B.K. (2000). Influence of Soil Properties and Organic Inputs on Phosphorus Cycling Herbaceous Legume-based Cropping Systems in the West African Derived Savanna. Leuven, Belgium. Catholic University Leuven. PhD dissertation. Tropical Soil, Biology and Fertility [TSBF] (1993). Handbook of methods. JM Anderson and SJ Ingram (Eds). CAB International, Oxon, UK. Tucker, M. R. (1999). Essential plant nutrients, their presence in North Carolina soils and role in plant nutrition. Agronomic Division. NCDA and CS. [Online] Available: http://www.ncagr.gov/agronomi/pdffiles/essnutr.pdf. Accessed on 15th September 2010. Turner BL, Papházy MJ, Haygarth PM, McKelvie ID (2002) Inositol phosphates in the environment. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 357: 449–469. Turner B. L., Newman S., Cheesman A. W., Reddy K. R. (2007). Sample pretreatment and phosphorus speciation in wetland soils. Soil Science Society of America Journal, 71: 1538– 1546. Twomlow, S.J. and Ncube, B. (2001). Improving soil management options for women farmers in Malawi and Zimbabwe: Proceedings of a Collaborators Workshop on the DFID-supported project ‘Will Women Farmers Invest in Improving their Soil Fertility Management? Experimentation in a risky environment’, 13–15 September 2000, ICRISAT, Bulawayo, Zimbabwe. International Crops Research Institute for the Semi-Arid Tropics, Bulawayo, Zimbabwe. Uematsu, H. and Mishra, A. (2010). Can Education Be a Barrier to Technology Adoption? Selected Paper prepared for presentation at the Agricultural & Applied Economics Association 2010 AAEA, CAES, & WAEA Joint Annual Meeting, Denver, Colorado, 25- 27. Undersander, D. (2003). Sorghums, Sudangrasses, and sorghum-Sudan hybrids. Univ. of Wisconsin Focus on Forage 5:5, Madison. http://www.uwex.edu/CES/crops/uwforage/SorghumsFOF.htm. Accessed on 26th May 2013. USDA National Nutrient Database for Standard Reference, Release 24. 2011. 182 University of Ghana http://ugspace.ug.edu.gh United States Department of Agriculture Agricultural Research Service [USDA-ARS] (2012). National Genetic Resources Program. Germplasm Resources Information Network - (GRIN). National Germplasm Resources Library, Beltsville, MD, USA. [Online] Available at http://www.ars-grin.gov/cgi-bin/npgs/html/taxon.pl?35092. Van Beuningen, L.T. and Busch, R.H. (1997). Genetic diversity among North American spring wheat cultivars: III Cluster analysis based on quantitative morphological traits. Crop Science, 97 (37): 981–988. van der Velde, M., Folberth, C., Balkovic, J., Ciais, P., Fritz, S., Janssens, I. A., et al. (2014). African crop yield reductions due to increasingly unbalanced nitrogen and phosphorus consumption. Global Change Biology, 20, 1278–1288. Van Harten AM (1998) Mutation Breeding: Theory and Practical Applications. Cambridge University Press, Cambridge, UK, 353. Van Kauwenbergh, S.J. (2006). Fertilizer Raw Material Resources of Africa. Africa Fertilizer Summit. Abuja, Nigeria. IFDC R-16. Vance, C. P. (2001) Symbiotic nitrogen fixation and phosphorus acquisition: Plant nutrition in a world of declining renewable resources. Plant Physiology, 127, 390–397. Van Vuuren, D.P., Bouman, A.F. and Beusen, A.H.W. (2010). Phosphorus demand for the 1970- 2100 period: a scenario analysis nod resource depletion. Global Environmental Change, 20: 428- 439. Vavilov, N.I. (1992). Origin and geography of cultivated plants. In: V.F. Dorofeyev (ed.), translated by D. Love. Cambridge University Press, Great Britain. Veihe, A. (2002). The spatial variability of erodibility and its relation to soil types: a study from northern Ghana. Geoderma, 106 (1-2): 101-120. Veneklaas E. J., Lambers H., Bragg J., Finnegan P. M., Lovelock C. E., Plaxton W. C., et al. (2012). Opportunities for improving phosphorus-use efficiency in crop plants. New Phytologist, 195: 306–32010. Verde, B. and Matusso, J. (2014). Phosphorus in Sub-Sahara African soils – Strategies and options for improving available soil phosphorus in smallholder farming systems: A review. Academic research journal of agricultural science and research, 2 (1):1-15. Verheye W. (2010). Growth and production of sorghum and millets. In: in Soils, Plant Growth and Production Volume II in Encyclopedia of Life Support Systems (EOLSS), Developed under the Auspices of the UNESCO, Vol. 2 Paris: Eolss Publishers; 183 University of Ghana http://ugspace.ug.edu.gh Vom Brocke, K., G. Trouche, E. Weltzien, C. P. Barro-Kon-dombo, E. Goze´, and