Nepal J Biotechnol. 2 0 2 2 D e c ; 1 0 (2): 85-90 Research article DOI: https://doi.org/10.54796/njb.v10i2.242 ©NJB, BSN 85 Molecular Genetic Diversity in Bambara groundnut [Vigna subterranea (L.) Verdc] Assessed by Microsatellite Markers. Nwakuche Chinenye Onwubiko1 , Michael Ifeanyi Uguru2 , Grace Ovute Chimdi3 1Department of Crop Science and Technology, Federal University of Technology, Owerri 1526, Nigeria 2Department of Crop Science, University of Nigeria, Nsukka 410001, Nigeria 3Department of Agricultural Technology, Federal Polytechnic Bauchi, Bauchi 0231, Nigeria Received: 17 Mar 2022; Revised: 26 Nov 2022; Accepted: 03 Dec 2022; Published online: 31 Dec 2022 Abstract Bambara groundnut is a valuable leguminous crop with many landraces. A study was carried out to establish genetic diversity and phylogenetic relationship, among 33 Bambara groundnut accessions based on simple sequence repeat (SSR) markers. The nine microsatellite markers amplified a total of 27 alleles with a mean of 6.00 alleles per locus. Marker P 36 had the highest number of polymorphic bands while makers P131 and P68 were monomorphic. Genetic distance among the accessions based on Jaccard’s similarity coefficient ranged from 0.84 to 1.00. Cluster analysis resolved the accessions into five major groups with subgroups. Each group had a combination of distinct accessions from different geographical origin. A substantial level of intra-accession polymorphism was obtained among the evaluated collection of Bambara groundnut. The significant genetic diversity observed can support the selection of appropriate parental genotypes for the improvement of Bambara groundnut through various breeding programmes. Keywords: Accessions, Bambara groundnut, genetic diversity, microsatellite markers. Corresponding author, email: nwakuche.onwubiko@futo.edu.ng Introduction Variation exists in crops; it can be interspecific or intraspecific. Naturally, variation occur due to mutation and it is maintained over the years by processes of evolution and natural selection. The reality of natural variation in both wild and domesticated crops species is evident on existence of recognisable diverse forms of crop species, which is the basis for selection and for crop improvement. However not all recognisable variation in crops is genetic. Variations triggered by the influence of the environment occurs, in which differences observed in the expression of some characters among crop species were not intrinsic. On the other hand, variation in crops because of the action of gene(s) is genetic. Genes regulate structure (size, shape, colour), physiological processes and functions, adaptability, phenology, and expression of characters (1). Conventionally assessment of genetic variability in crops is achieved through field screening in which morphological descriptors are used to discriminate between and within Species. This approach is popular among workers because it is easy to carry out and is relatively cost-effective (2-4). However, it has some limitations because the expression of some characters (especially polygenic) is environment specific. Therefore, the obtained result may not be reliable. Molecular characterisation on the other hand is the most reliable and effective technique to establish variations that exist in crop species. It uses molecular descriptors or markers that are not affected by the environment to discriminate among species at DNA level and can be used to screen the entire population at any stage of crop development. The use of molecular markers to establish genetic diversity has enabled breeders to identify the presence of allelic variations in the genes controlling different agronomic characters in crops (5), to differentiate between homozygote and heterozygote genotypes (6), and to manipulate important agronomic traits in crops (7). Bambara groundnut is a leguminous food security crop. Its seed is a complete food and many forms of this crop has been reported by workers who made collections from major areas in west Africa eco-regions where this crop is grown (25). Unfortunately, the diversity reported by these workers where based on field morphological characterization of accessions. (8-11), There are few documented reports on assessment of diversity on Bambara groundnut based on molecular markers (12 13, 3, 14. 