BREEDING FOR GRAIN QUALITY TRAITS IN COWPEA [VIGNA UNGUICULATA (L) WALP] BY MUHAMMAD LAWAN UMAR (10325389) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF DOCTOR OF PHILOSOPHY PLANT BREEDING DEGREE WEST AFRICA CENTER FOR CROP IMPROVEMENT SCHOOL OF AGRICULTURE COLLEGE OF BASIC AND APPLIED SCIENCES UNIVERSITY OF GHANA LEGON December, 2014 University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh ii ABSTRACT This study was conducted in the Northern guinea and Sudan savannas of Zaria and Kano, Nigeria to: (i) identify cowpea production constraints and assess farmer and consumer perceptions and preferences on grain quality traits in cowpea, (ii) assess the variability of grain nutritional values (protein, iron and zinc contents) of cowpea accessions from Africa and other parts of the world, (iii) determine the mode of inheritance of iron and zinc contents in cowpea grains and (iv) determine the relationship between zinc concentration and yield components. Farmers pointed out inadequate improved cowpea seed at planting time (rainy season) and lack of resistant varieties to pod sucking and pod boring insects as the major constraints to cowpea production in the areas. Farmers preferred a cowpea variety that combines high yield with rough texture, white or brown colour. Consumers‘ grain quality preferences were consistent with those of farmers, except for women who prioritized cooking time and oil consumption. Significant genetic variations were observed in: zinc content (1.01 to 329.15 mg/kg); iron content (10.01 to 386.3 mg/kg); protein content (1.72 to 29.93%) and other physicochemical properties of grain. Many of these variations can be generated by conventional breeding methods to address the nutritional needs in developing countries. In addition, 3 accessions: TVu-13088, TVu-13495 and TVu-9725 that combined the largest number of desirable nutritional attributes were identified which can be nominated for anti-nutritional factor testing prior to recommendation for infant diets formulation and other use. For nutrient enhancement, five accessions each were identified for protein, zinc and iron contents. Genetic diversity of 169 cowpea accessions using 119 SNP markers clustered the accessions into two main groups on genetic distances (0.00 to 0.212) with small genetic differentiation (0.26 to 0.45) between African and USA cowpea accessions. This indicates that the entire genetic diversity in the African germplasm might already have spread over cowpea- growing regions in the world as a whole, though not completely within any single region. The University of Ghana http://ugspace.ug.edu.gh iii Polymorphism Information Content (PIC) values ranged from 0.2366 (7344_500 SNP) to 0.427 in two SNP markers (4749_1972 and 14929_258). Weak negative correlation existed between iron content and fat (r= -0.18, P < 0.007), iron content and carbohydrate (r= -0.18, P < 0.007) but iron content was positively correlated with protein (r= 0.26, P < 0.001) content. Fat content was negatively correlated with ash content (r= -0.13, P < 0.05) and protein content (r= -0.85, P < 0.001). Ash content correlated positively with protein (r= 0.14, P < 0.05) and negatively with carbohydrate (r=-0.23, P < 0.001). Protein was negatively correlated with carbohydrate (r= -0.29, P < 0.01). Zinc concentration showed weak negative significant (r= 0.03, P < 0.05) correlation with number of pods per plant. No significant correlation was observed between zinc concentration and grain weight. Similarly negative correlation was observed between zinc content and number of pod per plant implying an increase in zinc content may lead to decrease in pod yield. Generation mean analysis of the six basic generations was significant (P < 0.001) among the generations. The significant variation of generation mean performance with transgressive segregants among the progenies and two fold increases in mean zinc concentration of F2 population over the high zinc content parent (82.54 mg/kg > 37.70 mg/kg), implies that cowpea can be enhanced with essential micronutrients using conventional approach. Additive [a] and additive by dominance [ad] model explained the inheritance of iron and zinc content in cowpea grain. An indication of maternal effect was observed in iron content inheritance indicating non suitability to make selection in early generation of selfing. Both seed weight and plant height are predominantly under complementary gene actions in this study, suggesting the possibility of considerable amount of heterosis for seed weight and fodder yield which are the determinants for choosing cowpea varieties,. The study revealed that: Variability of grain nutritional traits (Zn,Fe and Protein) among cowpea was observed. Genetic pattern of iron and zinc content elucidated and Possibility of enhancement in cowpea using conventional approach is realized. University of Ghana http://ugspace.ug.edu.gh iv DEDICATION To the entire family of late Alhaji Salihu Nuhu Umar Bagobiri, Tudun wada Dankadai. University of Ghana http://ugspace.ug.edu.gh v ACKNOWLEDGEMENT This study was made successful by a grant from Alliance for a Green Revolution in Africa (AGRA). I gratefully appreciate the support of my supervisory team: Professor I. K. Asante, Professor F. Kumaga, Dr. M. E. Yeboah and Professor M.F. Ishiyaku. The untiring efforts of Professor I. K. Asante and Professor M. F. Ishiyaku toward the successful completion of this work by following up through phone calls and email are well acknowledged My profound gratitude to the administrative staff of West Africa Centre for Crop Improvement (WACCI) for their support and understanding; they really made me feel at home. I would also like to thank Ahmadu Bello University (ABU), Zaria and Institute for Agricultural Research (IAR) for granting me study leave to pursue this program at the University of Ghana, Legon and use of its facilities (laboratory, research farm and screen house). The mentorship of Dr. J. D Ehlers of Bill and Melinda Gates Foundation (BMGF) toward proposal development, Single Nucleotide Polymorphism (SNP) markers selection is well acknowledged. I am also grateful to the genotyping Support System (GSS) of the Generation Challenge Program (GCP) via Drs. N.N. Diop and C. He and Biotechnology research program of the IAR for their genotyping support. The use of cowpea accessions from cowpea breeding unit of the IAR, Samaru and International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria is sincerely acknowledged. Most importantly I thank my wife, Mansura Said Salihu; my daughters; Safiyyah and Rukaiyyah and the entire family for their patience and prayers for the successful completion of this course of study. University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENTS DECLARATION ............................................................................................................................. i ABSTRACT .................................................................................................................................... ii DEDICATION ............................................................................................................................... iv ACKNOWLEDGEMENT .............................................................................................................. v TABLE OF CONTENTS ............................................................................................................... vi LIST OF TABLES .......................................................................................................................... x LIST OF FIGURES ..................................................................................................................... xiii LIST OF ABREVIATIONS ......................................................................................................... xv CHAPTER ONE ............................................................................................................................. 1 1 GENERAL INTRODUCTION ................................................................................................ 1 CHAPTER TWO ............................................................................................................................ 6 2.0 LITERATURE REVIEW ..................................................................................................... 6 2.1 Cowpea [Vigna unguiculata L. (Walp)] ............................................................................... 6 2.2 Cowpea grain quality traits ................................................................................................... 7 2.2.1 Grain size ....................................................................................................................... 8 2.2.2 Grain coat colour............................................................................................................ 9 2.2.3 Grain coat texture ........................................................................................................... 9 2.3 Inheritance of Grain Quality Traits ..................................................................................... 10 2.3.1 Inheritance of Grain Size ............................................................................................. 10 2.3.2 Inheritance of grain coat colour ................................................................................... 10 2.3.3 Inheritance of grain coat texture .................................................................................. 11 2.3.4 Inheritance of protein content ...................................................................................... 11 2.3.5 Inheritance of iron and zinc contents ........................................................................... 12 2.4 Association between Nutritional Quality traits ................................................................... 12 2.5 Physiological mechanisms of iron and zinc uptake ............................................................ 13 University of Ghana http://ugspace.ug.edu.gh vii 2.6 Marker Assisted Selection (MAS) for cowpea improvement ............................................. 14 2.6.1 Current trends in the use of marker technologies for cowpea breeding ...................... 14 2.6.2 Quantitative trait loci mapping .................................................................................... 15 2.6.3 Concept and principles of QTL.................................................................................... 16 2.6.4 Methods used in QTL detection ................................................................................... 17 CHAPTER THREE ...................................................................................................................... 19 3.0 APPRAISAL OF FARMER AND CONSUMER PREFERENCES FOR COWPEA IN THE SUDAN SAVANNA ZONE OF NIGERIA .................................................................... 19 3.1 Introduction ......................................................................................................................... 19 3.2 Materials and Methods ........................................................................................................ 21 3.2.1 Description of the study area ....................................................................................... 21 3.2.2 Site mapping and selection .......................................................................................... 22 3.2.3 Sampling Procedure ..................................................................................................... 22 3.2.4 Data collection techniques ........................................................................................... 23 3.3 Data analysis ....................................................................................................................... 24 3.4 Results ................................................................................................................................. 25 3.4.1 Constraints to cowpea production in the study areas ................................................... 25 3.4.2 Farmer‘s varietal preferences ....................................................................................... 26 3.4.3 Farmers‘ preference ranking for grain quality traits .................................................... 28 3.4.4 Farmers‘ perception on grain quality attributes ........................................................... 29 3.4.5 Problems associated with Cowpea Marketing ............................................................. 32 3.5 Discussion ........................................................................................................................... 33 3.6 Conclusion and Recommendations ..................................................................................... 38 CHAPTER FOUR ......................................................................................................................... 40 4. VARIATION AMONG COWPEA ACCESSIONS FOR GRAIN QUALITY TRAITS AS REVEALED BY MORPHOLOGICAL, BIOCHEMICAL AND SINGLE NUCLEOTIDE POLYMORPHISM (SNP) MARKERS ................................................................................... 40 4.2 Materials and Methods ........................................................................................................ 42 4.2.1 Source of cowpea accessions ....................................................................................... 42 4.2.2 Soil sampling and analyses .......................................................................................... 45 4.2.3 Phenotypic characterization of the accessions ............................................................. 47 4.2.3.1 Physicochemical analyses of cowpea grain .......................................................... 47 University of Ghana http://ugspace.ug.edu.gh viii 4.2.3.2 Iron and zinc content determination ..................................................................... 50 4.2.4 Molecular analysis ....................................................................................................... 52 4.2.4.1 Characterization of cowpea accessions using SNP markers ................................. 52 4.3 Data analysis ....................................................................................................................... 53 4.4 Results ................................................................................................................................. 55 4.4.1 Soil physicochemical properties of the trial site .......................................................... 55 4.4.2 Variations among cowpea accessions at morphological and molecular level ............. 57 4.4.2.1 Descriptive analysis of growth habit and grain quality traits for 200 cowpea Accessions......................................................................................................................... 57 4.4.2.2 Variation in iron, zinc and protein contents within 153 African cowpea accessions ........................................................................................................................................... 63 4.4.2.3 Variation in physicochemical properties of the United States and three Asian cowpea accessions ............................................................................................................ 69 4.4.3 Principal Component Analysis .................................................................................... 71 4.4.4 Correlations among nutritional quality traits of 200 cowpea accessions ..................... 72 4.4.5 Allelic variation based on SNP markers ...................................................................... 73 4.4.6 Genetic distance ........................................................................................................... 77 4.4.7 Comparison between Morphological, Biochemical and Molecular Characterization . 80 4.5 Discussion ........................................................................................................................... 80 4.6 Conclusion .......................................................................................................................... 86 CHAPTER FIVE .......................................................................................................................... 88 5 INHERITANCE OF IRON AND ZINC CONTENT AND OTHER GRAIN QUALITY TRAITS IN COWPEA ................................................................................................................. 88 5.1 Introduction ......................................................................................................................... 88 5.2 Material and Methods ......................................................................................................... 90 5.2.1 Parent Materials ........................................................................................................... 90 5.2.2 Population development............................................................................................... 91 5.2.3 Evaluation of Iron and Zinc concentration from the two sets of six generations ........ 92 5.2.4 Data collection ............................................................................................................. 92 5.3 Data Analysis ...................................................................................................................... 94 5.4 Results ................................................................................................................................. 96 5.4.1 Generation Means for Zinc Concentration, Grain weight and Plant Height ................ 96 University of Ghana http://ugspace.ug.edu.gh ix 5.4.2 Generation means for seed iron concentration, seed weight and plant height in a cross between low iron (TVu-1) and high iron (TVu-999) parents ............................................. 104 5.4.3 Yield performance among F3 population evaluated under screen house conditions . 111 5.4.4 Correlation between Zinc concentration and yield related traits in cowpea evaluated under screen house conditions ............................................................................................ 116 5.5 Discussion ......................................................................................................................... 116 5.6 Conclusion and recommendations .................................................................................... 124 CHAPTER SIX ........................................................................................................................... 126 6 GENERAL DISCUSSION ...................................................................................................... 126 6.1 Participatory Rural Appraisal (PRA) ................................................................................ 126 6.2 Variation among cowpea accessions for grain quality traits ............................................ 127 6.3 Inheritance of iron and zinc concentration and other seed quality traits in cowpea ......... 128 6.4 Challenges ......................................................................................................................... 129 6.5 Recommendation .............................................................................................................. 130 BIBLIOGRAPHY ....................................................................................................................... 131 APPENDICES ............................................................................................................................ 150 University of Ghana http://ugspace.ug.edu.gh x LIST OF TABLES Table 3.1: Number of farmers and consumers interviewed in selected cowpea growing and consuming communities ............................................................................................................... 23 Table 3.2: Cowpea production constraints faced by farmers in Sudano-Sahelian zone of Nigeria ....................................................................................................................................................... 25 Table 3.3: Farmers varietal preference ranking by community .................................................... 27 Table 3.4: Farmers preference ranking of cowpea variety traits .................................................. 28 Table 3.5: Farmers priorities for traits in new cowpea varieties ................................................... 28 Table 3.6: Consumers‘ perception on grain quality attributes ...................................................... 29 Table 3.7: Consumers preferences ranking for grain quality traits ............................................... 30 Table 3.8: Consumer perceptions on grain quality attributes ....................................................... 31 Table 3.9: Ranking for consumer preferences for most frequently consumed diets .................... 32 Table 4.1: List, origin and source of 200 cowpea accessions selected in this study .................... 43 Table 4.1: (Cont‘d) List, origin and source of 200 cowpea accessions selected in this study ...... 44 Table 4.2 Scoring system used to characterize the 200 cowpea accessions (Mahalakshmi et al, 2007) ............................................................................................................................................. 52 Table 4.3: Soil physicochemical properties of the evaluation site in 2013 and 2014 rainy seasons ....................................................................................................................................................... 56 Table 4.4 Different set of accessions representing top 10, least 10, overall mean, standard deviation and ranges of iron, zinc and protein contents of 153 African cowpea accessions ........ 68 Table 4.5: Mean physicochemical properties of the United State and three Asian cowpea accessions ...................................................................................................................................... 69 Table 4.5: (Cont‘d) Mean physicochemical properties of the United State and three Asian cowpea accessions ........................................................................................................................ 70 University of Ghana http://ugspace.ug.edu.gh xi Table 4.6: Principal Component Analysis showing the contribution of grain quality traits to the total variation among the cowpea accessions ............................................................................... 72 Table 4.7: Pearson‘s correlation coefficients among nutritional quality traits of 200 cowpea accessions ...................................................................................................................................... 73 Table 4.8: Allelic frequency, gene diversity and polymorphic information content (PIC) .......... 75 Table 5.1: Number of progenies generated from two set of crosses for iron, zinc and grain coat colour studies ................................................................................................................................ 96 Table 5.2: Analysis of variance for the zinc concentration, seed weight and plant height in a cross between TVu-14845 and TVu-15251 ........................................................................................... 96 Table 5.3: Means of families derived from a cross involving TVu-14845 (low zinc) and TVu- 15251 (high zinc) cowpea parents .............................................................................................. 100 Table 5.4 Estimate of gene effects (±SE means) for zinc, 100 grain weight and plant height in a cross between TVu-14845 and TVu-15251 ................................................................................ 103 Table 5.5 Genetic variances, heritability and effective factors for zinc content seed weight and plant height in a cross involving TVu-14845 and TVu-15251 ................................................... 104 Table 5.6 Mean squares for iron concentration, 100-grain weight and plant height in a cross between TVu-1 and TVu-999 ..................................................................................................... 104 Table 5.6: Mean squares for iron concentration, 100-grain weight and plant height in a cross between TVu-1 and TVu-999 ..................................................................................................... 106 Table 5.7 Means of families derived from a cross involving TVu-1 (low iron) and TVu-999 (high iron) cowpea parents ................................................................................................................... 107 Table 5.8 Estimate of gene effects (±SE means) for zinc, 100 grain weight and plant height in a cross between TVu-1 and TVu-999 ............................................................................................ 109 University of Ghana http://ugspace.ug.edu.gh xii Table 5.9 Genetic variances, broad and narrow sense heritability and number of effective factors in a cross involving TVu-1 and TVu-999 ................................................................................... 110 Table 5.10 Phenotypic variances of the generations derived from zinc and iron set of crosses . 110 Table 5.11 Distribution means of yield related traits among 148 F3 cowpea populations evaluated under screen house conditions .................................................................................................... 111 Table 5.12 Phenotypic correlation coefficients between zinc concentration and yield related traits measured from F3 population derived from a cross between TVu-14845 and TVu-15251........ 116 University of Ghana http://ugspace.ug.edu.gh xiii LIST OF FIGURES Figure 3.1: Problems associated with cowpea marketing ............................................................. 33 Figure 4.1 a: Cowpea descriptors (IBPGR, 1983; Mahalakshmi et al, 2007) showing the distribution of seed coat texture. ................................................................................................... 58 Figure 4.1 b: Distribution of seed coat colour pattern in cowpea accessions (Saunders, 1959) ... 59 Figure 4.1 c: Distribution of seed size based on 100 seed weight (g) Cowpea descriptors (Mahalakshmi et al, 2007) ............................................................................................................ 60 Figure 4.1 d: Distribution of cowpea accessions based on shattering or non-shattering Mahalakshmi et al, 2007 ............................................................................................................... 61 Figure 4.1 e: Distribution of growth habit among 200 cowpea accessions (Mahalakshmi et al, 2007) ............................................................................................................................................. 62 Figure 4.1 f: Distribution of maturity periods of the cowpea accessions used Mahalakshmi et al, 2007............................................................................................................................................... 63 Figure 4.2 a: Frequency distribution of African cowpea accessions based on grain iron concentration as determined by Atomic absorption spectrophotometry....................................... 64 Fig 4.2 b: Frequency distribution of African cowpea accessions based on zinc concentration as determined by Atomic absorption spectrophotometry (AAS) ...................................................... 65 Figure 4.2 c: Frequency distribution of African cowpea accessions based on percentage grain protein content .............................................................................................................................. 66 Figure 4.3: Hierarchical dendrogram of 169 cowpea accessions by using similarity coefficients based on the Nei‘s (1983) original genetic distance calculated from119 SNP data using the UPGMA method ........................................................................................................................... 78 Figure 5.1: Population mean distribution for zinc concentrations among six basic generations derived from low zinc content (TVu-14845) and high zinc content (TVu-15251) parents .......... 97 University of Ghana http://ugspace.ug.edu.gh xiv Fig 5.2: Distribution of 100-grain weight in six generations derived from crosses involving small seeded parent (TVu-14845) and large seeded parent (TVu-15251) ............................................. 98 Fig 5.3: Distribution of plant height (PLHT) in six generations derived from crosses involving short parent (TVu-14845) plants and tall parent (TVu-15251) plants .......................................... 99 Figure 5.4: Population mean distribution for iron concentrations among six basic generations derived from low iron content (TVu-1) and high iron content (TVu-15251) parents ................ 105 Figure 5.5: Frequency distribution of number of pod per plant among 148 F3 Population. ....... 112 Figure 5.6: Frequency distribution of mean pod length per pod among 148 F3 Population ....... 113 Figure 5.7: Frequency distribution of number of seed per pod among 148 F3 Population ......... 114 Figure 5.8: Frequency distribution of seed weight (g) among 148 F3 Population derived from a cross between TVu-14845 and TVu-15251 grown under screen house condition ..................... 115 University of Ghana http://ugspace.ug.edu.gh xv LIST OF ABREVIATIONS AATF African Agricultural Technology Foundation AAS Atomic absorption spectrophotometry AOAC Association of Analytical Chemist ASCN Agricultural Seed council of Nigeria CIM Composite interval mapping CHO Carbohydrates DAP Days after planting DNA Deoxyribonucleic acid GCP Generation Challenge Program IAR Institute for Agricultural Research IBPGR International Board of Plant Genetic Resources IITA International Institute of Tropical Agriculture IM Interval mapping NGICA Network for Genetic Improvement of cowpea in Africa NACGRAB National Center for Genetics Resources and Biotechnology NPC National Population Commission SNP Single Nucleotide Polymorphism PIC Polymorphism Information Content QTL Quantitative trait loci TVu Tropical Vigna unguiculata University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE 1 GENERAL INTRODUCTION Cowpea (Vigna unguiculata Walp (L.) is of vital importance to the livelihood of millions of people in West and Central Africa. It provides nutritious grain and an inexpensive source of protein for both rural and urban consumers (Joseph et al., 2011). In addition, cowpea contributes to the sustainability of cropping systems and soil fertility improvements in marginal lands by providing ground cover and plant residues, fixing nitrogen, and suppressing weeds (Ajeigbe et al., 2010) Cowpea is grown in about 14.5 million hectares globally, with over 6.5 million metric tons (Fatokun et al., 2012). Africa alone accounts for about 83% of the world production, with Nigeria being the world‘s largest producer (45.76%), followed by Niger (15%), Brazil (12%), and 5 % for Burkina Faso (Fatokun et al., 2012). Cowpea is grown in all parts of Nigeria, though large production is concentrated in the northern region (Akande & Balogun, 2009). Cowpea production has been increasing at an average rate of 5% annually, with 3.5% annual growth in area and 1.5% growth in yields. Fatokun et al. (2012) reported that area under cowpea cultivation increased from less than 10% to nearly 20% between 1990 and 2007. The increase in production has been largely due to the expansion of areas under cultivation. In spite of this, a huge gap still remains between the potential yield and actual yield due to some production constraints such as pest and diseases, which is estimated to be account for more than 80% of yield losses in cowpea production (Fatokun, 2009) The major driving factor for cowpea production, as for most market crops, is to supply rural surpluses to the urban areas where there is demand. Kormawa et al. (2000) reported that about 72% of Nigerian urban households consumed cowpea more than any other grain legume, University of Ghana http://ugspace.ug.edu.gh 2 such as Groundnut, Soybean and Bambara groundnut. On average, a household purchased and consumed 5 kg of cowpea grain per week, spending about two hundred and forty (N240= US$1.5) naira. Nigeria produces 2.1 million tons of cowpea on 7 million hectares and consumes around 2.7 million tons. This creates a national deficit of over 500,000 tons which is made up through imports of US $125 million (Ishiyaku, 2013) Consumers‘ preferences for grain quality traits vary from region to region. For example red to black colour cowpea with various range of grain texture is preferred in Latin America, particularly Cuba and parts of the Caribbean (IITA, 1983). In West Africa, the most preferred types of cowpeas are large sizes, white or brown grains with rough grain coat. Similarly, medium grain sizes with brown or red grains with smooth grain coat are preferred in East Africa (Langyintuo et al., 2004). In Nigeria, rough grain is most preferred due to its ease of dehulling and greater swelling capacity, which is used for processed food such as ―akara‖ and ― moin moin‖ (Dovlo et al., 1976). Similarly, white rough, large grain and brown rough, large grain were mostly preferred in the Northern and Southern part of the country. However, Hussain et al. (1984) reported that the choice of cowpea varieties by Nigerian women is guided predominantly by the cooking time, swelling capacity, taste and colour. Good appearance, taste, mouth feel, different flavor and nutritional qualities are also important components of grain legume quality (Negri et al., 2001). Therefore, an efficient method to evaluate these attributes, in addition to physical characteristics of cowpea grain, would be useful for plant breeders to improve these aspects of grain quality. Cowpea is a single crop species whose varietal requirements in terms of plant type, grain type, cropping system, maturity and use pattern are extremely diverse from region to region, thus making breeding programs for cowpea more complex than for most other crops (Singh et al., University of Ghana http://ugspace.ug.edu.gh 3 1997). Consequently these varying preferences show the need to develop varieties with different characteristics, as no single variety can be suitable for all regions. Therefore, there is need to combine both consumers and farmer‘s perceptions on cowpea grain quality traits, their needs and preferences into this research. Engaging end-users would help to understand their needs and preferences, which will enable us identify the most appropriate cowpea varieties for improvement. To achieve that, a Participatory rural appraisal (PRA) needs to be conducted to understand and include end-users‘ needs in the breeding objectives. Iron and zinc deficiency are among the most common and widespread micronutrient deficiency that affects more than half of the human population (White & Broadley, 2009). Asia and Africa are the most affected countries with iron and zinc deficiency (Gómez-Galera et al., 2010). A collaborative study by USAID (2006) showed the prevalence of stunted, wasted and underweight children suffering from iron and zinc deficiency in the dry and moist savannas of Nigeria. This calls for the need to identify cowpea varieties with high iron and zinc, for setting breeding program to address this problem. Selection for nutritional quality traits such as iron and zinc using conventional breeding methods can be very difficult due to lack of discrete phenotypic classes in the progeny, subjective testing methods requiring taste panels and costly biochemical evaluation procedures. (Asante, 2012) Similarly, environmental influences in the expression of such quality traits further complicate the selection processes. Therefore, complementing the conventional breeding approaches with application of molecular markers will facilitate the selection of grain quality traits. This is because markers are less subjective, have no environmental effects on the phenotype and accelerate the identification and selection of several genomic regions involved in the University of Ghana http://ugspace.ug.edu.gh 4 expression of complex traits to ‗assemble‘ the best-performing genotype within a single, or across related, populations (Ribaut et al., 2010) Studies on variability in the level of iron and zinc have been reported on cowpea germplasm (Asante et al., 2009; Pereira et al., 2014). Nevertheless, there is no report on the genetic studies of iron and zinc in cowpea. There is therefore, an urgent need to characterize the cowpea germplasm available to improve grain quality traits. This will facilitate the selection of the appropriate genotypes for breeding. For example, the popular and most recent varieties: Sampea 1-14 and landraces like: Kanannado and Biyu local have appreciable physical grain quality attributes: white, brown and rough texture and large grained. Such accessions will be suitable recurrent parent for introgression. However, the most sustainable way and readily accessible to common man is the crop that has been fortified with those essential elements. This is because fortified crops do not require recurring expenditure, robust distribution system and very careful implementation, as there is no fear of overdose, which can be harmful (Nestel et al., 2006). Identifying cowpea germplasm with high micronutrient contents that will be amenable to introgression of iron and zinc without affecting their yield will pave way to fortifying this important crop. It is therefore, important to screen a number of cowpea accessions and select promising parents with good grain quality traits particularly those with higher iron, zinc and protein contents. The grain quality traits in cowpea are many, however, this study focused on grain morphological traits and nutritional quality traits, particularly, protein, iron and zinc. Each of these qualities comprises many attributes whose standards are determined by both physicochemical properties and consumers‘ confirmation. The main objective of this study was to identify and select cowpea varieties with good grain quality traits for future cowpea breeding program. The specific objectives were to: University of Ghana http://ugspace.ug.edu.gh 5 1. Determine cowpea production constraints and grain quality traits that are mostly preferred by farmers and consumers of Sudano-Sahelian zone of Nigeria; 2. Assess the variability of grain protein content, iron concentration and zinc concentration in cowpea accessions from Africa and other parts of the world; 3. Determine the inheritance pattern of iron and zinc concentration in cowpea grains. 