Chantereau, J. (2010). Participatory variety development for sorghum in Burkina Faso: farmers’ selection and farmers’ criteria. Field Crops Research, 119: 183–194. Walkley, A. and Black, A.I. (1934). An examination of Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Science, 37:29-37 Wall, P. C., Neate, S.M., Graham, R.D., Reuter, D.J. and Rovira, A.D. (1994). The effect of rhizoctonia root disease and applied nitrogen on growth, nitrogen uptake and nutrient concentrations in spring wheat. Plant and Soil, 163: 111–120. Wang, X., Shen, J. and Liao, H. (2010). Acquisition or utilization, which is more critical for enhancing phosphorus efficiency in modern crops? Plant Science, 179:302–306. Wang, L., Lu, S., Zhang, Y., Li, Z., Du, X. and Liu, D. (2014) Comparative genetic analysis of Arabidopsis purple acid phosphatases AtPAP10, AtPAP12, and AtPAP26 provides new insights into their roles in plant adaptation to phosphate deprivation. Journal of Integrative Plant Biology, 56: 299 –314. Wang S. K., Bai Y. H., Shen C. J., Wu Y. R., Zhang S. N., Jiang D. A., et al. (2010). Auxin- related gene families in abiotic stress response in Sorghum bicolor. Functional Integrative Genomics 10, 533–546. Weaver, D.M. and Wong, M.T.F. (2011). Scope to improve phosphorus management and balance efficiency of crop and pasture soils with contrasting P buffer indices. Plant and Soil 349, 37-54. Westengen, O. T., Okongo, M. A., Onek, L., Berg, T., Upadhyaya, H., Birke-land, S. et al. (2014). Ethnolinguistic structuring of sorghum genetic diversity in Africa and the role of local seed systems. Proceedings of the National Academy of Sciences, USA, 111: 14100–14105. Whitehead, A. (2003). ‘Policy Discourses on Women’s Land Rights in Sub-Saharan Africa: The Implications of the Re-turn to the Customary’. Journal of Agrarian Change, 3 (1): 67–112. White, P.J. and Veneklaas, E.J. (2012). Nature and nurture: the importance of seed phosphorus content. Plant Soil, 357:1–8. Wissuwa, M, Kretzschmar, T, Rose, T.J. (2016). From promise to application: root traits for enhanced nutrient capture in rice breeding. Journal of Experimental Botany, 67: 3605- 3615. Wissuwa, M. and Ae, N. (2001). Genotypic variation for tolerance to phosphorus deficiency in rice and the potential for its exploitation in rice improvement. Plant Breeding, 120: 43–48. 184 University of Ghana http://ugspace.ug.edu.gh Wissuwa, M., Wegner, J. and Ae, N. (2002) Substitution mapping of Pup1: A major QTL increasing phosphorus uptake of rice from a phosphorus-deficient soil. Theoretical and Applied Genetics, 105: 890–897. Wissuwa M, Yano M, Ae N. (1998) Mapping of QTLs for phosphorus deficiency tolerance in rice (Oryza sativa L.). Theoretical and Applied Genetics, 97: 777–783 Wooten, D.R. (2001). The use of Sorghum propinquuto enhance agronomic traits in sorghum. M.S. Thesis. Texas A&M University, College Station, Texas. Wu, J. L., Wu, C., and Lei C. et al., (2005). Chemical and irradiation-induced mutants of indica rice IR64 for forward and reverse genetics. Plant Molecular Biology, 59 (1): 85–97. Xin, Z., Wang, M.L., Barkley, N.A., Burow, G., Franks, C., Pederson, G., Burke J. (2008). Applying genotyping (TILLING) and phenotyping analyses to elucidate gene function in a chemically induced sorghum mutant population. BMC Plant Biology, 8:103. Xin, Z., Li, W., Ming, Burow, G. and Burke, J. (2009). An Induced Sorghum Mutant Population Suitable for Bioenergy Research. BioEnergy Research. 2:10-16. Xin, Z., Gitz, D., Burow, G., Hayes, C., Burke, J.J. (2015). Registration of two allelic erect leaf mutants of sorghum. Journal of Plant Registrations, 9: 254–257. Xin, Z., Huang, J., Smith, A.R., Chen, J., Burke, J., Sattler, S.E., et al. (2017). Morphological Characterization of a New and Easily Recognizable Nuclear Male Sterile Mutant of Sorghum (Sorghum bicolor). PLoS ONE, 12(1):e0165195. Xu, Y. (2010). Molecular plant breeding. CAB International, Wallingford, UK. Yahaya, A. and Amoah, S.A. (2013). Bushfires in the Nandom District of the Upper West Region of Ghana: Perpetual Threat to Food Crop Production. Journal of the Environment and Earth Sciences, 3 (7) 10-14. Yamaguchi, M., Fujimoto, H.,Hirano, K., Nakamura, S.A.,Shinohara, K.O. et al. (2016). Sorghum Dw1, an agronomically important gene for lodging resistance, encodes a novel protein involved in cell proliferation. Scientific Reports, 6: 28366. Yan L., Liu S., Zhao S., Kang Y., Wang D., Gu T., et al. (2012). Identification of differentially expressed genes in sorghum (Sorghum bicolor) brown midrib mutants. Physiology of Plants, 146: 375–387. Yeshaneh, G. T (2015). Assessment of Farmers’ Perceptions about Soil Fertility with Different Management Practices in Small Holder Farms of Abuhoy Gara Catchemnt, Gidan District, North Wollo. American Journal of Environmental Protection, 3 (4):137-144. 