15), hence the need for this study whose primary target was to assess genetic diversity in Bambara groundnut based on microsatellite markers with a view to establishing phylogenetic relationship among the accessions. Nepal Journal of Biotechnology Publisher: Biotechnology Society of Nepal ISSN (Online): 2467-9313 Journal Homepage: www.nepjol.info/index.php/njb ISSN (Print): 2091-1130 https://orcid.org/0000-0003-3654-0861 mailto:nwakuche.onwubiko@futo.edu.ng mailto:nwakuche.onwubiko@futo.edu.ng Nepal J Biotechnol. 2 0 2 2 D e c ; 1 0 : 85-90 Onwubiko et al. ©NJB, BSN 86 Materials and method The plant materials were 33 accessions of Bambara groundnut obtained from the germplasm collection maintained at the gene bank of the International Institute of Tropical Agriculture (IITA), Ibadan (Table 1). DNA extraction Genomic DNA was extracted using the modified minipreparation protocol described by Dellaporta et al (16). Approximately 200 mg (0.2 gm) of lyophilized leaf sample was ground into fine powder with the aid of Genogrinder 2000. Table 1: Passport data of the 33 Bambara groundnut accessions used for the study. s/n Accession name Country of origin 1 TVSu-1483 Ghana 2 TVSu-1503 Nigeria 3 TVSu-1504 Nigeria 4 TVSu-1509 Nigeria 5 TVSu-1510 Nigeria 6 TVSu-1512 Nigeria 7 TVSu-1513 Nigeria 8 TVSu-1552 Nigeria 9 TVSu-1554 Nigeria 10 TVSu-1555 Nigeria 11 TVSu-1559 Nigeria 12 TVSu-1563 Nigeria 13 TVSu-1584 Nigeria 14 TVSu-1591 Togo 15 TVSu-1604 Togo 16 TVSu-1605 Togo 17 TVSu-1610 Togo 18 TVSu-1614 Togo 19 TVSu-1620 Togo 20 TVSu-1625 Togo 21 TVSu-1627 Togo 22 TVSu-1631 Togo 23 TVSu-1638 Mali 24 TVSu-1639 Mali 25 TVSu-1688 Togo 26 TVSu-1697 Togo 27 TVSu-1702 Togo 28 TVSu-1713 Zambia 29 TVSu-1766 Malawi 30 TVSu-1769 Malawi 31 TVSu-1788 Malawi 32 TVSu-1819 Cameroon 33 TVSu-1917 Cameroon To each tube 700 µL of hot (65oc) plant extraction buffer(PEB) [which contains 637.5 mL of double distilled water (ddH20), 100 mL of 1M Tris-HCl (pH 8.0), 100 mL of 0.5 M ethylenediaminetetraacetic acid (EDTA) (pH 8.0), 100 mL of 5M Nacl2 and 62.5mL of 20% sodium dodecylsulphate (SDS)] was added. One percent b- mercaptoethanol was added to the pre- warmed PEB just before use. The tubes were capped and inverted gently 6- 7 times to mix the sample with buffer. The solution was incubated at 65°C in water bath for 20 minutes with occasional mixing to homogenize the samples. After 20 minutes, samples were removed from the water bath and uncapped. The tubes were allowed to cool at room temperature for 2 minutes. After which 500 µL of 5M of potassium acetate (CH3COOK) was added to each tube and recapped. The tubes were mix inverted 6-7 times and incubated on ice for 20 minutes. After 20 minutes of incubation on ice, tubes were spun at 12,000 rpm for10 minutes at 4°C. The supernatant was transferred into new 1.5 mL eppendorf tubes using wider bore pipette tips (1000 µL) and making sure debris were not taken along with the supernatant. 700-µL chloroform isoamylalcohol was added to the supernatant and spun at 10,000 rpm for 10 minutes. The supernatant was transferred again to a new correspondingly labeled tubes and 700-µL ice-cold isopropanol was added to each tube and mixed by gently inverting the tubes 6-10 times. The tubes were allowed to stand undisturbed in a rack and stored in a freezer (-20°c) for at least 1 hour to precipitate the DNA. After 1-hour precipitation in the tubes, the DNA were centrifuged at 12,00 rpm for 10 minutes at 4°C. The supernatant in the freezer, was carefully discarded with great care to disallow the pellet from dislodging from the bottom of the tube. The tubes were allowed to drain inverted on clean paper towels for 1 hour. The DNA pellets were washed twice in 100µL, cold 70% ethanol for 20 minutes and air dried completely. After drying, 60µL of 1×TE [10mM Tris-HCL (pH 8.0), 1mM EDTA (Ph 8.0)] was added to the pellets, followed by 2µL of 10 ng/ml RNAse to remove the RNA. The DNA was then diluted to 10 ng/ul and then used for Polymerase Chain Reaction (PCR) PCR Amplification Nine SSR markers (Inqaba Biotech, South Africa) for