4. Determine the relationship between zinc concentration and yield components University of Ghana http://ugspace.ug.edu.gh 6 CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 Cowpea [Vigna unguiculata L. (Walp)] Cowpea is the most widely grown and most economically important indigenous African grain legume that provides an important source of protein to consumers (Langyintuo et al., 2003). Xu et al. (2011), and Huynh et al. (2013) reported that cowpea is a diploid crop with 11 chromosomes (2n= 2x=22) and 630 Mb genome size. Cowpea belongs to the genus Vigna, section Catiang, species unguiculata. It comprises four subspecies namely: unguiculata, stenophylla, dekindtiana and tenuis (Ng & Marechal, 1985). The subspecies unguiculata is the only cultivated cowpea, while the other three are wild relatives. The cultivated cowpea is grouped under subspecies unguiculata, which is further subdivided into four cultivar groups namely; cultivar group (cv-gr) unguiculata, biflora, sesquipedalis and testilis (Ng. & Marechal, 1985). The cultivar group unguiculata is the most diverse of the four and is widely grown in Africa, Asia and Latin America (Ehlers et al., 2002). The crop is also an important staple in many parts of the United States, South and Central America, the Caribbean, India, and Australia (Siemuri et al., 2011). Several studies have shown that cowpea was probably domesticated by African farmers (Faris, 1965) and assumed to have evolved in Africa, because wild cowpeas only exist in Africa and Madagascar (Steele, 1976). Although the center of diversity of wild Vigna species is in southeastern Africa, West Africa is a major center of diversity of cultivated cowpea (Padulosi & Ng, 1997). Evidence from molecular marker studies have shown that cowpea domestication occurred in northeastern Africa based on studies of amplified fragment length polymorphism (AFLP) markers by Coulibaly & Lowenberg-De Boer (2002) and could have occurred simultaneously with the domestication of sorghum and pearl millet in the third millennium B.C. (Steele, 1976). University of Ghana http://ugspace.ug.edu.gh 7 The wild cowpea, V. unguiculata ssp. unguiculata var. spontanea is the likely progenitor of cultivated cowpea (Ba et al., 2004). However, the most recent studies have indicated that the precise center of origin of cowpea is yet to be known (Boukar et al., 2010). 2.2 Cowpea grain quality traits Long-term cowpea genetic improvement through germplasm exchange have focused on developing high yielding varieties with tolerance to some biotic and abiotic constraints (Wang et al., 2006; Padi, 2007). These resulted in the development of cowpea varieties that are resistant to Striga spp, some insects and diseases (Singh & Mligo, 2007). Crude protein content and other mineral element analysis in cowpea were reported by many authors (Ajeigbe et al., 2008; Anele et al., 2011), with scanty literature on the variability of cowpea grains for iron and zinc concentrations. The need to breed for cowpea consumer preference traits has become necessary because: 1. There is the need to improve children and women‘s diets, thereby, improving food security at large. Presently, some state Governments in Nigeria are providing dietary supplements to school children and Federal Government approves formulation of sorghum-soya in feeding school children. Almost all boarding schools are fed with cowpea diets at least three times a week. 2. The earlier developed and released cowpea varieties in Nigeria have low iron and zinc content. 3. Cowpea farmers and processors, have their preferences for grain quality traits, which are market driven. 4. Cowpea consumers in Africa are willing to pay premium price for important grain quality traits (Langyintuo et al., 2004) University of Ghana http://ugspace.ug.edu.gh 8 The morphological appearance of cowpea grains especially colour, texture and size are the common determinants of cowpea marketability in Western and Eastern Africa (USAID- Nigeria, 2008). Consumer tastes and preferences are reflected in the market through price discounts and premiums that consumers pay for visible grain characteristics. In some cases, these visible indicators are proxies for some biochemical characteristic, such as cooking time and protein (Ng et al., 2011; Dugje et al., 2009) as well as iron and zinc contents. 2.2.1 Grain size Determining grain size in cowpea has historically relied on measuring the weight of 100 grains, taken at random. However, this method cannot distinguish different grain shape parameters such as grain thickness or grain plumpness from grain diameter. Poersch et al. (2013) noted that the traditional method of grain sizing using graded sieves can be just as effective for determining grain dimensions. This, however, still means that phenotyping grain size and especially grain plumpness can be time consuming and tedious in a breeding program. Selection of these traits using marker-assisted selection (MAS) could be a valuable, time saving venture if markers linked to the genes controlling these dimensions could be identified. Grain size in cowpeas is important because it directly influences productivity and, together with colour standards, determine grain quality for commercialization (Lopes et al., 2003). Variation in grain size ranges between less than 10 g per 100 grains and approximately 30 g (Ehlers & Hall, 1997) Consumers in West Africa prefer large cowpea grain. They show preference for a grain standard from medium to large (15 to 25 g) with a minimum tolerance limit that varies from state to state in Nigeria. It is, therefore, understandable that knowledge about the genetic factors responsible for inheritance of this trait is essential for breeding programs which present great genetic variability in germplasm collections of the species. University of Ghana http://ugspace.ug.edu.gh 9 2.2.2 Grain coat colour Physical characteristics are directly related to the way cowpeas are used in food preparation. For example, cowpea grain coat and eye colours are the major factors, limiting utilization of cowpeas as food. This is important, especially when the intended use for the grains requires decortications (removal of outer layer) to remove flecks. Poor milling and winnowing may still leave some flecks for which consumers have low acceptances (Langyintuo et al., 2004). In Cameroon, ‗kosai‘ (cowpea fritters) without black flecks are very popular while in Ghana black flecks have little impact on the use of cowpeas for ―Tubani‖ (steamed cowpea paste) or a mixture of cowpea with rice or ―gari‖ (cassava chips). Preferences for different culinary roles of cowpea in the different region or countries can have impacts on the preferences for cowpea grain characteristics. Therefore, knowledge of consumer‘s preferences for grain characteristics is essential to plant breeders for setting priority objectives in their breeding program. 2.2.3 Grain coat texture For wide adoption of cowpea in West Africa, new cowpea varieties must have features desired by consumers as well as farmers. Four types of grain coat texture have been identified in cowpea: smooth, rough, wrinkle and loose (Fatokun et al., 2012). Cowpeas with large white or brown grains with rough grain coat are preferred throughout West Africa, whereas in East Africa they prefer medium size, brown or red grains with smooth grain coat. But in some Latin American countries, particularly Cuba and part of Caribbean, black colour with various categories of grain coat texture are preferred (Obiegbuna et al., 2006). A rough grain coat is preferred in West and Central Africa, since it permits easy removal of the grain coat which is essential for indigenous food preparations (Singh & Ishiyaku, 2000). The preference for cowpea grain with rough grain coat in Nigeria is due to their ease of dehulling and greater expansion capacity. Such University of Ghana http://ugspace.ug.edu.gh 10 grains are used for cowpea diets such as ―akara‖ and ―moin moin‖ preparation (Dovlo et al., 1976). 2.3 Inheritance of Grain Quality Traits 2.3.1 Inheritance of Grain Size The genetic control of the grain size is complex due to environmental influence and number of genes involved. (Drabo et al., 1984) reported that seed weight was quantitatively inherited and small seed was partially dominant to large seed size. The gene action governing the inheritance of seed weight was predominantly additive but dominance and additive x additive epistatic effects were also significant. Five and eight genes were reported to control seed weight in cowpea by Lopes et al. (2003) and (Egbadzor et al., 2013), respectively. 2.3.2 Inheritance of grain coat colour Grain coat colour is considered as one of the useful phenotypic markers in cowpea breeding due to its stable expression and convenience for observation (Xu et al., 2011).Various grain coat colours such as white, cream, brown, maroon, black, red, blue and buff have been reported in cowpea (Asante, 1991; Mustapha, 2009). Similarly, Oluwatosin (2000) reported that a gene controlling red grain coat is monogenically dominant to that of cream colour. Two dominant genes which exhibit complementary epistasis were reported in bi-parental crosses consisting of Black x Cream parents and Brown x Cream parents (Oluwatosin, 2000). A general colour factor C is considered responsible for grain coat colour and its absence results in white grains (Spillman, 1912). In some genotypes, the C factor in combination with other genes conditions certain colours of grain coat. Spillman & Sando (1930) proposed that six major genes control grain coat colour in cowpea. Mustapha (2009) reported that grain coat colour is controlled by many genes, though University of Ghana http://ugspace.ug.edu.gh 11 some of the genes may be allelic. Recently Egbadzor et al. (2012) reported a link between flower colour, immature pod pigmentation and grain colour in crosses between some Ghanaian cowpea germplasm. Grain coat colour is one of the major cowpea consumers‘ determinants in Nigeria. It is, therefore, important to consider grain colour when selecting parental lines and among the segregating population. 2.3.3 Inheritance of grain coat texture Most of the genetic studies on the inheritance of grain coat texture in cowpea focused mainly on smooth and rough textures. A smooth testa was found to be dominant over rough testa (Drabo et al., 1984). Rough testa is controlled by at least two recessive genes (Franckowiak, 1973). Singh and Ishiyaku (2000) studied the segregation pattern of grain coat texture in bi- parental crosses involving smooth and rough parents and reported fitting ratios of smooth and rough close to 1:1 and 3:1 in the backcross involving rough parent (BC1F1) and (BC1F2) plants respectively. This indicated that rough grain coat is controlled by a recessive gene. Grain texture is an important morphological trait that guides consumers in selecting cowpea for processing. Therefore, incorporating consumer preferred grain texture in breeding for grain quality traits will ease adoption of new variety in Northern Nigeria. 2.3.4 Inheritance of protein content Proteins are major components of grain legumes and their nutritional and functional properties depend on the nature of soluble fractions (Vasconcelos et al., 2010). Cowpea is widely adapted and nutritious grain legume constitutes one of the main sources of plant protein in Nigeria. Dry grains for human consumption are the principal product of the plant, but leaves, fresh peas, and fresh green pods are consumed by many peasants who do not have access to broadly based diet (Ehlers & Hall, 2002). Understanding the genetic mechanism for the University of Ghana http://ugspace.ug.edu.gh 12 expression of protein content and genetic control of cowpea grain is therefore very essential for the improvement of that trait. Genetic variation for cowpea grain protein contents has been reported by Gupta et al. (2010) and Tchiagam et al. (2011a) Significant difference among cowpea protein content has also been reported by Asante et al. (1991); Ajeigbé et al. (2008) and Tchiagam et al. (2011b). Grain protein composition is genetically controlled but affected by environmental factors particularly nitrogen and sulphur availability (Tabe et al., 2000). 2.3.5 Inheritance of iron and zinc contents Narrow variations among cowpea lines of up to 60.3 mg/kg and 40.7 mg/kg for iron and zinc were reported by Boukar et al. (2010). However, significant variations in grain zinc and iron concentrations were reported in different edible crops. For example 4 fold differences in grains‘ iron and zinc content was reported in aromatic rice compared to popular cultivars (Cakmak et al., 1999; Welch & Graham, 1999). Significant genetic variation of micronutrient have been reported in maize (Bänziger & Long, 2000) with range of 9.6 to 63.2 mg/kg for grain Fe and 12.9 to 57.6 mg/kg of grain zinc. Similar variation was also reported in wheat grain with 3-4 fold increased iron and zinc content in wild type than the popular cultivars (Chhuneja et al., 2006; Rawat et al., 2009). 2.4 Association between Nutritional Quality traits Association studies between protein content and other physicochemical properties of cowpea was reported by many authors: For example Hussain & Basahy (1998) from Saudi Arabia, showed that ash content was positively correlated with protein content and negatively correlated with carbohydrate content. Hundred (100) seed weight was significantly negatively correlated (P < 0.05) with protein (PC) (r = -0.72) and iron content (r = -0.61) (Moura et al, 2012). A negative correlation between protein content and 100 seed weight was also reported by University of Ghana http://ugspace.ug.edu.gh 13 Asante et al. (2004). Fat content was negatively correlated with protein. Protein content showed very high negative correlation with carbohydrate content. This indicates that selection for high protein content will result in decrease in carbohydrate content and ash content, which will make the improved line to be nutritionally superior. The total fat content in cowpea is very low, its reduction due to increase in protein will not make much difference in overall nutritional quality of the varieties (Ajeigbe et al., 2008). 2.5 Physiological mechanisms of iron and zinc uptake Accumulation of minerals in grains is a complex phenomenon, which is most likely controlled by a number of genes. The movement of mineral elements from the soil to grains involves their mobilization from soils, uptake by roots, translocation to the shoot, redistribution within the plant and deposition in grains (Grusak & DellaPenna, 1999; White & Broadley, 2009). However, the genetic basis of most of these processes are unknown (Ding et al., 2010). Various mechanisms for soil mineral uptake and distribution in plants have been reported to be the result of tightly controlled homeostatic mechanisms that regulate metal uptake and distribution in plants, allowing adequate but nontoxic levels of these nutrients to accumulate in plant tissues. It has been reported that the embryo and the aleurone layer for cereal are the major depositories for grain zinc (Mazzolini et al., 1985;and Choi et al., 2007), while Ozturk et al (2006), showed that zinc is present in the outer part of the endosperm, which contain high amounts of protein. University of Ghana http://ugspace.ug.edu.gh 14 2.6 Marker Assisted Selection (MAS) for cowpea improvement 2.6.1 Current trends in the use of marker technologies for cowpea breeding Molecular genetic tools and genomic resources have been developed to augment conventional cowpea breeding methods. (Muchero et al., 2009a) and (Lucas et al., 2011) reported a significant improvement of cowpea in the United States, India, Brazil, and many countries in Africa and Asia. These integrated genomic resources include a 1536 SNP genotyping platform, an EST-derived SNP consensus genetic map, known syntenic relationships between cowpea, Medicago truncatula, Glycine max, Arabidopsis thaliana, and a cowpea EST sequence collection stored in HarvEST: Cowpea database (http://harvest.ucr.edu). A cowpea physical map has been partially anchored to the cowpea consensus genetic map using the same SNP markers (UCR cowpea group, unpublished) and is available on http://phymap.ucdavis.edu/cowpea. In addition, about 500 diverse cowpea accessions have been SNP-genotyped (UCR cowpea group) and a first draft of the cowpea genome, vs.0.02, has been assembled (www.harvest-blast.org). These resources enabled dissection of underlying genetic components of target agronomic traits using Quantitative Trait Locus (QTL) analysis and Association Mapping. The identified and confirmed QTLs facilitated cultivar improvement using marker-assisted breeding. The most detailed genetic linkage map of cowpea spans a total of 2670 cM with an average of 6.43 cM between markers was reported by Ouédraogo et al. (2002). This cowpea linkage map revealed the importance of associating between phenotypic and genotypic data leading to understanding the biological determinants of quantitative phenotypic variation. Molecular geneticists have identified genes controlled by one or a few loci with large effects. This is particularly true in cowpea, where genetic analyses have identified the biochemical basis of many important phenotypes such as drought stress parameters (Muchero et al., 2009b). However, University of Ghana http://ugspace.ug.edu.gh 15 understanding complex trait variation is sometimes difficult, due to the nature of genetic architecture of such traits, which may involve many loci of small effect that may interact with each other or with environment (Bernardo, 2008; Collard & Mackill, 2008). Genomic Selection (GS) technologies were recently introduced and provided an economically feasible way to survey genetic variation with a resolution that is now limited more by the linkage disequilibrium (LD) in a particular mapping population than by marker density. This phenomenon has motivated the assembly of large panels of genetic diversity as well as the creation of large inter-mated populations to manipulate LD and facilitate the association of genotype with phenotype (Lie-Zhao et al., 2006). Large and diverse populations increased the recombination frequency and the frequency of rare alleles in order to enhance the power to infer the effects of individual loci. 2.6.2 Quantitative trait loci mapping A quantitative trait loci (QTL) was proposed by Sax (1923), first coined by (Geldermann, 1975) and later used by many researchers to detect and locate QTL in segregating populations in crop and animal species. A QTL is defined as a region of a genome that is associated with an effect on quantitative trait. A QTL can be a single gene, or it may be a cluster of linked genes that affect the trait. QTL mapping had been reported in most crop plants for diverse traits including: yield, quality, disease and insect resistance, abiotic stresses like drought and other environmental constraints. Previous studies of grain-related traits in several crops have revealed loci controlling more than one related trait. In lentil for example, a flowering time locus was shown to be linked with the grain coat pattern locus (Slattery et al., 1982; Zhi-wen et al., 2005). Flowering times influence grain size in different crops (Gupta et al., 2006). Similarly, pre-anthesis changes in vegetative organs can affect the amount of assimilates that are partitioned to the grain while they are developing. Also post-anthesis processes can affect the time for maturation or grain filling, University of Ghana http://ugspace.ug.edu.gh 16 which could change the grain size (Moura et al., 2012). Loci controlling flowering time or other flower morphology traits have also been associated with grain weight or grain dimension loci in model legume crops (Burgess et al., 2007; Burgess et al., 2007; Greenup et al., 2009). The success of modern QTL mapping depends on the availability of dense DNA marker maps for the organism involved. In recent years, comprehensive DNA marker maps have been developed for many crops (Orjuela et al., 2010; Ouedraogo et al., 2012). This makes the identification of QTL to be feasible and successful in crop plants, using an appropriate mapping population (MP). Such population mostly involved bi-parental progenies such as backcrosses (BC), double haploid lines (DH), F2, or recombinant inbred lines (RILs) Crepieux et al. (2004); Crepieux et al. (2005). The QTL mapping literature has shown that a mapping population of N = 100–150 progenies is enough to identify QTL. Such population could be recombinant inbred lines (RIL), F2 or F2:3 or backcross population derived from two inbred lines coupled with reasonably good phenotypic data and genotypic data with markers spaced about 10 to 15 cM apart, are essential requirement for QTL identification (Flint-Garcia et al., 2003). However, Muranty (1996) suggested the use of progenies from several parents, to achieve a high probability of obtaining more than one allele at a putative QTL and also to have a more representative estimate of the variance accounted for by a QTL. 2.6.3 Concept and principles of QTL Quantitative characters are common features of natural variation in populations of all eukaryotes, including crop plants. These traits provided a conceptual base for partitioning the total phenotypic variance in terms of additive, dominance and epistatic effects. The basic principle of determining whether a QTL is linked to a marker is to partition the mapping population into different genotypic classes based on genotypes at marker locus, and apply correlative statistics to University of Ghana http://ugspace.ug.edu.gh 17 determine whether individuals of one genotype differ significantly with individuals of other genotype with respect to trait of interest. According to Moose & Mumm (2008), the success in using information about QTLs to increase genetic gain depends greatly on the suitable method used to determine the magnitude of QTL effects, precise estimation of QTL positions, stability of QTL effects across multiple environments, and whether QTLs are robust across relevant breeding germplasm. Prediction of QTL positions is enhanced by further fine mapping, which facilitates testing QTL effects and breeding values in additional populations. Methods for simultaneous detection and manipulation of QTL in breeding programs would thus enhance the applicability of MAS. 2.6.4 Methods used in QTL detection The study of quantitative traits has involved statistical techniques based on means, variances and covariance of relatives. The first models described for QTL detection with various progeny types, are single marker models (Beckmann & Soller, 1990; Srivastava et al., 2005). In these models, information from the phenotypic means and variances of genotypic classes at a single marker is used to test marker linked-effects and to estimate the putative QTL parameters. The most used method is based on analysis of variance (ANOVA) and considers contrasts among marker class means to test and estimate QTL effects (Tanksley et al., 1992). This approach has many shortcomings; it estimates effects of the QTL which are biased by the recombination between marker and QTL and provides no information about the likely position of the QTL. Weller & Soller (2004) described a method using approximate maximum likelihood analysis and moment‘s methods to estimate QTL parameters and the recombination fraction between marker and QTL. University of Ghana http://ugspace.ug.edu.gh 18 Interval mapping (IM) described by Lander and Botstein (1989) is an approach investigating the location of QTL between intervals of flanking markers by means of likelihood analysis. Furthermore, Hu & Xu (2009) compared the QTL detection powers obtained with random-effect models and fixed effects and found similar values for individual family sizes as low as 25 individuals. Statistical methods for the QTL analysis of bi-parental populations underwent successive improvements through the advent of interval mapping (Lander & Botstein, 1989) and its linearization, composite interval mapping and multiple-trait QTL mapping (Kang & Priyadarshan, 2007; Ding et al., 2010). University of Ghana http://ugspace.ug.edu.gh 19 CHAPTER THREE 3.0 APPRAISAL OF FARMER AND CONSUMER PREFERENCES FOR COWPEA IN THE SUDAN SAVANNA ZONE OF NIGERIA 3.1 Introduction Adoption of existing and newly generated technologies continues to be low, despite the time and resources committed by researchers in an effort to alleviate this problem (Chambers et al., 1989). Adoption of improved cowpea is still low in Nigeria (Takeshima et al., 2014), even though the National and International Research Institutes developed and released various cowpea varieties. Farmers are still growing cowpea landraces such as ―Kanannado‖ ―Danmisra‖ and ―Biyu local‖ (Kormawa et al., 2004). This has been attributed in part to low farmer involvement in the whole process of variety or technology development. The constraints to the distribution of improved grains are on both the supply and demand sides. Takeshima et al. (2014) reported that one less-studied issue is the farmers and consumers‘ preference for obtaining grains at planting time, when grain prices tend to be higher. While the formal grain sector is often constrained by the capacity for distributing improved grains to farmers, knowing whether farmer and consumer are willing to pay premium prices at planting time and during processing can provide potentially useful information to consider their preferences when setting breeding objectives. Knowledge of the range of plant, grain and processing traits are valuable for crop improvement programs and good market signals for the farmers (Langyintuo et al., 2003) The demand for improved cowpea varieties is likely to increase if, varieties are designed to include producers and consumers‘ preferred traits, in addition to farmers‘ traits, improved grains have provided 50% of the productivity gains in agriculture. The other 50% has come from University of Ghana http://ugspace.ug.edu.gh 20 improvement in management, including timeliness, best use of fertilizer, crop protection measures and equipment (Ayinde, 2005; USAID- Nigeria, 2008). Participatory Rural Appraisal (PRA) is one of the rapid and most important tools of identifying farmers and consumers‘ preferences. It reveals a number of traits that are not prioritized by breeders. Success stories of farmer and consumers‘ participatory breeding have been reported. These include: the consumers‘ preferences for sorghum-based clear beer in Tanzania (Makindara et al., 2013), pearl millet breeding in Mali (Weltzien et al., 1998) and nutritional quality and consumer acceptability of cowpea grain was recently reported by Ojiako & Kayode, 2014). High-yielding cowpea varieties with desired grain nutritional characteristics could make a great impact on the Nigerian cowpea industry because famers are likely to get good market for their product, while consumers will be willing to pay premium price for grain quality traits (Takeshima et al., 2014). This would stimulate farmers to substantially increase domestic cowpea production. It is therefore, essential for Nigerian cowpea breeders to incorporate the preferences of farmers and consumers in their breeding programs. However, information on the general criteria used by farmers for selecting cowpea varieties for cultivation is lacking. Also, farmers‘ perceptions on grain quality in relation to consumer demand needs to be considered. The objectives of this study were to: 1. identify farmers‘ production constraints and their perceptions on cowpea grain quality traits; 2. determine farmers and consumers‘ preferences and selection criteria for selecting cowpea varieties in Sudano-Sahelian zone of Nigeria; University of Ghana http://ugspace.ug.edu.gh 21 3.2 Materials and Methods 3.2.1 Description of the study area The study was conducted in the Sudan savanna and Sudano- sahelian transition zone of Kano state, Nigeria. Kano has an average annual rainfall between 500 mm to 1200 mm per annum, an average of 276 people /km2 with a total of 16 million inhabitants (NPC and ICF Macro, 2009). The average household size in Kano ranges between 8 and 10 members with average income of 4000 Naira (at N140 per US dollar $1). The major ethnic groups are Hausa and Fulani (Maiangwa & Ogungbile, 2008). Kano lies in the Sudan savanna, semiarid zone of Nigeria, on 11°34´ N and 8°44´ E latitude and longitude respectively. The Sudan ecology is characterized by a growing period of about 100 to 150 days. Annual rainfall, between 500 and 1000 mm, is erratic and restricted to 4 months (Inaizumi et al., 1999) . Three local government areas were selected for study on the cowpea production constraints. Two sub-communities each from Bunkure (11°41′41.55″N and 8°32′17.81″E), Wudil (11°49′11.15″N and 8°54′11.41″E) and Bichi (12°11′24.63″N and 8°18′45.73″E) (Table 3.1), were selected because they are among the major cowpea growing communities in Kano State. Cowpea consumers‘ preferences were assessed in Kano metropolis, which comprised Municipal, Tarauni districts and ―Dawanau‖ market. The commercial city of Kano has the largest immigrants‘ population from all parts of the country and other African countries. Similarly, the largest grain (Dawanau) market in West African region is located in the city. Therefore, consumer preferences for cowpea in Kano might be representative of the whole country and all similar zones across West Africa. University of Ghana http://ugspace.ug.edu.gh 22 3.2.2 Site mapping and selection Transect walk was conducted with 2 men and 2 women in each district of the study areas. During the exercise, cowpea based foods processing and selling points were identified. Problems and opportunities associated with cowpea production, processing and consumption were also identified. The communities were selected based on the cowpea growing and processing or selling points observed during transect walk. 3.2.3 Sampling Procedure The respondents are presented in Table 3.1 below according to their village areas, gender and whether producers or consumers. Three major cowpea-growing communities: Bunkure, Wudil and Bichi, with 2 village areas each were purposively selected. Five village areas representing consumers were purposely selected from Kano metropolis, from all the villages 50 consumers representing both male and female were selected except in Dawanau where 40 respondents were selected. Two hundred and forty (240) consumers selected from Kano metropolis were interviewed on their preferences for grain quality traits, using semi structured questionnaires (Appendix 3.2) University of Ghana http://ugspace.ug.edu.gh 23 Table 3.1: Number of farmers and consumers interviewed in selected cowpea growing and consuming communities Farmers Consumers Communities/LGA Male Female Male Female Total Bunkure LGA: a. Kofar gabar b. Lautaye 34 38 6 2 - - - - 40 40 Wudil LGA: a. Bakin Kasuwa b. Kausani 39 33 1 7 - - - - 40 40 Bichi LGA: a. Bichi town b. Badume 37 36 3 4 - - - - 40 40 Kano Metropolis: a. Kabara b. Kasuwa Rimi c. Unguwa uku d. Hotoro e. Dawanau - - - - - - - - - - 11 8 21 28 30 39 42 29 22 10 50 50 50 50 40 Total 217 23 98 142 480 LGA= Local Government Area 3.2.4 Data collection techniques Data for varietal selection criteria and grain quality preferences for farmers and consumers were collected using two techniques. These are (1) focus group discussions (FGD) using a checklist and (2) individual interview using structured questionnaires. The questionnaires and checklist designed were edited by a team of experts: comprising of plant breeder, socioeconomics, agronomist and extension agent. The questionnaires were first pre-tested to validate the relevance of the variables and the possible responses. The questionnaire was then revised to incorporate issues that emerged from the pre-testing before it was administered to the farmers and consumers. University of Ghana http://ugspace.ug.edu.gh 24 Focus group discussions were done with 15 farmers per community making a total of 30 per district. Agricultural extension officers with experience in working with cowpea farmers were recruited to organize the cowpea farmer groups in the communities for the FGD at each site. A standardized checklist of topics was used in the FGDs (Appendix 3.1). The topics discussed included general problems faced by cowpea farmers, grain quality preferences, marketing of cowpea and their perceptions on grain quality traits as it relates to consumer demands. Structured questionnaires (Appendix 3.2) were administered to individual farmers that were involved in the FGD from the 5 sub-communities. This enabled individual farmers, especially those who were not confident enough to contribute during the FGD, to freely express their views. Field observations and random measures of field sizes were used to validate the data. A different set of questionnaire (Appendix 3.3) was also administered to a group of consumers from Kano metropolis. This was to measure consumers‘ perceptions on various cowpea grain quality attributes and determines their preferences. Data were collated on the constraints to cowpea production, perceptions of farmer and consumers‘ on grain quality traits as well as their preferences for grain quality traits in cowpea. 