185 University of Ghana http://ugspace.ug.edu.gh Ye X. D, Al-Babili, S., Kloti, A. Zhang, J., Lucca, P., Beyer, P. and Potrykus, I. (2000). Engineering the provitamin A (beta-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science, 287:303-305. Yilma T. (2006). Modeling farm irrigation decisions under rainfall risk in the White Volta Basin of Ghana. A tool for policy analysis at the farm-household level. Bonn, Göttingen, Germany: University of Göttingen, 170. Yoneyama, K., Xie, X., Kusumoto, D., Sekimoto, H., Sugimoto, Y. and Takeuchi, Y. (2007b) Nitrogen deficiency as well as phosphorus deficiency in sorghum promotes the production and exudation of 5-deoxystrigol, the host recognition signal for arbuscular mycorrhizal fungi and root parasites. Planta, 227:125–132. Young, K.J. and Long, S.P., (2000). Crop ecosystem responses to climatic change: maize and sorghum. In: Reddy, K.R., Hodges, H.F. (Eds.), Climate Change and Global Crop Productivity. CABI Publishing, Wallingford, 107–131. Zhang, F., Shen, J., Zhang, J., Zuo, Y., Li, L., Chen, X. (2010). Rhizosphere processes and management for improving nutrient use efficiency and crop productivity: implications for China. Advances in Agronomy, 107: 1–32. Zhao, J., Paulo, M.J., Jamar D, Lou P, van Eeuwijk F, Bonnema, G., Vreugdenhil, D., Koornneef, M. (2007). Association mapping of leaf traits, flowering time, and phytate content in Brassica rapa. Genome, 50: 963–973. Zhao, J., Mantilla Perez, M.B., Hu, J. and Salas Fernandez, M.G., (2016). Genome-Wide Association Study for Nine Plant Architecture Traits in Sorghum. The Plant Genome, 9 (2). Zhu, J.M. and Lynch, J.P. (2004). The contribution of lateral rooting to phosphorus acquisition efficiency in maize (Zea mays) seedlings. Functional Plant Biology, 31: 949–958. Zohary, D., and M. Hopf. 2000. Domestication of Plants in the Old World, 3rd edition. Oxford: University Press. Zomer, R.J., Trabucco, A., Coe, R., Place, F. (2009). Trees on farm: analysis of global extent and geographical patterns of agroforestry. ICRAF Working Paper 89. Nairobi, Kenya: World Agroforestry Centre. Zhou, K., Xia, J., Wang, Y., Ma, T., & Li, Z. (2017). A Young Seedling Stripe2 phenotype in rice is caused by mutation of a chloroplast-localized nucleoside diphosphate kinase 2 required for chloroplast biogenesis. Genetics and Molecular Biology, 40 (3):630–642. Zougmoré, R. (2003). Integrated water and nutrient management for sorghum production in semi- arid Burkina. Tropical resources management papers, 45, Wageningen University and Research center, Wageningen, the Netherlands. 205. 186 University of Ghana http://ugspace.ug.edu.gh APPENDIX Appendix 1: PRA questionnaire Questionnaire number -------------------------- Name of interviewer --------------------------- Date----------------------------------------------- District--------------------------------------------- Community-------------------------------------- Section A: Background information 1. Name of respondent: ………………………………………………………………………………. 2. Age range: 18-30 [ ] 31-45 [ ] 45-60 [ ] Above 60 [ ] 3. Gender: Male [ ] Female [ ] 4. Religion: Muslim [ ] Christian [ ] Traditional [ ] Other [ ] (Please specify)……………............ 5. Level of education: Basic Education [ ] Secondary [ ] Tertiary [ ] None [ ] 6. Marital status: Married [ ] Single [ ] Divorced [ ] Widowed [ ] 7. Number of children: ………………………………………………………………………………………. 8. Main occupation…………………………………………………………………….. 9. What is the main source of income for your family? Farming [ ] Artisanship [ ] 187 University of Ghana http://ugspace.ug.edu.gh Trading [ ] Casual labourer [ ] Others [ ] 10. Do you own your farm land? Yes [ ] No [ ] Section B: Sorghum importance, production and constraints 11. Which crops you grow on your farm? Millet [ ] Sorghum [ ] Groundnuts [ ] Yam [ ] Cowpea [ ] 12. Which crop provides the most revenue? Millet [ ] Sorghum [ ] Groundnuts [ ] Yam [ ] Cowpea [ ] 13. How long have you been growing sorghum? Up to 5 years [ ] 6-10 years [ ] 11-15 years [ ] Over 15 years [ ] 14. Which sorghum varieties do you grow? Kapaala [ ] Naga white [ ] Belko [ ] Naga red [ ] Other [ ] Please specify …………………………………………………….. 