Bambara groundnut was used to perform the PCR reactions and analysis for genetic diversity among the Bambara groundnut accessions. The amplification was performed in a 25 µL reaction volume containing 10X buffer, 1.6 µL of 25 mM of Mgcl2, 2.0 of 5µ/µL Tag, 8.0µL sterile distilled water and 4.0 µL sample DNA using a PTC-200 Thermal cycle. The PCR reaction was carried out with the following protocol: initial denaturation at 94oC for 5mins followed by 45 cycles of 30 secs at 94oC, then 30mins at 65oC and continues in that order. The resulting amplicons were loaded on 1.5% agarose. Agarose gel electrophoresis of the PCR product 1.5% agarose was prepared, which was microwaved to dissolve the agarose and cooled down (560C). The gel Nepal J Biotechnol. 2 0 2 2 D e c ; 1 0 : 85-90 Onwubiko et al. ©NJB, BSN 87 was poured into the gel tray that was prefixed with comb. The gel tray was immersed into an electrophoresis tank containing 0.5 XTBE buffer. The comb in the gel was removed to expose the wells formed. The amplified DNA (10 µL) was loaded into the wells of the gel with the aid of a pipette. A standard DNA molecular size marker (1 kb DNA lambda) was also loaded as a check. The gel ran for 2 hours at 150V and 0.5mAmp. After electrophoresis, the gel was silver stained. Thereafter the gel was destained in distilled water for 10 minutes. and the DNA bands in the gel was observed under UV (Ultra violet) lamp and photographed using a digital camera. Each accession was scored (1) for presence and (0) for absence of polymorphic band for each primer. The band scoring data was used to calculate genetic similarity based on Jaccard’s similarity coefficients (17) as follows: GSij= a/(a+b+c), Where, GSij is the similarity between two accession i and j; a is the number of bands present in both I and j; b is the number of bands present in I but absent in j; and c is the number of bands present in j and absent in i. In addition, the Jaccard’s similarity coefficient was used to perform cluster analysis based on the Unweighted Pair Group with Arithmetic mean (UPGMA) and in constructing a dendrogram. The computer program NTSYS pc version 2.1 (18) was used for these analyses. Furthermore, the genetic diversity, allele frequency and polymorphic information content (PIC) were computed using PowerMarker (Version 3.25). Results and discussion Polymorphism in several genes controls all phenotypic variations within a species. In this respect, the assessment of genetic variability within crop species using molecular markers is of great importance to plant breeders (15, 19). The molecular analysis of genetic diversity in the evaluated accessions of Bambara groundnut as determined by the SSR markers amplified a total of 27 alleles. The PIC values, which is a measure of the allelic diversity of SSRs, ranged from a minimum of 0.001 to a maximum of 0.617 with a mean of 0.419. Seemingly the obtained PIC mean was relatively high when compared to the report from Basu et al (20), but however relatively lower than the report of Mohammed (21). Apparently the SSR makers used in this study were not Vigna subterranea specific but Vigna unguiculata. Concisely maker P36 had the highest Marker Index (MI) which is a measure of efficiency to detect polymorphism. Invariably maker P36 was more genetically efficient in distinguishing the phenotypical similarity that exist between the Bambara groundnut accessions. On the other hand, markers P131 and P68 had monomorphic phenotype. These two markers recorded the least number of alleles per locus. A similar result has been reported by other workers (14, 15, 22, 19). Generally, the 9 SSR markers (Table 2) showed the availability of a substantial level of polymorphism among the Bambara groundnut lines as revealed by both genetic distance and cluster analysis. The accessions were grouped into five groups based on Jaccard Neighbor- joining dendrogram. Each group had subgroups comprised of distinct genotypes from different eco- regions. This implies that variations among individual genotypes was mainly responsible for the observed genetic variation among the Bambara groundnut lines and not variations established between specific accession groups. Generally, reasonable