3.3 Data analysis Both quantitative and qualitative were collected, coded and analyzed using SPSS version 17.0 (SPSS, Inc., Chicago IL). Preference ranking was used to score production constraints, most preferred variety, quality traits, and perceptions on nutrient quality traits. Average scores and ranks were used for the focus group discussion. Likert scale (Responses; 1= most important; 2= important; 3= not so important; 4= not important) was used to measure the grain quality traits. The mean perception for grain quality traits was estimated using: ∑ University of Ghana http://ugspace.ug.edu.gh 25 Where; is the number of individuals who choose the i th response or attribute. the j th response or attribute. N= Total number of respondents The extent of agreement or disagreement on ranking the constraints to production was estimated using Kendal‘s coefficient of concordance (W) described by Siegel & Castellan-Jr. (1988). 3.4 Results 3.4.1 Constraints to cowpea production in the study areas Ten production constraints were identified as the major constraints to cowpea production in the communities and ranked by farmers. Farmers scored inadequate amount of improved cowpea seeds at planting time as the most important constraint limiting cowpea production in the areas. This was followed by pest, insects, diseases, drought, parasitic weeds, high cost of inputs, lack of farm implement, poor soil fertility, high cost of labour and lack of market (Table 3.2). There was 73.05% (Kendal‘s value W=0.7305) agreement among the farmers for these rankings. Table 3.2: Cowpea production constraints faced by farmers in Sudano-Sahelian zone of Nigeria Production constraint Mean Rank Rank Inadequate amount of improved cowpea seeds at planting time 1.38 1 Insects (such as aphid, pod sucking bugs and thrips) 2.90 2 Lack of formal seed system that make seed available 3.09 3 Diseases (damping off, bacterial blight, Fusarium wilt etc) 3.66 4 Drought (seedling and terminal - mostly affecting) 4.25 5 Parasitic weeds- (Striga and Alectra) 5.69 6 High cost of inputs (fertilizer and agrochemicals) 6.03 7 Lack of farm implements-(ploughs, thresher) 6.39 8 Poor Soil fertility (low Phosphorus level, flooding, iron toxicity) 7.08 9 High cost of labour 8.32 10 Constraint with smallest mean rank is the most important factor limiting cowpea production in the areas. University of Ghana http://ugspace.ug.edu.gh 26 Pod boring and sucking insects Farmers were not familiar with most of the pod boring and sucking insects, however, when they were shown the pictures of those insects and the nature of damage caused, they identified Lepidopterans, particularly, Maruca vitrata, Roptorus dentipes, Anoplocnemis cavipes and parasitoids (Hymenoptera, Braconidae, Aphidiinae) of cowpea aphids. About 78% of the total farmers interviewed reported that they had observed such insects in their fields. Cowpeas cultivated under irrigation were the most affected. Field visit and farmers‘ opinion during FGD identified insects belonging to the members of the Hymenoptera and lepidopterans to be the major constraints in Bunkure and Wudil, respectively. Majority of the farmers (85%) applied various insecticides at different formulations and different stages of cowpea growth, to control those insects. However, 53% of farmers thought that the varieties were susceptible to Lepidopterans insects, while 47% were on the opinion that the insecticides were ineffective. 3.4.2 Farmer’s varietal preferences Cowpea variety 277 popularly known as ―Danbazara‖ by farmers was the most cultivated variety at all locations. However, farmers‘ preferences varied amongst the communities. Farmers in Bunkure, Wudil and Bichi districts preferred ―Dankaka‖ (288), ―Kyambas‖ and ―Danwuri‖, respectively (Table 3.3). Eighty two (82%) percent of the farmers interviewed indicated their desire to have new cowpea varieties, if they would combine yield and quality attributes. During focus group discussions, Bunkure and Wudil farmers listed resistance to insects, high yield, good grain quality (good taste), and early maturity as their preferred traits. Good grain qualities were the top priorities of farmers at Bichi. They mentioned in addition to earliness, good appearance University of Ghana http://ugspace.ug.edu.gh 27 including grain coat colour, eye colour, texture, short cooking time and swelling capacity and high yield as their traits of interest. Table 3.3: Farmers varietal preference ranking by community Variety Bunkure Wudil Bichi Danrani (288) 1 * * Danbazara (277) 2 2 2 Danbazara (Help me) 3 * * Kanannado 4 * 3 Pegeot (Fiya fiya) * 3 * Kyambas * 1 * Balami * 4 * Cida gero * * 1 Badankami * * 4 *= variety is not commonly grown in a community Reasons for farmers’ preferences for cowpea varieties varied across communities (277, Kymbas and Danwuri) During focus group discussions, farmers from all the 3 districts cultivated variety Danbazara (277) mainly due to its good grain quality traits. The quality traits mentioned by the farmers included morphological attributes such as; white grain coat, rough texture, medium sized variety and cooking quality. This resulted in high demands for ―277‖ by the consumers. Other reasons for the wide adoption of ―277‖ included high yield performance under irrigation and rain fed conditions. Farmers in Wudil districts preferred ―Kyambas‖ to ―Danbazara‖ because of its adaptability to their niche and good grain quality attributes such as: medium size, white grain coat and rough texture. ―Danwuri‖ was the most preferred variety at Bichi districts, because of its earliness and good grain quality traits that are market driven. University of Ghana http://ugspace.ug.edu.gh 28 3.4.3 Farmers’ preference ranking for grain quality traits About 68% of the farmer ranked and agreed (Kendall's W = 0.685, p=0.000) with yield as their most important trait followed by grain quality traits and resistance to biotic and abiotic stresses (Table 3.4). Table 3.4: Farmers preference ranking of cowpea variety traits Traits Mean Ranking Overall Ranking High yields 1.21 1 Good grain quality-colour, texture, size and tasty 1.68 2 Resistance to insects and pest including weeds 2.96 3 Earliness 3.88 4 Means with smallest mean rank is the most important in each column Almost all the farmers interviewed indicated their willingness to adopt new varieties. Majority of the farmers preferred high yield (93%) and good grain quality traits (90%), especially grain coat colour and texture. Almost half of the farmers (58%) most of them from the transition zone of Bichi district, preferred for earliness in addition to grain quality traits (Table 3.5). Table 3.5: Farmers priorities for traits in new cowpea varieties Traits Yes (%) No (%) Order of preference High yields 92.54 7.46 1 Desirable grain quality-colour, texture, size and tasty 89.62 10.38 2 Resistance to insects and pest including parasitic weeds* 65.85 34.15 3 Earliness* 58.44 41.56 4 *= very important trait in the transition zone (Bichi district) though ranked least in the overall ranking. University of Ghana http://ugspace.ug.edu.gh 29 3.4.4 Farmers’ perception on grain quality attributes Farmers perceived grain colour (grain coat and eye colour), texture and size as very important grain quality traits. Cowpea grain shape and expansion ratio were regarded as important by most of the commercial farmers. Table 3.6: Consumers’ perception on grain quality attributes Grain Quality Attributes Very important (1) Important (2) Not so important (3) Not important (4) Rank score (x) Implication Colour 186 (77.50)* 38 (15.83) 11 (4.58) 5 (2.08) 1 Most Preferred Grain size 149 (62.08)* 56 (23.33) 24 (10.00) 11 (4.58) 1 Most Preferred Texture 109 (45.42) 83 (34.58) 41 (17.08) 7 (2.92) 2 Preferred Resistance to storage insects 86 (35.83) 98 (40.83) 37 (15.42) 19 (7.92) 2 Preferred Cooking time 54 (22.50) 96 (40.00) 64 (26.67) 36 (15.00) 2 Preferred Grain shape 66 (27.50) 95 (39.58) 35 (14.58) 44 (18.33) 2 Preferred Expansion capacity 61 (25.42) 99 (41.25) 55 (22.92) 25 (10.42) 2 Preferred Numbers in bracket indicate the percentage of farmers preferred trait, the higher the percentage the more important is the trait Consumers listed 3 most important traits in selecting for cowpea grain (Table 3.7). Both end users (75.64%) and processors (55.00%) preferred white to creamy colour grain. Majority of the end users prioritized large grains (72.49%), while processors preferred small to medium grain size (36.55% and 54.21%) respectively. Most of the processors preferred rough texture coat (86.25%), while end users showed not much preference for grain coat texture (56.19 % and 43.81%), respectively. University of Ghana http://ugspace.ug.edu.gh 30 Table 3.7: Consumers preferences ranking for grain quality traits Grain Quality Characteristics Percent (%) End user Processor Colour White /Cream*/** 75.60 55.00 Brown 24.40 34.70 Black - 6.30 Those with various speckle - 4.10 Grain size Large* 72.50 9.20 Medium** 24.50 54.20 Small 3.10 36.60 Texture Rough*/** 56.20 86.30 Smooth 43.80 13.75 *=preferred by end-users, **=preferred by processors and */** preferred by both Consumers‘ (processors and end users) preferences varied according to their need (Table 3.8). Forty five (45%) percent of the processors perceived cooking time to be the most important trait, followed by grain texture (68%), grain size (43%), oil consumption (42%) and ease of dehulling (52%) ranked as important grain attributes. They further indicated their preferences for white grains with brown eye for ―akara‖ preparation. On the other hand, most of the end users (75%) ranked protein as the most important. Almost all the women interviewed (66%) were aware that cowpea contained micronutrients content, due to medical personnel recommendation for them to be consuming cowpea during pregnancy. University of Ghana http://ugspace.ug.edu.gh 31 Table 3.8: Consumer perceptions on grain quality attributes Grain Quality Attributes Most preferred (1) Preferred (2) Not so Preferred (3) Not preferred (4) Mean score (x) Implication Protein* 181(75.42) 42(17.50) 15(6.25) 2 (0.83) 1 Most preferred Cooking time 108 (45.00) 67 (27.92) 17 (7.08) 14 (5.83) 1 Most preferred Texture 43 (17.92) 162 (67.50) 24 (10.00) 11 (4.58) 2 Preferred Grain size 53 (22.08) 105 (43.75) 55 (22.92) 27 (11.25) 2 Preferred Oil consumption 49 (20.42) 102 (42.50) 55 (22.92) 34 (14.17) 2 Not so preferred Dehulling 68 (28.33) 124 (51.67) 33 (13.75) 15 (6.25) 2 Preferred Micronutrients (Fe& Zn) * 66(27.5) 159 (66.25) 19 (7.92) 6 (2.50) 2 Important Colour 59 (24.58) 154 (64.17) 23 (9.58) 4 (1.67) 2 Preferred Expansion ratio 63 (26.25) 71(29.58) 81 (33.75) 25 (10.42) 3 Not so preferred Taste (Sweet/sweetness) * 51 (21.25) 64 (26.67) 94 (39.17) 31 (12.92) 3 Not so preferred  End users’ preferred trait, Number in the bracket is the percentage of consumer preference for certain trait Almost all (95%) the farmers interviewed grew cowpea for commercial purposes and kept about 35% of produce for home consumption. Cowpea was mainly sold in threshed grain form to traders and women processors (market women who processed cowpea into different diets). The market price for cowpea is mostly high towards planting time and during active growth of the cowpea growing season, farmers complained about storage insects during storage of excess cowpea. Six most consumed cowpea diets are presented in Table 3.9. The high value of Kendall‘s coefficient of concordance (W= 0.708, p= 0.000) indicated the 70% agreement of the consumers‘ ranking for ―Kosai‖ as the most preferred and consumed diets in Kano metropolis. This was followed by cowpea with rice cooked together, eaten with spices or stew (Wake da shinkafa); Cowpea steam paste (Moin moin); Cowpea dimples (Danwake), raw cowpea; and cowpea porridge (Table 3.9). University of Ghana http://ugspace.ug.edu.gh 32 Table 3.9: Ranking for consumer preferences for most frequently consumed diets Cowpea dishes Most Frequent Consumers Mean Rank Rank Akara (Kosai )* 62.70 1.38 1 Cowpea mixed with rice 59.70 2.44 2 Moin moin 48.59 3.59 3 Danwake 46.20 4.08 4 Raw cowpea 29.98 4.88 5 Cowpea porridge 42.66 6.12 7 Diet with smallest rank is the most prepared and most consumed, *= daily consumed 3.4.5 Problems associated with Cowpea Marketing Problems associated with cowpea marketing as identified by marketers and farmers that do not want to sell it at harvest time due to low price are presented in Figure 3.1. Both small scale cowpea farmers and marketers identified inadequate capital as the highest problem affecting markets followed by lack of adequate storage facilities leading to high cost of transportation to the homes at harvesting time and to market. University of Ghana http://ugspace.ug.edu.gh 33 Figure 3.1: Problems associated with cowpea marketing 3.5 Discussion Ranking of constraints to cowpea production by farmers in all 3 districts differed slightly among the communities; the challenges listed were similar across the districts. The highest ranked challenges were grain related issues such as; non-availability or inadequate amount of improved cowpea seed at planting time, lack of pod boring and sucking insect resistant varieties and lack of disease resistance among varieties. Agricultural Seed Council of Nigeria (ASCN) is an agency that ensures certified seeds are being sold to farmers at the right time. However, the efforts of 0 10 20 30 40 50 60 Inadequate capital Lack of storage facilities Bad roads Robbery Poor market facilities Number of farmers Market constraint P er ce n ta g e o f fa rm er s sc o re d a p ro b le m University of Ghana http://ugspace.ug.edu.gh 34 Nigerian government through ASCN appeared to be inadequate. There is need for the government to double its efforts through the agricultural transformation agenda towards solving farmers‘ problem. Majority of the low income farmers who are often associated with smaller harvests exhausted their grain stock during the lean season, and purchased grains for the coming planting season at higher price than affordable, which agrees with the findings of Takeshima et al. (2014). Non availability of cowpea grains at planting time was attributed to the lack of well collaborative work between scientists, grain companies and farmers. Therefore, there is need for farmers to work with other stakeholders especially extension agents for proper guidance. For example, most of the farmers interviewed reported that some cowpea varieties popularly called ―Danrani‖ coded: 288 and TVX-3236 became extinct despite their tolerance to drought and their high yield. The extinction was due to irregular visits by the extension officers and follow up from the researchers and lack of impact assessment from researcher‘s side. To my observation, the extinction from the farmers hand may be associated with natural disaster; such as insect inversion and climate change. This conformed to Bunkure Farmer‘s attestation when farmers lost their irrigated cowpea in 2012 (Field visit after FGD). Varaprasad & Sivaraj (2010) reported that climate change increased the risk of extinction for many species, and there may be loss of genetic variability even if the species survived. Therefore, further study is suggested to find out the causes of extinction and conservation efforts should be improved. Lack of insect resistant (pod borer for example) varieties, resulted in high cost of production leading to either small farm sizes or abandoned cowpea cultivation. Farmers had to apply various types of insecticides at different formulations to control various insects. The farmers need insect resistant varieties, as this will help them save some money by either reducing the number of insecticide applications or not spraying at all. Farmers in the 3 locations identified University of Ghana http://ugspace.ug.edu.gh 35 various insects that affected cowpea at both pre-flowering and post flowering stages particularly, flower thrips, pod boring and sucking insects. Pod boring insects especially Maruca vitrata was among the most damaging insects that causes up to 80% yield loss if not properly controlled (Fatokun, 2009). This was found in Bunkure district when they had an outbreak of such insects that resulted in 100% loss. Intensive screening of cowpea germplasm for Maruca resistance showed that there is no known cultivar with more than weak resistance (Murdock et al., 2008). Closely related cowpea species (Vigna vexillata) with resistance to Maruca were unfortunately genetically incompatible with the cultivated species. One of the farmers (Malam Aliyu) attested that he applied various formulations of different insecticides for more than15 times, yet no grain was harvested from the field. Inadequate varieties with combined farmers and consumers‘ attributes also limited production, though Danbazara (277) was adopted by the farmers in all locations. Currently, multidisciplinary teams of scientist in collaboration with African Agricultural Technology Foundation (AATF) are working toward solving that problem in Africa, using transgenic cowpea. There is also need for more collaboration toward solving farmers‘ problems of such nature. Farmers perceived colour, grain size and texture as the most important grain quality traits. Danbazara (277) was preferred by majority of the farmers in all the locations, because of its white colour, rough texture and medium size. These are the first traits considered by many farmers especially those that cultivate cowpea for commercial purpose. This result is not in agreement with the findings of Kormawa et al (2000) who reported brown grain cowpea as the most preferred by consumers over the white colour cowpea grain in the Nigerian markets. This would not be a surprise, because generally, in the study area the brown colour cowpea commands a University of Ghana http://ugspace.ug.edu.gh 36 higher price and producers prefer to sell them as exports to neighboring states and countries around, while the white colour grains are sold in the area of study. Over 80% of the farmers preferred yield and grain quality traits in all the districts visited (Bunkure, Wudil, Bichi and Kano metropolis) districts. However, farmers in Bichi area preferred grain quality traits and earliness. The latter was attributed to their proximity to cowpea consumers (Traders and processors) at Dawanau central market, while earliness was due to early cessation of rain fall. This is in conformity with the findings of Langyintuo et al. (2003) that even low income consumers were vigilant in identifying products that do not meet their standards, and are willing to pay a premium for products that match their preferences. Cowpea consumers identified grain colour (scored=64%), texture (scored=68%) and size (scored=44%) as the most important grain quality traits. These are the determinants of cowpea preference and indicators for marketability of cowpea in Nigeria in particular and Africa in general. Visible characteristics are directly related to the way cowpeas are used in diets preparations. For instance, cowpea grain colour or eye colours are important considerations when the intended use for the grains required decortication to remove flecks. Poor milling and winnowing may leave some flecks for which consumers showed low tolerance to it. This agrees with the findings of Langyintuo et al. (2004) and Cobbinah et al. (2011) that consumers in Africa, preferred ‗kosai‘ (cowpea fritters) without black flecks, while in Ghana black flecks have little impact on the use of cowpeas for ‗Tubani‘ (steamed cowpea paste) or a mixture of cowpea with rice. Small scale cowpea processors, who are mainly women, preferred short cooking time (45%), ease of dehulling (52%) and less oil consumption (43%). Long cooking time was identified as an undesirable characteristic of some cowpea cultivars due to relatively large amount University of Ghana http://ugspace.ug.edu.gh 37 of energy required to process cowpea grains leading to an increasing in the cost of diets preparation and reduces income (Yeung et al., 2009) Majority of the processors preferred small to medium grained cowpea due to its high expansion capacity and less oil consumption when preparing cowpea diets. They also perceived brown coloured grains are not suitable for ―Akara‖ and ―moin moin‖ making due to its grittiness and difficult in dehulling. This conforms to the findings of Yeung et al. (2009) that parenchyma cell of some cowpea cultivars clumps, creating a gritty and uncooked feeling when consumed. Cowpea varieties with smooth and wrinkled grain coat have thicker testa than those with rough grain coat, which in turn affects the rate of water absorption (Ojomo & Cheda, 1972). Similarly, cultivar with comparatively thick, smooth grain coats have a slow initial rate of water absorption, whereas thin grain coat cultivar have high initial rate and dehull better after soaking (Sefa-Dedeh et al., 1979). Over 75% of the consumers mostly women rated protein and micronutrient (Fe and Zn) contents as very important and important traits, respectively. There were significant differences between the perceptions of farmers and consumers for all grain quality attributes except for cooking quality, oil consumption and micronutrients content. This was attributed to the differences in diets and mode of cowpea processing. Processors prioritized cooking time, oil consumption during diets preparations and expansion capacity as either important or very important. While end users emphasized on the nutritional (Protein, Fe and Zinc) grain quality traits. All the farmers interviewed grew cowpea as a cash crop and recognized the importance of quality in the marketing of their produce. The specific preference for cowpea was generally similar for farmers, processors and consumers. For grain colour, white was the predominant colour (over 90%) for farmers and consumers. This is not surprising because University of Ghana http://ugspace.ug.edu.gh 38 almost all cowpea in the Kano market are white. However, there are niche markets for brown cowpea especially in Kura and Chiromawa town along Kano Zaria road. The study further revealed a higher preference for larger grain sizes of cowpea for end users and higher preference for small to medium sized grains for processors. These results agree with the findings of many researchers Madode et al. (2011); (Coulibaly et al. (2009) and Boys et al. (2007) in which they reported that consumers in West and Central Africa have their preferences for cowpea and varied from one location to another. 3.6 Conclusion and Recommendations The results of the PRA exercises in the Sudan savanna and Sudan-Sahelian transition zones of Nigeria have identified the major constraints to cowpea production and needs for cowpea smallholder farmers, processors and consumers. In all the 3 districts visited, farmers identified 10 constraints to cowpea production and emphasized on the first 5 (improved grain related issues): inadequacy of improved cowpea seeds at planting time, lack of insect resistant varieties, diseases resistant varieties and drought tolerance varieties. Cowpea varieties preferences varied across the districts, though adopted Danbazara (277) due to its high yield and good grain quality traits. This indicated the need to target for location and trait specific varieties to satisfy different farmers and consumers of different locations. The cowpea quality traits identified, were classified according to the category of cowpea users (producers, processors or end users). Quality traits were considered by both farmers and consumers to be important criteria for making choices for cowpea varieties. Majority of the processors preferred short cooking time, an important characteristic for women who are involved in food preparation, ease of removal of the hilum and testa and less oil consumption. End users, on the other hand, considered protein, micronutrients, grain colour and size. Specific cowpea University of Ghana http://ugspace.ug.edu.gh 39 characteristics desired by cowpea farmers, processors and consumers could increase profits for farmers and ultimately the processors as well. It is therefore, recommended that PRA exercises should be encouraged to identify the needs of the beneficiaries along the value chain for each set of players including farmers, processors, retailers and wholesalers. This calls for the need for better targeting varieties for site specific. Processing traits such as short cooking time and less oil consumption and consumers‘ traits such as protein and other nutrient should also be given priority in the selection of cowpea lines, for setting breeding objectives. This will help in setting breeding objectives toward problem solving, planning, implementation, monitoring and evaluation of breeding activities as well as impact assessment. Farmers‘ preference may vary according to sites and blanket recommendations could mask the preference choices in locations and lead to rejection of newly developed variety. University of Ghana http://ugspace.ug.edu.gh 40 CHAPTER FOUR 4. VARIATION AMONG COWPEA ACCESSIONS FOR GRAIN QUALITY TRAITS AS REVEALED BY MORPHOLOGICAL, BIOCHEMICAL AND SINGLE NUCLEOTIDE POLYMORPHISM (SNP) MARKERS 4.1 Introduction Cowpea is an important source of vegetable protein and is extensively grown throughout the world. Utilization of vegetable protein is gaining increasing attention due to the world‘s need for more low-cost dietary proteins, particularly for low income countries (Wang et al., 1997). The high cost and limited availability of animal proteins in the developing countries has directed interest towards several grain and legume proteins as potential sources of vegetable protein for food use (Sathe & Salunkhe, 1981). As a food source to over 200 million people in Africa (Fatokun et al., 2012), cowpea provides the cheapest protein supplement to the urban and rural poor in Africa. An extremely important issue to note is the sheer deficiency of urban diets in relation to the nutritional needs and the impact of this on people‘s health (Porter et al., 2003). This is particularly true in Nigeria given the general rising needs for cowpea to urban and rural poor (Ayinde, 2005). This has affected the protein and other essential element intake of most Nigerians. The eating habits of most poor people thus revolve round cowpea consumption in many forms; either through direct cooking of cowpea, processing it into ―akara” (cowpeas cake), ―moin-moin” or as component of other meals, as in cowpeas soup, rice and cowpeas, etc. This has made Nigeria to be the largest cowpea consumer in the world. To prepare such foods, a wide range of variation in cowpea grain quality traits is required in order to satisfy the various needs of the processors and end-users. The physical appearance of cowpea grain such as grain coat colour, eye colour, texture and shape; short cooking time and eating qualities are very important characteristics for cowpea University of Ghana http://ugspace.ug.edu.gh 41 consumers. Consumers in different parts of the world have specific preferences for cowpea. For example, cowpea variety (coded as 288) with medium grains size, short cooking time and white in colour is preferred by the processors in Northern Nigeria, while Ghana and Cameroon consumers prefer white, large grained and rough cowpea types (Langyintuo et al., 2004). The diversity in crop varieties is essential for agricultural development for increasing food production, poverty alleviation and promoting economic growth. Various studies have reported a wide variability in cowpea physicochemical properties, genetic variations for proteins (Kachare et al., 1988; Ajeigbe et al., 2008), nutritional composition, cooking time in cowpea (Onyenekwe et al., 2000), mineral components of cowpea grains (Anele et al., 2011) and nutritional grains quality traits in cowpea (Boukar et al., 2010) Despite much effort to assess the variability of cowpea mineral elements by many authors (Oluwatosin, 1998; Asante et al., 2009; Mamiro, 2011), large gaps still remain in the assessment of cowpea diversity for grain micronutrients particularly, iron and zinc. Further screening of cowpea accessions with molecular markers and physicochemical properties will facilitate breeding for enhanced nutritional quality traits in cowpea. Cowpea germplasm collection and characterization will be initiated to explore the important nutritional grain quality traits to satisfy the growing need of such traits. Quantification of phenotypic divergence in a crop is considered a necessary pre-breeding tool (Acquaah, 2007). Genetic variation are started with the assessment of morphological variation, and followed with molecular studies to facilitate the selection process. University of Ghana http://ugspace.ug.edu.gh 42 The specific objectives of this study were to: 1. Assess the genetic variation among cowpea accessions from Africa and other parts of the world using morphological, biochemical and molecular traits 2. Determine the association between nutrient quality traits and physicochemical properties of cowpea grain 3. Identify promising accessions for future planning for development and deployment of nutritionally (iron and zinc) enhanced cowpea varieties. 4.2 Materials and Methods 4.2.1 Source of cowpea accessions A total of 200 cowpea accessions from 24 countries (Table 4.1) were collected from IITA, IAR and WACCI breeding unit, Ghana. Most of the accessions were from the primary center of diversity especially the African countries. The accessions consisted of: 4 landraces, 195 core collection genotypes including wild types and farmer‘s preferred variety (Table 4.1). University of Ghana http://ugspace.ug.edu.gh 43 Table 4.1: List, origin and source of 200 cowpea accessions selected in this study S/N Accession Origin Source S/N Accession Origin Source S/N Accession Origin Source 1 TVu-13088 Benin IITA 38 TVu-16566 Guinea IITA 75 TVu-4886 Niger IITA 2 TVu-16343 Benin IITA 39 TVu-16574 Guinea IITA 76 TVu-5040 Niger IITA 3 TVu-16393 Benin IITA 40 TVu-16575 Guinea IITA 77 TVu-5444 Niger IITA 4 TVu-16399 Benin IITA 41 TVu-13457 Kenya IITA 78 TVu-5495 Niger IITA 5 TVu-16400 Benin IITA 42 TVu-13468 Kenya IITA 79 TVu-5562 Niger IITA 6 TVu-8673 Benin IITA 43 TVu-13495 Kenya IITA 80 TVu-6238 Niger IITA 7 TVu-8742 Benin IITA 44 TVu-1801 Kenya IITA 81 Ghana Ghana WACCI 8 TVu-8745 Benin IITA 45 TVu-2651 Kenya IITA 82 TVu-4841 Niger IITA 9 TVu-8751 Benin IITA 46 TVu-347 Kenya IITA 83 TVu-1 Nigeria IITA 10 TVu-8758 Benin IITA 47 TVu-433 Kenya IITA 84 TVu-10408 Nigeria IITA 11 TVu-15164 Cam IITA 48 TVu-8454 Kenya IITA 85 TVu-1172 Nigeria IITA 12 TVu-15187 Cam IITA 49 TVu-8455 Kenya IITA 86 TVu-1261 Nigeria IITA 13 TVu-15197 Cam IITA 50 TVu-11788 Malawi IITA 87 TVu-13094 Nigeria IITA 14 TVu-13867 C AR IITA 51 TVu-15086 Malawi IITA 88 TVu-13095 Nigeria IITA 15 TVu-14090 CAR IITA 52 TVu-15088 Malawi IITA 89 TVu-14012 Nigeria IITA 16 TVu-14109 CAR IITA 53 TVu-15107 Malawi IITA 90 TVu-14172 Nigeria IITA 17 TVu-15251 Chad IITA 54 TVu-15114 Malawi IITA 91 TVu-1449 Nigeria IITA 18 TVu-15323 Chad IITA 55 TVu-15141 Malawi IITA 92 TVu-1509 Nigeria IITA 19 TVu-16461 Chad IITA 56 TVu-15143 Malawi IITA 93 TVu-1551 Nigeria IITA 20 TVu-15204 Congo IITA 57 TVu-9929 Malawi IITA 94 TVu-15560 Nigeria IITA 21 TVu-15223 Congo IITA 58 TVu-14626 Mali IITA 95 TVu-15617 Nigeria IITA 22 TVu-15225 Congo IITA 59 TVu-14845 Mali IITA 96 TVu-15685 Nigeria IITA 23 TVu-15243 Congo IITA 60 TVu-7628 Mali IITA 97 TVu-15686 Nigeria IITA 24 TVu-7747 CD IITA 61 TVu-7638 Mali IITA 98 TVu-15694 Nigeria IITA 25 TVu-9446 Egypt IITA 62 TVu-7664 Mali IITA 99 TVu-15695 Nigeria IITA 26 TVu-9463 Egypt IITA 63 TVu-7966 Mali IITA 100 TVu-15719 Nigeria IITA 27 TVu-9725 Egypt IITA 64 TVu-8237 Mali IITA 101 TVu-15725 Nigeria IITA 28 TVu-1084 Ghana IITA 65 TVu-13096 Niger IITA 102 TVu-15762 Nigeria IITA 29 TVu-15884 Ghana IITA 66 TVu-14932 Niger IITA 103 TVu-15775 Nigeria IITA 30 TVu-15892 Ghana IITA 67 TVu-14939 Niger IITA 104 TVu-158 Nigeria IITA 31 TVu-15895 Ghana IITA 68 TVu-14967 Niger IITA 105 TVu-15811 Nigeria IITA 32 TVu-4489 Ghana IITA 69 TVu-4672 Niger IITA 106 Sampea 13 Nigeria IAR 33 TVu-7362 Ghana IITA 70 TVu-4701 Niger IITA 107 TVu-1586 Nigeria IITA 34 TVu-7999 Ghana IITA 71 TVu-4708 Niger IITA 108 TVu-15866 Nigeria IITA University of Ghana http://ugspace.ug.edu.gh 44 Table 4.1: (Cont’d) List, origin and source of 200 cowpea accessions selected in this study S/N Accession Origin Source S/N Accession Origin Source S/N Accession Origin Source 35 TVu-997 Ghana IITA 72 TVu-4747 Niger IITA 109 TVu-16414 Nigeria IITA 36 TVu-999 Ghana IITA 73 TVu-4776 Niger IITA 110 TVu-16483 Nigeria IITA 37 TVu-16521 Guinea IITA 74 TVu-4808 Niger IITA 111 TVu-16504 Nigeria IITA 112 TVu-18 Nigeria IITA 142 TVu-990 S. Africa IITA 172 Biyu local Nigeria IAR 113 TVu-301 Nigeria IITA 143 TVu-10454 Tanzania IITA 173 TVu-1985 USA IITA 114 TVu-320 Nigeria IITA 144 TVu-15355 Tanzania IITA 174 TVu-1986 USA IITA 115 TVu-3629 Nigeria IITA 145 TVu-15360 Tanzania IITA 175 TVu-202 USA IITA 116 TVu-3710 Nigeria IITA 146 TVu-15381 Tanzania IITA 176 TVu-2155 USA IITA 117 TVu-409 Nigeria IITA 147 TVu-2672 Tanzania IITA 177 TVu-232 USA IITA 118 SB_sata Nigeria IAR 148 TVu-7188 Tanzania IITA 178 TVu-233 USA IITA 119 TVu-43 Nigeria IITA 149 TVu-7293 Tanzania IITA 179 TVu-297 USA IITA 120 TVu-4536 Nigeria IITA 150 TVu-8330 Tanzania IITA 180 TVu-30 USA IITA 121 TVu-4545 Nigeria IITA 151 TVu-8612 Togo IITA 181 TVu-315 USA IITA 122 TVu-4558 Nigeria IITA 152 TVu-8619 Togo IITA 182 TVu-332 USA IITA 123 TVu-486 Nigeria IITA 153 TVu-117 Uganda IITA 183 TVu-337 USA IITA 124 TVu-50 Nigeria IITA 154 TVu-1184 Uganda IITA 184 TVu-374 USA IITA 125 TVu-526 Nigeria IITA 155 TVu-1185 Uganda IITA 185 TVu-384 USA IITA 126 TVu-59 Nigeria IITA 156 TVu-1272 Uganda IITA 186 TVu-387 USA IITA 127 TVu-746 Nigeria IITA 157 TVu-1283 Uganda IITA 187 TVu-393 USA IITA 128 TVu-857 Nigeria IITA 158 TVu-1330 Uganda IITA 188 TVu-408 USA IITA 129 TVu-875 Nigeria IITA 159 TVu-401 USA IITA 189 TVu-415 USA IITA 130 TVu-971 Nigeria IITA 160 TVu-1004 USA IITA 190 TVu-430 USA IITA 131 TVu-1706 Pakistan IITA 161 TVu-1016 USA IITA 191 TVu-456 USA IITA 132 TVu-11462 Philippines IITA 162 TVu-1030 USA IITA 192 Baban wake Nigeria IAR 133 TVu-11503 Philippines IITA 163 TVu-1036 USA IITA 193 TVu-497 USA IITA 134 TVu-14396 Senegal IITA 164 TVu-1037 USA IITA 194 TVu-566 USA IITA 135 TVu-10745 S. Leone IITA 165 TVu-1045 USA IITA 195 TVu-697 USA IITA 136 TVu-1171 S. Africa IITA 166 TVu-1251 USA IITA 196 TVu-13305 Zambia IITA 137 TVu-1236 S. Africa IITA 167 TVu-1494 USA IITA 197 TVu-15058 Zambia IITA 138 TVu-2397 S. Africa IITA 168 TVu-1560 USA IITA 198 TVu-16637 Zambia IITA 139 TVu-274 S. Africa IITA 169 TVu-1562 USA IITA 199 TVu-8407 Zambia IITA 140 TVu-36 S. Africa IITA 170 TVu-1616 USA IITA 200 Maikube Nigeria IAR 141 TVu-84 S. Africa IITA 171 TVu-1715 USA IITA IITA= International Institute of Tropical Agriculture, IAR= Institute for Agricultural Research, Zaria- Nigeria, WACCI= West Africa center for crop Improvement, breeding unit, SB_sata= Saka babba sata (landrace), Cam= Cameroon, CD= Cote’d voire, University of Ghana http://ugspace.ug.edu.gh 45 4.2.2 Soil sampling and analyses Soil physicochemical properties of the trial site and soil used for pot experiment were analyzed in order to know the approximate soil mineral content. This is because plant uptake soil mineral elements and translocate to other tissues including grains (White and Broadley, 2009). The availability of such elements in the soil may confound the end results of the iron and zinc concentration in the grain. Soil samples were collected from the Institute for Agricultural Research farm in Samaru. The samples were taken from 2 layers (0-15 cm and 15-30 cm) at 6 different locations within the experimental area as described by Anderson & Ingram, (1998). Un-decomposed plant materials were sorted out then the soil was air dried and the samples ground and passed through a 2 mm sieve to obtain the fine earth. The sieved soil samples were stored in thick polythene bags for laboratory analyses and subsequent screen house studies. Soil auger was driven gently into the soil until 0-15cm was completely buried. The auger was then removed and pushed out into a moisture can. This was repeated for 15-30 cm for different portions. The samples were than oven-dried at 105°C for 24 hours. The density of the soil was calculated as: Particle size distribution Particle size distribution was determined following Walkley & Black (1934). Forty grams (40 g) of the fine earth fraction of the soil was taken into a plastic bottle and mixed with 100 ml of 5% calgon (sodium hexametaphosphate) solution was added. The content of the bottle was then shaken on a mechanical shaker for 2 hours after which it was transferred into a 1.0 litre measuring cylinder and topped up to the mark with distilled water. The suspension was then agitated with a plunger and five minutes thereafter, the density of the suspension (silt and clay) was taken using a University of Ghana http://ugspace.ug.edu.gh 46 hydrometer. The hydrometer reading of the suspension was taken again after eight hours (clay). The suspension temperatures T1 and T2 were recorded during the 5 minute and 8 hour hydrometer readings. The contents of the cylinder after the eight hour reading were emptied onto a 47 µm sieve. The sand retained on the sieve was then washed off into a moisture can and dried at 105° C for 24 hours, after which the dry weight of the sand was recorded. Blank sample hydrometer readings at five minutes and eight hours respectively were also taken for the 5% calgon topped up to 1.0 liter. The particle size distribution was then determined using the formulae below. Percentage clay = x 100 Percentage sand = x 100 Percentage silt= ) The temperature for density of the soil particle suspensions was corrected by subtracting increase in weight of the particles from the blank hydrometer reading (Day, 1965). The correction was made because, for every 1°C increase in temperature, above 19.5°C, there is an increase of 0.3 in the density of the particles in suspension. The moisture content of the soil at field capacity was determined by saturating 500 g of soil sample with water and covered with a plastic and allowed to drain for 3 days. Sub samples were then taken and oven dried at 105°C. The gravimetric water content was determined as the difference in mass between moist soil and oven-dried soil per oven-dried soil. Soil chemical analysis The pH of the fine earth of the two soils was determined in a 1:1 soil to distilled water ratio. Ten gram (10 g) soil was weighed and 10 mL of distilled water added, stirred vigorously and allowed to stand for 30 minutes. A microprocessor pH 213 meter was calibrated, and then inserted into the supernatant of the soil solution and the pH read. Available iron and zinc contents University of Ghana http://ugspace.ug.edu.gh 47 in the soil were measured by atomic absorption spectrometry (AAS‐240FS from Varian Company) after digestion with acid. The organic carbon content of the soil was determined using the wet combustion method of Walkley and Black (1934). Total N of the soil and available phosphorus were determined by using the Kjeldahl digestion procedure as outlined by Anderson & Ingram (1998). 