This book is provided in digital form with the permission of the rightsholder as part of a Google project to make the world's books discoverable online. The rightsholder has graciously given you the freedom to download all pages of this book. No additional commercial or other uses have been granted. Please note that all copyrights remain reserved . About Google Books Google's mission is to organize the world's information and to make it universally accessible and useful. Google Books helps readers discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web at http://books.google.com/ Challenges and Opportunities for Enhancing Sustainable Cowpea Production Edited by CA. Fatokun, S.A. larawali, B.B. Singh, P.M. Kormawa, and M. lamb International Institute ofTropical Agriculture, Ibadan, Nigeria Digitized by Goog Ie AboutliTA The International Institute of Tropical Agriculture (I1TA) was founded in 1967 as an inter- national agricultural research institute with a mandate for improving food production in the humid tropics and to develop sustainable production systems. It became the first African link in the worldwide network of agricultural research centers known as the Consultative Group on International Agricultural Research (CGIAR), formed in 1971. I1TA's mission is to enhance the food security, income, and well-being of resource-poor people primarily in the humid and subhumid zones of sub-Saharan Africa, by conducting research and related activities to increase agricultural production, improve food systems, and sustainably manage natural resources, in partnership with national and international stakeholders. To this end, I1TA conducts research, germplasm conservation, training, and information exchange activities in partnership with regional bodies and national programs including universities, NGOs, and the private sector. The research agenda addresses crop improvement, plant health, and resource and crop management within a food systems framework and targetted at the identified needs of three major agroecological zones: the savannas, the humid forests, and the midaltitudes. Research focuses on smallholder crop- ping and postharvest systems and on the following food crops: cassava, cowpea, maize, plantain and banana, soybean, and yam. ISBN 978-131-190-8 Citation: Fatokun, C.A., S.A. Tarawali, B.B. Singh, P.M. Kormawa, and M. Tamo (editors). 2002. Challenges and opportunities for enhancing sustainable cowpea pro- duction. Proceedings of the World Cowpea Conference III held at the International Institute of Tropical Agriculture (I1TA), Ibadan, Nigeria, 4-8 September 2000. I1TA, Ibadan, Nigeria. Digitized by Google Contents Foreword Preface Acknowledgements Cowpea genetics and breeding VI Vll IX 1.1 Recent genetic studies in cowpea 3 B.B. Singh 1.2 Breeding cowpea for tolerance to temperature extremes and adaptation to drought 14 A.E. Hall, A.M. Ismail, J.D. Ehlers, K.D. Marfo, N. Cisse, S. Thiaw, and T.J. Close 1.3 Recent progress in cowpea breeding 22 B.B. Singh, J.D. Ehlers, B. Sharma, and F.R. Freire Filho 1.4 Breeding and evaluation of cowpeas with high levels of broad-based 41 resistance to root-knot nematodes J.D. Ehlers, W.C. Matthews, A.E. Hall, and P.A. Roberts 1.5 Breeding cowpea for resistance to insect pests: attempted 52 crosses between cowpea and Vigna vexillata C.A. Fatokun 1.6 Cowpea breeding in the USA: new varieties and improved germplasm 62 J.D. Ehlers, R.L. Fery, andA.E. Hall II Cowpea integrated pest management 2.1 The importance of alternative host plants for the biological 81 control of two key cowpea insect pests, the pod borer Maruca vitrata (Fabricius) and the flower thrips Megalurothrips sjostedti (Trybom) M. Tamo, D.Y. Arodokoun, N. Zenz, M. Tindo, C. Agboton, and R. Adeoti 2.2 Recent advances in research on cowpea diseases 94 A.M. Emechebe and s.T.D. Lagoke 2.3 Development of sex pheromone traps for monitoring the legume 124 podborer, Maruca vitrata (F.) (Lepidoptera: Pyralidae) M.C.A. Downham, M. Tamo, D.R. Hall, B. Datinon, D. Dahounto, and J. Adetonah 2.4 Evaluation of a novel technique for screening cowpea varieties 136 for resistance to the seed beetle Callosobruchus maculatus A.D. Devereau, L.E.N. Jackai, T.B. Olusegun, andA.N.J. Asiwe 2.5 Detection of fumonisin B 1 in cowpea seeds 147 Q. Kritzinger, T.A.S. Aveling, w.F.D. Marasas, G.s. Shephard, and N. Leggott iii Digitized by Google Challenges and opportunities for cowpea production 2.6 Breeding cowpea varieties for resistance to Striga gesnerioides and Alectra vogelii B.B. Singh III Biotechnology for cowpea 3.1 Isolation, sequencing, and mapping of resistance gene analogs from cowpea (Vigna unguiculata L.) B.S. Gowda, J.L. Miller, s.s. Rubin, D.R. Sharma, and MP. Timko 3.2 Regeneration and genetic transformation in cowpea J. Machuka, A. Adesoye, and 0.0. Obembe 3.3 Molecular cloning in cowpea: perspectives on the status of genome characterization and gene isolation for crop improvement MP. Timko 3.4 Potential role of transgenic approaches in the control of cowpea insect pests J.Machuka 3.5 Insecticidal activities of the African yam bean seed lectin on the development of the cowpea beetle and the pod-sucking bug o.G. Okeola, J.s. Machuka, and /.0. Fasidi IV Cowpea contributions to farming systems/agronomic improvement of cowpea production 4.1 Cowpea as a key factor for a new approach to integrated crop-livestock systems research in the dry savannas of West Africa S.A. Tarawali, B.B. Singh, s.c. Gupta, R. Tabo, F. Harris, S. Nokoe, S. Fernandez-Rivera, A. Bationo, ~~M Manyong, K. Makinde, and E.C. Odion 4.2 Cowpea rotation as a resource management technology for cereal-based systems in the savannas of West Africa R.J. Carsky, B. Vimlauwe, and 0. Lyasse 4.3 Advances in cowpea cropping systems research 0.0. Olufajo and B.B. Singh 4.4 Improving cowpea--cereals-based cropping systems in the dry savannas of West Africa B.B. Singh, and H.A. Ajeigbe 4.5 Cowpea varieties for drought tolerance B.B. Singh and T. Matsui 4.6 Soil fertility management and cowpea production in the semiarid tropics A. Bationo, B.R. Ntare, S.A. Tarawali, and R. Tabo iv Digitized by Google 154 167 185 197 213 223 233 252 267 278 287 301 4.7 Differential response of cowpea lines to application of P fertilizer G.o. Kolawole, G. Tian, and B.B. Singh 4.8 Farmer participatory evaluation of newly developed components of cowpea and cotton intercropping technology F.A. Myaka, J.C.B. Kabissa, D.F. Myaka, andJ.K. Mligo 4.9 Cowpea dissemination in West Africa using a collaborative technology transfer model J.o. Dlufowote and P. W. Barnes-McConnell V Cowpea postharvest and socioeconomic studies 5.1 The economics of cowpea in West Africa 0. Coulibaly and J. Lowenberg-DeBoer 5.2 Industrial potential of cowpea C. Lambot 5.3 Cowpea demand and supply patterns in West Africa: the case of Nigeria P.M Kormawa, ~~M Manyong, and J.N. Chianu 5.4 Potential adoption and diffusion of improved dual-purpose cowpea in the dry savannas of Nigeria: an evaluation using a combination of participatory and structured approaches /. Dkike, P. Kristjanson, S.A. Tarawali, B.B. Singh, R. Kruska, and ~~MManyong 5.5 Impact of cowpea breeding and storage research in Cameroon F. Diaz-Hermelo, A. Lanygintuo, and J. Lowenberg-DeBoer 5.6 Identifying cowpea characteristics which command price premiums in Senegalese markets M Faye, J.L. DeBoer,A. Sene, andM Ndiaye v Contents 319 329 338 351 367 376 387 407 424 Digitized by Google Foreword Cowpea is an important food legume and an essential component of cropping systems in the drier regions and marginal areas of the tropics and subtropics covering parts of Asia and Oceania, the Middle East, southern Europe, Mrica, southern USA, and Central and South America. It is particularly important in West Mrica with over 9.3 million hectares and 2.9 million tonnes annual production. With about 25% protein in its grains, cowpea is an important source of quality nourishment to the urban and rural poor who cannot afford meat and milk products. Cowpea haulms contain over 15% protein and constitute a valuable source of fodder. The International Institute of Tropi- cal Agriculture (I1TA) has the global mandate for cowpea improvement. In collabora- tion with the regional and national research programs, I1TA has developed a range of improved cowpea breeding lines combining multiple disease and insect resistance with early maturity and preferred seed types, and has distributed these to over 65 countries. From 1970 to 1988, the research concentrated on developing cowpea varieties for sole crop only. However, from 1989, cowpea breeding has been diversified to include sys- tematic improvement of local varieties as well as development of a range of improved dual-purpose cowpea varieties which can produce higher grain as well as fodder yields under sole cropping and in traditional intercropping systems. In 1996, I1TA decided to further broaden the objectives by including improvement of cowpea-cereals systems and involving crop-livestock integration rather than variety improvement alone. The specific objectives are to develop improved cowpea varieties and improved crop- ping systems with integrated pest management appropriate for adoption in the Sudan savanna and the Sahel. An integral part of I1TA's cowpea research is an active technology and information exchange system among cowpea researchers worldwide through cowpea international trials, special workshops, individual and group training, and periodic world cowpea conferences. I1TA organized the first World Cowpea Research Conference in 1984 and the second in 1995. The selected papers presented in these conferences were collated and published in two books: Cowpea research, production, and utilization (1985) and Advances in cowpea research (1997). Both books have become very popular among cowpea researchers the world over. In view of the rapid developments in cowpea research, the delegates of the second World Cowpea Conference felt that a gap of 10 years was too long and recommended that in future, the World Cowpea Conferences should be held every 5 years. I1TA agreed with the recommendation and organized the World Cowpea Conference III in September 2000. This book contains selected papers presented in the conference. It is hoped that this will be a useful supplement to the earlier two books and help to enhance collaborative work among cowpea researchers, leading to innovative approaches and improved technologies in the years to come. L. Brader Director General International Institute of Tropical Agriculture vi Digitized by Google Preface Cowpea researchers from different parts of the world came together to participate in the World Cowpea Research Conference III which took place from 4 to 8 September 2000 at the headquarters of the International Institute of Tropical Agriculture (I1TA), Ibadan, Nigeria. Two previous world cowpea research conferences had been held: the first in 1984 at Ibadan, Nigeria and the second in 1995 at Accra, Ghana. The interval between the first and second world cowpea research conferences was eleven years while that between the second and the third, was only five years. The shorter interval between the second and third conferences attests to an increase in the number of researchers focusing on cowpea. These conferences provide opportunities for cowpea researchers to interact and exchange scientific information resulting from their research activities. At the same time plans for the future are made. The conference featured both oral presentations and posters displays. Most of the oral paper presentations are included in this proceedings volume which is divided into five sections: (a) cowpea genetics and breeding, (b) cowpea integrated pest management, (c) biotechnology for cowpea, (d) cowpea contributions to farming systems, and (e) cowpea postharvest and socioeconomic studies. The reports presented indicate that appreciable progress has been made in cowpea research during the past five years and cowpea research has impacted positively on the productivity of the crop especially in sub-Saharan Africa. At the Accra meeting of 1995, a limited number of reports