15. Which is your most-preferred variety? ...................................................................................... 16. Why do you prefer this variety? ………………………………………………………………. 17. Why do you grow sorghum? Household use only [ ] Income only [ ] Both [ ] 18. What is the total size of your farm? ………………………………………………………… 19. What proportion of your total land area is allocated to sorghum cultivation? ......................... 20. What is your estimated grain yield? ........................................................................................... 21. What problems do you face with sorghum cultivation? 188 University of Ghana http://ugspace.ug.edu.gh ……………………………………………………………………………………………. 22. Which of these problems are you most concerned about? List them in order of importance. Production constraints Please tick all that apply Ranking Lack of sufficient rain Pests (insects, birds, etc.) Diseases High cost of inputs Land unavailability Declining soil fertility Weeds Post-harvest losses Lack of seed Other …………………. 23. Where do you obtain your sorghum seeds from? Self/saved seeds [ ] Local market [ ] Other farmers [ ] NGOs [ ] Family [ ] Others [ ] Please Specify 24. Do you receive agricultural information (new technologies, improved seeds, fertiliser subsidies, etc.) regularly? Yes [ ] No [ ] 25. If yes, what are the sources of such information? Agric. extension officers [ ] NGOs [ ] Radio [ ] Television [ ] Research institutions [ ] Section C: Farmers’ knowledge, attitudes and perception of soil fertility 1. Can you identify fertile and poor soils? Yes [ ] No [ ] 2. How can you tell if soil is fertile? Soil color [ ] Soil moisture [ ] Soil texture [ ] Crop performance [ ] 189 University of Ghana http://ugspace.ug.edu.gh 3. What causes soil infertility? Continuous cropping [ ] Insufficient rain [ ] Type of crop cultivated [ ] 4. How will you describe the land on which you grow sorghum in terms of fertility? Highly fertile [ ] Moderately fertile [ ] Poor [ ] 5. How can you improve the fertility of your farmland?................................................... 6. Do you apply fertiliser to sorghum? Yes [ ] No [ ] 7. If yes, why? ……………………………………………………………………………. 8. If no, why not? …………………………………………………………………………. 9. If yes, which type of fertiliser do you use? Organic [ ] Inorganic [ ] Both [ ] 10. If organic, which type do you use? Animal droppings/manure [ ] Compost/mulch [ ] 11. If inorganic, which type do you use? ……………………………………………………. 12. When do you apply the fertiliser to the crop? At planting [ ] One month after planting [ ] Two months after planting [ ] Whenever I obtain some [ ] 13. How many bags of fertiliser do you apply to your sorghum crop? 1 bag [ ] 2 bags [ ] 3 bags [ ] 4 bags [ ] Other [ ] Please specify……………………………… 14. How often do you apply fertiliser? Once in the growing season [ ] Twice in the growing season [ ] Other [ ] Please specify……………………………………... 15. What factors affect your use of fertiliser? Cost [ ] Availability [ ] Lack of access to credit [ ] Other [ ] Please specify…………………………………………………………. 190 University of Ghana http://ugspace.ug.edu.gh 16. Do you think soil infertility affects sorghum? Yes [ ] No [ ] 17. If yes, how does it affect the crop? Reduced yield [ ] Stunted growth [ ] Poor seedling establishment [ ] Discolouration of leaves [ ] Other [ ] 18. Which of these cropping systems do you practice? Mono-cropping [ ] Crop rotation [ ] Intercropping [ ] Mixed cropping [ ] 19. Do you practice any of these? Shifting cultivation [ ] Retention of crop residue [ ] Erosion control [ ] 5. If yes, which erosion control measures do you use on your sorghum farm? Cover cropping [ ] Stone bunding [ ] Earth bunding [ ] Ridging [ ] Grass strips [ ] Section D: Farmers’ preference for traits 1. What traits do you prefer in a sorghum variety? ................................................................................................................................................... 