4.2.3 Phenotypic characterization of the accessions Field trial was established in 2012, at the experimental station of the Institute for Agricultural Research, Samaru, Zaria-Nigeria (11° 09′ 31.38″ N and 7° 38′ 15.19″E). Two hundred cowpea (200) accessions were planted out in an augmented block design as described by Khairwal et al (2007) with 2 checks; Sampea-13 (early maturing) and Biyu local (late maturing), in a 9 m row per accession per plot. Data were taken on the days to 50% flowering, growth habit, flower colour, pod placement, 100 grain weight, grain coat colour, grain coat pattern and grain coat texture following Mahalakshmi et al. (2007) cowpea descriptors. 4.2.3.1 Physicochemical analyses of cowpea grain The proximate analysis was conducted on 200 cowpea accessions based on the procedure of the Association of Analytical Chemist (AOAC, 2000). During the analysis, moisture, fiber, ash, crude fats, proteins and carbohydrates contents were determined and expressed in percentages (%). Digestion of cowpea grain Two grams of the dried cowpea grain were taken at random from each accession, grinded with pestle and mortar. Half gram (0.5g) of the cowpea powder was taken and transferred to a digestion flask, containing 0.5 g digestion mixture, and 20 ml of concentrated sulfuric acid added to the mixture. The solution was heated until it became clear and frothing ceased. Then it was University of Ghana http://ugspace.ug.edu.gh 48 boiled briskly for another 2 hours by adding 20 ml mixture of perchloric acid and nitric acid; 10:10 ml (v/v), cooled and to the digest about 50 ml water was added in 5 ml portions with mixing. The digest was then transferred to a 100 ml volumetric flask and the volume made up to the mark. This was done 3 times for each sample. All the proximate values were reported in percentage (%) in accordance with the AOAC (2000) procedures. Moisture Content Determination Plastic dishes were washed and dried to a constant weight in an oven at 100°C. They were later removed and cooled in a desiccator and weighed (W1). Two grams of the grounded powder of each accession were placed in the weighed moisture dish (W2). The dish containing the sample was kept in an oven for about 3 hours, the sample was removed and cooled in the desecrator and weighed W3. The percentage (%) of moisture was calculated as 100 12 32   WW WW Determination of Ash content For ash content determination, crucibles were rinsed, dried in the oven, cooled in the desiccators and weighed (W1). Two grams of the grounded sample were placed in the crucibles and weighed (W2). They were transferred into the Muffle Furnace and incinerated at 55° C, then removed and cooled in the desiccator and then weighed (W3). The percentage (%) of Ash was calculated as 10032 W CC Determination of Fibre Two grams of the sample was placed in a beaker containing 1.2 ml of H2SO4 per 100 ml of solution and boiled for about 30 minutes, the residue was filtered and washed with hot water. The residue was then transferred to a beaker containing 1.2 gram of NaOH per 100ml of solution University of Ghana http://ugspace.ug.edu.gh 49 and boiled for about 30 minutes, the residue was washed with hot water and dried in an oven and weighed (C2), the weighed sample was incinerated in a Furnace at 55O°C, removed and allowed to cool, and weighed (C3). Fiber content was estimated from the loss in weight of the crucible and its content on ignition as shown below: The percentage (%) of Fibre was then calculated as: 10032 W CC Determination of Lipids (Fat) Lipid was determined using Soxhlet apparatus after 8 hours of refluxing. Oven dried digests were transferred to a desiccator and allowed to cool before the weight was taken. The percentage (%) Fat was calculated as: = Weight of fat × 100 Weight of sample Determination of Protein The nitrogen value, which is the precursor for protein of a substance, was determined by microkjeldahl method described by Hussain et al. (2010), involving digestions, distillation and finally titration of the sample. The nitrogen value was converted to protein by multiplying the factor of 6.25. One tablet of copper catalyst and 25 ml concentrated Sulfuric acid H2SO4 were added to the digest in the fume cupboard and heated till solution assumed a green colour. The digest was cooled and wash any black particle that appeared at the mouth and neck of the flask was washed down with distilled water. The digest was then transferred into 250 ml flask with several washings with distilled water. This was followed with distillation using the Markham distillation apparatus. After 15 minutes of steaming, 100 ml conical flask containing 10ml of boric indicator was placed under the condenser. Ten Mill (10 ml) of the digests was pipetted into the body of the apparatus via the small funnel aperture; washed down with distilled water followed by 10ml of 40% NaOH solution. Ammonium sulphate was then collected after 7 minutes of steaming. The solution in the receiving flask was then titrated using N/100 (0.01N) University of Ghana http://ugspace.ug.edu.gh 50 hydrochloric acid and calculated the Nitrogen content and hence the protein content of each sample. A blank solution was run along with each sample, until all the 200 samples were exhausted. Determination of Carbohydrate (CHO) content Carbohydrate content was determined by following the method described by Pearson (1976). In this method, carbohydrate was obtained by subtracting the sum of the percentages of moisture, ash, crude protein and crude fiber from 100: Carbohydrate content =100- (% of moisture + %Ash + %Protein + %Fibre) 4.2.3.2 Iron and zinc content determination The Harvest plus crop sampling protocol for micronutrients determination (Fatokun & Stangoulis, 2008) was used for this study. At 95% pod maturity, 100 g of matured pods were collected from 10 plants from each accession and the samples placed in clean, well labeled paper bags. The collected samples were dried at 60ºC for 3 days in a contaminant-free oven. The pods were threshed by gentle twisting of the pods with hand over a dust-free plastic tray. Grains of 5 g from each of the accession were collected based on quartering procedure of Harvestplus (Fatokun & Stangoulis, 2008). The Quartering procedure is done by spreading the grains in a circular form on a clean acid-washed tray. The circle was divided into four roughly equal parts and discarded the two diametrically opposite quarters, and remixed the remaining two parts. This was repeated until 5-10 g grains were obtained. The grain was ground using non-contaminating zirconium pestle and mortar. The milled samples were packaged in cleaned, paper envelopes and then digested with H2SO4 acid. Atomic Absorption Spectrophotometry (AAS) analysis was employed for both iron and zinc concentration following Benton-Jones (1989) technique, based on nitric perchloric acid digestion. University of Ghana http://ugspace.ug.edu.gh 51 A total of 0.25 mg of each sample was acid digested with 5 ml of a 2:1 mixture of 65% nitric acid (HNO3) and 70% perchloric acid (HClO4) in 50 ml Taylor digestion tubes for 2 hours. This was followed by a heat treatment of 200°C for another 2 hours and resuspension in 25 ml of deionized water. The concentration of iron and zinc grains of individual plant samples of each generation were then read using Atomic Absorption Spectrophotometer (AAS240SF, from Varian Company) machine. Readings were then evaluated against standard curves prepared from iron and zinc diluted to a concentration of 100 mg/kg and zinc diluted to 50 mg/kg. Measurement conditions were as recommended by the manufacturer for analysis of Zn and Fe in a cellulose based matrix and acquisition time of 60 seconds. Grain samples were scanned by pouring 0.2 ml of each sample into AAS cups sealed at one end by 4 μm Poly‐4 film. The actual iron and zinc concentration were estimated using the formular: ( ) – concentration of the blank solution Selection of parental lines Contrasting parental lines were selected, based on the field evaluation data and laboratory assessment of iron, zinc, protein and other physicochemical properties of the cowpea grain. Parental materials were selected using Mulamba & Mock (1978) rank summation index (Appendix 4.1) adopted from cowpea descriptors of IBPGR (1983) and Mahalakshmi et al. (2007) scoring system (Table 4.2). Conversely, these descriptors did not specify the clear range for low, medium and high protein, iron and zinc content. University of Ghana http://ugspace.ug.edu.gh 52 Table 4.2 Scoring system used to characterize the 200 cowpea accessions (Mahalakshmi et al., 2007) Traits Score 1 2 3 Maturity period Early (70-75days) Medium (80-95 days) late maturing (> 100 days) Growth habit Erect Semi-erect prostrating type Shattering Normal Low shattering High shattering 100 grain weight( g) 20g and above Between 15 & 19g less than 10g Protein content (mg/kg) 21-29%, 10-20% less than 10% Zinc content (mg/kg) 300 and above Between 100-290 below 100 Iron content (mg/kg) 100 and above 50-99 less than 50 Fibre content (%) 3.1- 4.9%, 2.1-3.0% 1.0-2.0% Carbohydrate (%) 71-78%, 60- 70% less than 59% Fat (%) 11-22%, 6-10% 1-5% Grain coat colour White Brown Others Grain coat texture Rough Smooth Wrinkle 1= scale for early maturing or high content; 2= medium maturing/content and 3= late maturing or low content 4.2.4 Molecular analysis 4.2.4.1 Characterization of cowpea accessions using SNP markers Thirty of the 200 accessions were discarded due to undesirable traits (susceptibility to insects and diseases, shattering etc) recorded against them. As a result, 170 cowpea accessions were selected based on the adopted Mahalakshmi et al. (2007) criteria outlined in Table (4.1) and rank summation index (Appendix 4.1), and genotyped using 119 SNP. Five seeds of each accession were planted in a pot (21cm height and 23 cm width) in the screen house of the Institute for Agricultural Research, Zaria, Nigeria in 2013. University of Ghana http://ugspace.ug.edu.gh 53 At 21 days after planting (DAP), the youngest and most fresh leaf sample was detached from the parent plant, placed on the Harris Cutting Mat, and gently pushed the Harris Uni-Core vertically through the leaf to take 4 discs from each genotype. The plunger was used to eject the sample disc from the Harris Uni-Core into the corresponding well of 96 well plate. The plates were well labeled according to the number of germplasm contained and blank well as suggested by the Kbioscience sampling protocol. Possible cross-contaminants were eliminated (Chum and Andre, 2009) by rinsing plunger and cutting end of the tool in a sodium hypochlorite (2% NaClO) solution and distilled water. The Harris cutting mat was periodically cleaned with the same decontaminants as the cutting tool, to further reduce the chance of cross-contamination. A perforated heat seal was placed on top of the storage plate containing the final number of samples disc. A conventional heat-sealer was used at 155°C for 2 seconds to seal up the plates. Similarly, a brief visual inspection was carried out to ensure that the wells were sealed. The plates were packaged with small desiccant bag to remove air. The second bag was then sealed up for additional protection. The plates were then shipped to Kbioscience in UK for genotyping. SNP Genotyping KASPar technology was used for genotyping at Kbioscience also known as LGC genomic laboratory in the United Kingdom. One hundred and nineteen (119) SNP markers (Appendix 4.2) were selected from the 1152 optimized SNP markers for cowpea, by the Generation challenge Program (GCP). Genotyping was conducted using KASP primer mix, KASP master mix and the extracted DNA samples, by following Kbioscience protocol as described by Thomson (2013). 4.3 Data analysis The 9 grain quality traits were recorded in 200 accessions. Descriptive statistic was performed on quality traits using MS excel (2010). The descriptive statistics generated were used University of Ghana http://ugspace.ug.edu.gh 54 to construct frequency graphs for each of the accession quality traits recorded. Principal component analysis (PCA) was performed using GenStat 15th edition, to determine the most significant variables associated with nutritional grain quality traits. PROC CORR for correlation analysis using Pearson‗s method was run to determine the association between nutritional quality traits (protein, iron and zinc) and physicochemical parameters. Power marker was used to determine, the number of allele per locus, allele frequency, gene diversity also known as expected heterozygosity (He). The allele frequency data was converted to bi-allelic and then estimated gene diversity at each locus according to Nei et al. (1983) and polymorphism information content (PIC) were also estimated using Power marker version 3.25 (Liu & Muse, 2005). Following the equation of Botstein et al. (1980) Where; Pij = is the frequency of j allele for I locus across summed of locus Where; Pi is the frequency of the j th allele for the i th marker, and summed over n alleles, it was used to represent the information value of a marker for detecting P th j th ith respectively. It depends on the number of detectable alleles and their frequency. Genetic distance (GD) for each pair of population was calculated using Nei et al. (1983) and a dendrogram using unweighted pair group method with arithmetic average (UPGMA) was generated using Power maker 3.25 (Liu & Muse, 2005) and viewed on Mega 5.0 software. University of Ghana http://ugspace.ug.edu.gh 55 4.4 Results 4.4.1 Soil physicochemical properties of the trial site The soil physicochemical properties are presented in Table 4.3. The level of iron and zinc content in the soil were generally low with less than 3 and 1 Cmol/kg for iron and zinc, respectively. Particle size distribution revealed that the soil was generally sandy-loam. Particle sizes were evenly distributed with approximately 1:1:4 ratios of clay silt and sand respectively. The surface soils were sandy loam, while the second layer (15-30 cm) was sandy-clay with small amount of loamy soil. The organic carbon content was low (ranged from 0.16 to 0.37), with adequate range (6.10-8.60 mol/kg) of cation exchange capacity (CEC) within the plots. The soil was generally acidic with pH ranging from 5.7 to 6.10. University of Ghana http://ugspace.ug.edu.gh 56 Table 4.3: Soil physicochemical properties of the evaluation site in 2013 and 2014 rainy seasons Source/ Location Particle size Distribution Corrected to 20 C (%) pH ratio 1:2.5 (%) [ Cmol 1kg] Clay silt sand Textural Class Water Cacl2 OC TN Ca Zn K Fe CEC R1 (1-33) 16 20 64 S L 5.90 4.80 0.37 0.350 3.20 0.42 0.23 2.65 6.80 R1(34-67) 24 20 56 SCL 6.20 4.60 0.27 0.104 4.00 0.16 0.22 1.48 8.60 R2 (68-102) 14 20 66 SL 6.10 4.80 0.31 0.350 4.40 0.39 0.18 2.57 8.20 R2(103-136) 22 18 60 SCL 5.90 4.50 0.21 0.082 4.00 0.20 0.16 1.30 7.80 R3 (137-169) 16 16 68 SL 5.70 4.60 0.31 0.105 2.40 0.38 0.23 2.65 6.10 Plant (168-200) 24 14 62 SCL 5.80 4.80 0.16 0.082 3.40 0.22 0.17 1.39 6.50 SL= sandy loam, SCL= sandy clay loam in the textural classification column, Numbers within the brackets in the first column are the accessions grown within a range, R1, R2 and R3= different portions where soil was collected and analyzed. University of Ghana http://ugspace.ug.edu.gh 57 4.4.2 Variations among cowpea accessions at morphological and molecular level 4.4.2.1 Descriptive analysis of growth habit and grain quality traits for 200 cowpea Accessions Frequency distribution of the six qualitative traits: grain coat texture, grain coat and eye colour pattern, 100 grain weight, shattering, growth habit and maturity period of the 200 cowpea accessions are presented in Fig. 4.1a to 4.1f, respectively. One hundred and thirty four of the accessions (67%) had smooth grain coat texture, while only 66 (33%) showed rough grain coat texture (Fig. 4.1a). Five descriptive grain coat colour pattern: Holstein, small eye, solid, Watson and Whippoorwill were observed (Fig. 4.1b). Sixty five (32.5%) of the accessions were Holstein, with definite eye margin extended over large portion of their grains, followed by 46 (23%) accessions with solid colour. Other colours observed included: 38 (19%) accessions having small eye, 29 (14.5%) Watson and 22 (11%) Whippoorwill. More than half (54%) had medium grain size followed by the large grained (35%) accessions and small grained cowpea accessions had the least (11 %) number (Fig. 4.1c). One hundred and seventy nine (89.5%) accession were normal (did not shatter), while 21 (10.5%) of the accessions showed a level of pod shattering before 10 % pod maturity and runs throughout the remaining period of maturity. Most of these accessions were small grained type (Fig. 4.1 d). Eighty six (43.0%) of the accessions showed semi erect growth habit, followed by 69 (34.5%) prostrating type and 45 (22.5%) erect type (Fig. 4.1 e) University of Ghana http://ugspace.ug.edu.gh 58 Ninety seven (48.5%) of the 200 accessions matured late, followed by 68 (34.0%) and 35 (17.5%) for the medium and early maturing respectively (Fig. 4.1 f). Figure 4.1 a: Cowpea descriptors (IBPGR, 1983; Mahalakshmi et al, 2007) showing the distribution of seed coat texture. 0 20 40 60 80 100 120 140 160 Rough Smooth 66 134 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 59 Figure 4.1 b: Distribution of seed coat colour pattern in cowpea accessions (Saunders, 1959) Holstein= seed colour has definite eye margin which extends over a large portion of the seed coat Small eye= Colour around the hilum is discontinuous, consisting of two separate elongated colour on either side of, or parallel to, the hilum. Solid = Seed colour extends more or less evenly over the entire seed coat. Watson = Edges of coloured area around hilum are not sharply demarcated but are broken up into separate fine spots especially at the micropylar end of seed. Whippoorwill = Seed coat has irregular areas of dark shade separated by lighter area. 0 10 20 30 40 50 60 70 Holstein Small eye Solid Watson Whippoorwill 22 29 46 38 65 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 60 Figure 4.1 c: Distribution of seed size based on 100 seed weight (g) Cowpea descriptors (Mahalakshmi et al, 2007) 0 20 40 60 80 100 120 small Medium large 22 108 70 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 61 Figure 4.1 d: Distribution of cowpea accessions based on shattering or non-shattering Mahalakshmi et al, 2007 0 20 40 60 80 100 120 140 160 180 200 Shattering Non shattering 21 179 D N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 62 Figure 4.1 e: Distribution of growth habit among 200 cowpea accessions (Mahalakshmi et al, 2007) 0 10 20 30 40 50 60 70 80 90 Erect Semi-erect Prostrating 86 69 E 45 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 63 Figure 4.1 f: Distribution of maturity periods of the cowpea accessions used Mahalakshmi et al, 2007 4.4.2.2 Variation in iron, zinc and protein contents within 153 African cowpea accessions The accessions were grouped into 3 groups namely; high, medium and low iron, zinc and protein contents, respectively (Fig. 4.2 a, b and c). Four of the accessions had high iron content (mean iron concentration >100 mg/kg), 45 accessions had intermediate iron content (50-99 mg/kg) and 104 showed low iron content of less than 50 mg/kg (Fig 4.2a). The four accessions with high iron content were TVu-1330, TVu-13495, TVu-8751 and 16400 with a mean grain iron concentration >100 mg/kg. Almost all the accessions (140) showed medium grain zinc concentration of 21-39 mg/kg and more than two-third (121) of the accessions had high protein content ranged between 21 and 29% (Fig 4.2b and C). 0 20 40 60 80 100 120 Early Medium Late 97 F 68 35 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 64 Figure 4.2 a: Frequency distribution of African cowpea accessions based on grain iron concentration as determined by Atomic absorption spectrophotometry 0 20 40 60 80 100 120 >100 mg/kg 50-99 mg/kg <50 mg/kg 45 104 4 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 65 Fig 4.2 b: Frequency distribution of African cowpea accessions based on zinc concentration as determined by Atomic absorption spectrophotometry (AAS) 0 20 40 60 80 100 120 140 160 >40 mg/kg 21-39 mg/kg <20 mg/kg 28 140 3 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 66 Figure 4.2 c: Frequency distribution of African cowpea accessions based on percentage grain protein content 0 20 40 60 80 100 120 140 21-29% 10-20% <10% 121 16 16 N u m b er o f a cc es si o n University of Ghana http://ugspace.ug.edu.gh 67 Variation in iron concentration, zinc concentration, protein contents and other physicochemical parameters are presented in appendix 4.2. Due to non-positive correlation of 3 important grain nutritional values (iron, zinc and protein), top 10 and least 10 of the accessions are selected each from iron, zinc and protein set and presented in Table 4.4. Mean grain iron concentration ranged from 1.01 to 325.15 mg/kg with a mean of 45.83 mg/kg. The highest iron concentration was recorded in TVu-1330 (329.15 mg/kg) followed by TVu-13495 (131.65 mg/kg) and TVu-16400 (123.53 mg/kg) and they originated from Zambia, Malawi and Benin. The accession from Ghana had the lowest iron concentration of 10.9 mg/kg (Table 4.4). Mean zinc content range from 10.01 mg/kg to 52.03 mg/kg with a mean of 26.66 mg/kg among 153 accessions (Table 4.5). The highest zinc concentration was recorded in TVu-15251 (52.03 mg/kg), followed by TVu-301 (51.8 mg/kg) and TVu-9725 (40.12 mg/kg) and they originated from Congo, Nigeria and Ghana respectively. TVu-15058 and TVu-1801 from Zambia and Malawi had the lowest zinc concentration of 10 and 13.46 mg/kg respectively (Table 4.4). The mean percentage of protein content was 20.9 %. It ranged from 1.7 % in TVu-544 from Nigeria to 29.9% for TVu-1185 from Uganda (Table 4.4). University of Ghana http://ugspace.ug.edu.gh 68 Table 4.4 Different set of accessions representing top 10, least 10, overall mean, standard deviation and ranges of iron, zinc and protein contents of 153 African cowpea accessions Accession Iron content Accession Zinc content Accession Protein content Top ten TVu-1330 329.15 TVu-15251 52.03 TVu-13495 29.93 TVu-13495 131.65 TVu-301 51.80 TVu-13088 29.89 TVu-16400 123.53 TVu-9725 40.12 TVu-15187 29.59 TVu-8751 108.65 TVu-7999 38.75 TVu-347 29.41 TVu-9725 99.42 TVu-13088 38.30 TVu-13468 29.37 TVu-999 98.05 TVu-15686 37.45 TVu-13867 29.30 TVu-4489 93.97 TVu-15107 36.62 TVu-13457 28.71 TVu-13088 92.01 TVu-15141 36.24 TVu-7362 28.59 TVu-15892 89.55 TVu-15719 36.15 TVu-8745 27.80 TVu-16521 87.68 TVu-15685 35.88 TVu-997 27.80 least ten TVu-15694 3.60 TVu-158 17.50 TVu-2397 2.69 Ghana 3.28 TVu-15164 17.21 TVu11462 2.60 TVu-1 1.50 TVu-8407 16.37 TVu-746 2.56 TVu-16483 1.50 TVu-14845 15.15 TVu-59 2.44 TVu-15694 1.13 TVu-13495 15.07 TVu-1706 2.38 Ghana 1.12 TVu-433 15.00 TVu-875 2.17 TVu-1 1.12 TVu-8742 14.87 TVu-971 2.17 TVu-16483 1.11 TVu-8454 14.54 TVu-1171 2.16 TVu-1184 1.10 TVu-15058 13.46 TVu-526 2.09 TVu-1185 1.01 TVu-1801 10.00 TVu-14845 1.72 Mean 45.83 26.66 20.98 STD 34.20 6.35 6.76 Range 1.01-329.15 10.0 - 52.03 1.72 - 29.93 STD= Standard deviation of the population; mean= population mean University of Ghana http://ugspace.ug.edu.gh 69 4.4.2.3 Variation in physicochemical properties of the United States and three Asian cowpea accessions The first three accessions (TVu-320, TVu-329 and TVu-3710) on the table 4.5 are from Philippines, they showed no significant variation among them in all the physicochemical properties analyzed. The U.S. accessions had mean iron content ranging from 1.04 mg/kg for TVu-30 to 58.36 mg/kg for TVu-1616 with a population mean of 17.73 mg/kg (Table 4.5). Zinc mean content varied significantly (16.25 to 386.53 mg/kg) among the accessions. TVu-332 had the least mean zinc content of 16.25 mg/kg, while TVu-1616 had the highest mean zinc content of 386.53 mg/kg (Table 4.5). Protein content among American accessions ranged from 1.21 % to 23.10 % shown by TVu-84 and TVu-8330, respectively. Table 4.5: Mean physicochemical properties of the United State and three Asian cowpea accessions Accessions Iron (mg/kg) Zinc (mg/kg) % Moisture % Fat % Ash % Protein % Fibre % CHO TVu-320 14.84 25.38 4.72 2.54 2.41 20.18 2.90 70.15 TVu-3629 15.84 23.19 6.74 2.58 3.65 18.16 2.55 68.87 TVu-3710 34.52 27.81 6.64 2.19 2.91 22.05 2.00 66.24 TVu-274 1.12 25.35 6.58 22.12 2.45 2.23 2.85 66.62 TVu-36 1.41 25.29 6.48 22.14 2.98 2.17 2.25 66.62 TVu-84 9.92 27.87 7.05 22.08 2.06 1.21 2.65 67.60 TVu-990 1.14 36.99 6.58 22.00 2.65 1.33 1.70 67.44 TVu-10454 17.80 23.43 6.00 2.48 2.32 22.16 1.85 67.04 TVu-1236 24.44 35.86 8.44 20.96 2.69 2.07 2.15 65.84 TVu-15360 4.80 38.59 7.00 2.69 2.22 22.70 1.05 65.39 TVu-15381 2.16 21.87 5.02 2.51 2.36 22.14 2.04 67.97 TVu-2672 1.13 27.77 7.04 2.02 3.99 22.28 2.70 64.67 TVu-7188 1.12 27.81 5.96 2.27 3.86 22.44 1.74 65.47 TVu-7293 1.13 23.92 6.30 2.82 2.54 22.74 2.05 65.60 TVu-8330 1.11 36.21 6.27 2.03 2.75 23.10 1.82 65.85 TVu-8612 14.31 31.44 4.38 2.21 2.01 22.84 1.64 68.56 TVu-8619 14.16 25.85 6.32 2.62 3.98 22.66 1.66 64.42 University of Ghana http://ugspace.ug.edu.gh 70 Table 4.5: (Cont’d) Mean physicochemical properties of the United State and three Asian cowpea accessions Accessions Iron (mg/kg) Zinc (mg/kg) % Moisture % Fat % Ash % Protein % Fibre % CHO TVu-117 1.12 24.46 6.57 2.07 2.17 22.46 1.72 66.73 TVu-1272 13.51 19.58 6.39 2.05 3.80 22.74 2.80 65.02 TVu-1251 14.22 22.18 3.15 2.84 2.96 22.18 2.95 68.87 TVu-1494 10.96 26.11 5.17 2.16 2.93 22.74 2.75 67.08 TVu-1560 17.26 22.73 5.35 2.12 2.67 22.42 1.30 67.44 TVu-1562 32.51 22.38 3.56 2.98 2.37 22.28 2.75 68.84 TVu-1616 58.36 386.53 4.13 2.65 2.03 22.48 2.75 68.71 TVu-1715 26.68 21.49 4.50 2.26 2.15 22.18 1.75 68.91 TVu-1985 33.37 26.66 4.34 2.56 2.56 22.54 1.10 68.00 TVu-1986 33.75 25.70 4.20 2.33 1.55 22.80 1.40 69.12 TVu-202 1.72 19.84 4.45 2.14 2.29 22.62 1.05 68.50 TVu-2155 10.73 164.80 4.04 2.51 2.37 23.10 1.24 67.98 TVu-232 20.95 45.02 4.39 2.49 2.00 22.64 2.90 68.48 TVu-233 35.17 36.19 5.92 2.12 2.55 22.18 4.82 67.23 TVu-297 13.60 18.24 4.50 2.27 2.18 22.42 2.60 68.63 TVu-30 1.04 21.47 5.78 2.82 2.75 22.84 1.46 65.81 TVu-315 14.31 18.38 4.41 2.65 2.28 22.16 3.80 68.50 TVu-332 24.09 16.57 4.18 2.57 2.88 22.80 2.75 67.57 TVu-337 27.79 19.55 4.43 2.64 2.56 22.18 1.15 68.19 TVu-374 8.98 24.93 0.81 6.90 2.20 22.18 2.10 67.91 TVu-384 8.87 23.97 0.25 6.73 2.50 19.42 2.90 71.10 TVu-387 32.40 19.70 0.89 6.59 2.43 21.16 2.45 68.93 TVu-393 25.27 21.13 0.16 7.17 2.69 22.10 1.14 67.88 TVu-408 1.19 20.00 1.69 7.52 2.99 20.92 1.98 66.88 TVu-415 43.93 20.97 0.70 7.13 2.42 22.12 2.05 67.63 TVu-430 42.88 19.41 1.01 6.28 3.99 21.84 2.15 66.88 TVu-456 41.34 26.14 0.92 6.30 5.23 21.80 2.95 65.25 TVu-566 27.93 26.62 0.64 6.18 2.80 22.64 2.80 67.74 TVu-697 26.96 25.75 1.24 6.06 2.84 22.12 1.86 67.74 TVu-13305 21.40 20.42 0.92 6.12 2.67 22.74 2.54 67.55 Mean 17.73 35.99 4.39 5.48 2.72 19.98 2.20 67.44 STD 14.3 56.38 2.26 5.99 0.67 6.41 0.76 1.43 Range 1.04-58.36 16.57- 386.53 0.16-8.44 2.02- 22.14 1.55- 5.23 1.21- 23.10 1.05-4.82 64.42- 71.10 All accessions are from USA except the first 3 that are from Philippines; CHO= carbohydrate; STD= standard deviation. University of Ghana http://ugspace.ug.edu.gh 71 4.4.3 Principal Component Analysis Principal component analysis was performed to identify the most contributing traits to the total variability observed. The results of the Principal component analysis for the 9 nutritional grain quality traits are presented in Table 4.6. The traits analyzed were: iron, zinc, protein, moisture content, fats content, ash content, fibre content, carbohydrate content and 100 grain weight. The first 4 principal components axis had eigenvalues greater than one (eigenvalues >1) and contributed to more than half (63.00%) of the total variation, hence, they were retained. The first principal component contributed 23.36% of the total variation, which were contributed by fat (0.5836), protein (0.6482), and grain iron concentration (0.3258). Fifteen point twenty four percent (15.24%) of the total variation was contributed by 100 grain weight (0.4057), carbohydrates (0.5313) and fibre (0.3562) clustered on the second principal component axis. Similarly, ash (0.4664), fibre (0.5313) and moisture (0.303) content were concentrated in the third principal component (PC3) axis, and contributed 12.69% of the total variation. In the fourth principal component (PC4) axis, iron concentration (0.4918), zinc concentration (0.7004) and fibre content (0.3813) contributed 11.72% to the total variance. University of Ghana http://ugspace.ug.edu.gh 72 Table 4.6: Principal Component Analysis showing the contribution of grain quality traits to the total variation among the cowpea accessions PC1 PC2 PC3 PC4 PC5 100 grain weight 0.0052 0.40579 0.25123 0.22905 0.79554 Ash 0.24767 -0.27133 0.46648 0.13522 0.06018 Carbohydrate -0.24476 0.55776 -0.26242 0.05335 -0.30477 Fat -0.58362 -0.23556 0.30046 -0.08529 0.09534 Fibre -0.03323 0.3562 0.53132 -0.38136 -0.22011 Moisture -0.10247 -0.472 -0.30303 -0.18679 0.23672 Protein 0.64827 0.09441 -0.16623 0.07909 0.03027 Iron 0.32586 -0.09898 0.26449 -0.49188 -0.11517 Zinc -0.00829 0.16358 -0.29734 -0.70044 0.37795 Eigenvalue 2.102 1.371 1.142 1.054 0.912 %variation 23.36 15.24 12.69 11.72 10.13 % cumulative Variance 23.36 38.60 51.28 63.00 73.13 4.4.4 Correlations among nutritional quality traits of 200 cowpea accessions The cowpea grain nutritional quality traits are presented in table (Table 4.7). Iron content did not correlate significantly with zinc, ash, fibre and 100 grain weight. Iron content was negatively correlated with Fat (r= -0.18, p < 0.007) and carbohydrate (r= -0.18, p < 0.007). On the other hand, iron content was positively correlated (r= 0.26, p < 0.001) with protein. fat content was negatively correlated with ash content (r= -0.13, p < 0.05) and protein content (r=-0.85, p < 0.001). Ash content correlated positively with protein (r= 0.14, p < 0.05) and negatively with carbohydrate (r=-0.23, p < 0.001). Protein was negatively correlated with carbohydrate (r= -0.29, p < 0.01). University of Ghana http://ugspace.ug.edu.gh 73 Table 4.7: Pearson’s correlation coefficients among nutritional quality traits of 200 cowpea accessions Iron (mg/kg) Zinc (mg/kg) % Moisture % Fat % Ash % Protein % Fibre % CHO 100 Swt (g) Iron 1 Zinc 0.039 1 Moisture -0.018 0.014 1 Fat -0.189*** -0.022 0.056 1 Ash 0.121 -0.082 -0.034 -0.137* 1 Protein 0.269*** 0.004 -0.154 -0.858*** 0.139* 1 Fibre 0.094 0.026 -0.120 0.024 0.008 -0.067 1 CHO -0.188*** 0.061 -0.127 -0.060 -0.239*** -0.299*** 0.098 1 100swt (g) -0.043 -0.000 -0.116 -0.017 -0.017 0.038 0.098 0.079 1 *, ** and *** =Significance at 0.05, 0.01 and 0.001 respectively; CHO= carbohydrate; swt= seed weight 4.4.5 Allelic variation based on SNP markers One of the 170 accessions was not genotyped due to low DNA quantity and quality, respectively. As a result of that, the genetic diversity of 169 cowpea accessions was assessed using 119 SNP markers. Summary statistics for number of alleles, gene diversity and polymorphic information content (PIC) are presented in Table 4.8. A total of 321 alleles with an average of 3 loci per allele were observed among the accessions. The gene diversity ranged from 0.269 to 0.535 in 7344_500 SNP and 14929_258 SNP with a mean of 0.4732. This tends to validate the homozygotic nature of the accessions used being a core collections. This is due to the fact that cowpea is a self- pollinating crop and the level of out crossing is minimal (less than 4%). The allele frequency of all the markers used was generally below 0.95, indicating that they are polymorphic markers. Polymorphism information content (PIC) values ranged from 0.2366 (7344_500 SNP) to 0.427 in University of Ghana http://ugspace.ug.edu.gh 74 4749_1972 and 14929_258 markers with an average of 0.3713 PIC. Twenty four of the 119 markers were found to be highly informative with high PIC values range from 0.4008 to 0.4273. The statistics describing the genetic diversity found at each locus were calculated. The average gene diversity for the whole sample was 0.46. The lowest gene diversity (0.26) was detected