were presented in the area of socioeconomic studies in cowpea. However, at this conference, a section was devoted to postharvest and socioeconomic studies and a number of papers were presented. There were presentations on the economics of cowpea in West Africa and cowpea supply prospects in Nigeria. The adoption of improved technologies that will enhance the productivity of cowpea by farmers in Ghana as well as the industrial potential of the crop in the subregion were also discussed. In addition, a session was specifically devoted to cowpea biotechnology during which reports were presented on the isolation and sequencing of resistance gene analogs from cowpea and the placement of these in the cowpea linkage map. Seed lectins obtained from some leguminous plants, particularly the African yam bean (Sphenostylis stenocarpa), were reported to have adverse effects on the important postflowering insect pests of cowpea such as the pod sucking bugs and the cowpea bruchid (Callosobruchus maculatus). The detection of the adverse effects of these lectins on some cowpea pests is an indication that there are potential candidate genes that can be used for the transformation of cowpea for resistance to these pests which are capable of causing extensive grain yield loss in the crop. When they become available, transgenic cowpea varieties with resistance to postflowering insect pests will boost the productivity of the crop in African farmers' fields. vii Digitized by Google The contributions of cowpea to the farming systems in the dry savanna regions of sub- Saharan Africa were also highlighted. Apart from contributing to soil fertility through nitrogen fixation and production of organic matter, cowpea fodder provides quality feed for livestock in the subregion. Ruminants fed cowpea fodder as supplement are known to gain weight appreciably. It is hoped that cowpea researchers will find the contents of this book useful and stimulating. C.A. Fatokun viii Digitized by Google Acknowledgements The authors wish to acknowledge the invaluable contributions made by the Rockefeller Foundation which provided funds in support of this publication and the Technical Centre for Agricultural and Rural Cooperation (CTA) for funding the participation of some researchers at this conference. We also wish to thank Dr Kenton Dashiell, former Director of the Crop Improvement Division at I1TA, for his commitment to the success of the World Cowpea Research Conference III. The following individuals are acknowledged for reviewing the manuscripts included in this proceedings volume: 1. Adu-Gyamfi, S.O. Ajala, K. Amegbeto, B. Asafo-Adjei, M.A. Ayode1e, R.1. Carsky, A Cherry, O. Coulibaly, B. Douthwaite, 1.D. Ehlers, AM. Emechebe, I. Fawole, M. Gedil,1. Gockowski, W.N.O. Hammond, R. Hanna, 1. Langewald, VM. Manyong, A Me1ake-Berhan, A Menkir, P. Oyekan, F. Schulthess, S. Schulz, B.B. Singh, M. Toko, L. Tripathi, 1. Wendt, and T.O. Williams. A special word of thanks to I1TA's Communications and Information Services staff: David Mowbray and Paul Philpot for monitoring the production process, Taiwo Owoeye, Yvonne Olatunbosun, Ayotunde Oyetunde, and Rose Umelo for their editorial support, Fatai Agboola for typesetting the manuscripts, and Godson Bright for designing the cover. ix Digitized by Google Digitized by Google Section I Cowpea genetics and breeding Digitized by Google Digitized by Google 1.1 Recent genetic studies in cowpea B.B.Singh1 Abstract A number of recent studies have added further information on the genetics of important traits in cowpea. These include inheritance of qualitative traits such as plant pigmentation; flower color; seed color; seed coat texture; resistance to rust, scab, smut, nematode, severe mosaic virus, Striga, Alectra, aphid, bruchid, heat; drought tolerance; and male sterility, and quantitative traits such as protein content, seed size, seed yield, and fodder quality. A few studies on linkage and mapping have also been conducted. The gene symbols from recent studies and earlier reports have been collated in a classified and trait-based gene index for easy reference. While reviewing the past genetic work, obvious gaps needing further studies have been indicated. Introduction The first comprehensive review of cowpea genetics was published in 1980 (Fery 1980) and subsequent supplements were published in 1985 (Fery 1985) and in 1997 (Fery and Singh 1997). This paper complements the earlier literature by reviewing some recent work on cowpea genetics and pointing out some gaps needing further research. Species relationship Cowpea is a variable species composed of wild perennials, wild annuals, and cultivated forms. Genetic variation in 199 germplasm lines of nine subspecies and two botanical varieties of wild and cultivated cowpea were evaluated by Pasquet (1999) using allozyme analysis to characterize the genepool. The allozyme data confirmed that perennial out- crossers are primitive and more remote from each other and from perennial out-inbreds. Within the large genepool, mainly made of perennial taxa, the cultivated cowpea form a genetically coherent group and are closely related to annual wild cowpea which may include the likely progenitor of cultivated cowpea. Cardinali et al. (1995) analyzed 32 accessions of cultivated and wild cowpea for phe- nolic content using HPLC to better characterize wild species of Vigna. The cultivated cowpea always contained three flavonoid aglycones: quercetin, kaempferol, and isorh- arnnetin. These were lacking in the wild relatives. They also observed that resistance to aphid in cultivated cowpea was related to high flavonoid levels. In a similar study, Son- nante et al. (1996) examined the isozyme variation in 25 accessions of wild and cultivated Vigna unguiculata, 49 accessions of seven wild species belonging to section Vigna, and 11 accessions of Vigna vexillata to assess genetic relationship within and among spe- cies. They observed that Vigna unguiculata was closer genetically to Vigna vexillata than to the species belonging to section Vigna. Venora and Padulosi (1997) carried out a karyotypic analysis of mitotic chromosomes of 11 wild taxa of Vigna unguiculata and 1. International Institute of Tropical Agriculture, Kano Station, PMB 3112, Kano, Nigeria. 3 Digitized by Google Cowpea genetics and breeding found a low degree ofkaryological variability. The results indicate that despite high mor- phological variability in cowpea, such diversity is not evident at the chromosomal level. Gomathinayagam et al. (1998) reported a successful cross between Vigna vexillata and Vigna unguiculata using embryo culture. They obtained 13 hybrid plants which showed intermediate morphological traits between the parents for leaf shape, pod color, and seed coat color. However, the stem and leaf types and pod hairiness of hybrid plants were like those of Vigna vexillata, the maternal parent. Elechrophoresis studies of the hybrid plants for peroxidase and esterase and cytological studies confirmed that they were true hybrids. However, the same cross has not been successful at I1TA (see Fatokun in this volume). Another report of an attempted wide cross between cowpea (Vigna unguiculata) and bambara groundnut (Vigna subterranea) was published by Begemann et al. (1997). The cowpea line TVu 13677 was crossed as a female parent with bambara groundnut variety TVsu-501. The crossed flowers produced an abnormally short pod ( 1 cm) with only one seed. The F I seed gave rise to a plant which had a longer growth period (80 days) compared to the female parent TVu 11677 (60 days). Tyagi and Chawala (1999) reported a successful cross between Vigna radiata and Vigna unguiculata using in vitro culture method. The above-mentioned studies may indicate that wide crosses are possible between Vigna unguiculata and other Vigna species, however, none of the authors have followed up with the hybrid populations, indicating that further work needs to be done to verify these reports. Genetics of plant pigmentation Because of the great diversity in pigmentation of cowpea stem, leaf, flower, peduncle, petiole, and pod, this trait has been studied by a large number of researchers from 1919 to date. However, since most of the studies involved pigmentation of one or the other plant parts at a time, there seems to be several gene symbols assigned for the same trait or similar traits. A summary of gene symbols assigned for pigmentation of different plant parts by previous workers is presented in Table 1, which clearly indicates the overlap and confusion. For example, seven gene symbols (Pp-l, Pp-2, pg, Pb, Pbr, Pu, and X) have been assigned to plant pigmentation covering the plant, petiole base, branch base, stem- pod-petiole, and all the vegetative parts, which obviously have overlaps. Similar overlaps and confusion about the gene symbols are evident for flower color, calyx color, and pod color (Table 1). None of the reports have endeavored to study plant pigmentation on a holistic basis, explaining the relationship between pigmentation in different plant parts. Even the recent studies reported here have not clarified the situation (Table 2). Joshi et al. (1994) reported that PI is a pleiotropic gene for pigmentation in axil, calyx, corolla, pod tip, and seed with localized genes conditioning coloration on individual parts. U guru (1995) showed that petal color is governed by one allelic pair WW, while pod and shoot colors appear to be determined pleiotropic ally by two allelic pairs, PrPr and GrGr. Calyx color was reported to be controlled by three duplicate genes and standard petal color is controlled by a single dominant gene. Biradar et al. (1997) reported three genes for calyx color, three genes for seed coat color, four genes for pod tip pigmentation, and four genes for flower color with some genes showing pleiotropic effects. Venugopal (1998) observed that 1-5 pairs of genes were involved in the inheritance of plant pigmentation in cowpea. Sangwan and Lodhi (1998) studied the inheritance of flower color and pod color. 4 Digitized by Google Recent genetic studies in cowpea Table 1. Gene index for plant pigmentation in cowpea. Trait Gene symbol References Plant pigmentation Purple plant Pp-l Yenugopal and Goud 1997* Purple plant Pp-2 Yenugopal and Goud 1997* Pale green plant pg Saunders 1960a Purple petiole base Pb Sen and Bhowal1961 Purple branch base Pbr Sen and Bhowal1961 Purple stem, pod, petiole Pu Sen and Bhowal1961 Anthocyanin in vegetative parts X Harland 1919b Flower color Purple flower pf Kolhe 1970 Pale flower L Harland 1919a Dark flower color 0 Harland 1919a Tinged flower G Harland 1920 Yellow strips on petals Ystp Kolhe 1970 Calyx color Brown calyx color vs green Bcy Kolhe 1970 Purple calyx color P Harland 1920 Purple calyx color Pv Sen and Bhowal1961 Purple calyx color E Harland 1920 Pod color Black pod vs white Bk Capinpin 1935* Brown pod vs straw Bp Saunders 1960b* Cocoa brown pod Cbr Krishnaswamy et al. 1945 Reddish (cerise) Ce Saunders 1960a* Green pod vs cream Gp Kolhe 1970 Green pod vs white Gnp Singh and Jindla 1971 * Purple pod Pp Mortensen and Brittingham 1952 Dark pod color k Mortensen and Brittingham 1952 light green pod Ig Krishnaswamy et al. 1945 Purple pod P Harland 1920 Purple pod, stem, petiole Pu Sen and Bhowal1961 Speckled pod Sk Saunders 1960a* Straw yellow pod-1 Sy-l Krishnaswamy et al. 1945* Straw yellow pod-2 Sy-2 Krishnaswamy et al. 1945* Red tip pod Pb Mortensen and Brittingham 1952 Purple pod with green sutures Pg Sen and Bhowal1961 Purple tip pod Pt Sen and Bhowal1961 Purple sutures on green pod Ps Sen and Bhowal1961 ·Symbols by Fery (1980). They observed that purple flower color is dominant over white flower and black pod color is partially dominant over white pod color with monogenic inheritance for both traits. The confusion about the genetics of plant pigmentation arises due to the fact that most of the published reports do not give specific details of the pigmentation pattern and pigmented parts. For example, purple flower color does not mean much because pigmentation in cowpea flowers may be restricted only to standard, wing, or keel petals