191 University of Ghana http://ugspace.ug.edu.gh 2. List them in order of importance. Preference criteria Tick all that apply Ranking Early maturing High yielding Adaptation to poor soils Drought tolerance Grain quality (taste, brewing, cooking, etc.) Tolerance to diseases Tolerance to pests (birds, striga, aphids, etc.) Resistance to lodging 192 University of Ghana http://ugspace.ug.edu.gh Appendix 2: p- values for Pearson’s correlation coefficient of quantitative traits among 550 genotypes Variables PL PE PW PH DTF NL SG HSW GYP PL 0 < 0.0001 < 0.0001 < 0.0001 0.944 < 0.0001 < 0.0001 < 0.0001 < 0.0001 PE < 0.0001 0 < 0.0001 0.178 0.655 < 0.0001 < 0.0001 0.229 < 0.0001 PW < 0.0001 < 0.0001 0 < 0.0001 0.010 < 0.0001 < 0.0001 < 0.0001 < 0.0001 PH < 0.0001 0.178 < 0.0001 0 0.304 < 0.0001 0.004 < 0.0001 < 0.0001 DTF 0.944 0.655 0.010 0.304 0 0.545 0.013 0.146 0.215 NL < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.545 0 < 0.0001 < 0.0001 < 0.0001 SG < 0.0001 < 0.0001 < 0.0001 0.004 0.013 < 0.0001 0 0.000 < 0.0001 HSW < 0.0001 0.229 < 0.0001 < 0.0001 0.146 < 0.0001 0.000 0 < 0.0001 GYP < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.215 < 0.0001 < 0.0001 < 0.0001 0 193 University of Ghana http://ugspace.ug.edu.gh Appendix 3: Members of individual clusters in Figure 4.5 Cluster Members of the cluster Cluster I Mut0984 Mut0988 Mut1456#1 Mut2461 Mut0676 Mut1315 Mut0064 Mut1148 Mut0181#2 Mut0767 Mut0640 Mut0641 Mut0696 Mut1156 Mut2204#1 Mut3312 Mut0662 Mut0321#1 Mut0758 Mut0937 Mut0701 Mut2246#1 Mut1408#1 Mut4791#1 Mut0118 Mut0452#1 Mut0182 Mut0058 Mut1206 Mut1202 Mut1721#2 Mut2245#1 Mut0457#1 Mut0454#1 Mut1875 Mut1300#1 Mut1249 Mut1457#1 Mut1455#1 Mut0933 Mut0657 Mut1401#1 Mut0518-2 Mut1663#1 Mut2098#1 Mut1204 Mut1508#2 Mut1768 Mut0509 Mut0459#1 Mut1820 Mut1865 Mut0814 Mut1610 Mut1767 Mut1727 Mut1725 Mut1453#1 Mut1154 Mut3315 Mut0062 Mut0579 Mut0839#2 Mut0992 Mut0986 Mut3161 Mut1773 Mut1817-2 Mut0683 Mut1666 Mut0437#1 Mut0006 Mut0626 Mut0753 Mut1027#1 Mut1555#2 Mut1716 Mut1149 Mut0589 Mut1146 Mut2157 Mut1248 Mut0516 Mut0935 Mut1044 Mut0623 Mut1399#1 Mut0116 Mut0577 Mut1253 Mut1766-2 Mut1204#2 Mut1201 Mut0651 Mut0176 Mut1821-2 Mut0453#1 Mut0398#1 Mut1986#1 Mut1113#1 Mut0257#1 Mut0122 Mut2854#1 Mut3708 Mut2309 Mut2711 Mut3466 Mut2038#1 Mut2095#1 Mut0251#1 Mut0258#1 Mut3008 Mut3766 Mut3966 Mut3222 Mut3267 Mut0883 Mut0811 Mut1294#1 Mut1252 Mut3011#1 Mut4260-2 Mut3857 Mut3962#1 Mut4016#1 Mut0538 Mut0397#1 Mut2088#2 Mut2037#1 Mut3710 Mut4512 Mut3155 Mut4061#1 Mut4365 Mut2853#2 Mut1303#1 Mut1609 Mut0885 Mut3813 Mut2616 Mut2663 Mut4406#2 Mut2410 Mut3010 Mut2094#1 Mut0060 Mut0536 Mut2853#1 Mut2852#1 Mut2311 Mut4461 Mut5159 Mut3709 Mut0817 Mut1297#1 Mut0256#1 Mut3812 Mut4015#1 Mut3611 Mut2666#2 Mut3862 Mut3814 Mut3472 Mut1937#1 Mut2155#1 Mut4258 Mut4158 Mut3964#1 Mut3658 Mut4261 Mut4214 Mut0369#1 Mut1608 Mut1251 Mut2619 Mut2900#1 Mut4308 Mut2250#1 Mut4361#2 Mut2660 Mut0396#1 Mut0880 Mut2149#1 Mut3711#2 Mut2571 Mut2517 Mut3815 Mut2362 Mut3412 Mut1100 Mut0301#1 Mut0329#1 Mut4161-2 Mut2901#1 Mut2714 Mut4060#1 Mut2807 Mut2857#1 Mut1879#1 Mut1938#1 Mut0630 Mut4107#1 Mut2810 Mut3269 Mut3470 Mut3217 Mut3760#2 Mut2093#1 Mut2040#1 Mut3053 Mut3003#1 Mut2359 Mut3612 Mut4967 Mut3863 Mut3808 Mut1047 Mut0124 Mut3564 Mut3108 Mut3565 Mut2518 Mut2808 Mut3915 Mut2576 Mut1302#1 Mut0007 Mut2661 Mut3715 Mut4013#1 Mut2952#1 Mut2950#1 Mut2513 Mut5095 Mut1930#1 Mut2198#1 Mut3042 Mut3809 Mut4310 Mut2306 Mut4404 Mut3415 Mut2858#1 Mut2091#1 Mut0542 Mut3763 Mut2575 Mut2572 Mut3657 Mut3468 Mut2806 Mut2955#1 Mut1611 Mut0123 Mut3765 Mut4625 Mut2809-2 Mut4062#1 Mut2954#1 Mut3864 Mut2759 194 University of Ghana http://ugspace.ug.edu.gh Mut0818 Mut0271#1 Mut2262 Mut3568 Mut2903#1 Mut4410 Mut4057#1 Mut3914 Mut2251#1 Mut4414 Mut4508 Mut4063#1 Mut3659 Mut2307 Mut4465 Mut4367 Mut3414 Mut4363#2 Mut4309 Mut2665 Mut2360 Mut3268 Mut3610 Mut2313 Mut4459#2 Mut2805 BTx623 Mut3310 Mut2904#1 Mut2248#1 Mut4017 Mut4012#1 Mut2712 Mut3469 Mut3514 Tx631 Cluster II Mut0982 Mut1405#1 Mut1614 Mut1989#1 Mut4211 Mut4156 Mut4509 Mut3262 Mut2354 Mut1668 Mut0575 Mut0262#2 Mut4602#1 Mut3318#2 Mut3364 Mut2756 Mut0008 Mut0671 Mut1613-2 Mut2034#2 Mut3368 Mut2899#1 Mut4110#2 Mut3859 Mut0517 Mut1353#1 Mut1094 Mut1960#2 Mut4559#1 Mut3092 Mut4507 Mut4366 Mut1509 Mut0664 Mut0588 Mut0889 Mut4259-2 Mut3761 Mut3366 Mut2457 Mut1726 Mut3311 Mut1092 Mut1510 Mut4558#2 Mut4511 Mut2573 Mut2715 Mut2463 Mut1352#1 Mut1935#1 Mut1557 Mut2617#2 Mut3362 Mut4264 Mut3213 Mut0985 Mut1838 Mut0178 Mut2662 Mut3712 Mut2850#1 Mut0317 Mut3916 Mut0936 Mut1665 Mut0179 Mut2358 Mut4219 Mut3058 Mut2574 Mut3419 Mut0644 Mut0596 Mut2199#1 Mut3160 Mut3609 Mut4014#1 Mut0641 Mut3663 Mut1724 Mut1348#1 Mut1607 Mut3009#1 Mut3110 