in one SNP (7344_500) marker for all accessions. Eighty eight (88) of the SNP markers showed gene diversity of > 0.45. The highest gene diversity was observed in 6 SNP markers (14929_258, 16462_1286, 6065_457, 13849_2039, 14462_1712 and 15534_890) with the gene diversity of 5 each (Table 4.8). University of Ghana http://ugspace.ug.edu.gh 75 Table 4.8: Allelic frequency, gene diversity and polymorphic information content (PIC) S/N Marker AF He PIC S/N Marker AF He PIC 1 10969_452 0.5250 0.4988 0.3744 31 13947_415 0.7564 0.3685 0.3006 2 4146_1588 0.7255 0.3983 0.3190 32 14030_764 0.5833 0.4861 0.3680 3 9673_1553 0.5732 0.4893 0.3696 33 14164_1877 0.7962 0.3246 0.2719 4 10115_384 0.6306 0.4659 0.3574 34 1426_521 0.7044 0.4164 0.3297 5 10661_873 0.5385 0.4970 0.3735 35 14462_1712 0.5032 0.5000 0.3750 6 10738_1400 0.5705 0.4901 0.3700 36 14497_540 0.5094 0.4998 0.3749 7 10811_937 0.5287 0.4984 0.3742 37 14542_452 0.5975 0.4810 0.3653 8 11367_1228 0.7278 0.3962 0.3177 38 14619_471 0.5649 0.4916 0.3707 9 11470_272 0.6855 0.4312 0.3382 39 14714_840 0.6923 0.4260 0.3353 10 11622_232 0.6139 0.4740 0.3617 40 14730_1034 0.5355 0.4975 0.3737 11 12119_480 0.5584 0.4932 0.3716 41 14769_1746 0.6730 0.4402 0.3433 12 12122_559 0.6234 0.4696 0.3593 42 14929_258 0.5000 0.5000 0.3750 13 12261_1773 0.7215 0.4019 0.3211 43 14965_280 0.6250 0.4688 0.3589 14 12393_305 0.5250 0.4988 0.3744 44 15054_315 0.6101 0.4758 0.3626 15 12505_1312 0.5535 0.4943 0.3721 45 15183_436 0.5063 0.4999 0.3750 16 12703_553 0.6452 0.4579 0.3530 46 15534_890 0.5032 0.5000 0.3750 17 12933_387 0.6266 0.4680 0.3585 47 15933_118 0.6289 0.4668 0.3578 18 12959_58 0.5226 0.4990 0.3745 48 16043_314 0.5506 0.4949 0.3724 19 1296_808 0.5316 0.4980 0.3740 49 16239_889 0.6139 0.4740 0.3617 20 13017_290 0.6667 0.4444 0.3457 50 16462_1286 0.5000 0.5000 0.3750 21 13034_542 0.5577 0.4933 0.3716 51 16914_262 0.5723 0.4895 0.3697 22 13207_784 0.7152 0.4074 0.3244 52 17023_955 0.5190 0.4993 0.3746 23 13506_333 0.5541 0.4941 0.3721 53 17450_1553 0.6772 0.4372 0.3416 24 13563_863 0.5975 0.4810 0.3653 54 17588_963 0.5732 0.4893 0.3696 25 13586_1058 0.7205 0.4028 0.3217 55 1799_940 0.5223 0.4990 0.3745 26 13707_697 0.6981 0.4215 0.3327 56 2185_132 0.5924 0.4829 0.3663 University of Ghana http://ugspace.ug.edu.gh 76 Table 4.8: (Cont’d) Allelic frequency, gene diversity and polymorphic information content (PIC) S/N Marker AF He PIC S/N Marker AF He PIC 27 13794_319 0.5472 0.4956 0.3728 57 2314_546 0.7468 0.3782 0.3067 28 13849_2039 0.5031 0.5000 0.3750 58 2339_52 0.5897 0.4839 0.3668 29 13863_519 0.5860 0.4852 0.3675 59 234_249 0.5786 0.4876 0.3687 30 13873_544 0.5570 0.4935 0.3717 60 2728_121 0.5478 0.4954 0.3727 61 2974_1109 0.7468 0.3781 0.3066 91 5428_339 0.5613 0.4925 0.3712 62 2997_519 0.7826 0.3403 0.2824 92 5435_569 0.6013 0.4795 0.3645 63 3098_224 0.6026 0.4790 0.3643 93 5448_461 0.7215 0.4019 0.3211 64 3485_771 0.6000 0.4800 0.3648 94 5503_54 0.5849 0.4856 0.3677 65 361_520 0.5125 0.4997 0.3748 95 5552_536 0.6646 0.4458 0.3465 66 3673_401 0.6433 0.4589 0.3536 96 5553_147 0.6415 0.4600 0.3542 67 3720_560 0.7613 0.3635 0.2974 97 5692_1408 0.5223 0.4990 0.3745 68 38_239 0.6346 0.4638 0.3562 98 593_329 0.5605 0.4927 0.3713 69 3838_830 0.7170 0.4058 0.3235 99 6065_457 0.5000 0.5000 0.3750 70 3885_1019 0.7898 0.3320 0.2769 100 6247_659 0.5438 0.4962 0.3731 71 3900_562 0.7170 0.4058 0.3235 101 6673_1242 0.5098 0.4998 0.3749 72 394_316 0.5094 0.4998 0.3749 102 6700_679 0.7179 0.4050 0.3230 73 395_895 0.5355 0.4975 0.3737 103 7344_500 0.8408 0.2678 0.2319 74 411_247 0.5125 0.4997 0.3748 104 7438_464 0.6250 0.4688 0.3589 75 4325_585 0.6456 0.4576 0.3529 105 7857_1368 0.5605 0.4927 0.3713 76 4339_822 0.6101 0.4758 0.3626 106 7906_1032 0.5935 0.4825 0.3661 77 4403_1123 0.5188 0.4993 0.3746 107 7993_539 0.5786 0.4876 0.3687 78 4462_114 0.5064 0.4999 0.3750 108 8011_481 0.5280 0.4984 0.3742 79 4558_472 0.5723 0.4895 0.3697 109 8118_1675 0.6226 0.4699 0.3595 80 4563_661 0.7290 0.3951 0.3170 110 8166_564 0.5871 0.4848 0.3673 81 4749_1972 0.5064 0.4999 0.3750 111 8193_441 0.6090 0.4762 0.3628 82 4778_497 0.5597 0.4929 0.3714 112 8306_119 0.5443 0.4961 0.3730 83 483_1152 0.7296 0.3946 0.3167 113 8605_2122 0.5063 0.4999 0.3750 84 4892_514 0.6625 0.4472 0.3472 114 8969_1386 0.5096 0.4998 0.3749 85 4904_278 0.5290 0.4983 0.3742 115 9134_1559 0.5163 0.4995 0.3747 86 5058_372 0.6646 0.4458 0.3465 116 9147_1655 0.5705 0.4901 0.3700 87 5239_234 0.5506 0.4949 0.3724 117 9678_835 0.8038 0.3154 0.2657 88 5294_469 0.7756 0.3480 0.2875 118 9880_545 0.5385 0.4970 0.3735 89 534_355 0.6968 0.4226 0.3333 119 9955_544 0.5316 0.4980 0.3740 90 5356_124 0.5597 0.4929 0.3714 Mean 0.6057 0.4633 0.3548 STD 0.0847 0.0486 0.0281 Range 0.50- 0.84 0.26- 0.50 0.23- 0.37 STD= standard deviation, He= gene diversity (expected heterozygosity) University of Ghana http://ugspace.ug.edu.gh 77 4.4.6 Genetic distance The cluster analysis using unweighted pairs average method (UPGMA) is presented in Fig. 4.3. Clustering was based on the calculated genetic distances and clustered the 169 cowpea accessions into 2 main groups ranging from 0.00 to 0.212 genetic distances. The dendrogram further sub-clustered each of the main clusters based on their related genetic distances regardless of their country of origin. For example, two accessions: TVu-1560 from USA and another TVu- 10745 from Sierra Leone had 0.00 genetic distance each and sub-clustered in the same branch. Moreover, the sub-clustering corresponded to biochemical data. For example two accessions TVu-320 and TVu-5040 from Nigeria and Niger with genetic distance of 0.17 each were sub- clustered within the main cluster one (Cluster I), both accession had low mean grain iron concentration of less than 50 mg/kg. Most of the cowpea accession grouped in the main cluster one, originated from Africa with few US cowpea accessions flanked with at least one African accession that showed similar genetic distance. The second main cluster (cluster II) had also various sub clusters comprising cowpea accessions with genetic distances ranging from 0.00 as in TVu-15381 and TVu-15360 to 0.2094 in ―Babban wake‖ a local landrace from Nigeria. The highest genetic distances of 0.212 and 0.209 were observed in TVu-14109 and ―Babban wake‖ from main cluster I and II respectively. Both accessions originated from Chad and Nigeria. . University of Ghana http://ugspace.ug.edu.gh 78 : Figure 4.3: Hierarchical dendrogram of 169 cowpea accessions by using similarity coefficients based on the Nei’s (1983) original genetic distance calculated from119 SNP data using the UPGMA method C lu st er 1 w it h s u b -c lu st er s University of Ghana http://ugspace.ug.edu.gh 79 Figure 4.3: (Cont’d) Hierarchical dendrogram of 169 cowpea accessions by using similarity coefficients based on the Nei’s (1983) original genetic distance calculated from119 SNP data using the UPGMA method. C lu st er I I University of Ghana http://ugspace.ug.edu.gh 80 4.4.7 Comparison between Morphological, Biochemical and Molecular Characterization Two hundred cowpea accessions were characterized morphologically and biochemically, while only 169 were genotyped. Thirty one accessions discarded due to undesirable traits associated with them. The 169 cowpea accessions screened for grain quality traits were grouped into three main groups based on morphological and biochemical data and into two main groups based on SNP marker characterization. Morphologically, the 169 accessions were grouped into three maturity groups, two shattering groups, two seed coat textural groups and five seed coat colour patterns. Biochemically, the accessions (169) were categorized into three based on the high, medium and low quantities. Biochemical data sub-grouped accessions into various sub-clusters with clear extreme values for high and low nutritional values. Similar trend was observed with molecular data. Both biochemical and molecular approaches were able to distinctly differentiate between the accessions and grouped them according to their nutritional value content and genetic distances, respectively. The three methods: morphological, biochemical and molecular were found to be effective in assessing variability and selection of cowpea grain nutritional value, they could be used concurrently in marker assisted selection. 4.5 Discussion The surface layer of the soil had high proportion of sandy soil and low organic matter content thereby making it physically fragile. This is a typical soil characteristic of the Northern guinea savanna zone of Samaru, Nigeria where cowpea is grown well. The particle size distribution corresponded to what was observed by Salako and Kang (2002) who characterized University of Ghana http://ugspace.ug.edu.gh 81 the soil of Samaru as sandy loam.. The low pH in water recorded showed that the soil is generally acidic in nature, thus favors the root growth of plants. The 200 cowpea accessions were classified into 3 maturity groups early, medium and late. Almost half (48%) of the accession were late maturing type and only 17% were early maturing. The late maturing group showed prostrating growth habit, while early and medium maturing accessions showed erect and semi erect growth habit, respectively. This may be due to their inherent genetic nature, environmental influence and variation in their response to day length, as most of the accessions are landraces from different countries. This conformed to the findings of Timko & Singh (2008) and Goenaga et al. (2008) who associated late maturing of cowpea with photosensitivity response, regardless of the planting time. Five grain eye patterns namely; Holstein, small eye, solid, Watson and Whippoorwill were identified from the 200 accessions. This conformed to classification of Saunders (1959), and reported by Asante (1991). For 153 African accession, (21accessions =13.73%) of the accessions were found to shatter starting from the first maturing pod to the end of their maturity period. Shattering might be due to the changes in environmental conditions such as low humidity or might be due to their inherent nature. However, the accessions were planted late (23rd August, 2012), compared with appropriate cowpea planting time in the Guinea savanna zone of Zaria. The high rainfall received in August and September coupled with early cessation of rainfall may have reduced the level of humidity and induced shattering. This result disagree with Ali et al. (2004) findings in Pakistan, who reported that high rates of nitrogen, excessive moisture and low humidity can result in excessive vegetative growth, delayed maturity and pod shattering. However, further research is suggested to find out the reason for shattering, because some shattering accessions like TVu- 13088 had high levels of iron, protein and carbohydrate contents. University of Ghana http://ugspace.ug.edu.gh 82 The significant variation observed among the accessions for percentage crude protein (1.72% - 29.93%) may be due to screening large sample size collected from diverse agro ecological zones of Africa and other parts of the world. This result is not in agreement with the previous results of Horax et al. (2004), Ajeigbe et al. (2008) and Gupta et al. (2010), who reported low variability and small range 20-26%. On the other hand, the results in this study are in close agreement with Tchiagam et al (2011b) who reported up to 31.78% crude protein in cowpea. This low range reported by the previous authors may be attributed to the small number of cowpea accessions and or varieties screened. For example, Ajeigbe et al. (2008) examined only 9 cowpea varieties from Nigeria, while Gupta et al. (2010) evaluated 23 genotypes from India. The wide variations observed in protein content may be attributed to either genetic or environment or both. The discrepancies in protein content is therefore, attributed to the combination of genotype and environment as reported by Wang & Daun (2004). Similarly, the high concentrations and substantial variation observed for zinc (10.01 mg/kg to 386.3) and iron (1.01 to 329.15) concentration could also be attributed to the large sample size from diverse agro ecologies. These results are not in agreement with the finding of Asante et al. (2009) who reported low variation in cowpea grain zinc concentration with small range of 4 to 13 mg/kg in 44 cowpea accessions. The low range could be attributed to the small number of the accessions and the environment of study. The variations in iron however, conformed to the findings of Asante et al (2009) in cowpea from Ghana and Ribeiro et al. (2012) in common bean from Brazil, respectively. The principal component analysis clustered the 9 grain quality traits into 4 principal component axes that accounted for 63% of the total variation. Grain size, iron concentration, zinc concentration, ash content and moisture contents showed low communalities estimate. This University of Ghana http://ugspace.ug.edu.gh 83 implied that they were the most variables among the 9 nutritional quality traits studied. Moreover, selection for these traits can be feasible for nutritional enhancement of cowpea. Similar, finding was reported in common beans by Ribeiro et al. (2012). The results showed that iron is not significantly associated with zinc content; this agrees with the findings of Blair et al. (2013)in common bean. On the other hand, the result disagrees with the findings of Oluwatosin (1998), who reported a negative correlation in cowpea trial conducted under field condition. This might be due to the environmental differences because, growth promoting bacteria can increase the uptake of iron and zinc by the plants (Barea et al., 2005) and the level of phytoavailability of the genotypes used by Oluwatosin (1998). Protein was negatively correlated with carbohydrate (r= -0.29, p < 0.01) and fat content (r= -0.85, p<0.001). This conformed to observations made by other authors (Omueti & Singh, 1987; Ajeigbe et al, 2008). Protein content was positively correlated with ash (r= 0.14, p < 0.05) and negatively correlated with fat (r= -0.13, p < 0.05) and carbohydrate (r= -0.23, p < 0.001) content indicating that selection for high protein will decrease carbohydrate content and increase ash content which will make the inferior parent superior and nutritional enhancement possible. On the other hand, iron content was positively correlated (r= 0.26, p<0.001) with protein which imply that iron could be selected indirectly by selecting for high protein cowpea accession. No significant correlation was observed between 100 seed weight and protein content and iron content, respectively. These correlations show that breeding for larger grain size (desirable) may not induce a reduction in the grain protein and iron content. Thus, direct selection to increase these two nutrients, via selection for 100 seed weight could be possible. Large grain size and seed coat colour are important consumer preference traits. These results disagree with the findings of University of Ghana http://ugspace.ug.edu.gh 84 Asante et al. (2004) and Moura et al. (2012) who reported negative correlations among these traits. Since traditional selection methods mainly depend on the phenotypic variations and morphological markers are easily influenced by the environment, and some of the morphological markers have epistatic effects (Tan et al., 2012), then selection for grain quality traits of cowpea require genetic markers based on individual nucleotide sequence variation. This improves the reliability and efficiency of the selection process. One hundred and nineteen (119) SNP markers were used to characterize 169 cowpea accessions. The allele frequency of all the markers used was generally below 0.95, indicating that they were all (100%) polymorphic. Similar results were obtained in cowpea (Ajibade et al., 2000; Gajera et al., 2014) and other crops like wheat (Nagaoka & Ogihara, 1997). It has been reported that the ability to resolve genetic variation may be more directly related to the number of polymorphisms detected by the marker system (Sivaprakash et al., 2004). The low level of polymorphism (0.24 to 0.38) detected in this results may be attributed to radiation of cowpea from the center of origin (Africa) to other parts of the world via domestication (Li et al., 2001; Tosti & Negri, 2002; Badiane et al, 2004; Diouf. & Hilu, 2005), in addition to its inherent nature of self-pollination mechanism (Badiane et al., 2012). The small genetic differentiation (0.26-0.45) observed among cowpea accessions comprising the African and non-African (USA and Asia) collections indicates that the entire genetic diversity in the African germplasm might already have spread over cowpea-growing regions in the world as a whole, though not completely within any single region. Another probable reason may be as a result of dispersal mechanisms that might have occurred as revealed by typical patterns of genetic relatedness between world cowpea collections relative to the two University of Ghana http://ugspace.ug.edu.gh 85 primary gene pools in Africa (Huynh et al., 2013). Moreover, the accessions used are core collection consisting of ~10% of total accessions, which between them capture most of the available diversity in the entire collection (Upadhyaya et al., 2010). These can be thoroughly evaluated and the information so derived can be utilized for improving the efficiency of breeding. This further agrees with (Brown, 1989) who reported that the entries to a core collection should be limited to approximately ~10 %, using the sampling theory of selectively neutral alleles, with a ceiling of 3000 per species. This level of sampling is effective in retaining 70% of alleles of entire collection Although only three accessions from Asia were included in this study, the majority of the accessions showed a kind of relatedness in either morphological data, biochemical data or molecular data, implying that most of the cowpea accessions from America might have moved from West Africa either as a result of slave-trading (Whit, 2007) or as a result of germplasm exchange (Ehlers & Hall, 1997). However, the results of this work further disagree with that of Huynh et al. (2013) who reported clear distinct clusters between African and American cowpea accessions. Another possible cause of relatedness may be due to maintenance methods used in conserving those accessions. Though cowpea is self-pollinating crop, some level of out crossing might probably occur between the accessions. This agrees with the previous findings of Moalafi et al. (2010) who reported 4 to 5% outcrossing rates in cowpea. Nevertheless, the effect of pollen movement on population structure is unknown (Kouam et al., 2012). Both domesticated cowpea and its wild progenitor are characterized by a flower structure that should promote inbreeding (Lush, 1979). The clustering of accessions based on their related genetic distances regardless of their country of origin could be due to residual heterozygosity that would be greater for landraces than for elite cultivars, and over time more out- crossing is likely to occur. This is in agreement University of Ghana http://ugspace.ug.edu.gh 86 with findings of Ortiz et al. (2008) in quinoa diversity studies in Peru. However, further work using molecular markers is recommended to check the possible redundancy that may exist among cowpea accessions. 4.6 Conclusion The description patterns of genetic distances and clustering of relatively similar individuals into groups and sub-groups reported in this study provides important insights that can improve the efficiency of cowpea germplasm conservation and breeding efforts. The information will enable rational planning by gene bank curators to help reduce duplicates among the accessions and form core collection of their accessions. For breeding programs, accessions clustered within a related group based on their relative genetic distances may exhibit common adaptive complexes of physiological traits coupled with a relatively restricted range of morphological and underlying genetic variation. Thus, crosses within related accessions are expected to produce a high frequency of relatively similar-looking progeny, while crosses between members of different cluster are expected to produce more variable progeny, perhaps with a relatively low average performance in early generations, in this case selection should be delayed to later generations. Breeding strategies involving series of back crosses using conventional method or 2-3 marker assisted back cross (MAB) with the identified markers would be required to increase the concentration of iron and zinc in cowpea grains. This study has revealed the existence of genetic diversity within cowpea genome that could be exploited to successfully develop and deploy nutrient-dense cowpea varieties. In addition, significant and positive correlations between iron content and protein content indicate the possibility of improving the concentrations of these elements in cowpea simultaneously. Based on the findings of this study, three cowpea accessions; TVu-13088, TVu-13495 and TVu- University of Ghana http://ugspace.ug.edu.gh 87 9725 with nutrient-enriched grains have been identified. They could be nominated for further testing in different environment and anti-nutritional factor analysis before recommending for use. Five accessions each for iron, zinc and protein content were identified as good candidates for further improvement;(1) Top 5 iron content accessions: TVu-1330, TVu-13495, TVu-16400, TVu-8751 and TVu-9725 (2) Top 5-zinc content accessions: TVu-1616, TVu-2155, TVu-15251, TVu-301 and TVu-232 (3) High protein content accessions: TVu-13495, TVu-13088, TVu- 15187, TVu-347 and TVu-13468 University of Ghana http://ugspace.ug.edu.gh 88 CHAPTER FIVE 5 INHERITANCE OF IRON AND ZINC CONTENT AND OTHER GRAIN QUALITY TRAITS IN COWPEA 5.1 Introduction Cowpea (Vigna unguiculata) is of vital importance to the livelihood of millions of people, especially in West and Central Africa. It provides nutritious grain and an inexpensive source of protein for both rural and urban consumers (Rangel et al., 2004). Cowpea grain contains up to 29% protein, 78% carbohydrate, 329 mg/kg iron and 386 mg/kg zinc (Chapter 4 above) and therefore has a tremendous potential to contribute to the alleviation of malnutrition among rural and urban consumers (Boukar et al., 2010). Humans require at least 22 mineral elements for their wellbeing (Welch & Graham, 2004; Graham et al., 2007). These can be supplied by an appropriate diet. Furthermore, it is estimated that over 60% and 30% of the world‘s population are iron (Fe) and zinc (Zn) deficient (White and Broadley, 2009). This situation is attributed to crop production in areas with low mineral phytoavailability and consumption of staple crops with inherently low tissue mineral concentrations (White & Broadley, 2005). The problem of iron and zinc malnutrition can be addressed through dietary diversification, mineral supplementation and increasing mineral concentrations in edible crops. However, strategies to increase dietary diversification, mineral supplementation and food fortification have not always been successful, due to concern of yield drag. For this reason, biofortification of crops through breeding varieties with an increased mineral elements concentration is recommended as an immediate strategy not only to increase mineral concentrations in edible crops but also to improve yields on infertile soils (White & Broadley, 2005). To set an enhanced breeding program, it is essential to know the proportion of phenotypic variation of a trait that is heritable (Kearsey & Pooni, 1996). This is because the University of Ghana http://ugspace.ug.edu.gh 89 selection efficiency of a trait is mainly dependent on the magnitude of genetic variation and heritability of such trait (Falconer & Mackay, 1996). Generation mean analysis provides information on the relative importance of additive and dominance effects in populations created from two inbred lines. It involves measuring the means of different generations (P1, P2, F1, F2, BC1P1, BC1P2) derived from two contrasting parents and interpreting the means in terms of the different genetic effects (Bernardo, 2002). This is because the actual means of single loci are unobservable, generation means estimate the pooled genetic effects across loci. Bi-parental mating system has been used to study gene action controlling important agronomic and quality traits in various crops. For example, Tchiagam et al. (2011b) reported 5 genes controlling sucrose accumulation in cowpea grain; Non-allelic interactions mainly additive x additive and additive x dominance were reported in rice grain size (Kiani et al., 2013). Dominance gene effect was also found to be more important than additive, in the inheritance of protein content in chickpea (Kumhar., 2013), while Tchiagam et al. (2011a) associated high percentage (%) of protein with recessive gene in cowpea. It is therefore, understandable that knowledge of the genetic factors responsible for the inheritance of iron, zinc and other agronomic traits is essential for breeding program. Moreover, information regarding genetic control of iron and zinc in cowpea is scarce. Therefore, the objectives of this study were to: 1. Determine the mode of inheritance of iron and zinc concentration in cowpea 2. Determine the mode of inheritance of grain size in cowpea 3. Determine the relationship between zinc concentration and yield related components University of Ghana http://ugspace.ug.edu.gh 90 5.2 Material and Methods 5.2.1 Parent Materials Six distinct cowpea accessions selected for inheritance of grain quality traits are presented in Table 5.1. Two parents: P1=TVu-14845 and P2 =TVu-15251 are characterized by having low zinc concentration (15.15 mg/kg) and high zinc concentration (52.03 mg/kg), respectively. The grains of the 2 parents are morphologically distinct: with white grain coat, rough texture, black eye for P1 and brown grain coat and rough texture for P2. Similarly, 2 parents P1=TVu-1 (smooth coat texture) and P2=TVu-999 (brown grain colour and smooth grain coat texture) were used for inheritance of cowpea grain iron concentration. Table 5.1 Cowpea parental materials, origin and contrasting characteristics Parents Origin Contrasting characteristics 1. TVu-1 A single plant selection from local variety from Nigeria Low iron content (1.50 mg/kg), smooth and brown grained and small seeded. 2. TVu- 15251 A single plant selection from local variety from Chad High iron and zinc (84.27 & 52.03 mg/kg), brown grain coat, rough texture and medium grained. 3. TVu- 999 A single plant selection from landrace from Ghana High iron content (98.05 mg/kg), Light brown coat and smooth texture and medium grained. 4. TVu- 14845 A single plant selection from local variety from Mali Low zinc content (15.15 mg/kg), white grain coat, rough texture, black eye with grey hilum and small grained. University of Ghana http://ugspace.ug.edu.gh 91 5.2.2 Population development The 6 basic generations (P1, P2, F1, F2, BC1P1 and BC1P2) were developed in the screen house of the Institute for Agricultural Research (IAR), Zaria (11° 17.0′ N and 7° 36.01′E). Pots (Height = 21cm, Width = 23cm.) were filled up with sandy-loam soil collected from the field, after the analysis of soil trace elements (iron and zinc). Three seed was plant per pot and weeds were hand removed from the pots and no fertilizer was applied. Two and three liters of water per pot were applied for every 24 hours at vegetative stage and flowering to maturity stages, respectively. Crosses were generated based on the combined procedures of Ehlers & Hall (1997) with slight modification. To ensure the success of the crosses and minimize contamination, capping method was employed throughout. The female plants were emasculated in the evening and pollinated in the morning. The process of emasculation was carefully done with sharply pointed forceps sterilized with alcohol between crosses to prevent contamination by unwanted pollen. Each cross was tagged immediately, names of parents that were involved in the cross and date of the cross. In the presence of adequate flowers from both parents, emasculation and pollination were simultaneously carried out twice a day; early morning and evening time when the sun was about to set. In a situation whereby the flowers of the paternal (P2) parent is ready and maternal (P1) parent is not, the paternal flowers were collected early in the morning, preserved in the fridge for 11 hours and then used to pollinate the maternal parent in the evening. Zinc set (TVu-14845 and TVu-15251) Crosses were made between low zinc (TVu-14845) and high zinc (TVu-15251) parents between 2013 and 2014. One hundred and seventy eight (178) F1s were generated and classified into 4 groups: for (1&2 portion) the 2 parents PTVu-14845 and PTVu-15251 were backcrossed to the 80 University of Ghana http://ugspace.ug.edu.gh 92 F1 plants (40 to each parent) and generated 75 BC1P1 TVu-14845 and 82 BC1P2TVu-15251, respectively. (2)The third portion (40 F1 grains) was advanced to harvest F2, seeds and retained the (4) fourth portion as F1. The F1 backcrosses to respective parents (BC1P1 TVu-14845 and BC1P2TVu-15251) were advanced to F2 generations backcrosses to get more seeds for elemental analysis (November, 2013 and February, 2014). Iron set (TVu-1 and TVu-999) A cross between TVu-1 and TVu-999 was started in March, 2013. The seeds from the direct crossing were classified into 4 groups. The 2 portions of the grains (80 grains) were backcrossed to their parents (PTVu-1 and PTVu-999) and generated BC1P1TVu-1 and BC1P1TVu-999, respectively. The third portion of the seeds from the direct crossing (F1) was advanced to F2, while the fourth portion was retained for subsequent evaluation (November, 2013 to February, 2014). 5.2.3 Evaluation of Iron and Zinc concentration from the two sets of six generations The 2 sets of the 6 basic generations (P1, P2, F1, F2, BC1P1 and BC1P2) derived from a cross between TVu-1 x TVu-999 (Iron) and TVu-14845 x TVu-15251(Zinc) were then evaluated for iron and zinc concentrations respectively, following the HarvestPlus protocol described in sub-section 4.2.3.2 of chapter four above. 5.2.4 Data collection Data were taken from 30 plants each from non-segregating populations (F1 and their parents), 60 plants each for backcrosses to respective parents (BC1P1 and BC1P2) and 150 plants for each of the F2 generations derived from 2 crosses: TVu-1 x TVu-999 and TVu-14845 x TVu- 15251). University of Ghana http://ugspace.ug.edu.gh 93 For agronomic data the following traits were measured on individual plants from each generation, in accordance with cowpea descriptors of International Board for Plant Genetic Resources (IBPGR, 1983) and Mahalakshmi et al (2007). Days to 50% flowering: This was recorded as the number of days from sowing to when half of the plants from each generation: 30 for non-segregating population, 60 for back crosses and 150 for F2 population had flowered. Plant height (cm): Plant height was taken as the perpendicular height of the plant from the top soil level (in the pot) to the end of the top most leaf of the plant. Hundred (100) grain weight (g): One hundred grains were randomly picked at 12% moisture content from individual plants of each generation and weighed. Grain coat colour: was assessed based on visual examination, by spreading sub-samples from individual plants from each generation on a white sheet of paper. The various colour observed were than recorded against each plant and generation. Grain coat texture: visual inspection and hand rubbing of the randomly picked grains, from individual plant of each generation was used to assess the grain coat texture. Elemental analysis: for iron and zinc concentrations from 2 sets of crosses (TVu-1 x TVu-999) and (TVu-14845 x TVu-15251) were estimated as described in sub-section 4.3.2.3 chapter four. Yield related traits data were recorded from the F2 population derived from a cross involving low zinc parent (TVu-14845) and high zinc content (TVu-15251) parent according to the procedure of Mahalakshmi et al. (2007) University of Ghana http://ugspace.ug.edu.gh 94 Number of pod per plants: The total harvestable pod per individual plant were counted after harvest Mean Number of Seed per pod: Due to the limited number of pods, five pods were taken at random, number of seed per pod was counted and mean was of the five pod was then estimated by dividing the total seed counted by the number of pod (5). Mean pod length (cm): A thread was used to measure the length of a pod and corresponding length was then measured on a ruler. The average pod length was then estimated from five randomly picked pods. Hundred (100) seed weight (g): One hundred seed was taken at random from the total seed threshed from individual plant. 