or a combination of two or all the three parts. A close examination (by the author) of several cowpea varieties has revealed very interesting and contrasting combinations of 5 Digitized by Google Cowpea genetics and breeding Table 2. Index of new gene symbols. Gene symbol bcm Bk-2 fa Dhp Gr PI pms ps Ptc Pt Pc Rdsl Rds2 Rtl Rt2 Vsm Character Resistance to black- eye cowpea mosaic virus Black pod color Fasciated plant Dehiscent pod Green shoot color Pleiotropic gene for axil, calyx, corolla pod tip, and seed colors Partial male sterility Photosensitivity Calyx pigmentation Calyx pigmentation Calyx pigmentation Resistance to drought Resistance to drought Rough seed coat texture Rough seed coat texture V-shaped mark on leaves Reference Arshad et al. (1998) Aliboh et al. (1996) Adu-Dapaah et al. (1999) Aliboh et al. (1996) Uguru (1995) Joshi et al. (1994) Singh and Adu-Dapaah (1998) Ishiyaku and Singh (2001) Biradar et al. (1997) Biradar et al. (1997) Biradar et al. (1997) Mai-Kodomi et al. (1999) Mai-Kodomi et al. (1999) Singh and Ishiyaku (2000) Singh and Ishiyaku (2000) Aliboh et al. (1996) plant pigmentation (Table 3). It has also been observed (by the author) that all the cowpea varieties with brown rough seed have no pigmentation on any plant part except for a faint purple tinge on the inner margins of the standard petal. Also all the cowpea varieties with white rough seed have purple pigmentation on the joints (bases of the branch, peduncle, petiole, and leaflets), which are always inherited as one gene. However, the pigmentation of whole stem, petiole, peduncle, and pod is independent of the pigmentation on the joints. It has also been observed that whenever the calyx is pigmented, the pod tips are also pig- mented and this is independent of other pigmentation. Another interesting pigmentation pattern is present in the cowpea variety Kamboinse local. It has dark purple pigmentation on the stem, petiole, peduncle, joints, calyx, and pod, but the flowers are completely white except for a purple dash in the back of the standard. The cowpea varieties listed in Table 3 represent a good set of differentials for different pigmentation patterns and efforts are under way to use them in planned genetic studies to elucidate inheritance pattern and interaction, if any, of specific plant pigmentations. Genetics of disease resistance Inheritance of, resistance to several cowpea diseases has been reported between 1995 and 2000. Vale et al. (1995) studied the inheritance of resistance to cowpea severe mosaic comovirus (CpSMV) using cowpea variety Macaibo as the resistant parent and Pitiuba as the susceptible parent. The F 1 plants were uniformly susceptible and F 2 segregated into a ratio of three susceptible to one resistant, indicating involvement of a single recessive gene pair for resistance. The authors have mentioned that Macaibo is immune to CpSMV Arshad et al. (1998) studied the inheritance of resistance to blackeye cowpea mosaic (BlCMV) in six cowpea varieties: IT86F-2089-5, IT86D-880, IT90K-76, IT86D-101O, IT86F-2065-5, and PBlCP3. The segregation pattern in F2, and backcross populations 6 Digitized by Google Recent genetic studies in cowpea Table 3. Pigmentation of different parts in selected cowpea varieties. Pigmentation in various plant parts Genetic type Stem jts Pet Ped Clx FL Pd Pdt Seed color s w k TVx 3236-0C-1 IT87D-941-1 + Brown rough TVx 3236-0C-2 - + - Brown rough IT98K-628-2 + + + + White rough IT90K-277-2 + White rough IT98K-598 + +2 +2 +2 - Brown smooth IT97K-1101-5 + +2 +2 +2 +2 +2 +2 +2 +2 +2 Black smooth Kamboinse local +2 +2 +2 +2 b +2 +2 +2 White rough IT86D-719 + + + +2 +2 +2 White rough I T95K-1491 + + + + - + + White smooth jts = joints, Pet = petiole, Ped = peduncle, Clx = calyx, FI = flower, Pd = pod, Pdt = pod tip, s = standard, w = wing, k = keel, b = a purple dash at the back of the standard petal. suggested that the resistance to BlCMV is controlled by single recessive gene pair in each cowpea line. They designated bcm as the gene symbol. Ryerson and Heath (1996) studied the inheritance of resistance to rust Uromyces vignae in cowpea cultivar Calico Crowder. The segregation pattern in F 2 generation and subse- quent progeny suggested the presence of multiple genes and also the presence of dominant and recessive resistance components. Rangaiah (1997) also reported the inheritance of rust (Uromyces vignae) resistance in cowpea in eight F 2 populations. He observed that a minimum of two genes control resistance to rust in cowpea. Nakawuka andAdipala (1997) screened 75 cowpea lines against scab of which 10 were resistant. These were then used to study the genetics of resistance to scab by Tumwegamire et al. (1998) using a half-dialle1 cross set. Broad-sense heritability for foliar resistance was 93.8% and for pod resistance it was 97% and 84.5%, respectively. This indicates major gene inheritance, which had earlier been reported by Abadassi et al. (1987) in TVx 3236. Genetics of resistance to nematodes Roberts et al. (1996) identified IT84S-2049 cowpea line from I1TA to be completely resistant to diverse populations of the root-knot nematodes, Meloidogyne incognita and M. javanica. The resistance in this variety was effective against nematode isolates that are virulent to the resistance gene Rk present in commercial cultivars in California such as CB5 and CB46. Systematic genetic studies indicated that the resistance in IT84S-2049 was conferred by a single dominant gene which was either allelic to Rk gene or a differ- ent gene very closely linked to Rk. Therefore, the symbol Rk2 was proposed to designate this new resistance factor. Rodriguez et al. (1996) screened nine cowpea varieties for resistance to the root-knot nematode Meloidogyne incognita. They observed that I1TA-3, Habana 82, Incarita-l, IT86D-364, IT87D-1463-8, Vinales 144, P902, and I1TA-7 were highly resistant whereas the local variety Cancharro was highly susceptible. Genetics of new mutants Singh and Adu-Dapaah (1998) reported a partial sterile mutant controlled by a single recessive gene pms. The mutant plants remained green for a longer period than the wild type and they had thick, leathery leaves with a few fleshy 1-3 seeded pods with gaps and 7 Digitized by Google Cowpea genetics and breeding about 77% viable pollen indicating partial male as well as partial female fertility. The homozygous recessive plants (pms pms) bred true for partial sterility. Adu-Dapaah et al. (1999) also reported a fasciated mutant, which was observed in an F 4 population of a cross TVu 3000 x IT82D-604. The mutant plants were both male and female sterile and exhibited crumpled petals and sepals, rosette branching, and abnormal stigmas ranging in number from zero to two. Genetic study showed that this trait was controlled by a single recessive gene, which was designated asia. Odeigah et al. (1996) reported several induced mutants of which four were male sterile and female fertile and two mutants were completely sterile. All the six mutants showed a monogenic recessive inheritance. Genetics of leaf, pod, and seed types Aliboh et al. (1996) studied the inheritance of inverted V-shaped marks on leaves, pod dehiscence, and dry pod color in crosses involving wild, weedy, and cultivated varieties of cowpea. The segregation pattern in F 2 and backcross generations indicated monogenic dominant inheritance for all the three traits. The gene symbols ~ sm, Dhp, and Bk-2 were assigned for the V-shaped leaf marks, pod dehiscence, and black dry pod color, respectively. Kehinde and Ayo-Vaughan (1999) and Singh and Ishiyaku (2000) reported inheritance of seed coat texture in cowpea and indicated the involvement of two pairs of genes for this trait. The crosses between smooth and rough seed texture segregated into three smooth: one rough. However, the crosses involving white rough seed x brown rough seed showed a complementary gene action. The F 1 was smooth and F 2 segregated into a nine smooth: seven rough ratio. This was supported by the backcross data. The gene symbols rt1 and rt2 were assigned for rough testa. Rough seed coat texture is an important trait in West and Central Africa because it facilitates removal of the seed coat for certain food preparations. Genetics of photosensitivity and drought tolerance Ishiyaku and Singh (2001) observed that the photosensitive cultivars not only flower early but also become extremely dwarfed when day lengths are less than 12.5 hours. The dwarfing under short-day length was observed to be a pleiotropic effect of the photosen- sitivity gene as it showed monogenic recessive inheritance that is completely associated with photosensitivity. The gene symbol ps was assigned to it. This is the first report indi- cating the effect of photoperiod on vegetative growth of plants. All earlier reports linked photosensitivity with reproductive stage only. Mai-Kodomi et al. (1999) reported simple inheritance of drought tolerance in cowpea. U sing a box screening method, they identified two types of shoot drought tolerance. Type 1 plants stayed green for a long time after withholding water and the whole plant died with continued dry conditions. In contrast, the Type 2 plants stayed alive for a much longer period, but the whole plant did not die with continued dry conditions. They mobi- lized moisture from the lower leaves to keep the growing tips alive for longer and so the plants dropped the lower leaves first and dried upward slowly such that when watering was resumed, they recovered. Both Type 1 and Type 2 drought tolerance are inherited as monogenic dominant traits. The F 1 crosses between them showed dominance of Type 1 and F 2 segregated into three Type 1 :one Type 2, suggesting that these are alleles at the same locus. The gene symbols Rds 1 (resistance to drought stress) and Rds2 were assigned for these traits. This is the first report of monogenic inheritance of drought tolerance in plants. The simple inheritance was observed probably because of simplified screening methods 8 Digitized by Google Recent genetic studies in cowpea and selective screening for shoot drought tolerance only. The details are further presented elsewhere in this volume (Singh). Menendez and Hall (1996) studied the heritability of carbon isotope discrimination (DELTA) which may be a useful selection criterion for drought adaptation in cowpea. Broad-sense heritability for DELTA in two crosses (TVx 309 x Prima and TVx 309 x CB 46) was 0.47 and 0.33, respectively, indicating an inter- mediate level of genetic variability for this trait. Ten cDNAs of genes that were induced by dehydration stress were cloned by differential screening from drought tolerant cowpea variety (Luchi et al. 1996). The clones were collectively named CPRD (cowpea clones responsive to dehydration). A dehydrin gene involved in chilling tolerance during seedling emergence has been identified (Ismail et al. 1997, 1999) and mapped using recombinant inbreds (Menendez et al. 1997). Genetics of quantitative traits and heterosis Damarany (1994) published information on heritability and genetic advance for 13 char- acters in cowpea. Broad-sense heritability for seed weight/plant was 94.4%, 85.9% for pods/plant, and 83.3% for 100 seed weight. Genetics of pod yield and its components were studied in F 2 and backcross populations of a cross involving two vegetable cowpea varieties, UCR 193 and IT8ID-1228-l4 by Pathmanathan et al. (1997). The broad-sense heritability for pod weight was 84% and the narrow-sense heritability was 75% indicating good genetic variability for effective selection. Menendez and Hall (1996) studied the heritability of harvest index in two crosses (TVx 309 x Prima and TVx 309 x CB 46). The broad-sense heritability for this trait was 38% and 58% in the two crosses, respectively. Sangwan and Lodhi (1995) studied heterosis for yield and yield components in 25 crosses involving 11 cowpea varieties. Better parent heterosis ranged from 28.8% to 84.0% for seed yield/ha. Heterosis up to 81.6% over better parent was observed for pod/plant, 35.6% for pod length, 20.4% for seed/pod, and 36% for seed weight/plant. Hybrid Fos-l