intra-accessions polymorphism was observed in the cluster analysis of the accessions. In previous studies some workers reported extensive genetic diversity (12, 15), while others observed considerable genetic diversity (13, 3), yet some others reported a low range of genetic diversity (23) in Bambara groundnut. Further in this study, divergent genotype was revealed by the cluster analysis using Unweighted Pair-Group Method with Arithmetic Average (UPGMA). TVSU 1554 from Nigeria was the only accession found in Group 5 among the five groups of the cluster analysis. Invariably this accession was dissimilar from other accessions evaluated. Divergent genotype(s) usually have good breeding value which can be used for crop improvement, in both direct selection and as parents for making crosses. Table 2. Description of the SSRs markers used in this study Marker Forward primer Reverse primer Gene bank ID P36 51-AAAATTGGAGAAAGGGGTTTTT-31 51-GATTTCGCCATATCCCCATC-31 GQ411715.1 P56 51-GCAATGGGTTCGTCGATACT-31 51-GCTCGATGCTTTTTGTTTCC-31 EU717407.1 P57 51-GGGAAACAAAAAGCATCGAG-31 51-CGCTACCCCAAAATACCAAA-31 EU717407.1 P61 51-GTCAGAGGCGAATTGAAAGC-31 51-AGGTCTTCCCGTTCCTTCAT-31 EU717373.1 P63 51-ATGAAGGAACGGGAAGACCT-31 51-CCTAAGGGCATATCGGTTGA-31 EU717373.1 P68 51-CAAGTCCCTCTATCCCCAAA-31 51-CAAGTCCCTCTATCCCCAAA-31 EU717348.1 P71 51-GTGTTGGGTTCAAAGCTGGT-31 51-CATCGGTCCACACAGTTGTC-31 EU717266.1 P131 51-CAAAGCCATTGCTGAAGACA-31 51-GGATGCTACACCGTTCGATT-31 HB823749.1 P184 51-GCCAGAGACTCTCACGTTCC-31 51-TGCATGGTCCCTGTTGTAGA-31 HB465729.1 Nepal J Biotechnol. 2 0 2 2 D e c ; 1 0 : 85-90 Onwubiko et al. ©NJB, BSN 88 Figure.1: Molecular dendrogram analysis of 33 accession of Bambara groundnut with unweighted pair group method with arithmetic method (UPGMA). The genetic distance from the molecular dendrogram analysis based on Jaccard’s similarity coefficient ranged from 0.84 to 1.00 (Figure 1). This implies that at 100% level of similarity, all the accessions were distinct from each other, while at 84% level of similarity all the accessions clustered to form a single accession. Invariably it means that each accession had at least one neighbour with more than 84% similarity. However, at 94% level of similarity, the accessions were grouped into five clusters. The pattern of clustering of the Bambara groundnut accessions in groups with the Unweighted Pair Group with Arithmetic mean (UPGMA) and Jaccard’s Neighbour-joining dendrogram was similar (Figure 2). Accessions were clustered in the same group based on genetic similarity and not on sources of collection or eco- region of origin. There were five heterogenic groups in the two-cluster analysis. The reason why accessions from the same geographical area could not form a distinct cluster was because they were genetically dissimilar. Each cluster, therefore, contained accessions with similar genetic characters. For example, group one of the Unweighted Pair Group with Arithmetic mean (UPGMA) grouping was the smallest group. It had two accessions, and both were collected from different geographical areas as can be seen in their identification numbers; TVSU 1483 was from Ghana, while TVSU 1631 from Togo. However, both were clustered together in group one, implying that the two accessions were duplicates or closely similar, and not two different accessions from two different localities as indicated in their identification numbers and places of collections. Another outstanding example was observed in group four. The accessions clustered in this group were TVSU 1584, TVSU 1591 and TVSU 1604. Accession TVSU 1584 was collected from Nigeria, while TVSU 1591 and TSVU 1604 were from Togo. A detailed analysis of this result surprisingly showed that these three accessions, that were collected from different geographical areas, were the most genetically similar lines among the evaluated Bambara groundnut collections. Similarly, in group 1 of Jaccard Neighbour-joining (JNJ) dendrogram, a comparable association was substantiated between TVSU 1697 and TVSU 1627 (from Togo) and TVSU 1503 (from Nigeria). These complex linkages suggest the possibility that these accessions were related. They either had similar genes or where from a common origin but were given different identification numbers. Apparently, these results showed the potentials of the SSR markers in detecting differences and establishing the extent of genetic relatedness in existence among the evaluated genetic materials. It also emphasizes the superiority of SSR markers in classifying collections of Bambara groundnut more precisely, as against the use of morphological markers in germplasm characterization (19). Nepal J Biotechnol. 