5.3 Data Analysis With the exception of degree of dominance, all other analyses were run using SAS software (version 9.3). The iron and zinc concentrations determined for the 2 set of crosses were subjected to analysis of variance to test for the equality of the means using proc ANOVA. Bartlett‘s homogeneity variance test was run for the F2 phenotypic values of each trait, using DIST macro of SASQuant, developed by Gusmini et al (2007). This is to verify the assumption of equal variance needed to pool data for a specific source. Genetic parameters, m (mid-parent), a (additive component) and d (dominance component), variance components and broad and narrow sense heritability were estimated using SAS program (SASQuant) developed by Gusmini et al. (2007). Hayman‘s mean separation analysis procedure (Hayman, 1958; Gamble, 1962) based on the following linear model: ddadaadamY   22. University of Ghana http://ugspace.ug.edu.gh 95 Y = the observed mean,  and  are the coefficients for a and d , respectively, m = mean of the F2 (i.e., the base population), a = pooled additive effects, d = pooled dominance effects, aa = additive x additive gene interaction effects, ad = additive x dominance gene interaction effects and dd = dominance x dominance gene interaction effects The notation of Gamble (1962) m, a, d, aa , ad , dd were used and weighted using the inverse of the variance. The best fitting model among the two models: model 1 [m ], [ a ] and [ d ] and the second model comprises of the epistatic effects, [ aa ], [ad ], [dd ] were used to determine the significant additive or dominant effect. Broad and narrow sense heritability were estimated as follows: Broad sense heritability (Hb 2) = [ VF2 – ( VP1 + VP2 + VF1 ) / 3 ] / VF2 Narrow sense heritability (Hn 2) = [ 2VF2 – ( VBC1P1 + VBC1P2 ) ] / VF2 Where, V = variance for P1, P2, F1, F2, BC1P1and BC2P2 generations. Gene factors controlling the iron, zinc, seed weight and plant height were calculated following the method of Wright (1952) using SASQuant program of Gusmini et al. (2007) The degree of dominance (deviation from the mid-parent value) and direction of dominance in the two sets (zinc and iron) and their respective 100 grain weight and plant heights were estimated by hand in accordance with the method of Falconer & Mackay (1996) as follows: D (degree of dominance) = d/a Where, d= heterozygote = means of F1-1/2 (P1+P2) Where means of P1, P2 and F1are used in the two crosses. University of Ghana http://ugspace.ug.edu.gh 96 5.4 Results 5.4.1 Generation Means for Zinc Concentration, Grain weight and Plant Height The segregating and non-segregating populations derived from the 3 set of crosses are presented in Table 5.2. The first two sets of crosses were developed from a single family each and used independently, in the inheritance studies of cowpea grain zinc concentration and iron concentration. As expected, the number of F2 plant was relatively higher than that of their corresponding back crosses (Table 5.1). Table 5.1: Number of progenies generated from two set of crosses for iron, zinc and grain coat colour studies Cross F1 F2 BC1P1 BC1P2 Population generated used for: TVu-14845 x TVu-15251 178 221 86 94 Inheritance of zinc content, seed size and plant height TVu-1 x TVu-999 189 234 88 91 Inheritance of iron content, seed size and plant height The results of analysis of variance (ANOVA) for the six basic generations are presented in Table 5.2. Highly significant (P < 0.001) difference were detected among generation means for grain zinc concentration, 100 grain weight and plant height, respectively. Table 5.2: Analysis of variance for the zinc concentration, seed weight and plant height in a cross between TVu-14845 and TVu-15251 Source of variation DF Zinc (mg/kg) SWT(g) PLHT(cm) Generations 5 476.903*** 490.0*** 1220.655*** Error 354 69.249 17.089 76.956 ***; Significant at P<.0001; SWT= 100 grain weight (g); PLHT= Plant height (cm); DF = degree of freedom University of Ghana http://ugspace.ug.edu.gh 97 Population mean distribution for zinc content is presented in Fig 5.1. Distribution of zinc content in the two backcrosses (BC1P1 and BC1P2), was skewed toward high zinc parent (TVu- 15251). Individual backcrosses and some F2 plants produced high grain zinc content, higher than the superior parent (TVu-15251), indicating transgressive segregation among some individual F1 and F2 populations. Figure 5.1: Population mean distribution for zinc concentrations among six basic generations derived from low zinc content (TVu-14845) and high zinc content (TVu-15251) parents 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 P1 F1 BC1P1 BC1P2 F2 P2 Zinc (mg/kg) Z in c co n ce n tr at io n ( m g /k g ) University of Ghana http://ugspace.ug.edu.gh 98 The distributions of seed weight among segregating and non- segregating populations derived from a cross involving small seeded parent (TVu14845) and large seeded parent (TVu- 15251) are presented in Figure 5.2. The seed weight in parent two (TVu-14845) was almost two times heavier than that of parent one (TVu-14845), indicating the distinctness of the two parents. As expected, F2 showed wider variability and the two backcrosses to parent one and two was skewed toward the low and high seed weight parents (BC1P1 to TVu-14845 and BC1P2 to TVu- 15251), respectively. Fig 5.2: Distribution of 100-seed weight (swt) in six generations derived from crosses involving small seeded parent (TVu-14845) and large seeded parent (TVu-15251) University of Ghana http://ugspace.ug.edu.gh 99 The distribution of mean plant height among segregating and non- segregating populations generated from a cross between short parent plant (TVu14845) and tall parent plant (TVu-15251) are shown in Figure 5.3. The two plants showed contrasting height having means of 13.40 cm and 30.30 cm for parent one and two respectively. More variability and transgressive segregation was observed among F2 population. Fig 5.3: Distribution of plant height (PLHT) in six generations derived from crosses involving short parent (TVu-14845) plants and tall parent (TVu-15251) plants University of Ghana http://ugspace.ug.edu.gh 100 Means for parental, F1, F2 and backcross generations derived from a cross involving TVu-14845 and TVu-15251 cowpea parents are listed in Table 5.3. The parent with high zinc concentration (TVu-14845) had significantly higher mean performance compared to the low zinc content parent (TVu-15251). The means of grain zinc concentration of non-segregating population (F139.90) and segregating populations (F2 37.30; BC1P1 40.70 and BC1P2 37.50) were significantly higher than that of the mid-parent (33.95). Similar trend was equally observed in 100 seed weight and plant height Table 5.3. Table 5.3: Means of families derived from a cross involving TVu-14845 (low zinc) and TVu- 15251 (high zinc) cowpea parents Traits Generation [Zinc] mg/kg 100 SWT (g) Plant height (cm) P1 30.20 5.80 13.40 P2 37.70 11.40 30.30 F1 39.90 10.30 18.60 F2 37.30 13.30 29.00 BC1P1 40.70 7.30 30.70 BC1P2 37.50 11.40 25.50 MP 33.95 8.60 21.85 d = F1- 0.5 (P1- P2) D = d/a 5.95 0.1725 1.7 0.1976 - 3.25 0.1487 MP= mid-parent mean; [Zinc] = seed zinc concentration; D= degree of dominance; means of F1, P1 and P2 and d = heterozygote; d =heterozygote University of Ghana http://ugspace.ug.edu.gh 101 The degree of dominance for zinc content, grain weight and plant height were estimated according to Falconer & Mackay (1996) to determine the direction of dominance in each trait. Degree of dominance for zinc concentration Degree of dominance = d/a = 5.95/33.95 = 0.1752 (17.52 %) Where d= F1- ½ (P1 + P2) = 5.95 from the table above and a= ½ (P1 + P2) = 33.95 Dominance in seed weight Degree of dominance = d/a = 1.70/ 8.60 = 0.1976 (19.76 %) Where d = F1 - ½ (P1 + P2) = 1.70 and a = ½ (P1 + P2) = 8.60 Dominance in plant height Degree of dominance = d/a = - 3.25/21.85 = - 0.1487 (14.87 %) a = ½ (P1 + P2) P1 P2 39.9 0 F1 -a +a ½ (P1 + P2) P1 P2 -a +a 30.20 37.70 33.95 5.80 8.60 11.40 F1 10.30 a= ½ (P1 + P2) P1 P2 -a +a 13.40 21.85 30.30 F1 18.60 d d University of Ghana http://ugspace.ug.edu.gh 102 The nature of gene action involved in the control of the zinc concentration, 100 grain weight and plant height were investigated by the analysis of generation means and presented in Table 5.4. The generation means analysis indicated that the additive-dominance model was adequate to explain the gene action involved in the inheritance of zinc concentration in cowpea grain. The negative dominance x dominance [dd] and positive dominance suggest the presence of dominance effects toward the high zinc content parent plant. For grain weight involving a cross between TVu-14845 and TVu-15251 (Table 5.4), fitted the first model with highly significant additive [a] and dominance [d] effects, both having negative signs. The dominance [dd] gene action was also significant (P < 0.05) and positive in the direction of the high zinc content parent (TVu-15251). Positive significant additive [a], additive x dominance [ad] gene action and negative dominance by dominance [dd] gene actions were detected for plant height (Table 5.4). The significant and negative dominance x dominance [dd] gene action would suggest the presence of dominance effects at heterozygous loci for shorter parent plants. The negative sign for the dominance [d] and dominance x dominance [dd] gene action indicated the contribution of alleles from the shorter parent (TVu-15251) plant. University of Ghana http://ugspace.ug.edu.gh 103 Table 5.4 Estimate of gene effects (±SE means) for zinc, 100 grain weight and plant height in a cross between TVu-14845 and TVu-15251 Genetic component Zinc concentration (mg/kg) 100 grain weight (g) Plant height (cm) M 37.30 + 0.76 *** 13.29 + 0.44*** 28.99 + 0.94*** A 3.21 + 2.31 -4.10 + 0.87*** 5.21 + 2.06** D 13.07 + 9.19 -14.30 + 4.18 *** -6.65 + 9.04 Aa 7.13 + 7.66 -15.93 + 3.52*** -3.42 + 7.90 Ad 6.96 + 2.94* -1.31 + 1.03 13.64 + 2.64*** Dd -15.72 + 15.34 16.52 + 6.52* -28.21 +14.31* *, ** and *** are significant difference at P< 0.05, P<0.01 and P<0.001 respectively; a =additive gene effect; d = dominance gene effect; aa = additive by additive gene effect; ad = additive by dominance gene effect; dd = dominance by dominance gene effect. The magnitude of the phenotypic variances for zinc concentration, 100 grain weight and plant height were generally higher than the genotypic variances (Table 5.5) showing a greater influence of the environment on these traits. Similarly, broad sense heritability, narrow sense heritability and number of effective factors are presented in Table 5.5. High broad sense heritability: hb 2=0.79, 0.86 and 0.93 were recorded in grain zinc concentration, 100 grain weight and plant height, respectively. Overestimated narrow sense heritability (1.03 and 1.22) was observed in 100 grain weight and plant height, respectively. Number of effective factors based on Wright method was 0.2 each for zinc concentration and 100 grain weight, while 0.3 was observed for plant height (Table 5.5). University of Ghana http://ugspace.ug.edu.gh 104 Table 5.5 Genetic variances, heritability and effective factors for zinc content seed weight and plant height in a cross involving TVu-14845 and TVu-15251 Trait measured (VG) (VA) (VD) (VE) (VP) (h 2 b) (h2n) EF Zinc mg/kg 68.51 1.71 66.80 18.32 86.83 0.79 0.02 0.2 100 gw (g) 25.41 36.11 -10.70 4.22 29.63 0.86 1.22 0.2 Plant height (cm) 122.41 136.64 -14.23 9.80 132.22 0.93 1.03 0.3 VG= Genotypic variance; VA=Additive variance; VD =Dominance variance; VE=Environmental variance; VP= Phenotypic variance; hb 2 = Broad sense heritability and h2N= Narrow sense heritability; gw = grain weight; EF = effective factor [Wright (1952)] 5.4.2 Generation means for seed iron concentration, seed weight and plant height in a cross between low iron (TVu-1) and high iron (TVu-999) parents Analysis of variance results for the six basic generations showed highly significant difference (P<0.001) for iron concentration, 100 grain weight and plant height (Table 5.6) Table 5.6 Mean squares for iron concentration, 100-grain weight and plant height in a cross between TVu-1 and TVu-999 Source of variation DF Iron (mg/kg) SWT(g) PLHT(cm) Generations 5 82941.057*** 139.215*** 1144.157*** Error 199 10605.52 26.585 209.740 ***= significant at P < 0.001; SWT= 100 grain weight and PLHT= Plant height Distribution of generation means of iron concentrations among segregating and non- segregating population derived from a cross involving low iron (TVu-1) and high iron (TVu-999) parents is presented in Figure 5.4. Parent two (TVu-999) had higher iron concentration (422.89 mg/kg) than that of parent one TVu-1 (383.83 mg/kg). Population means distribution for iron concentration showed that back cross (BC1P2) to high iron content parent (TVu-999) produced higher grain iron content than that of back cross with low iron content parent (TVu-1). University of Ghana http://ugspace.ug.edu.gh 105 Figure 5.4: Population mean distribution for iron concentrations among six basic generations derived from low iron content (TVu-1) and high iron content (TVu-15251) parents Analysis of variance results for the six basic generations showed highly significant difference (P<0.001) for iron concentration, 100 grain weight and plant height respectively (Table 5.6) 0.00 100.00 200.00 300.00 400.00 500.00 600.00 700.00 P1 F1 BC1P1 BC1P2 F2 P2 Iron (mg/kg) Ir o n c o n ce n tr at io n ( m g /k g ) University of Ghana http://ugspace.ug.edu.gh 106 Table 5.6: Mean squares for iron concentration, 100-grain weight and plant height in a cross between TVu-1 and TVu-999 Source of variation DF Iron (mg/kg) SWT(g) PLHT(cm) Generations 5 82941.057*** 139.215*** 1144.157*** Error 199 10605.52 26.585 209.740 ***= significant at P < 0.001; SWT= 100 grain weight; PLHT= Plant height and DF=degree of freedom. Means for parental, F1, F2 and backcross generations for the cross involving TVu-1 and TVu-999 cowpea are presented in Table 5.7. As expected, the parent with high iron concentration (TVu-999) had higher mean (419 mg/kg) iron concentration compared to the low iron parent TVu-1 with 384 mg/kg. The means for iron concentration of both F1 (300.02) and F2 (353.00) were lower than the low iron parent (TVu-1). The BC1P1 was closer to the low iron content of the parent, while BC1P2 population had significantly higher mean concentration of iron than the high iron content parent (TVu-999) and mid-parent. For 100 grain weight, both F1 and F2 produced large grains that resembled the weight of the largest grained parent (TVu-999). Backcross to respective parents yielded large grain, though, a bit larger in BC1P2. For plant height, the BC1P1 population produced the tallest plant with 60 cm population mean. This is taller than the tallest means of the parent plant and the mid-parent means. University of Ghana http://ugspace.ug.edu.gh 107 Table 5.7 Means of families derived from a cross involving TVu-1 (low iron) and TVu-999 (high iron) cowpea parents Traits Generation Iron content (mg/kg) 100 grain weight (g) Plant height (cm) P1 384.01 5.90 40.70 P2 419.00 12.40 21.30 F1 300.02 12.60 31.80 F2 353.00 11.20 25.00 BC1P1 332.01 9.10 60.00 BC1P2 458.00 8.30 24.10 MP Heterozygote (d) D = d/a 401.51 -101.50 0.2527 9.15 3.45 0.3770 31.00 0.80 0.0258 MP= mean of mid parent; d = heterozygote; D = degree of dominance Degree of dominance for grain iron content, grain weight and plant height Degree of dominance for iron content Degree of dominance for iron content = d/a = -101.48/ 401.50 = 0.2527 (25.27 %) a =½ (P1 + P2) P1 P2 -a +a 300.02 401.50 419.00 F1 384.01 d University of Ghana http://ugspace.ug.edu.gh 108 Degree of dominance for seed weight Degree of dominance = d/a = 3.45/9.15 = 0.3770 (37.70 %) Degree of dominance for plant height Degree of dominance = d/a = 0.80/31 = 0.0258 (2.58 %) The genetic effect of iron concentration and 100 seed weight are presented in Tables 5.8. The generation means analysis indicated negative additive [a] gene action is significant to explain the nature of gene action for iron concentration. Additive by dominance [ad] dominance gene action was also significant but in the negative direction, suggesting the contribution of alleles from the low iron parent (TVu-1). The heterozygote (d = 300.02 < 384.01 low iron content parent) on the dominance scale further, indicates the low iron parent is dominant over high iron content parent (TVu-999). Significant negative additive x additive [aa] and positive additive x dominance [ad] and dominance x dominance [dd] gene actions were identified as important for grain weight a =½ (P1 + P2) P1 P2 -a +a 12.60 9.15 12.40 F1 5.90 a =½ (P1 + P2) P1 P2 -a +a 31.80 0.80 21.30 F1 40.70 d d University of Ghana http://ugspace.ug.edu.gh 109 inheritance in cowpea. This indicates duplicate gene action in grain weight inheritance in a cross between small grained (TVu-1) and large grained (TVu-999) parents. Table 5.8 Estimate of gene effects (±SE means) for zinc, 100 grain weight and plant height in a cross between TVu-1 and TVu-999 Components Iron concentration (mg /kg) 100 grain weight(g) Plant height (cm) m 353.07 + 8.15*** 11.16 + 0.68*** 24.97 + 1.68*** a -125.20 + 29.39*** 0.81 + 0.81 3.30 + 3.16 d 65.31 + 110.77 -6.31 +5.14 4.01 + 16.63 aa 167.60 +91.35 -9.73 + 4.33* 3.22 + 13.04 ad -107.70 +35.27** 4.07 + 1.09*** -6.40 + 4.65 dd -344.80 + 188.95 18.25 + 7.57* 19.46 + 26.54 *, ** and *** are significant difference at P< 0.05, P<0.01 and P<0.001 respectively; a =additive gene effect; d = dominance gene effect; aa = additive by additive gene effect; ad = additive by dominance gene effect; dd = dominance by dominance gene effect. The magnitude of the phenotypic variances for all the traits (iron content, 100 grain weight and plant height) were generally higher than their corresponding genotypic variances (Table 5.9) indicating greater influence of the environment on these traits. Negative additive variance [VA] and negative dominance variance [VD] were recorded in seed iron concentration and 100 grain weight respectively. Exaggerated negative narrow sense heritability (-2.4) was also recorded in cowpea iron concentration. Broad sense heritability estimates were generally medium to high; 0.54, 0.89 and 0.70 for iron content, grain weight and plant height respectively. Almost one (0.8) effective factor was recorded for iron concentration in cowpea grain, and 0.2 effective factors each for 100 grain weight and plant height respectively. University of Ghana http://ugspace.ug.edu.gh 110 Table 5.9 Genetic variances, broad and narrow sense heritability and number of effective factors in a cross involving TVu-1 and TVu-999 Trait measured (VG) (VA) (VD) (VE) (VP) (h 2 b) (h2n) EF Iron mg/kg 3606.50 -15923 19529 3028.30 6634.80 0.54 -2.40 1.0 100 gw (g) 41.14 72.66 -31.52 4.93 46.07 0.89 1.58 0.2 Plant height (cm) 197.60 263.27 -65.68 84.07 281.67 0.70 0.93 0.2 VG= Genotypic variance; VA=Additive variance; VD =Dominance variance; VE=Environmental variance; VP= Phenotypic variance; hb 2 = Broad sense heritability and h2N= Narrow sense heritability; gw = grain weight; EF= effective factor [Wright (1952)] The phenotypic variances of the six generations derived from zinc and iron set of crosses are presented in Table 5.10. The two backcrosses (BC1P1 and BC1P2) had larger phenotypic variances in both crosses. Backcross one to high zinc parent (BC1PTVu-14845) had the largest phenotypic variance (129.69) to both parents. Similar trend was also detected in a cross between low and high iron (TVu-1 x TVu-15251) parents. The largest phenotypic variances observed for BC1P2 to their respective high zinc and iron parents indicate that dominance is for the lower iron and zinc content parents. Table 5.10 Phenotypic variances of the generations derived from zinc and iron set of crosses Generation Zinc (TVu-14845 x TVu-15251) Iron (TVu-1 xTVu-999) P1 18.33 249.96 P2 7.08 2743.80 F1 23.94 5923.20 F2 86.83 11655.00 BC1P1 42.26 6741.00 BC1P2 129.69 24147.00 University of Ghana http://ugspace.ug.edu.gh 111 5.4.3 Yield performance among F3 population evaluated under screen house conditions Means number of pod per plant; pod length; seed per pod and 100 seed weight per individual plants of 148 F3 population and their respective parents (TVu-14845 and TVu-15251) evaluated under screen house condition are presented in Table 5.11. Number of pod per plant among F3 populations ranges from 7 to 15 pods per plant. For 100 seed weight, high zinc content parent (TVu-15251) produced relatively large seed (11.596 > 5.575) compared with low zinc content parent. Similar trend was observed for number of seeds per pod, with low zinc parent (TVu-14845) having the highest number of seed per pod (11) compared to the high zinc content parent (TVu-15251) with 6 seeds per pod. Table 5.11 Distribution means of yield related traits among 148 F3 cowpea populations evaluated under screen house conditions. Generation Number of pod per plant Pod length Number of seed per pod 100 seed weight (g) P1 4 9.731 11 5.575 P2 8 6.825 6 11.596 F3 8 10.185 10 14.448 Range 7 – 15 1.3 - 16.5 1 – 15 3.28 - 26.13 P1= TVu-14845, P2= TVu-15251 The distribution of number of pod per plant (Figure 5.5) showed that high pod yielding parent plant (TVu-1521) produced 2 fold pod yield increase (8 > 4 pods per plant) compared to the low pod yielding parent (TVu-14845). The distribution of pod per plant among F3 plants skewed toward high pod yielding parent (TVu-15251), with most individual plants producing 7-8 pods per plants, indicating transgressive segregation. University of Ghana http://ugspace.ug.edu.gh 112 Figure 5.5: Frequency distribution of number of pod per plant among 148 F3 Population. Majority of the plants (134) produced long pod, ranging from 6.0 -16.5 cm, though some seed chambers were empty seeded (Figure 5.6). This also indicates the skewedness of the population toward parent two (TVu-14845) that produced long pods. 0 10 20 30 40 50 60 70 P1 P2 F3 F3 F3 F3 F3 F3 F3 F3 Plant pod/plant N u m b er o f p o d p er p la n t University of Ghana http://ugspace.ug.edu.gh 113 Figure 5.6: Frequency distribution of mean pod length per pod among 148 F3 Population The distribution of mean number of seed per pod per plant among F3 population showed a transgressive segregation with 19 plants producing more number of seed per pod than the best seed producing parent (TVu-14845) and 5 plants produced few number of seed per pod lower than the low producing seed per pod plants (TVu-15251). Majority of the F3 population (62 individual plants) produced 6 -10 seeds per pod (Fig 5.7). 0 5 10 15 20 25 30 35 40 45 Plant pod length M ea n P o d l en g th p er p la n t University of Ghana http://ugspace.ug.edu.gh 114 Figure 5.7: Frequency distribution of number of seed per pod among 148 F3 Population The distribution of seed weight among 148 F3 is presented in Figure 5.8. The populations also revealed significant number of (43) of individual plants showed transgressive segregation producing large seeds per plant that ranged from 19.82 to 26.13 g. 0 10 20 30 40 50 60 70 80 P1 P2 F3 F3 F3 Plant seed/pod N u m b er o f se ed p er p o d University of Ghana http://ugspace.ug.edu.gh 115 Figure 5.8: Frequency distribution of seed weight (g) among 148 F3 Population derived from a cross between TVu-14845 and TVu-15251 grown under screen house condition 0 10 20 30 40 50 60 70 80 P1 P2 F3 F3 F3 F3 F3 Plant 100 SWT 29 14 31 H u n d re d s ee d w ei g h t p er p o d University of Ghana http://ugspace.ug.edu.gh 116 5.4.4 Correlation between Zinc concentration and yield related traits in cowpea evaluated under screen house conditions The zinc content and yield related traits (number of pod per plant; pod length; seed per pod; and 100 seed weight) were included in the correlation analysis (Table 5.12). Weak negative significant (P < 0.05) correlation was recorded between zinc concentration and number of pod per plant. Pod size was positively correlated with number of seed per pod and 100 seed weight respectively. But no significant correlation was observed between zinc concentration and 100 seed weight. However, zinc content showed negative association with all yield related traits studied. Table 5.12 Phenotypic correlation coefficients between zinc concentration and yield related traits measured from F3 population derived from a cross between TVu-14845 and TVu- 15251 Pod/plant Pod length Seed/pod 100 seed weight (g) Zinc (mg/kg) Pod /plant 1 Pod length 0.0071 1 Seed/pod -0.0069 0.419*** 1 100 seed weight 0.0335 0.272*** 0.137 1 Zinc -0.1806* -0.062 -0.114 -0.074 1 *;***= significant difference at 0.05 and 0.001 respectively 5.5 Discussion Differential performance of genotypes in segregating populations indicates the presence of heritable variation. The large degrees of variation among the different populations for iron and zinc concentrations in the two different crosses strongly suggest the existence of genetic differences for these traits which can be utilized in breeding for high iron and zinc cowpea. Similar results have been reported for genotypic differences of zinc in rice grain (Graham et al., University of Ghana http://ugspace.ug.edu.gh 117 1999; Cakmak et al., 2004; Gregorio et al., 2000), maize grain (Bänziger & Long, 2000) and cowpea grain by Asante et al. (2009) and Carvalho et al. (2012) The transgressive segregates observed in some individual F2 and backcross populations in zinc concentration and the deviation of F1 population toward the high zinc content parent with mean zinc concentration (39.90 > 37.70) higher than that of the high zinc content parent, suggests that high zinc content parent (TVu-15251) contributed more of the alleles. Also the high zinc content parent was partially dominant (0.17 degree of dominance) to low zinc content. Similar results were observed in common beans by Islam et al. (2004) and Blair et al. (2008). Similarly, the two fold increase in cowpea grain zinc concentration observed among some F2 individual (82.54 mg/kg) compared with the highest zinc content parent (37.70 mg/kg) conform to a similar report of up to 37.3% increases zinc concentration in common bean by Ribeiro et al. (2014), when the crossing and selection among segregation progenies in early hybrid generations were done in a controlled environment in Zaria. The marginal over dominance (0.17) observed for zinc concentration in cowpea in favor of the high zinc content parent, could be due to allelic effects, where allele affects the components in opposite directions and there is degree of dominance on the scale in which the components combine to give fitness (Wallace, 1968). Moreover, different stages of life cycle (maturity period) observed among the population, coupled with seasonal variation encountered by the same F3 population could also be another reason for over dominance (Falconer & Macky, 1996). This population undergoes seasonal variation from optimum to stressed (extremely cold) temperatures (November 18th, 2013 to February, 2014). For iron concentration in cowpea, F1 population produced seeds with low iron concentration, lower than that of the lowest iron content parent (TVu-1), implying that the low University of Ghana http://ugspace.ug.edu.gh 118 iron content parent is dominant over the high iron content parent (TVu-999). However, transgressive segregation was observed among some F2 population with two fold iron concentration increase over the high iron content parent (1070.23 mg/kg > TVu-999=419 mg/kg). My preliminary findings indicate an increase in iron concentration. This is the first attempt to study the genetic pattern of iron and zinc in cowpea; the results need to be validated. Furthermore, this result implies that cowpea breeding for enhanced nutrition could be possible. However, similar reports of 94.0% increases in iron concentration in the common bean grain have been reported by Blair et al. (2008) and Ribeiro et al. (2012). The deviation from the mid-point to less than the concentration of low iron content parent implies that low iron content parent (TVu-1) contributes more alleles to the inheritance of iron concentration. The marginal over dominance could be due to seasonal variation leading to changes in the environmental differences. For example maximum temperature before flowering, rainfall after flowering, presence of zinc in 30 – 60 cm soil depth and relative humidity (RH) after flowering could attribute to significant variation. This conforms to the findings of Joshi et al. (2010) who reported that environmental conditions were significant and accounted for around 59% of variation in grain mineral concentration in wheat. The low iron concentration observed among F1 seeds which resemble the low iron content parent is an indication of significant maternal effect. This further implies that iron concentration is dependent on the seed coat and selection should begin on the F3 seeds (embryos in the F2 generation) when segregation was verified F3. Thus, a breeding program should investigate the existence of maternal effect on iron concentrations in germplasm subjected to selection. The embryo generation should be considered for the selection process and the progression of University of Ghana http://ugspace.ug.edu.gh 119 segregating populations in genetics and breeding programs. Similar findings were reported in common bean by Rosa et al. (2010). On the other hand, the expression of low zinc concentration was overshadowed by the paternal parent in the F1 seeds derived from low and high zinc parents. The phenotypes of these seeds were similar in terms of zinc concentration and represented the expression of the genotype of the F1 generation. Therefore, F3 seeds showed embryos in the F2 generation. For this reason, selecting for a high zinc concentration in the cowpea seeds should begin with the F2 seeds, a generation where ample genetic variability is observed. Similar findings were reported in copper concentration in common bean (Samineni et al., 2011) In addition to additive [a] and additive by dominance [ad] gene effects observed in the two set of crosses for zinc and iron concentrations, epistatic gene effects had high contributions in controlling zinc content in cowpea grain. The observed positive dominance [d] and negative dominance by dominance [dd] gene effects for both iron and zinc contents indicate the duplicate gene action. This is in accordance with Mather & Jinks (1982) who stated that gene action is considered to be duplicating when dominance [d] and dominance by dominance [dd] gene effects have different signs (positive and negative or vice versa). Furthermore, the observed negative dominance by dominance [dd] in both iron and zinc population implies that unidirectional dominant and reductive alleles were involved in dominant phenotype (Ribeiro et al., 2012) Effective factors; 0.2 and 0.8 estimated for zinc and iron concentrations, respectively indicate that in addition to minor genes, one important gene is involved in the control of iron concentration in cowpea seed, while many genes interact in the inheritance of zinc concentration. These minor genes could be in duplicate recessive form, where when one gene exists in recessive form, the zinc concentration may not be inherited. These results do not agree with the findings of University of Ghana http://ugspace.ug.edu.gh 120 Gelin et al. (2007), who did not identify a QTL for iron in common bean, perhaps due to limited map coverage. However, Foster-Hartnett et al. (2002) and Cichy et al. (2005) reported a monogenic inheritance for grain zinc concentration in common bean and recently, Blair et al. (2008) and Blair et al. (2011) reported 6 QTL for zinc concentration in the same common beans using different genotypes with that of Foster-Hartnett et al. (2002) and Cichy et al. (2005). These discrepancies could be due to species differences (Phaseolus vulgaris and Vigna unguiculta) and environmental differences as well. This is because within common bean alone discrepancies exist between Foster-Hartnett et al. (2002) and Blair et al. (2008) findings, which may be due to genotypic differences and environmental effects. Further genetic study to explore the inheritance pattern of zinc in cowpea grain is therefore recommended. The high broad sense heritability observed for both iron and zinc concentrations indicate that a progress could be made. This result did not agree with the finding of Oluwatosin (1998) and Joshi et al. (2010) who reported low heritability for iron and zinc concentrations in cowpea and wheat grains respectively. The discrepancies between these results and that obtained by Oluwatosin (1998) could be due to the genetic make-up of the cowpea genotypes used and environments were the evaluation was conducted. In this study, the populations were evaluated under confined area (Screen house), with known status of soil iron and zinc content, while Oluwatosin‘s trials were conducted under field conditions in three locations without details of soil nutritional content. The same multilocation evaluation applied for wheat trial by Joshi et al. (2010). The overestimated narrow sense heritability values recorded for iron content as well as grain weight and plant height could be alluded to epistasis or developmental factors or true micro- environmental variations or sampling error. Similarly, values in early generation tend to be University of Ghana http://ugspace.ug.edu.gh 121 overestimated due to an upward bias from repulsions phase of linkage. This linkage is broken in a later generations due to recombination and low degree of dominance (Samineni et al., 2011) However, Kearsey & Pooni (1996) reported that developmental factor such as plant height and grain weight cannot be separated from the genuine within the family variations. Earlier reports also indicated that it could be caused by environmental influence and some errors during sampling. Similar findings were reported in other traits in chickpea (Samineni et al., 2011) and rice (Kiani et al., 2013). Signs associated with different estimates of epistasis indicate the direction in which gene effects influence the population means. Mather and Jinks (1982) proposed the association or dispersion of genes in the parents based on signs associated with epistatic gene effects such as additive by additive [aa] and additive by dominance [ad]. These signs were in opposite directions and significant in the control for 100 seed weight in a cross between low iron (small seeded) and high iron (medium seeded) parents. A negative sign for any of these parameters indicates an interaction between increasing and decreasing alleles, thus providing some evidence for the existence of dispersion in the parental genotypes which hinders early selection for such traits. Similarly, signs of these parameters were in opposite direction for seed weight which suggests the contribution of alleles from the larger grain parent in the expression of the dominant phenotype. On the other hand, significant negative additive [a] and negative dominance [d] gene actions were observed with epistasis interaction of negative [aa] in a cross between TVu-14845 and TVu- 15251 respectively. Having the same negative signs indicates large influence of recessive parent. (Mather & Jinks, 1982) Positive or negative form of additive × additive [aa] interaction show association and dispersion of alleles in parents, respectively. Therefore, negative and significant values of additive University of Ghana http://ugspace.ug.edu.gh 122 by additive [aa] interaction for 100 grain weight in the two crosses showed alleles dispersion in parents for 100 seed weight per plant. The contribution of additive gene effect relative to dominance genes toward cowpea grain weight was identified, characterizing an additive allelic interaction. This result is vital for breeding programs. When an additive allelic interaction is predominant, selection is facilitated, because superior parent Tvu-14845 and TVu-999) will produce superior ancestry, as observed in the segregating population of this result. Similar results were reported by many authors (Lopes et al., 2003; Adeyanju et al., 2012; Egbadzor et al., 2013) for Vigna unguiculata. The negative dominance [d] gene action for seed weight further indicates reductive alleles involving dominant phenotype otherwise increasing alleles including dominant phenotype. This result shows an increase or contribution of alleles from the large seeded parent from each of the cross. Plant height showed negative significant additive [a] and positive additive by dominance [ad] gene action in a cross between short and tall plants of TVu-14845 and TVu-15251. This implies that plant height was predominantly under complementary gene action in this study. These results are in agreement with the findings of Adeyanju et al. (2012) who reported complementary and duplicate gene action for plant height in two different crosses of cowpea. The occurrence of duplicate gene action would limit selection in early generations for plant height. It is therefore good to make a selection in later generations when reasonable homozygosity is attained. The occurrence of high narrow sense heritability in grain size indicates that the selection for grain size can be made in early generations. High narrow sense heritability observed for 100 seed weight, conformed to Tchiagam, et al. (2011b) but did not agreed with other authors (Adeyanju et al., 2012; Egbadzor et al., 2013). These differences could be attributed to the types University of Ghana http://ugspace.ug.edu.gh 123 of cowpea genotype used, environmental conditions and more importantly developmental factor such as grain size which varied within a pod of a plant. To improve grain size in cowpea, it is important to pay attention to the nature of the genotype, true micro-environmental variations and sampling error, especially, when dealing with duplicate structures like pod size and grain size (Kearsey & Pooni, 1996). Selection for seed weight and plant height in cowpea is possible. The fixation of these characters could be observed in advanced generations. The significant variation in mean number of pods per plant, seeds per pod, pod length and 100 seed weight among F3 population with various degree of segregation in all the yield related traits observed, conform to the findings of many authors (Ariyo, 1995; Rangaiah et al., 1999; (Romanus et al., 2007). Negative significant correlation observed between zinc concentration and number of pod per plant, implies that increase in number of pods per plant may lead to decrease in zinc concentration in cowpea seeds. It further suggests that selection for high pod loads per plants may indirectly lead to selection for low zinc content. This could be due to environmental factors such as availability of soil mineral elements that may lead to plant uptake and subsequent accumulation in the seed and other parts of the plants. This conform to similar findings of Joshi et al. (2010) who observed high grain concentration in wheat grains, grown in the soil that contained high zinc content and replicated in a low zinc content soil. The correlation further shows that breeding for higher pod yield (desirable for farmers) can induce a reduction in the grain zinc content. Thus, indirect selection to increase either of the traits, via selection for one, should be avoided. Similar findings were reported in cowpea by Moura et al. (2012) The non-significant correlation observed between zinc concentration and seed weight in cowpea, conformed to similar reports of Moraghan & Grafton, 2001) in common bean. The strong University of Ghana http://ugspace.ug.edu.gh 124 positive correlation observed between pod length and number of seeds per pod and 100 seed weight; suggest that pod size increases with increase in number and size of the seeds contained within a pod. The correlation also indicates the reliability of these traits in selecting for large seeded plants. 