x Col, Fos-2 x EC 4216, and EC 4216 x C28 were most promising. Arvindhan and Das (1996) reported 215% heterosis for seed yield in the cross CS 55 x C04. Bhoret al. (1997) studied PI' F l' and F 2 populations of 14 crosses and observed 63.8% better parent heterosis for seed yield in the cross V240 x VCM8. They further observed that the heterosis was 4.3% for plant height and 91.52% for days to maturity. They observed that progeny derived from crosses showing high heterosis also showed high inbreeding depression indicating the importance of nonadditive gene action. Bhushana et al. (2000) estimated heterosis for several traits in 36 hybrids. They observed a midparental heterosis of 171.5% for number of secondary branches/plant, 11.5% for pods/plant, 105.3% for seed yield/plant, 75.5% for primary branches/plant, 30.31 % for pod length, and 20% for 100 seed weight. They also observed -15.9% heterosis for days to 50% flowering. Heterosis for fodder yield was reported by Ponmariammal and Das (1996) and Arvindan and Das (1996) and the highest heterosis (121 %) was recorded for the hybrid UPC9201 x C05. High values for heterosis indicates good genetic diversity among cowpea varieties used in these studies indicating the possibility of isolating high yielding transgressive segregates from hybrid populations. However, the estimates for heterosis in most cases is from space planted F 1 hybrids, which may not be a true index of performance undernormal plant popu- lations used for commercial crops. Therefore, there is a need to estimate heterosis under recommended plant population for maximum yield of cowpea. This will, of course, require 9 Digitized by Google Cowpea genetics and breeding making a large number of pollinations to obtain enough F 1 seeds to test under normal density. Linkage and mapping The first report oflinkage in cowpea was published by U guru and Ngwuta (1995). From the genetic analysis of F l' F 2' and F 3 populations derived from the crosses involving cowpea variety AN-14-D with purple calyx, purple petal, and purple pod and varieties An-36-F and AE-36-W which were nonpigmented. They observed linkage between the three traits with calyx and petal color most tightly linked (0.576 ± 0.009 cm). Another report on linkage was published by Githiri et al. (1996) who identified genes on four linkage groups as indicated below: Linkage group 1. Sw ----'?- Fbc 41 ± 4.8 2. Pus ----'?- Pub ----'?- Cbr 4±1.5 30±5.7 3. Pod ----3- Ndt ----3- Hg ----3- Bpd 26 ± 28 26 ± 2.8 24 ± 9.5 4. Put ----3- Bk 19 ± 2.4 Trait involved Sw = swollen base, Fbc = cream flower bud Pus = purple stem, Cbr = cocoa brown pod color, Pub = purple pods Pd = purple peduncle, Ndt = nondeterminate, Hg = erect plant Bpd = branched peduncle Put = purple pod tips, Bk = grey black pod They used F 2 data from four crosses to estimate the recombination frequencies, which need further confirmation using backcross and F 3 data. Kehinde et al. (1997) studied the segregation pattern of 12 loci in F 2 and backcross populations and identified five linkage groups. Linkage group 1 comprised of five genes, Pg (nodal pigmentation), Pi (purple flower), Pc (smooth seed coat), Na (narrow eye), and Br (brown seed coat) with the probable order Pg----'?-Na----'?-Br----'?-Pc ----'?-Pf. The second linkage group was Bpd (branched peduncle)----'?-Bp (brown dry pod}---3Dhp (pod dehiscence). The third linkage group con- sisted of Crt (crinkled leaf) ----3-Pt (sessile leaf). The hastate leaf (Ha) and septafoliate leaf (spt) showed independent segregation from others showing different linkage groups (4 and 5). A DNA marker based (RFLP and RAPD analysis) genetic map of cowpea was first reported by Fatokun et al. (1993). This contained 92 markers with a span of 717 cM of the genome from a cross between IT84S-2246-4 and TVNu 1963. Recently, Menendez et al. (1997) published another genetic map consisting of 181 loci, compris- ing 133 RAPDs, 19 RFLPs, 25 AFLPs, three morphological or classical markers, and a biochemical marker (dehydrin). These markers identified 12 linkage groups spanning 972 cM with an average distance of 6.4 cM between markers. Myers et al. (1996) identi- fied one RFLP marker, to be tightly linked to the aphid resistance gene (Raci). Recently, Ouedraogo et al. (2001) have identified three AFLP markers tightly linked to the Striga resistance gene Rsg 2-1 and six AFLP markers linked to the Striga resistance gene Rsg 4-3 setting the stage for marker-assisted selection (MAS) in cowpea. However, a lot of work is needed to saturate the genetic map of cowpea and identify more markers before routine MAS can be practised. 10 Digitized by Google Recent genetic studies in cowpea References Abadassi, IA., B.B. Singh, TA.O. Ladeinde, SA. Shoyinka, andA.M. Emechebe. 1987. Inheritance of resistance to brown blotch, Septoria leaf spot and scab in wild Vigna (Vigna vexillata). Indian Journal of Genetics 47: 299-303. Adu-Dapaah, H.K., B.B. Singh, and C.A. Fatokun. 1999. A fasciated mutant in cowpea (Vigna unguiculata (L.). Acta Agronomica Hungarica 47: 371-376. Aliboh, VO., O.B. Kehinde, and I. Fawole. 1996. Inheritance ofleafmark, pod dehiscence and dry pod color in crosses between wild and cultivated cowpeas. African Crop Science Journal 5(2): 283-288. Arshad, M., M. Bashir,A. Sharif, and B.A. Malik. 1998. Inheritance of resistance in cowpea (Vigna unguiculata [L.] Walp) to blackeye cowpea mosaic potyvirus. Pakistan Journal of Botany 30(2): 263-270. Arvindhan, S. and L.D.V Das. 1996. Heterosis and combining ability in fodder cowpea for green fodder and seed yield. Madras Agricultural Journal 83: 11-14. Begemann, F., I Heller, and I Mushanga. 1997. An experiment to cross bambara groundnut and cowpea. Pages 135-137 in Bambara groundnut. Proceedings of Workshop on conservation and improvement of Bam bar a groundnut, 14-16 November 1995, Harare, Zimbabwe. Bhor, T.I, N.S. Kute,A.D. Dumbre, and N.D. Sarode. 1997. Heterosis and inbreeding depression in cowpea. Indian Journal of Agricultural Research 31: 122-126. Bhushana, H.O., K.P. Viswanatha, P.A Runachala, and G.K. Halesh. 2000. Heterosis in cowpea for seed yield and its attributes. Crop Research (Hisar) 19: 277-280. Biradar, B.D., IV Goud, and S.S. Pati!o 1997. Differential expression of pleiotropic genes for pigmentation in cowpea (Vigna unguiculata [L.] Walp). Crop Research (Hisar) 14(2): 233- 242. Capinpin, IM. 1935. A genetic study of certain characters in varietal hybrids of cowpea. Philippine Journal of Science 57: 149-164. Cardinali, A., V Linsalata, P. Perriono, V Lattanzio, R Brouillard, M. Jay, andA. Scalbert. 1995. Pages 375-376 in Chemotaxonomy of wild Vigna species as potential sources of resistance to insects. Polyphenols 94: 17th International Conference, Palma de Mallorca, Spain. Damarany,A.M. 1994. Estimates of genotypic and phenotypic correlation, heritability and potency of gene set in cowpea (Vigna unguiculata [L.] Walp). Assuit Journal of Agricultural Science 25: 1-8. F atokun, C.A., D. Darinsh, D.1. Menancio-Hautea, and N.D. Young. 1993. A linkage map for cowpea (Vigna unguiculata) [L.] Walp.) based on DNA markers. Pages 256-258 in Genetic maps, edited by S.J. O'Brien. Locus Maps of Complex Genomes. VI Edition. Cold Spring Harbor Laboratory Press, USA. F ery, RL. 1981. Genetics of Vigna. Pages 311-394 in Horticultural reviews, edited by J. Janick. A VI Publishing, Westport, CT, USA. F ery, RL. 1985. The genetics of cowpea. A review ofthe world literature. Pages 25--62 in Cowpea research, production and utilization, edited by S.R. Singh and K.O. Rachie. John Wiley and Sons, Chichester, UK. Fery, RL. and B.B. Singh. 1997. Cowpea genetics: a review of recent literature. Pages 13-29 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell and L.E.N. Jackai. Copublication ofilIA and JIRCAS. I1TA, Ibadan, Nigeria. Githiri, S.M., P.M. Kimani, and RS. Pathak. 1996. Linkage relationships among loci controlling morphological traits in cowpea (Vigna unguiculata [L.] Walp.) Euphytica 92(3): 307-311. Gomathinayagam, P., S.G. Ram, R Rathnaswamy, and N.M. Ramaswamy. 1998. Interspecific hybridization between Vigna unguiculata (L.) Walp. and V vexillata (L.). A. Rich, through in vitro embryo culture. Euphytica 102(2): 203-209. 11 Digitized by Google Cowpea genetics and breeding Harland, S.C. 1919a. Inheritance of certain characters in the cowpea (Vigna sinensis). Journal of Genetics 8: 101-132. Harland, S.C. 1919b. Notes on inheritance in cowpea. Agricultural News, Barbados 18: 20. Harland, S.C. 1920. Inheritance of certain characters in the cowpea (Vigna sinensis). Journal of Genetics 10: 193-205. Ishiyaku, M.F. and B.B. Singh. 2001. Inheritance of shortday-induced dwarfing in photosensitive cowpea. African Crop Science Journal 9(2): 1-8. Ismail, A.M., A.E. Hall, and T.J. Close. 1997. Chilling tolerance during emergence of cowpea associated with a dehydrin and slow electrolyte leakage. Crop Science 37: 1270-1277. Ismail, A.M., A.E. Hall, and T.J. Close. 1999. Allelic variation of a dehydrin gene co-segregates with chilling tolerance during seedling emergence. Proceedings of National Academy of Science 96: 13566-13570. Joshi, S.S., R. Sreekantaradhy, K.G. Shambulingappa, D.P. Jagannatha, and C.V Jayaramu. 1994. Inheritance of a few qualitative characters in cowpea (Vigna unguiculata [L.] Walp.) Crop Research (Hisar) 8(2): 330-336. Kehinde, O.B., G.O. Myers, and I. Fawole. 1997. Analysis of genetic linkage in the cowpea Vigna unguiculata. Pertanika Journal of Tropical Agricultural Science 20(1): 75-82. Kehinde, O.B. and MA. Ayo-Vaughan. 1999. Genetic control of seed coat texture in cowpea, Vigna unguiculata (L.) Walp. Tropical Agricultural Research and Extension 2(1): 7-9. Kolhe, A.K. 1970. Genetics studies in Vigna sp. Poona Agricultural College Magazine 59: 126- 137. Krishnaswamy, N., K.K. Nambiar, andA. Mariakulandai. 1945. Studies in cowpea (V unguiculata [L.] Walp). Madras Agricultural Journal 33: 145-160. Luchi, S., K. Yamaguchi-Shinozaki, T. Urao, T. Terao, and K. Shinazaki. 1996. Novel drought inducible genes in the highly drought-tolerant cowpea: cloning of DNAs and analysis of the expression of the corresponding genes. Plant and Cell Physiology 37: 1073-1082. Mai-Kodomi, Y., B.B. Singh, O. Jr. Myers, J.H. Yopp, J.P. Gibson, and T. Terao. 1999. Inheritance of drought tolerance in cowpea. India Journal of Genetics and Plant Breeding 59: 317-323. Menendez, C.M. andA.E. Hall. 1996. Heritability of carbon isotope discrimination and correlations with harvest index in cowpea. Crop Science 36(2): 233-238. Menendez, C.M,A.E. Hall, and P. Gepts. 1997. A genetic linkage map of cowpea developed from a cross between two inbred domesticated lines. Theoretical and Applied Genetics 95: 1210- 1217. Mortensen, JA. and W.H. Brittingham. 1952. The inheritance of pod color in the southern pea, Vigna sinensis. Proceedings oftheAmerican Society of Horticultural Science sg.451-456. Myers, G.O., CA. Fatokun, and N.D. Young. 1996. RELP mapping of an aphid resistance gene in cowpea. Euphytica 91: 181-187. Nakawuka, C.K. and E. Adipala. 1997. Identification of sources and inheritance of resistance to Sphaceloma scab in cowpea. Plant Disease 81: 1395-1399. Odeigah, P.G.C., A.O. Osanyin Peju, and G.O. Myers. 1996. Induced male sterility in cowpea. Journal of Genetics and Plant Breeding 50: 171-175. Ouedraogo, J.T., V Maheshwari, D.K. Berner, C.A. St-Pierre, F. Belize, and M.P. Timko. 2001. Identification of AFLP markers linked to resistance of cowpea to parasitism by Striga gesnerioides. Theoretical and Applied Genetics 102: 1029-1036. Pasquet, R.S. 1999. Genetic relationship among subspecies of Flgna unguiculata (L.) Walp. based on allozyme variation. Theoretical and Applied Genetics 98: 1104-1119. Ponmariammal, T. and L.D. Vijendra Das. 1996. Heterosis for fodder yield in cowpea. Madras Agricultural Journal 83: 658--659. Rangaiah, S. 1997. Inheritance of resistance to Uromyces phaseoli in Vigna unguiculata (L.) Walp. Crop Improvement 24(2): 251-252. 12 Digitized by Google Recent genetic studies in cowpea Roberts, P.A., W.C. Matheswand, and J.D. Ehlers. 