2 0 2 2 D e c ; 1 0 : 85-90 Onwubiko et al. ©NJB, BSN 89 Figure 2: Jaccard Neighbour-joining dendrogram illustrating genetic diversity and relationships among 33 Bambara groundnut accessions. Clustering of accessions of Bambara groundnut collected from different geographical areas in the same group may have arisen due to duplication of genotypes caused by high rate of seed exchange between farmers from diverse ethnic and agro-geographical areas. In fact, there was a report from a previous study on a similar trend of association between Bambara groundnut accessions collected from different geographical areas, and in that study, it was concluded that these accessions were either related, or the genotypes were the same (22). A detailed examination of the result from the dendrogram showed that accessions from Nigeria were more dispersed among the clusters of accessions than accessions from other African countries. These accessions were found in four out of the five genetic groups. This observation may have some implication on the origin of Bambara groundnut, in that it supports the report of earlier studies on the origin of this crop; Nigeria may have been a regional centre of diversity of Bambara groundnut (24), which other studies confirmed by the existence of genuinely wild state of the crop in this area (25, 26). Conclusion Molecular analysis of genetic diversity of 33 accessions of Bambara groundnut based on SSR maker was reported. Genetic distances result and the cluster analysis revealed high level of polymorphism, indicating the existence of a wide range of genetic diversity among the accessions. This result can facilitate selection of appropriate genotypes for the development of improved lines of Bambara groundnut through various breeding programs. This study has contributed to broadening the genetic base of Bambara groundnut. The usefulness of this report in the effective utilisation, management, and conservation of Bambara groundnut germplasm is undoubtable. Author’s contribution MIU conceptualized the research proposal and supervised the research activities. NCO performed the lab works, scoring, data analysis and interpretation. NCO and GOC wrote the first draft of the paper. All authors read and approved the final manuscript. Competing Interests No competing interests were disclosed. Funding This research work was not funded by any organization. Acknowledgment The authors thank the staff of the International Institute of Tropical Agriculture (IITA), Ibadan for providing us with seeds of the accessions of Bambara groundnut used for this study. Nepal J Biotechnol. 2 0 2 2 D e c ; 1 0 : 85-90 Onwubiko et al. ©NJB, BSN 90 References 1. Ma Q, Wenheng, Z, Xiang Q. Evolution and developmental genetics of floral display—A review of progress. 2017.Volume 55 Issue 6, Pages 487-515. 2. Afolayan GO, Aladele, SE, Danquah EU, Bley E. Marker assisted selection: A potent tool for sorghum improvement. 38th annual GSN conference 2014 October, Edo, Benin State, Nigeria. Pp 193-196. 3. Ntundu WH, Bach IC, Christiansen JL, Andersen SB. Analysis of genetic diversity in Bambara groundnut (Vigna subterranea (L.) Verdc.) landraces using amplified fragment length polymorphism (AFLP) markers. African Journal Biotechnology 2004.3(4):220–225 4. Afuape SO. Morphological diversity of Sweet potato (Ipomoea batatas (L.) Lam) landraces as expressed in contrasting agro- ecologics in Nigeria. Nigerian Journal of Genetics. 2014 28:55-77. 5. Ejeta G, Knoll JE,. Marker-assisted selection in sorghum. 2007 pp.187-205 In: R. Vashney and R. Tuberosa (eds) Genomic Assisted Crop Improvement: Vol 2, Genomic Applications in Crops. Springer-Verlag, Berlin. 