5.6 Conclusion and recommendations Substantial variation of grain nutritional traits among cowpea observed: Fe (1.0 - 329mg/kg), Zn (10.01 - 386 mg/kg) and Protein (1.72-29.93%). Cowpea genotypes with less than 10% protein content cleared the consumers‘ perception that all cowpea have high proteins. Differential performance of genotypes among segregating populations and large degrees of variation among the different populations for iron and zinc concentration strongly suggest the existence of genetic variability that can be utilized for cowpea enhancement. The marginal over dominance and transgressive segregates observed in some individual of F1, F2 and backcross populations in zinc and iron concentration with two fold increase mean zinc concentration over the high zinc content parent suggests that progress could be made at early generation selection for zinc and later generation for iron. Genetic pattern of iron and zinc content elucidated and possibility of enhancement of Fe and Zn in cowpea is realized. The results of this study indicated that the inheritance of iron and zinc accumulation in cowpea grains was found to be epistasis with mainly additive-dominance gene effects. Inheritance of iron and zinc is polygenically controlled. This can be explained by the fact that both iron and zinc have similar mechanisms of uptake and accumulation to grain. This is promising for plant breeders who are interested in micronutrient enhancement, given that if they have similar mechanism for uptake and inheritance it may be easy to select for these traits both University of Ghana http://ugspace.ug.edu.gh 125 phenotypically and through marker assisted selection. Plant height was predominantly under complementary gene action in both crosses. The weak negative correlation between zinc concentration and number of pods per plant may decrease the pod yield of a cowpea genotype; as such direct selection for zinc content by selecting high yield genotype should be avoided. For selection of a cowpea genotype to be effective in at least maintaining the yield of that genotype, it is important that yield components should not be significantly correlated with zinc content. The number of seed per pod, 100 seed weight and pod length which showed relatively non significantly correlated, could be considered the best predictors of seed yield in cowpea. Five promising accessions could be used as donor parents in future study or consumptions: (1) Iron content: TVu-1330, TVu-13495, TVu-16400, TVu-8751 and TVu-9725 (2) Zinc content: TVu-1616, TVu-2155, TVu-15251, TVu- 301 and TVu-232 (3) Protein: TVu-13495, TVu-13088, TVu-15187, TVu- 347 and TVu-13468 University of Ghana http://ugspace.ug.edu.gh 126 CHAPTER SIX 6 GENERAL DISCUSSION 6.1 Participatory Rural Appraisal (PRA) The participatory rural appraisal conducted revealed that farmers identified cowpea production constraints that did not differ significantly between Sudan-savanna zone (Bunkure and Wudil) and Sudano-Sahelian transition zone of Bichi local government area. Farmers in both Sudan savanna and Sudano-sahelian transition zones emphasized on seed related issues as the main production constraints: inadequate supplies of improved cowpea varieties especially at planting time, lack of insect resistant variety, diseases and drought tolerant varieties. As part of the management strategies, farmers adopt use of various insecticides at different formulations, which affect the environment and alter the ecosystem thereby, reducing the beneficial insects that aid pollination. This implies that breeding emphasis should focus more on host plant resistance in addition to ongoing pod borer resistant cowpea project in Africa by AATF. Furthermore, tackling biotic constraints like that of insects needs a multidisciplinary team of cowpea breeders, entomologist, pathologists, soil scientists, agronomist and socio-economist. The team could work more successfully if farmers‘ participation is encouraged. Farmers and consumers identified seed quality traits as important criteria for making choices for cowpea varieties. Majority of the processors mainly women preferred reduced cooking time, ease of removal of the hilum and testa and less oil consumption. End users, on the other hand, considered protein, grain colour and size as their most preferred traits. Women consumers had more knowledge of nutritional values (iron and zinc content) of cowpea than men. Different cowpea characteristics desired by different cowpea users (farmers, processors and consumers) requires the involvement of various cowpea stakeholders to actively participate in the University of Ghana http://ugspace.ug.edu.gh 127 process of variety development. This will lead to the development of cowpea varieties for specific locations and particular needs of consumers. Farmers in the study areas indicated their willingness to pay premium price for improved cowpea varieties that will satisfy their needs. Farmers in Bunkure and Wudil preferred insect resistant varieties in addition to grain attributes, while Bichi farmers were more concerned about early maturing variety that will escape terminal drought. Consumers especially women showed their willingness to pay for premium price for nutrient added value cowpea. These preferences for cowpea pose challenges to breeder to think of forming a multidisciplinary team that will include medical personnel, nutritionist and food industry toward solving consumer‘s problems. 6.2 Variation among cowpea accessions for grain quality traits The significant variation observed among the accessions for percentage crude protein, zinc content and iron content, implies that genetic variability for nutritional value particularly iron and zinc exist in cowpea and could be exploited. Many of these variations can be generated by conventional breeding methods to address the nutritional needs in developing countries. Modifying rank summation index of Mulamba and Mock (1978) identified three accessions: TVu- 13088, TVu-13495 and TVu-9725 that contain the largest number of desirable nutritional attributes. These accessions can be nominated for anti-nutritional testing and yield stability testing across cowpea growing areas before recommendation for release for infant diets formulation and general use. For nutrient specific, five accessions each for high protein content, high zinc content and high iron content were identified which can also be recommended for use after validation. The small genetic differentiation (0.26-0.45) between African cowpea accessions and USA accessions provides important insights that can be used to improve the efficiency of cowpea germplasm preservation and selection for parental lines when initiating a breeding work. The University of Ghana http://ugspace.ug.edu.gh 128 information will enable rational planning by gene bank curators to help reduce duplicates among the accessions and form core collection of their accessions. Clustering of accessions based on closely related genetic distances implies that, accessions that are within a related genetic distance may exhibit common adaptive complexes of physiological traits coupled with a relatively restricted range of morphological and underlying genetic variation. Thus, crossing related accessions is expected to produce high frequency of relatively similar-looking progeny, while crossing between members of different cluster is expected to produce more variable progeny, perhaps with a relatively lower average performance in early generations, in this case selection should be delayed to later generations. Breeding strategies involving series of back crosses using conventional method is therefore recommended. The negative correlation observed between protein and carbohydrate as well as protein and fat content indicates that selection for high protein will decrease carbohydrate content and fat content. Therefore, indirect selection for these traits should be avoided. On the other hand, the positive significant correlation observed between protein content and iron content, protein content and ash content implies the possibility of improving iron content by selecting for high protein or ash content. The non-significant correlation observed between 100 seed weight and protein content and iron content, respectively, indicates that direct selection for large grain size through selecting protein or iron is possible. 6.3 Inheritance of iron and zinc concentration and other seed quality traits in cowpea Iron and zinc content in cowpea grain fitted into the additive [a] and additive by dominance [ad] model. The resemblance of F1 to low iron content parent might be due to dominance, overdominance or epistasis. It may also be an indication of maternal inheritance of iron content in cowpea grain, as the expression of high iron content was masked in F1, and started University of Ghana http://ugspace.ug.edu.gh 129 manifesting in the subsequent generations (F2 & F3 seeds). This implies that iron concentration may be dependent on the seed coat, and selection for high iron in the cowpea grain should begin from the F3 seeds (embryo in F2 generation). On the other hand, the non-resemblance of the F1 population to the maternal parent (low zinc content) suggest that selection for high zinc content can begin with the F2 seeds. Seed weight is predominantly under complementary gene action in this study, suggesting the possibility of considerable amount of heterosis for seed weight. The implication of this to breeding is that selection is facilitated as superior (large seeded) parent produced superior progenies (transgressive segregates) observed in this study. Seed weight in a cross between TVu- 14845 and TVu-15251 showed significant negative additive [a] and negative dominance [d] suggesting large influence of recessive parents. High narrow sense heritability estimates for grain weight further indicates that selection can be made in early generation. Plant height is under the control of duplicate gene action in a cross between short and tall plant and complementary in a cross between tall and short plant respectively. The prevalence of duplicate gene action would limit selection in early generations for plant height. It may therefore be necessary to delay selection in early selfing generations until homozygosity is reached. Therefore, it is recommended to make maternal parent plant to be taller when making crosses, as heterosis is expected there, due to complementary gene action. 6.4 Challenges Integrating preferences of different cowpea stakeholders (farmers, processors, and vendors) into breeding objectives needs a multidisciplinary team of scientists. Few polymorphic markers (25 SNP for zinc and 10 out of 25 for iron) identified, constrained the identification of QTL associated with iron and zinc, despite the development of mapping population. University of Ghana http://ugspace.ug.edu.gh 130 6.5 Recommendation PRA should be considered as pre-breeding and part of breeding activity, as it will be used to set breeding objectives that will solve farmers and consumer‘s needs. This will facilitate adoption of new varieties when developed. This will help in proper designing, planning, implementation, monitoring and evaluation of breeding activities and impact assessment. Farmers‘ preference varied according to agro-ecological savanna and blanket recommendations could mask the preference choices in a particular savanna. This calls for the need for breeding varieties that are trait and site specific, in addition to high yielding. Variability study have identified some promising accessions that combine reasonable (within the recommended dietary intake level) amount of protein, iron and zinc content and other set for nutrient specific. I therefore, recommend further anti-nutritional testing and yield stability testing before recommending to consumers as alternatives to animal protein which is currently beyond the reach of many consumers. The study also raises several questions of a physiological and biochemical nature of minerals uptake by the plant, indicating that processes that culminate in the mineral storage in cowpea seeds are not well studied. Additional studies are needed to understand the integration of all those processes and their implications to micronutrient enhancement in crops like cowpea. Further research to explore more polymorphic SNP markers that could be used to identify QTL associated with iron and zinc concentrations and subsequent MAS is also encouraged. This will facilitate the development and deployment of enhance cowpea with iron and zinc content. University of Ghana http://ugspace.ug.edu.gh 131 BIBLIOGRAPHY Acquaah, G. (2007) Principles of plant Genetics and Breeding. Blackwell Publishing, Malden, USA. Adeyanju, A.O., Ishiyaku, M.F., Echekwu, C.A.&, et al. (2012) Generation mean analysis of dual purpose traits in cowpea ( Vigna unguiculata [ L .] walp ). African Journal of Biotechnology, 11 (46): 10473–10483 Ajeigbe, H. A. Singh, B. B., Adeosun, J.O. & Ezeaku, I.E. (2010) Participatory on-farm evaluation of improved legume- cereals cropping systems for crop-livestock farmers : Maize- double cowpea in Northern Guinea Savanna Zone of Nigeria. African Journal of Agricultural Research, 5 (16): 2080–2088 Ajeigbe, H. A., Ihedioha, D. & Chikoye, D. (2008) Variation in physico-chemical properties of seed of selected improved varieties of Cowpea as it relates to industrial utilization of the crop. African Journal of Biotechnology, 7 (20): 3642–3647 Ajibade, .S. R., Weeden, N. F. & Chite, S.M. (2000) Inter simple sequence repeat analysis of genetic relationships in the genus Vigna. Euphytica, 111: 47–55 Akande, S.R., & Balogun, M.O. (2009) Multi-locational evaluation of cowpea grain yield and other reproductive characters in the forest and southern guinea savanna agroecologies of Nigeria. Electronic Journal of Environmental, Agricultureal and Food chemistry, 8 (7): 2009 Ali, Y., Aslam, Z., Hussain, F. & Shakur, A. (2004) Genotype and environmental interaction in cowpea ( Vigna unguiculata-L ) for yield and disease resistance. International Journal of Environmental Science and Technology, 1 (2): 119–123 Anderson, J. M. & Ingram, J.S.I. (1998) Tropical Soil Biology and Fertility: A Handbook of Methods. UK: CAB International Anele, U.Y., Südekum, K.H., Hummel, J., Arigbede, O.M., Oni, A.O., Olanite, J.A.., Böttger, C., Ojo, V.O. & Jolaosho, A.O. (2011) Chemical characterization, in vitro dry matter and ruminal crude protein degradability and microbial protein synthesis of some cowpea (Vigna unguiculata L. Walp) haulm varieties. Animal Feed Science and Technology [online], 163 (2-4): 161–169. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0377840110003871 [Accessed 13 October 2014] AOAC (2000) Official methods of Analysis of the Association of Analytical Chemist. International, 17th Ed., Volume II: Food Composition; Additives; Natural contaminants., II: 2000 University of Ghana http://ugspace.ug.edu.gh 132 Ariyo, O.J. (1995) Genetic Variability , Correlations and Path Coefficient Analysis of Components of Seed Yield in Cowpea ( Vigna unguiculata ). Pertanika Journal of Tropical Agricultural Science, 18 (1) Asante, I.K. (1991) Seed coat inheritance in the cowpea (Vigna unguiculata (L.) Walp ). Ghana Journal of Science, 39: 77–81 Asante, M.D. (2012) Genetics of grain quality traits in rice. PhD thesis submitted to the West African Center for Crop Improvement, College of Basic and Applied Sciences, University of Ghana, Legon Asante, I. K., Adu-Dapaah, H. & Acheampong, A.O. (1991) Determination of Some Mineral Components of Cowpea ( Vigna unguiculata ( L .) Walp ) Using Instrumental neutron activation analysis . Ghana Journal of Science, 39 (1-6): 0–5 Asante, I. K., Adu-Dapaah, H. & Addison, P. (2004) Seed weight and protein and tannin contents of 32 cowpea accessions in Ghana. Tropical Science, 44: 2004 Asante, I.K., Asare, C., Annan, K. & Amuzu, N. (2009) Chemical characterization of cowpea (Vigna unguiculata (L.) Walp) accessions and some of their wild relatives. African Journal of Science and Technology (AJST) Science and Engineering Series, 10 (2): 99–107 Ayinde, I.A. (2005) Market dyneamics for cowpeas : Weight and measurement issues issues in Kuto market, Abeokuta, Nigeria. Report submitted as part of the market study on: Building a food marketing Policy evidence base in Nigeria Ba, F. S., Pasquet, R. S. & Gepts, P. (2004) Genetic diversity in cowpea [ Vigna unguiculata ( L .) Walp .] as revealed by RAPD markers. Genetic Resources and Crop Evolution, 51: 539– 550 Badiane, F.A., Gowda, B.S., Cissé, N., et al. (2012) Genetic relationship of cowpea ( Vigna unguiculata ) varieties from Senegal based on SSR markers. Genetics and molecular Research, 95 (1): 292–304 Badiane, F.A., Diouf, D., Sané, D., Diouf, O., Goudiaby, V. & Diallo, N. (2004) Screening cowpea [ Vigna unguiculata (L.) Walp.] varieties by inducing water deficit and RAPD analyses. African Journal of Biotechnol., 3: 174–178 Bänziger, M., & Long, J. (2000) The potential of increasing the iron and zinc density of maize through plant breeding. Food and Nutrition Bulleting, 21: 397–400 Barea, J.M., Pozo, M.J., Azcon, R. & Azcon-Aguilar, C. (2005) Microbial co-operation in the rhizosphere. Journal of Experimental Botany, 56 (417): 1761–1778 Beckmann, J. S. & Soller, M. (1990) Toward a unified approach to genetic mapping of eukaryotes based on sequence tagged microsatellite sites. Nature Biotechnology, 8: 930–932 University of Ghana http://ugspace.ug.edu.gh 133 Benton-Jones, J. (1989) Plant analysis techniques. Benton-Jones Laboratories, Georgia. p. 24 Bernardo, R. (2002) Breeding for quantitative traits in plants. Stemma Press , Woodbury, MN. Bernardo, R. (2008) Molecular Markers and Selection for Complex Traits in Plants: Learning from the Last 20 Years. Crop Science [online], 48 (5): 1649. Available from: https://www.crops.org/publications/cs/abstracts/48/5/1649 [Accessed 15 July 2014] Blair, M. W., Astudillo, C., Grusak, M. A., Graham, R. & Beebe, S.E. (2008) Inheritance of seed iron and zinc concentrations in common bean (Phaseolus vulgaris L.). Molecular Breeding [online], 23 (2): 197–207. Available from: http://link.springer.com/10.1007/s11032-008- 9225-z [Accessed 25 October 2014] Blair, M.W., Astudillo, C., Rengifo, J., Beebe, S. E. & Graham, R. (2011) QTL analyses for seed iron and zinc concentrations in an intra-genepool population of Andean common beans (Phaseolus vulgaris L.). Theoretical and Applied Genetics. [online], 122 (3): 511–521. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21113704 [Accessed 25 October 2014] Blair, M.W., Paulo, I., Astudillo, C. & Grusak, M.A. (2013) A legume biofortification quandary: variability and genetic control of seed coat micronutrient accumulation in common beans. Frontiers in Plant Science [online], 4 (July): 1–14. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3725406&tool=pmcentrez&rend ertype=abstract [Accessed 28 October 2014] Botstein, D., White, R. L., Skolnick, M. & Davis, R.W. (1980) Construction of a genetic linkage map in man using restriction fragment length polymorphisms. American Journal of Human Genetics, 32: 314–331. Boukar, O. Muranaka, S., Franco, J. & Fatokun, C. (2010) ―Protein and mineral composition in grains of elite cowpea lines.‖ In Innovative research along the cowpea value chain, Proceedings of the Fifth World Cowpea Conference on improving livelihoods in the cowpea value chain through advancement in science, Edited by O. Boukar, O. Coulibaly, C.A. Fatokun, K. Lopez, and M. Tam. 2010. pp. 88–99 Boys, K., Faye, M., Fulton, J. & Lowenberg-deboer, J. (2007) The economic impact of cowpea research in Senegal : an ex-post analysis with disadoption. Agricultural Economics, 36 (2007): 363–375 Brown, A.H.D. (1989) ―The case for core collections. Pages 136 – 156 in The use of plant genetic resource. Cambridge, UK: Cambridge University Press.‖ In In the use of Plant genetic resource. Cambridge, UK: Cambridge University Press. pp. 136–156 Burgess, K.S., Etterson, J.R. & Galloway, L.F. (2007) Artificial selection shifts flowering phenology and other correlated traits in an autotetraploid herb. Heredity, 99: 641–648 University of Ghana http://ugspace.ug.edu.gh 134 Cakmak, I., Kalayci, M., Ekiz, H., Braun, H.J., Kilinc, Y. & Yilmaz, A. (1999) Zinc de ® ciency as a practical problem in plant and human nutrition in Turkey : A NATO-science for stability project. Field Crops Research, 60: 175–188 Cakmak, I., Torun, A., Millet, E., Feldman, M., Fahima, T., Korol, A. B., Nevo, E. Braun, H.J. & Ozkan, H. (2004) Cakmak I, Torun A, Millet E, Feldman M, Fahima T, Korol AB, Nevo E, Braun HJ, Ozkan H (2004). Triticum dicoccoides: an important genetic resource for increasing zinc and iron concentration in modern cultivated wheat. Soil Science Plant Nutrition., 50: 1047–1054 Carvalho, A.U., Mateus de Sousa, N., Farias, D.F., Rocha-Bezerra, C.B., Pareira da Silva, R.M., Viana, M.P., Gouveia, S.T., Sampaio, S.S., Barbosa de Sousa, M., Gomes de Lima, G. P., Selene, M.M., Barros, C. C. & Filho, F.R. (2012) Nutritional ranking of 30 Brazilian genotypes of cowpeas including determination of antioxidant capacity and vitamins. Journal of Food Composition and Analysis [online], 26 (1-2): 81–88. Available from: http://linkinghub.elsevier.com/retrieve/pii/S088915751200018X [Accessed 22 October 2014] Chambers, R., Pacey, A. & Thrupps, L.A. (1989) Farmer first. Farmer innovation and agricultural research. Intermediate Technology Publications, London . Thrup. (Ed. Lokndon: Intermediate Technology Publicationa Chhuneja , P., Dhaliwal, H. S., Bains, N.S. & Singh, K. (2006) Aegilops kotschyi and Aegilops tauschii as sources for higher levels of grain iron and zinc. Plant Breeding, 125: 529 – 531 Choi, E.Y., Graham, R. & Stangoulis, J. (2007) Semi-quantitative analysis for selecting Fe- and Zn-dense genotypes of staple food crops. Journal of Food Composition and Analysis, 20: 496–505 Cichy, K. A., Forster, S., Grafton, K. F. &, Hosfield, G.L. (2005) Inheritance of seed zinc accumulation in navy bean. Crop Sci ence, 45:864–870. Doi :10.2135/cropsci2004.0104, 45: 864–870 Cobbinah, F. A., Addo-Quaye, A. A. & Asante, I.K. (2011) Characterization , evaluation and selection of cowpea (Vigna unguiculata ( L .) Walp ) Accessions with desirable traits from eight regions of Ghana. ARPN Journal of Agricultural and Biological Science, 6 (7): 21– 32 Collard, B.C.Y & Mackill, J.C. (2008) Marker-assisted selection — an approach for precision plant breeding in the twenty-first century. Philosophical Transactions of the Royal Society of Botany, 363: 557–572 Coulibaly, O. & Lowenberg-De Boer, J. (2002) ―The economics of cowpea in West Africa.‖ In in Challenges and opportunities for enhancing sustainable cowpea production, edited by C.A. Fatokun, S.A. Tarawali, B.B. Singh, P.M. Kormawa, M. Tamò. Proceedings of the University of Ghana http://ugspace.ug.edu.gh 135 3rd World Cowpea Conference 4-8 September, 2000, IITA, Ibadan, Nigeria. 2002. pp. 351–366 Coulibaly, O., Alene, A.D., Manyong, V., Sanogo, D., Abdoulaye, T., Chianu, J., Kamara, A., Tefera, H. & Boukar, O. (2009) Situation and outlook for cowpea and soy bean Sub- Saharan Africa: report of the the Tropical Legumes II Project in West and Central Africa Crepieux, S., Lebreton, C., Flament, P., & Charmet, G. (2005) Application of a new IBD-based QTL mapping method to common wheat breeding population: Analysis of kernel hardness and dough strength. Theoretical and Applied Genetics, 111: 1409–1419 Crepieux, S., Lebreton, C., Servin, B. Charmet, G. (2004) Quantitative Trait Loci (QTL) detection in multicross inbred designs: Recovering QTL identical-by-descent status information from marker data. Genetics, 168: 1737–1749 Day, P.R. (1965) Particle fractionation and particle size analysis. In C.A. Black ed., Methods of Soil analysis, Madison, WI: American Society of Agronomy, pp. 545–567 Ding, G., Yang, M., Hu, Y., Liao, Y., Shi, L., Xu, F. & Meng, J. (2010) Quantitative trait loci affecting seed mineral concentrations in Brassica napus grown with contrasting phosphorus supplies. Annals of Botany [online], 105 (7): 1221–34. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2887070&tool=pmcentrez&rend ertype=abstract [Accessed 13 October 2014] Diouf, D. & Hilu, K.W. (2005) Microsatellites and RAPD markers to study genetic relationship among cowpea breeding lines and local varieties in Senegal. Genetic Resources and Crop Evolution, 52: 1057–1067 Dovlo, F.E., Williams, C.E. & Zoaka, L. (1976) Cowpeas: Home preparation and use in West Africa. International Research and Development Council, Ottawa, Canada. Annual report Drabo, I., Redden, R., Smithson, J. B. & Aggarwal, V.D. (1984) Inheritance of seed size in cowpea (Vigna unguiculata (L.) Walp.). Euphytica [online], 33 (3): 929–934. Available from: http://link.springer.com/10.1007/BF00021923 Dugje, I.Y., Omoigui, L.O. and Ekeleme, F., Kamara, A.Y. & Ajeigbe, H. (2009) Farmers ’ Guide to Cowpea Production in West Africa. An IITA, Publication Egbadzor, K. F., Amoako-Attah, I., Danquah, E. Y., Offei, S. K., Ofori, K. & Opoku-Agyeman, M.O. (2012) Relationship between flower, immature pod pigmentation and seed testa of cowpea. International Journal of Biodiversity and Conservation [online], 4 (12): 411– 415. Available from: http://www.academicjournals.org/IJBC/abstracts/abstracts/abstracts2012/Sept/Egbadzor et al.htm [Accessed 17 October 2014] University of Ghana http://ugspace.ug.edu.gh 136 Egbadzor, K. F., Dadoza, M., Danquah, E. Y., Yeboah, M., Offei, S. K. & Ofori, K. (2013) Genetic control of seed size in cowpea (Vigna unguiculata (L.) Walp). International Journal of Agriculture Sciences [online], 5 (2): 367–371. Available from: http://bioinfopublication.org/viewhtml.php?artid=BIA0001956 Ehlers, J. D., R. L. Fery, & Hall, A.E. (2002) ―Cowpea breeding in the USA: new varieties and improved germplasm.‖ In C. A. Fatokun, S. A. Tarawali, B. B. Singh, P. M. Kormawa, and M. Tamo, ed. Challenges and Opportunities for Enhancing Sustainable Cowpea Production. Ibadan, Nigeria: International Institute of Tropical Agriculture. pp. 62–77 Ehlers, J.D. & Hall, A.E. (1997) Cowpea (Vigna unguiculata (L.) Walp.). Field. Field Crops Res, 53: 187–204 Falconer, D. S., & Mackay, T.F.C. (1996) Introduction to quantitative genetics. London, UK .: Longman, Faris, D.G. (1965) The origin and evolution of the cultivated forms of Vigna sinensis. Canadian Journal of Genetic and Cytology, 7: 4333–452 Fatokun, C.A. (2009) Designer (Cowpea) Plants. Available on http://r4dreview.org/2009/03/designer- cowpea-plants/ Fatokun, C. A. & Stangoulis, J. (2008) Cowpea Sampling Protocol, In: HarvestPlus Crop sampling protocols for micronutrient analysis Fatokun, C., Boukar; O., Kamara; A., Coulibaly, O., Alene, A., Boahen, S., Bandyopadhyay, R., Kumar, L., Kouyate, K., Omoigui, L., Gimba, S., Garba, G., Dioffo, S., Abaidoo, R., Muranaka, S., Touré, M., Adamou, M., Ishiyaku, M. F., Mligo, J. and Amane, M. (2012) Enhancing Cowpea Productivity and Production in Drought-Prone Areas of Sub- Saharan Africa. In: Abate T. (ed.). 2012. Four Seasons of Learning and Engaging Smallholder Farmers: Progress of Phase 1. PO Box 39063, Nairobi, Kenya. International Crops Research Flint-Garcia, S.A., Thornsberry, J.M. & Buckler, E.S. (2003) Structure of linkage disequilibrium in plants. Annual Review of Plant Biologyiology [online], 54: 357–74. Available from: http://www.ncbi.nlm.nih.gov/pubmed/14502995 [Accessed 12 September 2014] Foster-Hartnett, D., Mudge, J., Larsen, D., Danesh, D. & Yan, H.H. (2002) Comparative genomic analysis of sequences sampled from a small region on soybean (Glycine max) molecular linkage group G . Genome, 45: 634–645 Franckowiak, J.D. (1973) ―Importance of seed coat characteristics inheritance in cowpea improvement. Proc.‖ In First IITA Grain Legume Improvement Workshop 1973,IITA, Ibadan. 5pp. 1973. p. 5 University of Ghana http://ugspace.ug.edu.gh 137 Gajera, H.P., Domadiya, R.K., Patel, S.V. & Golakiya, B.A. (2014) Appraisal of RAPD and ISSR Markers for Genetic Diversity Analysis among Cowpea ( Vigna unguiculata L .) Genotypes. Journal of Crop Science and Biotechnology, 17 (2): 79–88 Gamble, E.E. (1962) Gene effects in corn (Zea mays L.). Separation and relative importance of gene effects for yield. Canadian Journal of Plant Science., 42: 339 –348 Geldermann, H. (1975) Investigations on inheritance in quantitative characters in animals by gene markers methods. Theoretical and Applied Genetics, 46: 319–330 Gelin, J.R., Forster, S., Grafton, K.F., McClean, P., & Rojas-Cifuentes, G.A. (2007) Analysis of seed-zinc and other nutrients in a recombinant inbred population of navy bean (Phaseolus vulgaris L.). Crop Science, 47: 1361 – 1366 Goenaga, R., Quiles, A., & Glilaspe, A.G. (2008) Assessing Yield Potential of Cowpea Genotypes Grown Under Virus Pressure. Horticultural Science, 43 (3): 673–676 Gómez-Galera, S., Rojas, E., Sudhakar, D., Zhu, C., Pelacho, A. M., & Capell, T. (2010) Critical evaluation of strategies for mineral fortification of staple food crops. Transgenic Research, 19: 165 – 180 Graham, R.D., Senadhira, D., Beebe, S. E., Iglesias, C. & Ortiz-Monasterio, I. (1999) Breeding for micronutrient density inedible portions of staple food crops: Conventional approaches. Field Crops Research, 60: 57 – 80. Gregorio, G. B., Sendhira, D., Htut, H. & Graham, R.D. (2000) Breeding for trace mineral density in rice . Food and Nutrition Bulletin, 21: 382–386 Grusak, M. A.& DellaPenna, D. (1999) Improving the nutrient composition of plants to enhance human nutrition and health. Annual Review of Plant Physiology and Plant Molecular Biology, 50: 133 – 161 Gupta, P., Singh, R., Malhotra, S., Boora, K. S. & Singal, H.R. (2010) Characterization of seed storage proteins in high protein genotypes of cowpea [ Vigna unguiculata ( L .) Walp .]. Physiology and Molecular Biology of Plants, 16 (1): 53–58 Gupta, P.K., Rustgi, S. & Kumar, N. (2006) Genetic and molecular basis of grain size and grain number and its relevance to grain productivity in higher plants. Genome / National Research Council Canada = Genome / Conseil national de recherches Canada, 49: 565– 571 Gusmini, G. and Wehner, T.C., & Donaghy, S.B. (2007) SASQuant: a SAS software program to estimate genetic effects and heritabilities of quantitative traits in populations consisting of 6 related generations. The Journal of heredity [online], 98 (4): 345–50. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17602177 [Accessed 30 October 2014] University of Ghana http://ugspace.ug.edu.gh 138 Hayman, B.I. (1958) The separation of epistatic from additive and dominance variation in generation means. Heredity, 12: 371–390 Horax, R., Hettiarachchy, N.S., Chen, P. & Jalaluddin, M. (2004) Preparation and characterization of protein isolate from cow pea ( Vigna unguiculata (L.) walp). Journal of food science, 69: 114–118 Hu, Z. & Xu, S. (2009) PROC QTL-A SAS Procedure for Mapping Quantitative Trait Loci. International Journal of Plant Geenomics. Hindawi Publishing., 2009: 1–3 Hussain, J., Riaz, U., Rehman, N., Khan, A., Muhammad, Z., Khan, F., Hussain, S.T. & Anwar, S. (2010) Endogenous transitional metal and proximate analysis of selected medicinal plants from Pakistan. Journal of Medicinal Plants Research, 4 (3): 267–270 Hussain, M. A.& Basahy, A.Y. (1998) Nutrient composition and amino acid pattern of cowpea Vigna unguiculata (L.) Walp, Fabaceae) grown in Gizan area of Saudi Arabia. International Journal of Food Sciences and Nutrition, 49 (2): 117–124 Hussain, M. A., Akinyele, I. O., Omololu, A. & Akinlosotu, A. (1984) Medical problems associated with the consumption of cowpeas: Perception of Nigerian women. Paper presented at the World Cowpea Research Conference, IITA, Ibadan. Nov. 4-9, 1984 . Medical problems associated with the consumption of cowpeas: Perception of N. In 1984. p. 12 Huynh, B., Close, T. J., Roberts, P. A., Hu, Z., Wanamaker, S., Lucas, M. R., Chiulele, R., Cissé, N., David, A., Hearne, S., Fatokun, C., Diop, N. N. & Ehlers, J.D. (2013) Gene Pools and the Genetic Architecture of Domesticated Cowpea. The Plant Genome [online], 6 (3). Available from: https://www.crops.org/publications/tpg/abstracts/6/3/plantgenome2013.03.0005 [Accessed 24 October 2014] IBPGR (1983) Cowpea Descriptors, International Board for Plant Genetic Resources. pp. 1– 30 IITA (1983) International Institutes of Tropical Agriculture Annual Report. Inaizumi, H., Singh, B. B., Sanginga, P. C., Manyong, V. M. & Adesina, A.A. (1999) Adoption and impact of dry-season dual-purpose cowpea in the semiarid zone of Nigeria [online]. Available from: http://books.google.com/books?hl=en&lr=&id=y8C4AIoRaWoC&oi=fnd&pg=PP2&dq=Ad option+and+impact+of+dry-season+dual- purpose+cowpea+in+the+semiarid+zone+of+Nigeria&ots=WjSNJe14TK&sig=PIOCw3k4 M9tgqcKics18NArbwMA University of Ghana http://ugspace.ug.edu.gh 139 Ishiyaku, M.F. (2013) Agricultural Biotechnology-from laboratory to field: the Cowpea Experience. Paper presented at the Open Forum for Agricultural Biotechnology (OFAB), National Assembly, Abuja, Nigeria. 22 October, 2013 Islam, F.M., Beebe, S. Munoz, M., Tohme, J., Redden, R. J. & Basford, K.E. (2004) Using molecular markers to assess the eVect of introgres- sion on quantitative attributes of common bean in the Andean gene pool. Theoretical and Applied Genetics, 108: 243–252 Joseph, O. A., Henriëtte, L. K. & Taylor, J.R.N. (2011) Evaluation of the functional quality of cowpea-fortified traditional African sorghum foods using instrumental and descriptive sensory analysis. LWT - Food Science and Technology [online], 44 (10): 2126–2133. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0023643811002118 [Accessed 10 October 2014] Joshi, A. K., Crossa, J., Arun, B., Chand, R., Trethowan, R., Vargas, M. & Ortiz-Monasterio, I. (2010) Genotype×environment interaction for zinc and iron concentration of wheat grain in eastern Gangetic plains of India. Field Crops Research [online], 116 (3): 268–277. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0378429010000067 [Accessed 27 September 2014] Kachare, D. P., J. K. Chavan, & Kadam, S.S. (1988) Nutritional quality of some improved cultivars of cowpea. Plant Foods for Human Nutrition, 38 (2:): 155–162 Kang, M.S. & Priyadarshan, P.M. (2007) Breeding major food staples. Blackwel publishingl Kearsey, M. J., & Pooni, H.S.J. (1996) The genetical Analysis of quantitative traits. Chapman and Hall, London Khairwal, I.S., Yadav, S.K., Upadhyaya, H.D., Kachhawa, D. and Nirwan, B., Bhattacharjee, R. & Rajpurohit, B.S. (2007) Evaluation and identification of promising pearl millet germplasm for grain and fodder traits. Science Agriculture and Technology, e-journal. Open Access Journal published by ICRISAT., 5 (1): 1–6 Khodambashi, M., Bitaraf, N. and Hoshmand, S. (2012) Generation Mean Analysis for Grain Yield and Its Related Traits in Lentil. Journal of Agricultural Science and Technology, 14: 609–616 Kiani, S. H., Kazemitabar, S.K., Jelodar, N.A.B. & Ranjbar, G.A. (2013) Genetic Evaluation of Quantitative Traits of Rice ( Oryza sativa L .) Using Generation Mean Analysis. International Journal of Agriculture and Crop Sciences, 5 (19): 2329–2336 Kormawa, P. M., Chiunu, J. N. & Munyong, V.M. (2000) ―Cowpea Demand and Supply Pattern in West Africa:‖ In The case of Nigeria, Proceedings of the World Cowpea Conference III held at the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria. 2000. p. 504 University of Ghana http://ugspace.ug.edu.gh 140 Kormawa, P. M., Ezedinma, C. I. & Singh, B.B. (2004) Factors influencing farmer-to-farmer transfer of an improved cowpea variety in Kano State, Nigeria. Journal of Agriculture and Rural Development in the Tropics and Subtropics, 105 (1): 1–13 Kouam, E.B., Pasquet, R.S., Campagne, P., Tignegre, J.