1996. New resistance to virulent root-knot nematodes linked to Rk locus in cowpea. Crop Science 36: 889-894. Rodriguez, I., M.G. Rodriguez, L. Sanchez, and A. Iglesias. 1996. Expression of resistance to Meloidogyne incognita in cowpea cultivars. Revista de Proteccion Vegetal 11: 63-65. Ryerson, D.E. and M.C. Heath. 1996. Inheritance of resistance to the cowpea rust fungus in cowpea cultivar Calico Crowder. Canadian Journal of Plant Pathology 18(4): 384-391. Sangwan, RS. and G.P. Lodhi. 1995. Heterosis for grain characters in cowpea (Vigna unguiculata [L.] Walp.). Legume Research 18: 75-80. Sangwan, RS. and G.P. Lodhi. 1998. Inheritance of flower and pod color in cowpea (Vigna unguiculata [L.] Walp). Euphytica 102(2): 191-193. Saunders, A.R 1960a. Inheritance in the cowpea 2: seed coat color pattern; flower, plant and pod color. South African Journal of Agricultural Science 3: 141-142. Saunders,A.R 1960b. Inheritance in the cowpea 3: mutations and linkages. South African Journal of Agricultural Science 3: 327-348. Sen, N.K. and J.G. Bhowal. 1961. Genetics of Vigna sinensis (L.) savio Genetics 32: 247-266. Singh, B.B. and H.K. Adu-Dapaah. 1998. A partial male sterile mutant in cowpea. African Crop Science Journal 6: 97-101. Singh, B.B. and M.F. Ishiyaku. 2000. Genetics of rough seed coat texture in cowpea. Journal of Heredity 91: 170-174. Singh, K.B. and L.N. Jindla. 1971. Inheritance of bud and pod color, pod attachment and growth habit in cowpeas. Crop Science 11: 928-929. Sonnante, G., A.R. Piergiovanni, Q. Ng, and P. Perrino. 1996. Relationship of Vigna unguiculata, Vigna vexillata and species of section Vigna based on isozyme variation. Genetic Resources and Crop Evolution 43: 157-165. Tumwegamire, S., P.R Rubaihayo, and E. Adipala. 1998. Genetics of resistance to Sphaceloma scab of cowpea. African Crop Science Journal 6(3): 227-240. Tyagi, D.K. and H.S. Chawala. 1999. Effect of season and hormones on crossability barriers and in vitro hybrid development between Vigna radiata and Vigna unguiculata. Acta Agronomica Hungarica 47: 147-154. Uguru, M.1. 1995. Inheritance of color patterns in cowpea (Vigna unguiculata [L.] Walp.). Indian Journal of Genetics and Plant Breeding 55(4): 379-383. Uguru, M.1. and AA. Ngwuta. 1995. Genetics and linkage relationships of anthocyanin genes in vegetable cowpea. Biologiches Zentralbaltt. 114: 273-278. Pathmanathan, u., RP. Ariyanayagam, and S.O. Haque. 1997. Genetic analysis of yield and its components in vegetable cowpea (Vigna unguiculata [L.] Walp). Euphytica 96(2): 207-213. Vale, C.C. do, JA. Lima, do Vale, C.C. 1995. The inheritance of immunity in Vigna unguiculata Macaibo to cowpea severe mosaic virus. Fitopatologia Brasileira 20(1): 30-32. Venugopal, R 1998. Inheritance in cowpea (Vigna unguiculata [L.] Walp. V) pod characters. Crop Research (Hisar) 15(1): 77-84. Venugopal, Rand J.V Goud. 1996. Inheritance in cowpea (Vigna unguiculata [L.] Walp). III Floral characters. Mysore Journal of Agricultural Sciences 30(1): 14-20. Venugopal, Rand J.V Goud. 1997. Inheritance of pigmentation of cowpea. Current Science 3: 141-142. Venora, G. and S. Padulosi. 1997. Karyotypic analysis of wild taxa of Vigna unguiculata (L.) Walpers. Caryologia 50: 125-138. 13 Digitized by Google 1.2 Breeding cowpea for tolerance to temperature extremes and adaptation to drought A.E. Hall" A.M. Ismail" J.D. Ehlers" K.O. Marfo2, N. Cisse3, S. Thiaw3, and T.J. Close1 Abstract Cowpea exhibits incomplete emergence when soil temperatures are below 19°C. Chilling tolerance at emergence appears to be conferred by a dominant gene encod- ing a dehydrin protein. Seed immunoblot assays facilitate breeding for this trait. Cowpea can exhibit floral bud suppression and low pod set when nighttemperatures are higher than 17°C. Heat-tolerance genes enhanced flowering, pod set, and grain yield under hot subtropical conditions but with no difference between tolerant and susceptible lines in hottropical conditions. In glasshouse studies, heat-tolerant lines had high yields under both long and short days but heat-susceptible lines only exhibited low yields in long days. Delayed leaf senescence can enhance drought adaptation of early cowpea cultivars by enabling them to produce a greater second pod flush if the first flush is damaged by drought. Genetic studies demonstrated that combining the delayed leaf senescence and heat tolerance traits could breed cultivars with enhanced yield stability. Introduction In subtropical zones, such as the San Joaquin Valley of California, cowpea is sown in the spring (Hall and Frate 1996). Early sowing can result in high grain yields if it enables the crop to escape hot summer weather that can hinder reproductive development (Hall 1992). If sowing is too early, however, and the soil is cooler than 19°C, chilling damage can cause slow and incomplete emergence (Ismail et al. 1997). This paper will discuss research showing that breeding cowpea for both chilling tolerance at emergence and heat tolerance at flowering can partially solve these problems for subtropical zones. We also will discuss studies of whether genes that confer heat tolerance during reproductive development in subtropical zones have any adaptive value for cowpea grown in tropical zones in West Africa. Cowpea in the Sahelian (annual rainfall of about 200 to 500 mm) and dry savanna (annual rainfall of about 500 to 700 mm) zones of West Africa can experience both heat and drought stress (Hall et al. 1997 a). Cowpea cultivars that begin flowering early can escape drought in some locations and years and produce useful yields of grain. Unfortunately, the early cowpea cultivars tend to be very sensitive to droughts that occur during early stages of reproductive development (Thiaw et al. 1993). A delayed-leaf-senescence (DLS) trait has the potential to enhance the drought adaptation of cowpea in the dry savanna 1. Department of Botany & Plant Sciences, University of Cali fomi a, Riverside, CA 92521, USA. 2. Savanna Agricultural Research Institute, PO Box 483, Tamale, Ghana. 3. Centre National de Recherches Agronomiques, BP 53 Bambey, Senegal. 14 Digitized by Google Breeding cowpea for tolerance to temperature extremes and adaption to drought zone and wetter part of the Sahelian zone. In California, the DL S trait had been shown to enhance the ability of early flowering cowpea to recover after an early drought and produce a compensatory second flush of pods in some field conditions (Gwathmey and Hall 1992). The DLS trait also had been shown to enhance the second flush of pods in one tropical location where lines were tested in the wetter part of the Sahelian zone (Hall et al. 1997b). We will discuss research on whether there is an interaction between the DLS and heat-tolerance genes because they have contrasting effects on the partitioning of carbohydrate in the plant. Chilling tolerance during emergence Warm-season annual crops exhibit slow and incomplete emergence when subjected to cool soils. The threshold soil temperature where cowpea exhibits incomplete emergence is quite high at about 19°C (Ismail et al. 1997). Soil temperatures below 19°C often occur in spring in the San Joaquin Valley of California where cowpea is grown (soil and air temperature data for many locations in California can be obtained at the web site www.ipm.ucdavis.edu). A cowpea line with chilling tolerance was found and it was hypothesized that the chilling tolerance is due to two independent and additive factors (Ismail et al. 1997). The factors are a specific dehydrin protein with a dominant nuclear effect and a maternal effect associated with slow electrolyte leakage from seed under chill- ing conditions. Slower electrolyte leakage indicates greater plasma membrane integrity. The dehydrin protein has been purified and partially characterized (Ismail et al. 1999a). The hypothesis concerning the contribution to chilling tolerance during emergence of the dehydrin protein has been confirmed using near isogenic lines and it was shown that the maternal electrolyte leakage effect is not cytoplasmically inherited (Ismail et al. 1999b). The phenotypic expression of the dehydrin has been mapped (Menendez et al. 1997) and the structural gene encoding the dehydrin maps to the same location (Ismail et al. 1999b). The dehydrin protein can be readily manipulated by classical breeding. An immunoblot assay of a chip taken from a single seed is used to detect the presence of the dehydrin protein and the seed still retains its ability to germinate (Ismail et al. 1999b). Using this assay technique, cowpea lines that combine the dehydrin and chilling tolerance during emergence with heat tolerance during reproductive development have been developed. Heat tolerance during reproductive development Six genetically similar pairs of lines that either have or do not have heat tolerance during reproductive development were bred at the University of California (UCR). A pedigree breeding approach was used with field screening for flower production and pod set in a very hot field environment (average maximum and minimum daily air temperatures in a weather station shelter of 43 and 24°C, respectively, for the first 60 days after sowing) as described by Hall (1992). The performance of these six pairs of lines has been evaluated in eight field environments in the subtropical zone of California that have contrasting temperatures but similar high levels of solar radiation and optimal management with complete irrigation (Ismail and Hall 1998). A subset of the data from this study is presented in Table 1. All of the heat-susceptible lines had much lower grain yields in the environments with average night temperatures higher than 17°C at flowering. The heat-tolerant lines had 394 to 554 kg/ha greater average grain yields than 15 Digitized by Google Cowpea genetics and breeding Table 1. Grain yields of six pairs of heat-tolerant and heat-susceptible cowpea lines grown with complete irrigation at three locations in the subtropical zone of California, USA, over 2 years with high levels of solar radiation and optimal management. Riverside with Coachella early sowing Shafter valley 1995&1996 1995 1996 1995&1996 --kglha-- Heat-tolerant lines 3086 3357 2492 894 Heat-susceptible lines 2976 3310 2098 340 Significance NS NS ** **** Daily minimum air temperature oct 14.6 16.4 17.4 23.7 ** and **** are significant at the 0.01 and 0.0001 levels whereas NS is not significant at the 5% level. +Average for the 3-week period beginning one week prior to the start of flowering. Source: From Tables 2 and 3 in Ismail and Hall 1998. the heat-susceptible lines in the hotter environments, but similar grain yields in the cooler environments. One ofthese heat-tolerant lines has been released as California Blackeye No. 27 (CB27) for use as a dry grain cultivar in California (Ehlers et al. 2000). Heat-tolerant lines were much shorter compared with heat-susceptible lines and this effect was more pronounced in hotter environments (Ismail and Hall 1998). In California, the semi -dwarf cultivar CB27 has a greater yield advantage over current standard height cultivars, such as CBS, when grown on rows 51 to 76 cm apart rather than the wide rows (about 102 cm apart) that are used by some growers (Ismail and Hall 2000). Row spacing in this study was 10 cm between plants in the row. In the relatively high night temperatures experienced in tropical zones, the heat-tolerant lines developed in California experience even more dwarfing than in subtropical zones. We have studied whether the heat-tolerance genes shown to be effective in the subtropical zone of California are also effective in tropical zones of West Africa where there is sub- stantial cowpea production. Daily minimum air temperatures were substantially higher in the savanna and Sahelian zones (Table 2) than the threshold of 17°C for causing damage to flower development and pod set of the heat-susceptible lines indicated by the studies of Ismail and Hall (1998) in Table 1. In all of the trials in West Africa, however, there was no significant difference in grain yield between the averages of the six heat-tolerant and six heat-susceptible lines (Table 2). The contrasting performance of the 2 sets of lines may be explained by the longer day lengths experienced by plants in California compared with West Africa. Controlled- environment studies had shown that high night temperatures could be more damaging to cowpea in long-day than in short-day conditions (Mutters et al. 1989). Studies in which the heat tolerance of contrasting cowpea lines were evaluated in greenhouses with high night temperature and either long or short days (Ehlers and Hall 1998) have provided some explanations for the contrasting performance of the lines in different field conditions. A subset of the data is presented in Table 3. Heat-tolerant California lines had greater grain yields than heat-susceptible California lines in hot long-day greenhouse conditions (Table 3) as had been observed in hot long-day field conditions in Shafter in 1996 and in the Coachella Valley in both 1995 and 1996 (Table 1). In contrast, in hot, short-day greenhouse conditions, heat-tolerant and heat-susceptible California lines had similar 16 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) ..... 