Mut4312 Mut3157 Mut3517 Mut0646 Mut1876#1 Mut0275#2 Mut4212 Mut3810 Mut4510 Mut2713 Mut4364 Mut1200 Mut2097#1 Mut1564 Mut3112 Mut4554#2 Mut3363 Mut3959#2 Mut3707 Mut1771-2 Mut1661 Mut1562 Mut2856#2 Mut2620 Mut2811 Mut4409 Mut3516 Mut2205#2 Mut0576 Mut0180 Mut3007#1 Mut4549#1 Mut3057 Mut5850 Mut3512 Mut0460#1 Mut1101 Mut0568 Mut4213 Mut3105 Mut3563 Mut4405 Mut0881 Mut2757 Mut4411#2 Mut3520 Mut3162 Mut3515-2 195 University of Ghana http://ugspace.ug.edu.gh Cluster III Mut1504 Mut4768#1 Mut0674 Mut0394#1 Mut1254 Mut3094 Mut4105 Mut0121 Mut0764#2 Mut1043#1 Mut0682 Mut1305#1 Mut1984#1 Mut4557#1 Mut3963#1 Mut5115 Mut0120 Mut0647 Mut0456#1 Mut1658#2 Mut2039 Mut3004#1 Mut4106#1 Mut2716 Mut1458#1 Mut0697 Mut1884 Mut0393#1 Mut0882 Mut4159 Mut2363 Mut3860 Mut1505 Mut1199 Mut2203#1 Mut1877#1 Mut1561 Mut3567 Mut2409#2 Mut4460 Mut0755 Mut0932 Mut1669 Mut0323#1 Mut1150 Mut4313-2 Mut5398#1 Mut3434 Mut0643 Mut1774 Mut0672 Mut0005 Mut1991#1 Mut2570 Mut3960#1 Mut3662 Mut0001 Mut1046 Mut4848 Mut1931 Mut2147#1 Mut3762 Mut4463 Mut3218 Mut0514 Mut0931 Mut2247-2 Mut0059 Mut0395#1 Mut4155 Mut2949#1 Mut5199 Mut0700 Mut0678 Mut2142#1 Mut0878 Mut1250 Mut4112 Mut2404 Mut5588 Mut0512 Mut2205#1 Mut0056 Mut0580-2 Mut1992#1 Mut4311-2 Mut3158 Mut3911#2 Mut1454#1 Mut0652 Mut1147 Mut0316#1 Mut3764 Mut2902#1 Mut2514 Mut3221 Mut1769 Mut0679 Mut1350#1 Mut2041#1 Mut4218 Mut4108#1 Mut4059#1 Mut2761 Mut1041#1 Mut0458#1 Mut0002 Mut0561 Mut1864 Mut3115 Mut4058#1 Mut3220 Mut0934 Mut1818-2 Mut0813 Mut0061#2 Mut3005 Mut2906#1 Mut2364#2 Grinkan 196 University of Ghana http://ugspace.ug.edu.gh Appendix 4: Procedures for soil characterization and analysis Bulk density Core samplers of approximately 75mm long and 75 mm internal diameter were driven into the soil until the top edge of the sampler was a few a millimeters below the soil surface and the inner cylinder was filled. The sampler was carefully removed in order to preserve the samples. The two cylinders were separated, retaining the ‘undisturbed’ soil in the inner cylinder. Metal disks were placed at each end of the cylinder and placed in a plastic bag. The process was repeated with other cores in close proximity so obtain 3 replicates. Soil samples were transported to the laboratory carefully and oven-dried at 105oC for 24 hours. The weight of the oven-dried samples was noted. The volume of the core sampler was estimated using the diameter and height. Bulk density was calculated using the following formula. Bulk density (Mg/m3) = Mass of dry soil (g)/ Volume of core (cm3) Particle size distribution One hundred ml of 5% calgon (sodium hexametaphosphate) was added to 40 g of fine earth fraction of soil in a plastic bottle. The content of the bottle was thoroughly mixed on a mechanical shaker for 2 hours. The mixture was then transferred to a measuring cylinder and topped to the 1 litre mark. The suspension was then agitated for 5 minutes after which the density was measured using a hydrometer. Measurement of the density of the suspension was repeated after 8 hours. Temperature of the suspension during the 5 minute and 8 hour hydrometer readings was recorded as T1 and T2. The contents of the cylinder were emptied into a sieve of pore size 47 µm. The soil sample (sand) retained on the sieve was washed and dried for 24 hours at 105oC. Subsequently, the dry weight of the sand was recorded. Blank sample hydrometer readings at 5 minutes and 8 197 University of Ghana http://ugspace.ug.edu.gh hours were also recorded for the 5% calgon. Particle size distribution was then determined using the formula below; % clay and silt = (5 minute reading – correction for temperature) / oven dry sample x 100 % clay = (8 hour reading – correction for temperature) / oven dry sample x 100 % silt = % clay and silt - % clay % sand = (oven dry weight of particles retained on the sieve)/ oven dry mass of sample x 100 The effect of temperature on density of the soil particles was accounted for using the relation provided by Day (1965). For every 1 oC increase in temperature above 19.5 oC, there is an increase of 0.3 in the density of the particles in suspension. Correction for temperature = blank hydrometer reading – increase in weight of particles Soil moisture at field capacity A 500 g soil sample was saturated with water in a filter paper-lined perforated plastic container of 15 cm in diameter and 20 cm in height. The plastic container was covered to prevent evaporation and allowed to drain for 3 days. The soil sample was oven-dried at 105 oC for 24 hours. The moisture content of the soil sample at field capacity was determined as the difference in mass between moist soil and oven-dried soil. Soil pH Ten millilitres of distilled water was added to 10 g of soil was to form a suspension. The suspension was stirred vigorously and allowed to stand for 30 minutes. The pH was measured by inserting the microprocessor pH 213 meter into the supernatant of the soil solution. 