-B., Thoen, K., Gaudin, R., Ouedraogo, J.T., Salifu, A.B. and Muluvi, G.M. & Gepts, P. (2012) Genetic structure and mating system of wild cowpea populations in West Africa Genetic structure and mating system of wild cowpea populations in West Africa. BMC plant Biology, 12 (113): 1–13 Kumhar, B. L., Singh, D., Bhanushally, T. B. & Koli, N.R. (2013) Gene Effects for Yield and Yield Components in Chickpea (Cicer arietinum L.) under Irrigated and Rainfed Conditions. Journal of Agricultural Science [online], 5 (3): 1–13. Available from: http://www.ccsenet.org/journal/index.php/jas/article/view/20101 [Accessed 8 November 2014] Lander, E. & Botstein, D. (1989) Mapping Mendelian factors under- lying quantitative traits using RFLP linkage maps. Genetics, 121: 185 – 199. Langyintuo A.S., Lowenberg-Deboer J., Faye D., Lambert D., Ibro G., Moussa B., Kergna A., Kushwaha S., Musa S., & Ntoukam, G. (2003) Cowpea supply and demand in West and Central Africa. Field Crops Research, 82: 215 – 231 Langyintuo, A. S., Ntoukuam, G., Murdock, L., Lowenberg-Deboer, J. & Miller, D.J. (2004) Consumer preferences for cowpea in Cameroon and Ghana. Agricultural Economics [online], 30 (3): 203–213. Available from: http://doi.wiley.com/10.1016/j.agecon.2002.12.001 [Accessed 12 November 2014] Li , L.T., Chen, O. S., Ward, & D. M., Kaplan, J. (2001) CCC1 is a transporter that mediates vacuolar iron storage in yeast. Journal of Biological Chemistry, 276: 29515–29519. Lie-Zhao, L., Meng, J. and Lin, N., Chen, L., Tang, Z., Zhang, X. & Li, J. (2006) QTL Mapping of Seed Coat Color for Yellow Seeded Brassica napus. Acta Genetica Sinica, 33 (February): 181–187 Liu, J. & Muse, S. V. (2005) PowerMarker: Integrated analysis environment for genetic marker data. Bioinformatics, 2 (9): 2128–2129 Lopes, F.C., Gomes, R.L., Rodrigues, F. & Filho, F. (2003) Genetic control of cowpea seed sizes. Scientia Agricola [online], 60 (2): 315–318. Available from: http://www.scielo.br/scielo.php?pid=S0103- 90162003000200016&script=sci_arttext&tlng=pt [Accessed 4 November 2014] Lucas M.R., Diop N. N., Wanamaker, S., Ehlers J.D., Roberts, P.A. & Close, T.J. (2011) Cowpea Soy bean synteny clarified through an improved genetic map. Plant Genome, 4: 218–225 University of Ghana http://ugspace.ug.edu.gh 141 Lush, W.M. (1979) Floral morphology of wild and cultivated cowpea (Vigna unguiculata). Economic Botany, 33 (4): 442 Madode, Y.E., Houssou, P.A., Linnemann, A.R. and Nout, M .J.R., Hounhouigan, D.J., & Van- Boekel, M.A. (2011) Ecology of Food and Nutrition Preparation , Consumption , and Nutritional Composition of West African Cowpea Dishes. Ecology of Food and Nutrition, 50 (2011): 115–136 Mahalakshmi, V., Ng Q., Lawson, M. & Ortiz, R. (2007) Cowpea [Vigna unguiculata (L.) Walp.] core collection defined by geographical, agronomical and botanical descriptors. Plant Genetic Resources, 5: 113–119 Maiangwa, M.G. & Ogungbile, A.O. (2008) Seed varieties and Quantities Of Chemical Fertilizers used in Agricultural Production In Northern Nigeria. Global Journal of Agricultural Sciences. 7 Makindara, R., Hella, J.P., Erbaugh, J.M. & Larson, D.W. (2013) Consumer preferences and market potential for sorghum based clear beer in Tanzania. Journal of Brewing and Distilling [online], 4 (1): 1–10. Available from: http://academicjournals.org/journal/JBD/article-abstract/AD5438E258 [Accessed 3 December 2014] Mamiro, P. (2011) Nutritional quality and utilization of local and improved cowpea varieties in some region of Tanzania. African Journal of food Agriculture Nutrition and Development, 11 (1): 4490–406 Mather, K. & Jinks, J.L. (1982) Biometrical genetics . London.: Chapman and Hall. Mazzolini, A. P., Pallaghy, C. K. & Legge, G.J.F. (1985) Quantitative microanalysis of Mn, Zn and other elements in mature wheat seed. New Phytologist 100, 483 – 509. New phythologist, 100: 483–509 Moalafi, A. I., Asiwe, J. A. N. & Funnah, S.M. (2010) Germplasm evaluation and enhancement for the development of cowpea ( Vigna unguiculata ( L .) Walp dual-purpose F 2 genotypes. African Journal of Agricultural Research, 5 (7): 573–579 Moose, S.P. & Mumm, R.H. (2008) Molecular plant breeding as the foundation for 21st century crop improvement. Plant Physiology, 147: 969 – 977. Moraghan, J.T. & Grafton, K. (2001) Genetic diversity and mineral composition of common bean seed. Journal of the Science of Food and Agriculture, 81: 404 – 408 Moura, J.D.O., Rocha, M.D.M., Lúcia, R., et al. (2012) Path analysis of iron and zinc contents and others traits in cowpea. Crop Breeding and Applied Biotechnology, 12: 245–252 University of Ghana http://ugspace.ug.edu.gh 142 Muchero, W., Diop, N. N., Bhat, P. R., Fenton, R. D., Wanamaker, S., Pottorff, M., Hearne, S. and Ndiaga, C., Fatokun, C., Ehlers, J. D., Roberts, P. A. & Close, T.J. (2009) A consensus genetic map of cowpea [Vigna unguiculata (L) Walp.] and synteny based on EST-derived SNPs. Proceedings of the National Academy of Sciences of the United States of America [online], 106 (43): 18159–18164. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2761239&tool=pmcentrez&rend ertype=abstract Muchero, W., Ehlers, J. D., Close, T. J. & Roberts, P.A. (2009) Mapping QTL for drought stress- induced premature senescence and maturity in cowpea [Vigna unguiculata (L.) Walp.]. TAG. Theoretical and applied genetics. Theoretische und angewandte Genetik [online], 118 (5): 849–63. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19130034 [Accessed 24 October 2014] Mulamba, N. N., & Mock, J.J. (1978) Improvement of yield potential of the Eto Blanco maize ( Zea mays L.) population by breeding for plant traits. Egyptian Journal of Genetic and Cytology, 7: 40–51 Muranty, H. (1996) Power to test for quantitative trait loci detection using full sib families in different schemes. Heredity, 76 (1996): 156–165 Mustapha, Y. (2009) Inheritance of seed coat colour pattern in cowpea [ Vigna unguiculata (l.) Walp]. Bayero Journal of Pure and Applied Sciences, 2 (2): 70–74 National Population Commission (NPC) [Nigeria] and ICF Macro (2009) National Demographic and Health Survey. Nigerian Demographc Health Survey 2008, pp. 187–196 Negri, V., Floridi, S. & Montanari, L. (2001) Organoleptic and chemical evaluation of Italian cowpea (Vigna unguiculata subsp. unguiculata cv gr. unguiculata (L) Walp.) landraces from a restricted area. Italian Journal of Food Science, 13: 383–390 Nei, M., Tajima, F., & Tateno, Y. (1983) Accuracy of estimated phylogenetic trees from molecular data. Journal of Molecular Evolution, 19: 153–170 Nestel, P., Bouis, H. E., Meenakshi, J. V. & Pfeiffer, W. (2006) Biofortification of staple food crops. Journal of Nutrition, 136: 1064–1067 Ng, N.Q. & Marechal, R. (1985) Cowpea taxonomy, origin and germplasm. In: S. R. Singh and K.O. Rachie (ed), Cowpea research, production and utilization. Chichester, UK,: John Wiley and Sons, Ng, Z-X., Chai, J-W. & Kuppusamy, U.R. (2011) Customized cooking method improves total antioxidant activity in selected vegetables. International Journal of Food Sciences and Nutrition [online], 62 (2): 158–63. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21250903 [Accessed 30 November 2014] University of Ghana http://ugspace.ug.edu.gh 143 Obiegbuna, J. E., Ameh, A. & Morah, G.N. (2006) A comparative study of the effects of steaming and rehydration on the acceptability of akara from cowpea and African yambean meals. Journal of Sustainable Agriculture and the Environment, 8: 65–72 Ojiako, F. O. & Kayode, R.M.O. (2014) Nutritional quality and consumer acceptability of cowpea seeds (Vigna unguiculata L. Walp) treated with natural and synthetic insecticides against bruchid infestation. Journal of Stored Products Research, 56: 9–15 Ojomo, O.A. & Cheda, H.R. (1972) Physicochemical properties of cowpea (V. unguiculata) influencing varietal differences in culinary values. Journal of West African Sciences Association 17(1): 3. Journal of West African Sciences Association, 17 (1): 3 Oluwatosin, O.B. (1998) Genetic and Environmental Variabilit y in Starch , Fatt y Acids and Mineral Nutrients Composition in Cowpea ( Vigna unguiculata ( L ) Walp ). Journal of Science Food and Agriculture, 78: 1–11 Oluwatosin, O.B. (2000) Inheritance of genes controlling seed coat color in cowpea (Vigna unguiculata [L] Walp). Journal of Genetics and Breeding, 54: 259–262 Omueti, O. & Singh, B.B. (1987) Nutritional attributes of improved varieties of cowpea (Vigna unguiculata[L.] Walp.). Food Science and Nutrition, 41F: 103–112 Onyenekwe, P. C., Njoku, G. C. & Ameh, D.A. (2000) Effect of cowpea (Vigna unguiculata) processing methods on flatus causing oligosaccharides. Nutrition Research, 20: 349–358 Orjuela, J., Garavito, A., Bouniol, M., Arbelaez, J., Moreno, L., Kimball, J., Wilson, G., Rami, J- F., Tohme, J., McCouch, S. & Lorieux, M. (2010) A universal core genetic map for rice. Theoretical and Applied Genetics , 120, 563-572. A universal core g. Theoretical and Applied Genetics., 120: 563–572 Ortiz, O., Frias, G., Ho, R., Cisneros, H., Nelson, R., Castillo, R., Orrego, R., Pradel, W., Alcazar, J., & Bazán, M. (2008) ―Organizational learning through participatory research: CIP and Care in Peru‖,. Agriculture and Human Values, 25: 419–431 Ouedraogo, J.T., Ouedraogo, M., Gowda, B.S., et al. (2012) Development of sequence characterized amplified region (SCAR) markers linked to race-specific resistance to Striga gesnerioides in cowpea (Vigna unguiculata L.). African Journal of Biotechnology [online], 11 (62): 12555–12562. Available from: http://www.academicjournals.org/ajb/abstracts/abs2012/2Aug/Ouedraogo et al.htm [Accessed 17 October 2014] Ouédraogo, J. T., Gowda, B. S., Jean, M., Close, T. J., Ehlers, J. D., Hall, A. E., Gillaspie, A. G., Roberts, P. A., Ismail, A. M., Bruening, G., Gepts, P. & Timko, M.P. and Belzile, F.J. (2002) An improved genetic linkage map for cowpea ( Vigna unguiculata L .) Combining AFLP , RFLP , RAPD , biochemical markers , and biological resistance traits. Genome, 45: 175–188 University of Ghana http://ugspace.ug.edu.gh 144 Ozturk, L., Yazici, M. A., Yucel, C., Torun, A., Cekic, C., Bagci, A., Ozkan, H., Braun, H.J., Sayers, Z. & Cakmak, I. (2006) Concentration and localization of zinc during seed development and germination in wheat. Physiologia Plantarum, 128: 144–152 Padi, F.K. (2007) Early generation selection for high yielding cowpea genotypes in additive series intercropping systems with sorghum. Annals of Applied Biology [online], 151 (3): 391– 400. Available from: http://doi.wiley.com/10.1111/j.1744-7348.2007.00179.x [Accessed 13 October 2014] Padulosi, S. & Ng, N.Q. (1997) ―Origin, taxonomy, and morphology of Vigna unguiculata (L.) Walp.‖ In In: Singh BB,Mohan Raj DR, Dashiell KE, Jackai LEN (eds) Advances in Cowpea Research. Copublication International Institute of Tropical Agriculture (IITA) and Japan International Research Center Agric Sci (JIRCAS). Sayce, Devon, UK,. pp. 1–12 Pearson, D. (1976) The Chemical Analysis of Foods, 7th ed. Churchill Living Stone, London. p. 103. Pereira, E. J., Carvalho, L.M.J., Cardoso, S.N., Dellamora-ortiz, G.M., Carvalho, J.L.V., Viana, D.S., Freitas, S.C. & Rocha, M.M. (2014) Effects of cooking methods on the iron and zinc contents in cowpea (Vigna unguiculata) to combat nutritional deficiencies in Brazil. Food and Nutrition Research, 58 (20694): 1–7 Poersch, N.L., Ribeiro, N.D., Rosa, D.P., et al. (2013) Genetics of the concentration of copper in common bean seeds - doi: 10.4025/actasciagron.v35i3.16520. Acta Scientiarum. Agronomy [online], 35: (3): 301–306. Available from: http://periodicos.uem.br/ojs/index.php/ActaSciAgron/article/view/16520 [Accessed 28 November 2014] Porter, P.M., D.R. Huggins, C.A. Perillo, S.R. Quiring, & Crookston, R.K. (2003) Organic and other management strategies with two- and four-year crop rotations in Minnesota. Agronomy Journal, 95: 233–244 Rangaiah, S., Nehru, S. D. & Mahadevu, P. (1999) Genetic studies in two cross derivatives of cowpea (Vigna unguiculata (L.) Walp.). Mysore Journal of Agricultural Science. Mysore Journal of Agricultural Science, 33: 125 – 129 Rangel, A., Saraiva, K., Schwengber, P., Narciso, M., Domont, G., Ferreira, S. & Pedrosa, C. (2004) Biological evaluation of a protein isolate from cowpea (Vigna unguiculata) seeds. Food Chemistry [online], 87 (4): 491–499. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0308814604000226 [Accessed 30 September 2014] Rawat, N., Tiwari, V. K., Singh, N., Randhawa, G.S., Singh, K., Chhuneja, P. & Dhaliwal, H.S. (2009) Evaluation and utilization of Aegilops and wild Triticum species for enhancing iron and zinc content in wheat. Genetic Resources and Crop Evolution, 56: 53–64 University of Ghana http://ugspace.ug.edu.gh 145 Ribaut, J-M., de Vicente, M.C. & Delannay, X. (2010) Molecular breeding in developing countries: challenges and perspectives. Current Opinion in Plant Biology [online], 13 (2): 213–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20106715 [Accessed 13 October 2014] Ribeiro, D. N., N.D., Maziero, S.M., Prigol, M., et al. (2012) Mineral concentrations in the embryo and seed coat of common bean cultivars. Journal of Food Composition and Analysis [online], 26 (1-2): 89–95. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0889157512000440 [Accessed 28 October 2014] Ribeiro, N.D., Jost, E., Maziero, S.M., Storck, L. & Rosa, D.P. (2014) Selection of common bean lines with high grain yield and high grain calcium and iron concentrations. Rev. Ceres, Viçosa, v. 61, n.1, p. 077-083, jan/fev, 2014, pp. 77–83 Romanus, K.G., Hussein, S. & Mashela, W.P. (2007) Combining ability analysis and association of yield and yield components among selected cowpea lines. Euphytica [online], 162 (2): 205–210. Available from: http://link.springer.com/10.1007/s10681-007-9512-5 [Accessed 13 October 2014] Rosa, S. S., Ribeiro, N. D., Jost, E., Reiniger, L. R. S., Rosa, D. P. & Cerutti, T. (2010) Potential for increasing the zinc content in common bean using genetic improvement. Euphytica, 175 (2): 207–213 Salako, F. K., Tian, G. & Kang, B.T. (2002) Indices of root and canopy growth of leguminous cover crops in the savanna zone of Nigeria. Tropical Grasslands, 36: 33–46 Samineni, S., Gaur, P.M., Colmer, T.D., et al. (2011) Estimation of genetic components of variation for salt tolerance in chickpea using the generation mean analysis. Euphytica [online], 182 (1): 73–86. Available from: http://link.springer.com/10.1007/s10681-011-0496- 9 [Accessed 2 December 2014] Sathe, S. K. & Salunkhe, D.K. (1981) Functional Properties of Great Northern bean (Phaseolus vulgaris), Proteins, Emulsion, Foaming, Viscosity and Gelation Properties J. Food Sci 46: 71-76. Journal of food Science, 46: 71 – 76 Saunders, A.R. (1959) Inheritance in the cowpea ( Vigna sinensis Endb.). I: colour of the seed coat. South African Journal of Agriculture, 3: 141–162 Sax, K. (1923) The Association of Size differences with seed-coat pattern and pigmentation in Phaseolus vulgaris. Genetics, 8: 552–560 Sefa-Dedeh, S., Stanley, D.W. & Voisey, P.W. (1979) Effect of storage time and conditions on the hard to cook defects in cowpeas ( Vigna unguiculata ). Journal of Food Science, 44: 79 Siegel, S. & Castellan-Jr., N.J. (1988) Nonparametric Statistics for the Behavioral Sciences, International Edition. Internatio. New York: McGraw-Hill Book Company University of Ghana http://ugspace.ug.edu.gh 146 Siemuri, E.O., Akintunde, J.K. & Ajiboye, A.E. (2011) Acid-hydrolysed local Cowpea ( Vigna unguiculata ) as culture media for Laboratory Yeast ( Saccaromyces cerevisiae ). International Research Journal of Microbiology, 2 (July): 187–191 Singh, B.B. & Ishiyaku, M.F. (2000) Genetics of rough seed coat texture in cowpea. The Journal of heredity, 91 (2): 170–174 Singh, B.B. & Mligo, J.A. (2007) Registration of ‗Vuli-1‘ cowpea.pdf. Crop Science, 47: 473– 438 Singh, B.B., Ehlers, J.D., Sharma, B., Freire Filho, F.R. (1997) ―Recent progress in cowpea breeding.‖ In Challenges and opportunities for enhancing sustainable cowpea production. C.A. Fatokun, S.A. Tarawali, B.B. Singh, P.M. Kormawa, and M.Tamo (eds). Publication of International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria. pp. 22–24 Sivaprakash, K.R., Prasanth, S.R., Mohanty, B.P. & Parida, A. (2004) Genetic diversity of black gram andraces as evaluated by AFLP markers. Current Science, 86: 1411–1415 Slattery, H.D., Atwell, B.J. & Kuof, J. (1982) Relationship between Colour , Phenolic Content and Impermeability in the Seed Coat of various Trifolium subterraneum L genotypes. Annals of Botany, 50: 373–378 Spillman, W.J. (1912) Genetics of vigna. Horticulture Review., 2: 311–394 Spillman, W. J. & Sando, W.J. (1930) Mendelian factors in cowpea Vigna species. Michigan Academy of Science, Arts and Letters Papers., 11: 249–283 Srivastava, P. S., Narula, A. & Srivastava, S. (2005) Plant Biotechnology and Molecular Markers. New Delhi: Anamaya Publishers Steele, W.M. (1976) Cowpea, Vigna unguiculata (Leguminosae-Papillionatae). In: Simmon. London: Longman, Tabe, L., Hagan, N. & Higgins, T.J.V. (2000) Plasticity of seed protein composition in response to nitrogen and sulphur availability. Current Opinion in Plant Biology, 5: 212 – 217 Takeshima, H., Nagarajan, L., Salau, S.&, et al. (2014) Farmers ’ Preferences on Seed Purchase Timing - Rice , Cowpea , and Maize Growers in Nigeria, Policy Note No . 31. IFRI, Nigeria Tan, H., Manman, T., Qian, L. Yongpeng, Z. J. & Huanxiu, L. (2012) A Review of Molecular Makers Applied in Cowpea ( Vigna unguiculata). Journal of life sciences, 6 (2012:): 1190– 1199 University of Ghana http://ugspace.ug.edu.gh 147 Tanksley, S.D., Ganal, M.W., Price, J.P., de Vincente, M.C., Bonicrbale, M.W., Broun, P., Fulton, T.M., Giovannoni, H., Grandillo, S., Martin, G.B., Messeguer, R., Miller, J.C., Miller, L., Paterson, A.H., Pineda, O., Röder, M., Wing, R.A., Wu, W. & Young, N.D. (1992). (1992) High density molecular linkage maps of tomato and potato genomes; biological inferences and practical application. Genetics, 132: 1141–1160 Tauris, B., Borg, S., Gregersen, P.L., et al. (2009) A roadmap for zinc trafficking in the developing barley grain based on laser capture microdissection and gene expression profiling. Journal of experimental botany [online], 60 (4): 1333–47. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2657541&tool=pmcentrez&rend ertype=abstract [Accessed 13 October 2014] Tchiagam, J. N., Bell, J. M., Ngakeu, D. F., Njintang, N. Y. & Youmbi, E. (2011a) Diallel analysis of cowpea (Vigna unguiculata (L.) Walp.) for some physical properties of seed under the Sudano-guinean conditions. Agriculture and Biology Journal of North America [online], 2 (4): 698–707. Available from: http://www.scihub.org/ABJNA/PDF/2011/4/ABJNA-2-4-698-707.pdf [Accessed 13 October 2014] Tchiagam, J. N., Bell, J. M., Ngakeu, D. F., Njintang, N. Y. & Youmbi, E. (2011b) Genetic analysis of seed proteins contents in cowpea ( Vigna unguiculata L . Walp .). African Journal of Biotechnology, 10 (16): 3077–3086 Thomson, R.A. (2013) Genetic Analysis of postharvest physiological deterioration in cassava ( Manihot esculenta Crantz) storage roots. PhD. Thesis submitted to the University of Ghana, Legon in partial fulfillment of the requirement for the award of Doctor of philosophy degre Timko, M.P. & Singh, B.B. (2008) Cowpea , a Multifunctional Legume. Genomics of Tropical Crop Plants. P.H. Moore, R.M. (ed.). Springer Tosti, N. & Negri, V. (2002) Efficiency of three PCR-based markers in assessing genetic variation among cowpea (Vigna unguiculata ssp. unguiculata) landraces. Genome, 45: 656 – 660 Upadhyaya, H.D., Yadav, D., Dronavalli, N., Gowda, C.L.L. & Singh, S. (2010) Mini core germplasm collections for infusing genetic diversity in plant breeding programs. Electronic Journal of Plant Breeding, 1 (4:): 1294–1309 USAID (2006) Regional Partnerships to enhance Bean and Cowpea consumption and production in Africa and Latin America. 2006 Research and Training Highlights of the Bean / Cowpea Collaborative Research Support Program. USAID- Nigeria (2008) A study of the cowpea value chain in Kano state, Nigeria, from a pro- poor and gender perspective University of Ghana http://ugspace.ug.edu.gh 148 Varaprasad, K. S. & Sivaraj, N. (2010) Plant genetic resources conservation and use in light of recent policy developments. Electronic Journal of Plant Breeding, 1 (July): 1276–1293 Vasconcelos, I.M., Maia, F.M.M., Farias, D.F., et al. (2010) Protein fractions, amino acid composition and antinutritional constituents of high-yielding cowpea cultivars. Journal of Food Composition and Analysis [online], 23 (1): 54–60. Available from: http://linkinghub.elsevier.com/retrieve/pii/S088915750900163X [Accessed 13 October 2014] Walkley, A. & Black, I.A. (1934) An Examination of Degtjareff Method for Determining Soil Organic Matter and a Proposed Modification of the Chromic Acid Titration Method., 37: 29–37 Wallace, B. (1968) Further data (1963) on the over-dominance of induced mutations. Genetics, 48: 633 – 651 Wang, G., McGiffen-Jr, M.E., Ehlers, J.D. & Marchi, E.C.S. (2006) Competitive ability of cowpea genotypes with different growth habit. Weed Science, 54: 775–782 Wang, N. & Daun, J.K.. (2004) Effect of variety and crude protein content on nutrients and certain antinutrients in field peas (Pisum sativum). Journal of Sciences Food and Agriculture., 84: 1021– 1029. Wang, N., Lewis, M.J., Brennan, J.G. & Westby, A. (1997) Effect of processing methods on nutrients and anti-nutritional factors in cowpea. Food Chemistry, 58: 59–68 Welch, R.M. & Graham, R.D. (1999) A new paradigm for world agriculture : meeting human needs Productive , sustainable , nutritious. Field Crops Research, 60: 1–10 Welch, R.M. & Graham, R.D. (2004) Breeding for micronutrients in staple food crops from a human nutrition perspective. Journal of experimental botany [online], 55 (396): 353–64. Available from: http://www.ncbi.nlm.nih.gov/pubmed/14739261 [Accessed 4 September 2014] Weller, J.I. & Soller, M. (2004) An analytical formula to estimate confidence interval of QTL location with a saturated genetic map as a function of experimental design. Theoretical and Applied Genetics, 109: 1224–1229 Weltzien, R. E., Whitaker, M. L., Rattunde, W. H. F., Dhamotharan, M. & Anders, M.M. (1998) Participatory approaches in pearl millet breeding. In: making the most of new varieties for small farmers. Oxford and IBH publishing Whit W.C. (2007) ―Soul food as cultural creation, in: A. Bower (Ed.), African American foodways: explorations of history and culture, Urbana : University of Illinois Press. pp. 45- 58 Soul food as cultural creation, in: A. Bower (Ed.), African American foodways: University of Ghana http://ugspace.ug.edu.gh 149 exploration.‖ In African American foodways: explorations of history and culture, Urbana. pp. 45–58 White, P. J. & Broadley, M.R. (2005) Biofortifying crops with essential mineral elements. Trends in Plant Science, 10 (12:): 586–593 White, P.J. & Broadley, M.R. (2009) Biofortification of crops with seven mineral elements often lacking in. New Phythologist, 182: 49–84 Wright, J.N. (1952) System of mating. Genetics, 6: 111–178 Xu, P., Hu, T., Yang, Y., et al. (2011) Mapping Genes Governing Flower and Seedcoat Color in Asparagus Bean ( Vigna unguiculata ssp . sesquipedalis ) Based on Single Nucleotide Polymorphism and Simple Sequence Repeat Markers. HortSciencence, 46 (8): 1102–1104 Yeung, H., Ehlers, J.D., Waniska, R.D., et al. (2009) Rapid screening methods to evaluate cowpea cooking characteristics. Field Crops Research [online], 112 (2-3): 245–252. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0378429009000872 [Accessed 27 September 2014] Zhi-wen, L., Ting-dong, F., Jin-xing, T.&, et al. (2005) Inheritance of seed colour and identification of RAPD and AFLP markers linked to the seed colour gene in rapeseed (Brassica napus L.). TAG. Theoretical and applied genetics. Theoretische und angewandte Genetik [online], 110 (2): 303–10. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15565377 [Accessed 17 October 2014] University of Ghana http://ugspace.ug.edu.gh 150 APPENDICES Appendix 3.1: Checklist for Focus Group Discussion (FGD) guiding questions Section A: Meeting with Farmers/growers 1. How do you access agricultural information? (Be brief) 2. What are the social structures (relationships & membership of farmers to Researchers, Agricultural organization & Seed Companies)? 3. Identify the dominant crop grown in the area and why? 4. When does the cowpea grain is considered as the most important food crop in a year? (Calendar) 5. What are the problems that constrained cowpea production in the area? (Draw a table and allow farmers to rank in order of their preference) 6. What are the percentages of the area under cowpea cultivation, farmers that grow cowpea for food, feed or cash? (Draw a table) 7. Which of the varieties preferred by the consumers? Name & Seed size, colour & texture 8. Where do farmers get cowpea seed for planting? (Help to set research objectives) 9. How many varieties of cowpea do you grow? (Improved or local) 10. What attributes of cowpea do you prefer (early/late, high grain or fodder yield, Size, colour, texture, taste)? 11. Why do you cultivate cowpea; for food, feed or income? Which of the variety do you preferred most and why? University of Ghana http://ugspace.ug.edu.gh 151 12. Do you process cowpea? [If yes] What are the attributes do you like most and why? any special preference traits that you need in a new variety? 14. In what form do you market your cowpea? Grain, prepared products etc. Section B: Meeting with the Processors/ Consumers 15. Which of the cowpea attributes do you prefer most and why? Seed; sizes, colour, coat, texture, sweet or sweet less grain 16. Which of the cowpea consumable products (moin-moin, Akara etc) do you like best? and why? 17. How frequent do you consume your favourite cowpea products and of what quantity? 18. What difficulties do you encounter after eating any of the cowpea consumable products? Flatulence, stomach disorder etc 19. What are the constraints to cowpea processing? Time to imbibe water when soaking, hard-to- cook etc, 20. How does seed coat texture, colour and size of the cowpea affect coast of production, processing and marketing of the prepared products? Oil consumption, time to cook, milling difficulties etc 21. Which of the consumable products preferred most by the consumers? 22. What other importance of cowpea do you know apart from quenching hunger? Nutritional content of cowpea, present of micronutrients etc…… University of Ghana http://ugspace.ug.edu.gh 152 Appendix 3.2: Farmers‘ Perceptions and Preferences for Cowpea Grain Quality Traits Questionnaire 2012 This inquiry is undertaken by a Plant Breeding student to collect information. I assure you that the information to be provided will be recorded completely, accurately, and be treated as CONFIDENTIAL Section A: Farmer’s identification 1. State/savanna…………………………… 5. L.G.A./District……………………… 2. Extension area…………………………… 6. Village………………………………… 3. Farmer‘s Name……………………………. 7. Farmer‘s sex ………………… (Assign code 1 for male and code 2 for female) 4. Farmer‘s age………………………… (Assign code 1 for below 40 and code 2 for above 50 years) Section B: Constraint to production and farmers’ preferences 1. What is the main constraint to cowpea production in this area? (a) Striga (b) Drought (c) Insects (d) Diseases (e) others………………… 2. Nature of the major constraint to cowpea production in your area (a) Drought: (a) terminal (b) erratic rainfall (b) Major insect (a) Aphid (b) Thrips (c) Maruca (c) Major disease caused by: (a) Virus (b) Fungi (c) Nematode (d) Others……………………………………………………. 3. If drought is the main constraint, at which stage is becoming more devastating? (a) seedling (b) pre-flowering (c) during flowering (d) During grain filling 4. Do you have access to quality planting seed? (a) Yes (b) No University of Ghana http://ugspace.ug.edu.gh 153 5. Is the seed affordable, at the required time? (a) Yes (b) No 6. If yes, how do you get access to the seeds? (a) Agro-dealers (b) Government- subsidy (c) NGOs (d) Others……………………………… 7. Why do you cultivate cowpea? (a) Food (b) Feed (c)Income (d) All 8. Which percent do you sell? (a) below 50% (b) Above 50% 9. Which of the grain attributes do consumer pay premium price in your locality? (a) White & large seeded (b) White & small seeded (c) Brown & large seeded (d) Others specify……………………………………………….. 10. Do you consider end users‘ preferences when sourcing a variety? (a) Yes (b) No (c) sometimes 11. Which of the attributes of cowpea do you like most? (a) earliness (b) lateness (c) high grain yield (d) high fodder yield (e) a, c & d University of Ghana http://ugspace.ug.edu.gh 154 Appendix 3.3: Cowpea Processing and Consumption Constraints and Preferences for Grain Quality Traits This inquiry is undertaken by a Plant Breeding student to collect information. I assured you that the information to be provided will be recorded completely, accurately, and be treated as CONFIDENTIAL Section A: Processor/Consumer’s identification 1. State/savanna zone……………………….. 5. L.G.A./District……………………… 2. Extension area…………………………… 6. Village area ……………………… 3. Name……………………………. 7. Sex………………………………………….. Assign code 1 for male and code 2 for female) 4. Interviewee‘s age………………………….. (Assign 1 for 15-30yeras, 2 for 30-40yrs and 3 for 50 years and above) 12. Does seed coat colour determine your acceptability of cowpea?(a)Yes (b) No 13. If yes, which colour is the most acceptable by the consumers? (a) Red (b) White (c) Brown (d) others…………………………. 14. Does grain texture affect processing/ preparation time? (a) Yes (b) No 14. How does the texture affect preparation time? (a) Delays imbibitions when soaking (b) Difficulty in dehulling (c) Difficulty in milling (d) others specify…………............................................. 14. Which of the texture has short time to preparation? (a) Smooth (b) Rough 15. Do you store cowpea grain for a long time? (a) Yes (b) No 16. Which of the texture hasten hard-to-cook phenomenon when cowpea is stored for a long time? (a) Rough (b) smooth 17. Which of the grain colour do you prefer most? (a) White (b) Red (c) Brown (d) others specify…………………… University of Ghana http://ugspace.ug.edu.gh 155 18. Which of the cowpea consumable products do you like most? (a) Akara/kossai (b) Dumplings (c) Cowpea with rice (c) Moin moin (d) others…………………………………… 19. How frequent do you eat cowpea or cowpea products? Cowpea consumable products everyday 3 days in a week Once in a week Occasionally (> 2 week) Not at all cooked cowpea Cowpea dumpling (Danwake) Cowpea mixed with rice Moin moin Akara / Kossai Cowpea for soup Others specify……………………….. (Score in order of your preference 1=most, 2=moderately and 3=least preferred) 20. If you are not frequently eating cowpea/ cowpea products why? (a) Flatulence (b) stomach disorder (c) others specify…………………………………………. 21. How frequent do you feed your children (under age 5) with cowpea/ cowpea product? (a) Frequently (a) Not frequently (c) Not at all University of Ghana http://ugspace.ug.edu.gh 156 22. Do you encourage pregnant and nursing mothers to eat cowpea/ its products? (a) Yes (b) No 23. Which of the cowpea grain do you like must? (a) Sweet (b) Sweet less University of Ghana http://ugspace.ug.edu.gh 157 Appendix 4.1 Selection criteria for cowpea Parental lines (Mulamba and Mock, 1978) Accession Mat urity Grai n size Zinc Iron % prote in % Fibr e % CH O % F at Gro wth habit Sha tteri ng Grain colour Grain Text ure sum total score Remarks TVu-1 2 2 3 3 1 3 1 3 2 1 2 1 2.0 (24) S TVu-15225 2 2 3 2 1 3 2 3 3 1 2 2 2.2 (26) NS TVu-13088* 1 3 2 2 1 2 2 3 1 1 3 2 1.9 (23) NS TVu-16400 2 1 3 1 2 3 1 3 1 1 2 2 1.8 (22) S Babban wake 3 1 3 3 1 3 1 2 3 1 1 1 1.9 (23) NS TVu-999 2 2 3 2 1 3 2 3 2 1 3 3 2.3 (27) S TVu-8742 1 2 3 3 1 2 2 3 1 1 2 2 1.9 (23) S Sampea-13 1 2 3 3 1 2 2 2 1 1 1 1 1.7 (20) S TVu-14845 2 2 3 3 3 2 2 3 1 1 1 1 2.0 (24) S TVu-875 2 2 3 3 3 2 2 3 2 1 1 2 2.2 (26) NS TVu-997 2 2 3 1 1 3 2 3 3 1 2 2 2.1 (25) NS TVu-15251 2 2 3 2 1 1 2 3 3 1 1 1 1.8 (22) S Number in the bracket is the sum total, the lower the sum total score the better the cowpea lines; S= selected for making crosses; NS= not selected University of Ghana http://ugspace.ug.edu.gh 158 Scales: 1= High protein or Fe or Zn content or / early maturing or erect or white colour, 2= Medium protein or Fe or Zn content or shattering or brown or Semi erect and 3= Low protein or Fe or Zn content, prostrating, late maturing or small grained. a. Maturity period – 1= early (70-75days), 2= Medium (80-95 days) and 3= late maturing (above 100 days) b. Growth habit – 1- Erect, 2= Semi-erect and 3= prostrating type c. Shattering- 1= Normal, 2= Low shattering and 3= High shattering d. Grain size (100 grain weight g) - 1= Greater than 20g, 2= 15-19g and 3= less than 10g e. Protein content - 1= 21-29%, 2=10-20% and 3= less than 10% f. Zinc content (mg/kg)- 1= greater than 50, 2= 21-40 and 3= below 20 g. Iron content (mg/kg)- 1= greater than 100, 2= 50-99 and 3= less than 50 h. Fibre content (%)- 1= 3.1- 4.9%, 2= 2.1-3.0% and 3= 1.0-2.0% i. Carbohydrate- 1= 71-78%, 2= 60- 70% and 3= less than 59% j. Fat – 1= 11-22%, 2= 6-10% and 3= 1-5% k. Grain coat colour- 1= White, 2= Brown and 3= others l. Grain coat texture – 1= Rough and 2= Smooth University of Ghana http://ugspace.ug.edu.gh 159 Appendix 4.2: List of SNP markers used for characterizing 169 cowpea accessions S/N DNA \ Assay S/N DNA \ Assay S/N DNA \ Assay S/N DNA \ Assay S/N DNA \ Assay 1 10969_452 26 13707_697 51 16914_262 76 4339_822 101 6673_1242 2 4146_1588 27 13794_319 52 17023_955 77 4403_1123 102 6700_679 3 9673_1553 28 13849_2039 53 17450_1553 78 4462_114 103 7344_500 4 10115_384 29 13863_519 54 17588_963 79 4558_472 104 7438_464 5 10661_873 30 13873_544 55 1799_940 80 4563_661 105 7857_1368 6 10738_1400 31 13947_415 56 2185_132 81 4749_1972 106 7906_1032 7 10811_937 32 14030_764 57 2314_546 82 4778_497 107 7993_539 8 11367_1228 33 14164_1877 58 2339_52 83 483_1152 108 8011_481 9 11470_272 34 1426_521 59 234_249 84 4892_514 109 8118_1675 10 11622_232 35 14462_1712 60 2728_121 85 4904_278 110 8166_564 11 12119_480 36 14497_540 61 2974_1109 86 5058_372 111 8193_441 12 12122_559 37 14542_452 62 2997_519 87 5239_234 112 8306_119 13 12261_1773 38 14619_471 63 3098_224 88 5294_469 113 8605_2122 14 12393_305 39 14714_840 64 3485_771 89 534_355 114 8969_1386 15 12505_1312 40 14730_1034 65 361_520 90 5356_124 115 9134_1559 16 12703_553 41 14769_1746 66 3673_401 91 5428_339 116 9147_1655 17 12933_387 42 14929_258 67 3720_560 92 5435_569 117 9678_835 18 12959_58 43 14965_280 68 38_239 93 5448_461 118 9880_545 19 1296_808 44 15054_315 69 3838_830 94 5503_54 119 9955_544 20 13017_290 45 15183_436 70 3885_1019 95 5552_536 21 13034_542 46 15534_890 71 3900_562 96 5553_147 22 13207_784 47 15933_118 72 394_316 97 5692_1408 23 13506_333 48 16043_314 73 395_895 98 593_329 24 13563_863 49 16239_889 74 411_247 99 6065_457 25 13586_1058 50 16462_1286 75 4325_585 100 6247_659 University of Ghana http://ugspace.ug.edu.gh 1 University of Ghana http://ugspace.ug.edu.gh