'I Table 2. Grain yields of the same six pairs of heat-tolerant and heat-susceptible cowpea lines used for the study in Table 1 when grown under rainfed conditions at three locations in the dry Savanna zone (northern Ghana) and three environments in the Sahelian zone (peanut basin of Senegal) with optimal management. Northern Ghana Nyankpala Damongo Manga 1998 1998 1998 kglha Heat-tolerant lines 1241 869 564 Heat-susceptible lines 1351 895 559 Significance in all cases differences between sets of lines were not significant at the 5% level Air temperature DC' 25.7 20.7 22.5 Sowing date 29 July 28 June 2 August Senegal Bambey Thilmakha 1 998 1 999 1998 ---kglha---- 1165 1101 1839 1894 24.4 23.5 5 August 14 July 1370 1369 25.3 29 July 'Average daily minimum air temperature for the three-week period beginning one week prior to the start of flowering, using Louga data for Thilmakha. O:l ~ rt) ~ 01:i 8 ~ ~ 0' .... 6" iii" Ql ::J @ 6" l ~ c: ~ ~ " ~ ffi II> ::J c.. II> 2- "tl g. ::J 6" Q- o c: l Cowpea genetics and breeding grain yields and they were high (Table 3). In hot, short-day field conditions in Africa also heat-tolerant and heat-susceptible California lines had similar grain yields but they were moderate (Table 2). The moderate level of the grain yields may be explained by the fact that these California lines are not well adapted to West Africa. For example, the Califor- nia lines are highly susceptible to wet and dry pod rots. The heat-tolerant parents (prima and TVu4552) used in developing the heat-tolerant California lines have been shown to have high pod set under hot, short-day conditions. Prima was shown to have higher grain yield than IT84S-2246 due to its greater pod set in studies in hot growth chambers with l2-hour days (Craufurd et al. 1998). In hot, short-day conditions in screenhouses at Kano, West Africa, TVu4552 has exhibited much greater pod set and grain production than many other cowpea accessions (personal communication B. B. Singh, March 2001). The genes in Prima, TVu4552, and the California lines that confer heat tolerance during pod set probably can enhance pod set in tropical conditions but they need to be combined with additional genes that confer local adaptation. Also effects of these genes on grain yield may not be as large in hot, short-day tropical environments as has been observed in hot subtropical zones. Some of the African cultivars and lines that were studied in greenhouses by Ehlers and Hall (1998) also had heat tolerance in that they exhibited high grain yields in hot, short-day conditions (Table 3). These heat-tolerant materials included landraces that evolved in the hot Sahelian zone (58-57 and Suvita 2) and cultivars (Mouride and TN88-63) and breeding lines (B89-600 from Senegal and the IT lines developed by I1TA at Kano, Nigeria) that had been selected based on grain yield in hot tropical conditions. The heat-susceptible African lines in Table 3 also include some that had been selected for high grain yield in hot, short-day conditions (Melakh, N'diambour, and Bambey 21). These data suggest that selecting for high grain yield in hot parts of Africa is not always effective in incorporating Table 3. Grain yields of contrasting cowpea cultivars and lines in a greenhouse with high night temperature (day/night 36/27 ·C) at Riverside, CA, USA under summer (long-day) or spring (short-day) conditions. Long days Short days ___ glplant ___ _ Heat-tolerant California lines (4) 26 (22 to 32) 47 (42 to 54) H8-9-3, H8-14-13, 518-2, H8-8-4+ Heat-susceptible California cultivars and lines (7) 1 (0 to 3) 40 (36 to 48) H14-10-10, H8-8-31, H35-5-6, H8-14-18, CBS, CB46, H8-14-19-1 Heat-tolerant African cultivars and lines (10) 1 (0 to 6) 44 (38 to 53) Mouride, B89-600, IT89KD-252, IT89KD-245, IT88DM-400, IT89KD-1 07-5, TN88-63, 58-57, IT89KD-355, Suvita 2+ Heat-susceptible African cultivars and lines (12) 2 (0 to 9) 21 (10t030) Melakh, IT82D-889, N'diambour, IT84S2049, Bambey 21, TVx12-01 e, IT86D-719, KN-1, Bambey 23, IT82D-375, Sumbrisogle, IT85F-2614+ +Cultivars and lines listed in rank order with the first having the highest yield in short days. Source: From Table 6 in Ehlers and Hall 1998. 18 Digitized by Google Breeding cowpea for tolerance to temperature extremes and adaption to drought heat tolerance. Field screening for heat tolerance is difficult in the Sahelian and savanna zones because of biotic stresses, such as flower thrips, that damage floral development and pod set in a manner that is similar to the effects of heat stress. Progress has been made, however, in screening cowpea for reproductive-stage heat tolerance in Africa by growing them in screenhouses during the dry season in Kano, Nigeria where daily minimum air temperatures vary from 24 to 27°C and daily maximum air temperatures vary from 38 to 42 °C (Singh 1998). Methods for screening to detect reproductive-stage, heat-tolerance genes that are more efficient than field screening have been sought. Ismail and Hall (1999) have suggested that measurements of plasma membrane thermo stability based upon electrolyte leakage from leaf disks has the potential to be used for screening for reproductive-stage, heat-tolerance genes. Recent studies by S. Thiaw andA.E. Hall indicate, however, that effective screen- ing for plasma membrane thermostability may require that plants be grown in long-day conditions. Also it may be necessary to put leaf disks in aerated solutions when measuring electrolyte leakage and the differences in electrolyte leakage between genotypes differing in reproductive-stage heat tolerance may be small. The overall conclusion is that the reproductive-stage, heat-tolerance genes discovered by Hall and associates can be effective in subtropical conditions and may be effective in the tropics under either long-day or short-day conditions providing, other stresses do not damage plant growth and development. Also note that with sowing prior to late June in the Sahelian and dry savanna zones of West Africa, day lengths may be long enough to enhance the detrimental effects of heat on reproductive development. Empirical breeding studies indicate that heat-tolerance genes may be useful in West Africa. For example, Marfo has bred a cultivar for northern Ghana, SuI 518-2, using a heat-tolerant line from California as one of the parents with some initial screening for heat tolerance but the heat-tolerance of SuI 518-2 under Ghanaian conditions has not yet been confirmed. Heat tolerance interaction with delayed-leaf-senescence Reproductive-stage, heat-tolerance genes cause greater partitioning of carbohydrates to pods (Ismail and Hall 1998), whereas, the delayed-leaf-senescence (DLS) trait is associ- ated with greater partitioning of carbohydrate to stem bases (Gwathmey et al. 1992) and also probably to roots. Studies were conducted to test: (l) whether the DLS trait reduces first-flush yields and thus reduces the beneficial effects of the heat-tolerance trait; and (2) whether the heat-tolerance trait enhances senescence after the first flush of pods is produced thereby reducing the expression of the DL S trait (Ismail et al. 2000). A cross was made between a heat-tolerant parent and a DLS parent and then four sets of lines were selected that either have or do not have the DLS and heat-tolerance traits (Ismail et al. 2000). It was shown that the DLS trait can be effectively selected beginning with F 3 families pro- viding a field nursery is used that has a senescence inducing soil environment. There is a tendency for senescence inducing soil conditions to develop in fields where cowpea has been grown for several years, even with alternate year rotation to other crop species, due possibly to the build up of a soilborne disease (Ismail et al. 2000). Individual plants with DLS were selected from families where most plants exhibited DLS. Selection for heat tolerance was done in field nurseries in extremely hot field and greenhouse environments and involved selecting plants for flower production and pod set (Hall 1992). Performance of the four sets of lines in a hot field environment is described in Table 4. The heat-tolerance trait enhanced grain yield by a substantial amount, 886 kg/ha, whereas 19 Digitized by Google Cowpea genetics and breeding presence of the DLS trait did not have a significant effect on the first-flush grain yield of the heat-tolerant lines (there was a nonsignificant decrease of 295 kg/ha). If this decrease in yield is real, it does not represent a large penalty in that the DLS trait has the potential to increase second-flush grain yield by up to 2000 kg/ha (Ismail and Hall 1998). Performance of the four sets of lines was evaluated in a soil environment where there was substantial death of non-DLS lines after producing the first flush of pods (73% of the plants died compared with < 1% for DLS lines). The presence of the heat-tolerance genes did not have a significant effect on the proportion of plants that died (nonsignificant increases of 10 percentage points in non-DLS lines and I percentage point in non-DLS lines occurred as shown in Table 5). The overall conclusions are that (1) the DLS trait can greatly enhance plant survival after the first flush of pods is produced and may only cause a small decrease in first-flush grain yield; and (2) the heat-tolerance trait can substantially increase first-flush grain yield and may only slightly enhance the tendency for premature plant death in non-DLS lines with no effect on lines having the DLS trait. Table 4. First-flush grain yields of four sets of lines with and without heat-tolerance and with and without the delayed-leaf-senescence trait. Delayed-Ieaf- Senescent senescence lines lines Average kglha Heat-tolerant lines 3168 3463 3316 Heat-susceptible lines 2248 2613 2430 Average 2708 3038 The heat-tolerance effect was very highly significant whereas the delayed-leaf-senescence and interaction effects were not significant at the 5% level and the CV was 19.3%. Note: Data are average values for 10 lines per set from an experiment at Shafter, CA, USA in 1998 (Ismail etal. 2000). Table 5. Percentage of plants that died after producing the first flush of pods for four sets of lines with and without heat-tolerance and with and without the delayed-leaf-senes- cence trait. Heat-tolerant lines Heat-susceptible lines Delayed-leaf - senescence lines 1 o Percentage of plants that died Senescent lines 78 68 The delayed-leaf-senescence effect was very high Iy significant whereas the heat-tolerance and interaction effects were not significant at the 5% level. Note: Data are average values for 10 lines per set from an experiment at Riverside, CA, USA in 1998 (Ismail et al. 2000). Conclusions Cowpea cultivars can be bred that combine chilling tolerance at emergence with heat tolerance during flowering and pod set. These cultivars could have enhanced yield stability in subtropical zones such as those in California. Cowpea cultivars can be bred that combine early flowering with heat tolerance during flowering and pod set and 20 Digitized by Google Breeding cowpea for tolerance to temperature extremes and adaption to drought delayed-leaf-senescence. These cultivars might have enhanced yields and yield stability in subtropical zones and tropical zones, such as the drier part of the savanna zone and the wetter part of the Sahelian zone in West Africa. References Craufurd, P.Q., M. Bojang, T.R Wheeler, and RJ. Summerfield. 