6. Collard, B.C.Y., Grams, R. A., Bovill, W.D., Percy, C.D., Jolley, R., Lehmensiek A., Wildermuth, G., and Sutherland, M.W. Development of molecular markers for crown rot resistance in wheat: mapping of QTL for seedling resistance in a ‘2-49’ x ‘Janz’ population. Plant Breed 2005. 124:532–537 7. Salgotra RS, Stewart CN. Functional Markers for Precision Plant Breeding. International Journal of Molecular Sciences, 2020, 21, 4792. 8. Drabo I, Sereme P, Dabire C. Burkina Faso: In: Heller J, Begemann, F, and Mushonga J. (Eds.), pp 11-18. Bambara Groundnut (Vigna subterranea L. Verdc.). Promoting the Conservation and Use of underutilised and Neglected Crops. Proceeding of Workshop on Conservation and Improvement of Bambara Groundnuts (Vigna subterranea L. Verdc.), 1995.14-16 November, Harare, Zimbabwe 9. Massawe FJ, Azam-Ali SM, Roberts JA. The use of molecular markers to explore phenotypic variation between and within landraces of Bambara groundnut. Journal of Experimental Biology (Supplement). 2000. 51: 71. 10. Ntundu WH. Genetic diversity of Bambara groundnut (Vigna subterranea (L.) Verdc) in Tanzania, 2002. PhD Thesis, the Royal Veterinary and Agricultural University, Copenhagen, Denmark. 11. Ofori K, Kumaqa FK, Tonyiqah A. Morphological characterization and agronomic evaluation of Bambara groundnut (Vigna subterranea [L.] Verdc.) Germplasm in Ghana. International Plant Genetic Resources Newsletter/FAO Bioversity, 2006.145: 23-2 12. Massawe, FJ, Robert JA, Azam-Ali SN, Davey MR. Genetic diversity in Bambara groundnut (Vigna subterranea (L.) Verdc.) landraces assessed by random amplified polymorphic DNA (RAPD) markers. Genetic Resources and Crop Evolution. 2003.50:737–741 13. Amadou HI, Bebeli PJ, Kaltsikes PJ. Genetic diversity in Bambara groundnut (Vigna subterreneae) germplasm revealed by RAPD markers. Genome.2001 44:995-999. 14. Somta P, Chankaew S, Rungnoi O, Sioles G. Genetic diversity of the Bambara groundnut (Vigna subterranea (L.) Verdc.), as assessed by SSR markers. Genome. 2011.54:11. 898-910. 15. Alaka CO, Okanlola GO, Akanbi AG, Olalekan OJ. Genetic variability in Bambara groundnut (Vigna subterranea (L.) Verdc) using molecular markers. 38th annual GSN conference 2014 October Edo Benin State Nigeria. Pp 241 – 247. 16. Dellaporta, S.L., wood, V.P and Hicks, J.B. (1983) A plant DNA mini preparation. Vision 11. Plant Mol. Boil. Rept 1:19-21. 17. Jaccard P. Nouvelles researchers sur la distribution florale. Bull. Soc. Vaud. Sci. 1908.44:22-270. 18. Rohlf FJ. NTSYSpc: Numerical taxonomy and multivariate analysis system. 2000. Exeter Software, New York. 19. Siise A. Massawe, FJ. Microsatellites based marker molecular analysis of Ghanaian Bambara groundnut (Vigna subterranea (L.) Verdc.) landraces alongside morphological characterization. Genetic Resources and Crop Evolution. 2013.60:777–787 20. Basu S, Roberts J, Azam-Ali S, Mayes S. Development of microsatellite markers for Bambara groundnut (Vigna subterranea [L.] Verdc.) – an underutilized African legume crop species. Molecular Ecology Note. 2007a.7: 1326-1328. 21. Mohammed SM (2014) Pre-breeding of Bambara Groundnut (Vigna subterranea [L.] Verdc.), PhD thesis, University of KwaZulu-Natal, Pietermaritzburg Campus, South Africa 22. Ntundu WH, Shillah, SA, Marandu WYF, Christiansen JL, Morphological diversity of Bambara groundnut (Vigna subterranea (L.) Verdc.) landraces in Tanzania. Genetic Resources and Crop Evolution. 2006..53:367–378 23. Pasquet RS, Scwedes S, Gepts P. Isozyme diversity in Bambara groundnut Crop Science. 1999.39: 1226–1236. 24. Tanimu B, Aliyu L. Country Report on Bambara groundnut production in Nigeria. In J. Heller, F. Begemann, and J. Mushonga (Eds.), Bambara Groundnut Vigna Subterranea (L.) Verdc: Proceedings of the Workshop on Conservation and Improvement of Bambara Groundnut (Vigna Subterranea (L.) Verdc.): 1995. 14-16 November, Harare, Zimbabwe Bioversity International.pp.89. 25. Hepper FN. 1970. Bambara groundnut (Voandzeia subterranea). Review article. Field Crop Abstract. 23 (1):1-6. 26. Goli AE, Begemann F, Ng, NQ. Characterization of IITA’s Bambara groundnut collection. In: Proceedings of the workshop, conservation and improvement of Bambara groundnut (Vigna subterranea (L.) Verdc.). 1995. 14–16 November, Harare Zimbabwe