198 University of Ghana http://ugspace.ug.edu.gh Soil organic carbon This was determined using the wet combustion method (Walkley and Black, 1934). Soil sample was prepared by sieving through a 0.5 mm sieve. In an Erlenmeyer flask, 10 ml of 0.167 M dichromate (K2Cr2O7) solution and 20 ml of sulphuric acid was added to 0.5 g of the sieved soil. The flask was swirled after which the mixture was allowed to stand for 30 minutes. The unreduced K2Cr2O7 in the sample was titrated with 0.2 M ferrous ammonium sulphate solution. The bright green end point was reached after the addition of 10 ml of orthophosphoric acid and 2 ml of barium diphenylamine sulphonate indicator. A standardized titration of the K2Cr2O7 with ferrous ammonium sulphate was also done. The amount of organic carbon was calculated as the difference between the number of moles of K2Cr2O7 present in the standardized titration and the number of moles of the unreduced K2Cr2O7. Total nitrogen Total N of the fine earth was determined by using the Kjeldahl digestion procedure (Anderson and Ingram, 1998). In a digestion tube, 5 ml of concentrated sulphuric acid was added to1 g of fine earth. The mixture was heated for 30 minutes after which 2 ml of hydrogen peroxide was added. The temperature was increase to 360 o C and maintained till the mixture was a colourless solution. Distilled water was added to the cooled digest to the top of a 100 ml volumetric flask. Twenty millilitres of the digest was transferred to a distillation flask followed by the addition of 10 ml of 40 % NaOH. After distillation, the ammonia released was condensed and 199 University of Ghana http://ugspace.ug.edu.gh Total and available phosphorus This was determined by the Bray1method (Bray and Kurtz, 1945). Fifty millilitres of Bray’s extractant (0.03 M NH4F and 0.025 M HCl) was added to 5 g of soil sample. The mixture was shaken for 5 minutes and allowed to stand. The supernatant was passed through filter paper in order to obtain a clear solution. For colour development, a drop each of P-nitrophenol and ammonium hydroxide was added to 1 ml of the clear sample solution in a volumetric flask. Total phosphorus was determined by digesting biochar (0.2 g) with 25ml of a mixture of HNO3 and 60% HClO4 in a ratio of 1:1:5. Distilled water was added to the filtered digest to a volume of 100 ml. The amount of phosphorus in the digest was determined as described by Murphy and Riley (1962) and the P concentration calculated. Exchangeable bases In an extraction bottle, 50 ml of 1 M ammonium acetate was added to 5 g of fine earth. The solution was then filtered through Whatman filter paper. Calcium and magnesium levels in the extract was determined using an atomic absorption spectrophotometer while sodium and potassium levels were estimated on a flame photometer. Appendix 5: Determination of phosphorus in leaf and grain samples Leaf and grain samples were analysed for P content were milled separately in a blender. Total digestion was carried on one gram of ground plant material using the Bhargava and Raghupathi (1984). Aliquots of the resultant solutions were used for the determination of P concentration. P concentration (total amount of phosphorus) in the plant samples in mg/g was determined using the Murphy and Riley (1962) method. The same protocol was used for determining P concentration in grain samples. 