1998. Heat tolerance in cowpea: effect oftiming and duration of heat stress. Annals of Applied Biology 133: 257-267. Ehlers, J.D. andA.E. Hall. 1998. Heat tolerance of contrasting cowpea lines in short and long days. Field Crops Research 55: 11-21. Ehlers, J.D.,A.E. Hall, P.N. Patel, P.A. Roberts, and W.C. Matthews. 2000. Registration of Calif or- nia Blackeye 27. Crop Science 40: 854-855. Gwathmey, C.O. andA.E. Hall. 1992. Adaptation to midseason drought of cowpea genotypes with contrasting senescence traits. Crop Science 32: 773-778. Gwathmey, C.O.,A.E. Hall, and MA. Madore. 1992. Pod removal effects on cowpea genotypes contrasting in monocarpic senescence traits. Crop Science 32: 1003-1009. Hall, A.E. 1992. Breeding for heat tolerance. Plant Breeding Reviews 10: 129-168. Hall,A.E. and CA. Frate. 1996. Blackeye bean production in California. University of California Division of Agricultural Science Publications 21518.23 pp. Hall, A.E., B.B. Singh, and J.D. Ehlers. 1997a. Cowpea breeding. Plant Breeding Reviews 15: 215-274. Hall,A.E., S. Thiaw,A.M. Ismail, and J.D. Ehlers. 1997b. Water-use efficiency and drought adap- tation of cowpea. Pages 87-98 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropi- cal Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Ismail,A.M. andA.E. Hall. 1998. Positive and potential negative effects of heat-tolerance genes in cowpea. Crop Science 38: 381-390. Ismail,A.M. andA.E. Hall. 1999. Reproductive-stage heat tolerance, leaf membrane thermostabil- ity and plant morphology in cowpea. Crop Science 39: 1762-1768. Ismail, A.M. and A. E. Hall. 2000. Semidwarf and standard-height cowpea responses to row spac- ing in different environments. Crop Science 40: 1618-1623. Ismail, A.M., A.E. Hall, and T.J. Close. 1997. Chilling tolerance during emergence of cowpea associated with a dehydrin and slow electrolyte leakage. Crop Science 37: 1270-1277. Ismail,A.M.,A.E. Hall, and T.J. Close. 1999a. Purification and partial characterization of a dehydrin involved in chilling tolerance during seedling emergence of cowpea. Plant Physiology 120: 237-244. Ismail, A.M., A.E. Hall, and T.J. Close. 1999b. Allelic variation of a dehydrin gene co-segregates with chilling tolerance during seedling emergence. Proceedings of the National Academy of Science 96: 13566-13570. Ismail, A.M., A.E. Hall, and J.D. Ehlers. 2000. Delayed-leaf-senescence and heat-tolerance traits mainly are independently expressed in cowpea. Crop Science 40: 1049-1055. Menendez, C.M., A.E. Hall, and P. Gepts. 1997. A genetic linkage map of cowpea (Vigna unguic- ulata) developed from a cross between two inbred, domesticated lines. Theoretical and Applied Genetics 95: 1210-1217. Mutters, RG., A.E. Hall, and P.N. Patel. 1989. Photoperiod and light quality effects on cowpea floral development at high temperatures. Crop Science 29: 1501-1505. Singh, B.B. 1998. Screening for heat tolerance. Page 42 in Project 11 Cowpea--cereals systems improvement in the dry savannas. Annual Report 1998, I1TA, Ibadan, Nigeria. Thiaw, S., A.E. Hall, and D.R. Parker. 1993. Varietal intercropping and the yields and stability of cowpea production in semiarid Senegal. Field Crops Research 33: 217-233. 21 Digitized by Google 1.3 Recent progress in cowpea breeding B.B. Singh \ J.D. Ehlers2, B. Sharma3, and F.R. Freire Filho4 Abstract Considerable progress has been made in breeding improved cowpea varieties in the last five years. The major breeding objectives were to develop high yielding cowpea varieties for sole cropping as well as intercropping with acceptable seed types and resistance to major diseases, insect pests, nematodes, and the parasitic plants Striga and Alectra and tolerance to heat and drought. Good progress was also made in breeding early maturing grain type, dual purpose, and fast growing fodder type cowpea varieties. The informal network of world cowpea researchers catalyzed by I1TA and the Bean/Cowpea Collaborative Research Support Program has been very effective in evaluating and selecting improved cowpea varieties for a wide range of environments. As a consequence, total world cowpea production has substantially increased. Importance Cowpea is an important food legume and an essential component of cropping systems in the drier regions of the tropics covering parts of Asia and Oceania, the Middle East, southern Europe, Africa, southern USA, and Central and South America. Being a fast growing crop, cowpea curbs erosion by covering the ground, fixes atmospheric nitrogen, and its decaying residues contribute to soil fertility. Cowpea is consumed in many forms: the young leaves, green pods, and green seeds are used as vegetables; dry seeds are used in various food preparations; and the haulms are fed to livestock as nutritious supplement to cereal fodder. In West and Central Africa, cowpea is of major importance to the liveli- hoods of millions of people providing nourishment and an opportunity to generate income. Trading fresh produce and processed food and snacks provide rural and urban women with the opportunity for earning cash income and, as a major source of protein, minerals, and vitamins in daily diets, it positively impacts on the health of women and children. The bulk of the diet of rural and urban poor Africa consists of starchy food made from cassava, yam, plantain and banana, millet, sorghum, and maize. The addition of even a small amount of cowpea ensures the nutritional balance of the diet and enhances the protein quality by the synergistic effect of high protein and high lysine from cowpea and high methionine and high energy from the cereals. This nutritious and balanced food ensures good health and enables the body to resist infectious diseases and slow down their development. World production of cowpea Singh et al. (1997) estimated a world total of about 12.5 millionha grown to cowpea with a production of 3 million tonnes (t). The exact statistics are still not available but there 1. International Institute of Tropical Agriculture (lIT A), Kano Station, PMB 3112, Kano, Nigeria. 2. Dept. of Botany and Plant Sciences, University of California, Riverside, CA92521-0124, USA. 3. Indian Agricultural Research Institute, Pusa, New Delhi 110012, India. 4. EMBRAPA, CPAMN, Teresina Piaui, Brazil. 22 Digitized by Google Recent progress in cowpea breeding seems to be an increase in the area as well as production since 1997. The available data on area, production, and average yield of cowpea in 11 selected countries (Table 1) totals 11.3 million ha and 3.6 million 1. The estimated area and production in over 50 other countries in Asia, Africa, and Central and South America that grow cowpea would make a world total of over 14 million ha and 4.5 million 1. Nigeria is the largest producer and consumer of cowpea with about 5 million ha and over 2 million t production annually. Each Nigerian eats cowpea and the per capita consumption is about 25 to 30 kg per annum. Niger Republic is the next largest producer with 3 million ha and over 650000 t production. Northeast Brazil grows about 1.5 million ha of cowpea with about 491 558 t production that provides food to about 25 million people. In Brazil as a whole, per capita consumption of cowpea is about 20 kg annually. In southern USA, about 40000 ha of cowpea is grown with an estimated 45000 t annual production of dry cowpea seed and a large amount of frozen green cowpeas. India is the largest cowpea producer in Asia and together with Bangladesh, Indonesia, Myanmar, Nepal, Pakistan, Sri Lanka, Thailand, and other Far Eastern countries, there may be over 1.5 million ha under cowpea in Asia. There is a need to make concerted efforts to collect accurate statistics on cowpea area and production in different countries. Progress in cowpea breeding Recent reviews by Singh et al. (1997) and Hall et al. (1997) have described progress in cowpea breeding in different regions of the world. The aim of this paper is to update both articles. The International Institute of Tropical Agriculture (I1TA) continues to be the center for cowpea research. However, recently, cowpea improvement programs at the University of California, Riverside (USA) and Empresa Brasileira de Pes qui sa Agropecu- aria (EMBRAPA), Brazil have been strengthened and expanded. Significant research on various aspects of cowpea improvement is also being done in Burkina Faso, India, Mali, Nigeria, and Senegal, and to a lesser extent in a number of other countries. A brief review of the progress made is presented. Breeding methods Singh (1996) reported the results of an experiment conducted to ascertain whether segregat- ing populations such as F 2' F 3' F 4' F 5' and others should be grown under intercrop or sole Table 1. Major cowpea growing countries in the world (1999-2000). Area under Production Yield Country cowpea (ha) (t) (kg/ha) Nigeria 5050100 2 108000 417 Niger 3800000 650000 171 Brazil 1 500000 491 558 324 Mali 512455 113000 220 Tanzania 145000 46000 317 Myanmar 105000 100000 952 Uganda 64000 64000 1000 Haiti 55000 38500 700 USA 40000 45000 1000 Sri Lanka 15000 12 120 808 South Africa 13 000 5600 430 Total 11 299 555 3669778 324 Source: FAOSTAT and national reports. 23 Digitized by Google Cowpea genetics and breeding crop for selecting high yielding lines for intercropping. Two crosses involving IT89KD-374 and IT89KD-288 as local improved parents and IT90K-48-l, which is resistant to aphid, bruchid, thrips, and Striga and Alectra, were made in 1990 and F 2 seeds from the two populations were subdivided into two sets each. One set was grown in sole crop with two insecticide sprays and the other set was grown under intercropping with millet, without insecticide spray in 1991. The F 3 progenies selected from these populations were grown in sole crop and intercrop, respectively, maintaining separate sole crop and intercrop streams in 1992. Likewise F 4 progenies were grown in separate streams in 1993, F 5 progenies in 1994, and F 6 progenies in 1995. The standard pedigree method was followed to select desirable plant/progenies while evaluating F 2 to F 6 generations. The promising F 6 prog- enies were bulk harvested in 1995 and multiplied in the dry season for a yield trial under intercrop and sole crop in the 1996 crop season. A total of 52 F 6 lines selected from the segregating progenies of the two crosses advanced in sole crop and intercrop streams were yield tested along with eight checks, including the original parents as well as best local and improved checks. The trial included sole crop and a combination of I-row millet with I-row cowpea intercropped with and without spray of insecticide. The grain and fodder yields of the breeding lines selected under intercropping were significantly better than those selected under sole crop averaged over the two crosses. The mean grain yield of all the lines derived from the sole crop was 1149 kg/ha in sole-crop sprayed and 190 kg/ha in intercrop with no spray, compared to 1328 kg/ha and 265 kg/ha, respectively, of the lines derived from intercrop. This indicated that selection under intercropping without spray is more effective for higher yield than selection under sole crop. This may be due to greater stress and selective pressure under intercropping. In a comparative study of different breeding methods, the mean performance of F 3 progenies derived from single seed descent method was better than that of progenies developed via single plant selection for yield and yield components (Mehta and Zaveri 1997). Also, the broad-sense heritability was higher in the population developed through the single seed descent selection method. Vishwanathan and Nadarajan (1996) conducted G x E analysis of several cowpea varieties and they observed IT86D-l 056 and C04 