200 University of Ghana http://ugspace.ug.edu.gh Appendix 6: Names of putative mutants (codes) used in PCA biplot of 200 genotypes as shown in Figure 5.3 Genotype Code Genotype Code Genotype Code Genotype Code Genotype Code Mut1727 M1 Mut4107#1 M41 Mut2572 M81 Mut2250#1 M121 Mut0058 M161 Mut3108 M2 Mut2620 M42 Mut4012#1 M82 Mut4060#1 M122 Mut0120 M162 Mut2463 M3 Mut3779 M43 Mut4108#1 M83 Mut4967 M123 Mut0056 M163 Mut0936 M4 Mut3003#1 M44 Mut2573 M84 Mut2950#1 M124 Mut0460#1 M164 Mut1455#1 M5 Mut3809 M45 Mut3115 M85 Mut3468 M125 Mut0056 M165 Mut1769 M6 Mut4625 M46 Mut2665 M86 Mut2954#1 M126 Tx430 Tx430 Mut1771-2 M7 Mut3568 M47 Mut2309 M87 Mut3262 M127 CSM63 CSM63 Mut2205#2 M8 Mut2857#1 M48 Mut3857 M88 Mut4459#2 M128 TxARG1 TxARG1 Mut1817-2 M9 Mut3105 M49 Mut4264 M89 Mut3469 M129 FeteritaGishesh FeteritaGishesh Mut1669 M10 Mut3760#2 M50 Mut0317 M90 Mut3659 M130 Tx631 Tx631 Mut0454#1 M11 Mut3808 M51 Mut2574 M91 Mut3268 M131 ICSV1049 ICSV1049 Mut1665 M12 Mut2858#1 M52 Mut3964#1 M92 Mut5115 M132 BN223 BN223 Mut0393#1 M13 Mut2955#1 M53 Mut4105 M93 Mut3859 M133 Tx2752 Tx2752 Mut1614 M14 Mut3708 M54 Mut4308 M94 Mut4017 M134 SC599 SC599 Mut0369#1 M15 Mut3766 M55 Mut2517 M95 Mut4366 M135 Tx436 Tx436 Mut1254 M16 Mut4156 M56 Mut3269 M96 Mut3466 M136 Framida Framida Mut1253 M17 Mut4260-2 M57 Mut3612 M97 Mut2716 M137 MR732 MR732 Mut0538 M18 Mut4512 M58 Mut4106#1 M98 Mut3267 M138 Mota Maradi Mota Maradi Mut0395#1 M19 Mut2616 M59 Mut2363 M99 Mut3860 M139 Segeolane Segeolane Mut3160 M20 Mut3318#2 M60 Mut2713 M100 Mut2410 M140 Sariaso Sariaso Mut3564 M21 Mut2852#1 M61 Mut2952#1 M101 Mut3434 M141 HoneyDrip HoneyDrip Mut3764 M22 Mut2899#1 M62 Mut2306 M102 Mut3213 M142 Grinkan Grinkan Mut3112 M23 Mut3092 M63 Mut3657 M103 Mut3862 M143 CE151262A1 CE151262A1 Mut2856#2 M24 Mut4015#1 M64 Mut3959#2 M104 Mut3916 M144 El-Mota El-Mota Mut4211 M25 Mut4313-2 M65 Mut4062#1 M105 Mut4361#2 M145 Wassa Wassa Mut3368 M26 Mut4511 M66 Mut4465 M106 Mut3419 M146 PI609567 PI609567 201 University of Ghana http://ugspace.ug.edu.gh Genotype Code Genotype Code Genotype Code Genotype Code Genotype Code Mut3709 M27 Mut2570 M67 Mut2409#2 M107 Mut3663 M147 Ajabsido Ajabsido Mut4214 M28 Mut2850#1 M68 Mut2313 M108 Mut3217 M148 MDK MDK Mut3412 M29 Mut3762 M69 Mut5398#1 M109 Mut4364 M149 KuyumaWSV387 KuyumaWSV387 Mut2854#1 M30 Mut4155 M70 Mut3960#1 M110 Mut3415 M150 KoroKollo KoroKollo Mut3008 M31 Mut4510 M71 Mut4063#1 M111 Mut3707 M151 Sepon82 Sepon82 Mut3011#1 M32 Mut2571 M72 Mut4409 M112 Mut3218 M152 Theis Theis Mut3813 M33 Mut3363 M73 Mut2360 M113 Mut3864 M153 SC35 SC35 Mut4218 M34 Mut3057 M74 Mut2248#1 M114 Mut0881 M154 Seguetana Seguetana Mut4259-2 M35 Mut3563 M75 Mut3222 M115 Mut3414 M155 Kadaga West Kadaga West Mut1864 M36 Mut4112 M76 Mut5850 M116 Mut2757 M156 Mut0988 M166 Mut3005 M37 Mut4311-2 M77 Mut3962#1 M117 Mut3514 M157 Mut0839#2 M167 Mut3712 M38 Mut2902#1 M78 Mut4061#1 M118 Mut2759 M158 Mut2091#1 M168 Mut3810 M39 Mut2359 M79 Mut2404 M119 Mut1725 M159 Mut2098#1 M169 Mut4161-2 M40 Mut4310 M80 Mut3158 M120 Mut0001 M160 Mut0122 M170 202 University of Ghana http://ugspace.ug.edu.gh Appendix 7: Distribution of 200 genotypes in three clusters – Figure 5.4 Cluster I Mut1727 Mut1864 Mut2571 Mut4062#1 Mut3514 Mut3764 Mut3318#2 Mut4264 Mut2463 Mut3810 Mut3363 Mut4465 Mut0056 Mut3112 Mut3092 Mut0317 Mut1769 Mut2620 Mut4112 Mut2409#2 Mut0056 Mut2856#2 Mut4313-2 Mut2574 Mut1771-2 Mut3568 Mut4311-2 Mut5398#1 Tx631 Mut4211 Mut2850#1 Mut4105 Mut2205#2 Mut2857#1 Mut2902#1 Mut3960#1 SC599 Mut3762 Mut4308 Mut4366 Mut1817-2 Mut3766 Mut4012#1 Mut4063#1 Mut3368 Mut4155 Mut4106#1 Mut3213 Mut1614 Mut4156 Mut3115 Mut4409 Mut4214 Mut4510 Mut2713 Mut3218 Mut0369#1 Mut4512 Mut3857 Mut5850 Mut4218 Mut3962#1 Mut3158 Mut2250#1 Mut4017 Cluster II Mut3108 Mut0538 Mut4259-2 Mut3760#2 Mut3563 Mut3612 Mut4060#1 Mut3268 Mut0936 Mut0395#1 Mut3005 Mut3808 Mut2359 Mut2363 Mut4967 Mut5115 Mut1455#1 Mut3160 Mut3712 Mut2858#1 Mut4310 Mut2952#1 Mut2950#1 Mut3466 Mut1669 Mut3564 Mut4161-2 Mut2955#1 Mut2572 Mut2306 Mut3468 Mut2716 Mut0454#1 Mut3709 Mut4107#1 Mut3708 Mut4108#1 Mut2360 Mut2954#1 Mut3267 Mut1665 Mut3412 Mut3779 Mut4260-2 Mut2665 Mut2248#1 Mut3262 Mut3860 Mut0393#1 Mut2854#1 Mut3809 Mut2616 Mut3964#1 Mut3222 Mut4459#2 Mut2410 Mut1254 Mut3008 Mut4625 Mut2852#1 Mut2517 Mut4061#1 Mut3469 Mut3434 Mut1253 Mut3813 Mut3105 Mut2570 Mut3269 Mut2404 Mut3659 Mut3862 Mut3916 Mut3663 Mut3415 Mut2757 Mut0001 Tx430 BN223 Framida Mut4361#2 Mut3217 Mut3864 Mut2759 Mut0058 TxARG1 Tx2752 MR732 Mut3419 Mut4364 Mut3414 Mut1725 Mut0120 ICSV1049 Tx436 Grinkan CE151262A1 SC35 KuyumaWSV387 Mut0839#2 Mut2098#1 MDK Mut0988 Sepon82 Mut2091#1 Mut0122 Cluster III Mut3011#1 Mut4015#1 Mut2573 Mut3959#2 Mut3707 CSM63 Segeolane El-Mota Mut3003#1 Mut4511 Mut2309 Mut2313 Mut0881 FeteritaGishesh Sariaso Wassa 203 University of Ghana http://ugspace.ug.edu.gh Mut2899#1 Mut3057 Mut3657 Mut3859 Mut0460#1 Mota Maradi HoneyDrip PI609567 Ajabsido KoroKollo Theis Seguetana Kadaga West 204 University of Ghana http://ugspace.ug.edu.gh 205