cowpea varieties to be the most stable. Singh (2000) showed that by testing and selection of varieties at known hot spots for different diseases, insect-pests, and StrigaiAlectra, the genotype x environment interaction can be minimized to ensure stable performance of improved vari- eties over a wider range of environments. He also showed that by simultaneously testing and selecting under sole crop with only two sprays of insecticide, sole crop without spray and intercrop without spray, high yielding varieties with stable performance with little or no insecticide could be identified (Singh 1 999a, 2000). Diallel analysis of six cowpea genotypes and their F 1 hybrids revealed additive gene action for most of the quantitative traits including green fodder and total dry matter (ponmariammal and Das 1996). I nterspecific crosses Gomathinayagam et al. (1998) reported successful crosses between Vigna vexillata and Vigna unguiculata using embryo culture. They grew the F 1 hybrids and harvested F 2 seeds that were planted and then backcrossed to T ~ unguiculata. However, the resulting backcross seeds looked closer to Vigna vexillata. Therefore, there is a need to further examine the progenies obtained from this cross before ascertaining whether this was a true hybrid. Tyagi and Chawla (1999) also reported successful crosses between Vigna radiata and 24 Digitized by Google Recent progress in cowpea breeding Vigna unguiculata using in vitro culture techniques. Gibberellic acid treatment sustained the pods for 9-10 days, which were then used for embryo culture. About 10% of total embryos cultured resulted in plantlet formation. However, the authors did not report further growth and culture of these plantlets and therefore, it is not certain whether the crosses were true hybrids. Extreme wide crosses have been possible in other crop species using large numbers of pollinations along with newer techniques and perseverance. For example, Knyast et al. (2000) successfully crossed oat (var. Seneca 60 hexaploid) with maize pollen and added maize chromosomes to oat genome. This involved pollinating 60000 oat spikelets by maize pollen 48 hours after emasculation. The spikelets were sprayed with 100 ppm 2- 4-D about 48 hours after pollination. A total of 4300 embryos were isolated and cultured on modified M.S medium 14 days after pollination. From these only 379 F J plantlets developed successfully and these were transferred to pots of which 135 plants survived and had retained one or two maize chromosomes in addition to the complete oat haploid genome. From these four fertile disomic and two fertile monosomic oat-maize addition lines were developed, which are now being used to widen the genetic base of barley and to breed improved varieties with completely new traits. This study indicates that a very large number of pollinations and application of new embryo culture techniques along with a lot of patience is needed to achieve success in wide hybridization. Therefore, there is a need to continue efforts to cross Vigna vexillata and other Vigna species with cowpea to broaden its genetic base using new emerging techniques. Mutations Adu-Dapaah et al. (1999) reported a fasciated mutant and Singh and Adu-Dapaah (1998) reported a partial sterile mutant, both of which originated spontaneously. The fasciated mutant does not have much breeding value but the partial sterile mutant can be used for facilitating hybridization in cowpea. John (1999) reported 50 Kr of gamma rays to be most effective for inducing mutations in cowpea and Odeigah et al. (1996) obtained several male sterile mutants using gamma rays, ethyl methane sulphonate (EMS), and sodium azide. Saber and Hussein (1998) reported induced mutants using gamma rays showing resistance to rust. Gunasekaran et al. (1998) treated seeds of the cowpea variety C04 with gamma rays and ethidium bromide and analyzed M J and M2 progenies for different agronomic traits. They observed a great deal of variation in M2 population for different traits and further noticed that gamma rays were more effective in inducing mutation than ethidium bromide. Disease resistance Latunde-Dada et al. (1999) studied the mechanism of resistance to anthracnose in TVx 3236 cowpea. In this variety the initially injected epidermal cells underwent a hypersensi- tive response restricting the growth of the pathogen. The phytoalexins "kievitone" and "phaseollidin" accumulated more rapidly in the stem tissue of TVx 3236 compared to the sucessible variety. Lin et al. (1995) screened 131 cowpea varieties by artificially inoculating with Cercospora cruenta (Mycosphaerella cruenta) from which 15 varieties were identified immune and seven resistant. Singh et al. (1997), Singh (1998), and Singh (1999a) developed several cowpea lines with resistance to Cercospora, smut, rust, Septoria, scab, Ascochyta blight, and bacterial blight (Table 2). Some of the varieties, which showed multiple resistance 25 Digitized by Google Cowpea genetics and breeding Table 2. Sources of resistance to major diseases in cowpea. Diseases Anthracnose Cercospora Smut Rust (Uromyces) Septoria Scab Ascochyta Bacterial blight Sources of resistance TVx 3236 IT89KD-288, IT97K-l021-15 IT97K-463-7, IT97K-478-10 IT97K-l069-8, IT97K-556-4 IT97K-556-4, IT95K-l 090-12 IT95K-l 091-3, IT95K-ll 06-6 IAR-48, IT97K-506-6 IT97K-l042-8, IT97K-569-9 IT97K-556-4, IT97K-l069-8 IT95K-238-3, IT97K-819-118 IT90K-277 -2, IT97K-l 021-1 5 IT96D-610, IT86D-719 TVu 12349, TVul1761, IT95K-398-14 IT90K284-2,IT95K-l090-12 IT97K-l 021-15, IT98K-205-8 IT98K-476-8, IT97K-819-118, IT95K-193-12 TVu 1234, IT95 K-l 090-12, IT98K-476-8, IT97K-l069-8 TVx 3236, IT95K-398-14 IT97K-l 021-1 5, IT95 K-1133-6 TVu 11761 IT95K-398-14, IT95K-193-12 IT81O-1228-14, IT95K-1133-6 IT97K-556-4, IT97K-l069-8, IT90K-284-2, IT91 K-93-1, IT91 K-118-20 were IT97K-102l-l5, IT97K-556-4, and IT98K-476-8. Wydra and Singh (1998) screened 90 cowpea breeding lines and identified IT90K-284-2, IT9lK-93-1O, and IT9lK-118-20 to be completely resistant to three virulent strains of bacterial blight. Eight varieties were resistant to two strains and two varieties were resistant to one strain. All the remaining variet- ies were susceptible to bacterial blight. Santos et al. (1987) screened 156 cowpea varieties under field infestation with smut and identified three highly resistant ones. Nakawuka and Adipala (1997) identified Kvu 46, Kvu 39, and Kvu 454 to be resistant to scab in Uganda. Rodriguez et al. (1997) found L-198 and CNx 377-lE to be resistant to Macrophomina. Uday et al. (1996) identified V-265 also to be resistant to Macrophomina. In an interesting study, Zohri (1993) artificially inoculated 16 cowpea varieties with Aspergillus flavus to monitor aflatoxin production. He found that two cowpea varieties from I1TA, IT82E-16 and IT8lD-l 032, did not support Aspergillus growth and therefore no aflotoxin production was observed on these varieties. This indicates the possibility of breeding for resistance to Aspergillus flavus in cowpea. Resistance to nematodes Several sources of resistance to nematodes were identified including some of the improved breeding lines with high yield potential (Rodriguez et al. 1996; Roberts et al. 1996,1997; 26 Digitized by Google Recent progress in cowpea breeding Fery and Dukes 1995a; Ehlers et al. 2000a; and Singh 1998). Some of the varieties with high yield and nematode resistance are IT849-2049, IT89KD-288, IT86D-634, IT87D- 1463, IT95K-398-l4, IT96D-772, IT96D-748, IT95K-222-5, IT96D-61O, IT87K-8l8-l8, and IT97K-556-4. Among these varieties, IT89KD-288 was found to be resistant to four strains of Meloidogyne incognita in USA (Ehlers et al. 2000a). Singh et al. (1996, 1998a) found IT89KD-288 to be high yielding and highly resistant to nematodes in the trials conducted at Kano (Nigeria), where nematode attack is very severe in the dry season plant- ing with irrigation. IT89KD-288 was taken by one farmer in 1994 and through farmer to farmer diffusion, it has become a popular variety because of its nematode resistance and high yield in the dry season. Cowpea cultivation in the dry season was not possible before because all the local cowpea varieties were susceptible to nematodes. Resistance to viruses Singh and Hughes (1998, 1999) reported several cowpea breeding lines to be completely resistant to cowpea yellow mosaic, blackeye cowpea mosaic, and cowpea aphid borne mosaic. Of these IT96D-659, IT96D-660, IT97K-1068-7, and IT95K-52-34 were most promising in terms of resistance and yield potential. Bashir et al. (1995) screened several cowpea varieties from I1TA and observed that IT86F 2089-5, IT86D-880, IT90K-284-2, IT90K-76, IT86D-101O, and IT87D-611-3 were immune to blackeye cowpea mosaic. Van-Boxtel et al. (2000) artificially screened 14 cowpea varieties with three isolates of blackeye cowpea mosaic and 10 isolates of cowpea aphid borne mosaic virus in order to identify lines with multiple strain resistance. They observed that cowpea breeding lines IT86D-880 and IT86D-1O 10 were resistant to all the three isolates of blackeye cowpea mosaic and five strains of cowpea aphid borne mosaic. IT82D-889, IT90K-277-2, and TVu 201 showed resistance to one or the other of the five remaining isolates and thus by using the abovementioned five cowpea varieties as parental lines, it is possible to breed new cowpea varieties with combined resistance to all the 13 strains of the viruses. The most important factors that constrain cowpea production in the northeastern region of Brazil are the virus diseases, caused mainly by cowpea severe mosaic virus (CSMV) of the group Comovirus, cowpea aphid borne mosaic virus (CABMV) of the group Potyvirus, cucumber mosaic virus (CMV) of the group Cucumovirus, and cowpea golden mosaic virus (CGMV) of the group Geminivirus (Lima and Santos 1988). Substantial efforts have been made in breeding for resistance to viruses and progress has been made. Lima and Nelson (1977) identified the cultivar Macaibo as having immunity to CSMV while Vale and Lima (1995) showed that inheritance of this resistance is conditioned by a recessive gene. Rios and Neves (1982) confirmed the immunity of Macaibo and a new source of resistance to CSMV in line FP 7733-2, from which the variety CNC 0434 was developed (Rios et al. 1982). This variety was recommended for cultivation in the state of Maranhao (EMBRAPA 1986). Lima et al. (1986), in a study that involved 248 genotypes, identified four new genotypes (TVu 379, TVu 382, TVu 966, and TVu 3961) as being immune to CSMVand CABMV Cultivars Cowpea 535, Dixiecream, Bunch Purple Hull, Lot. 7909- Purple, V-17, and TVu 612 were immune only to CABMV Lima et al. (1998), in another study that involved 44 genotypes, confirmed the immunity of genotypes TVu 379, TVu 382, TVu 966, and TVu 3961 to three strains of CSMV Santos and Freire Filho (1986) screened 450 genotypes for resistance to CGMV Of those genotypes, 57 were classified as highly resistant, among these being CNC 0434, TVu 612, CE-3l5 (TVu 2331), and 27 Digitized by Google Cowpea genetics and breeding BR l-Poty. Three lines from the EMBRAPA cowpea breeding program, TE87-98-8G, TE87-98-l3G, and TE87-108-6G and two lines introduced from I1TA, IT84S-2l35 and IT84S-l627, were found to be resistant to CABMV and immune to CMVby the Laboratory of Virology of the Center of Agrarian Sciences of the Federal University of Ceara. Two other lines from I1TA, IT85F -2687 and IT86D-7l6, were immune to both viruses (Rocha et al. 1996). These resistance sources have been used in cowpea improvement in Brazil. Several varieties that have been released commercially, and breeding lines that are still under evaluation were developed from crosses with the varieties CNC 0434, Macaibo, and TVu 612. Resistance to CSMV, CABMV, and CGMVhas already been incorporated in some of the released varieties like BR lO-Piaui (Santos et al. 1987), BR l2-Caninde (Cardoso et al. 1988), BR l4-Mulato (Cardoso et al. 1990), BR 17 -Gurgueia (Freire Filho et al. 1994), EPACE 10 (Barreto et al. 1988), Setentao (paiva et al. 1988), IPA 206 (lPA, 1989), and BR l6-Chapeo-de-couro (Fernandes et al. 1 990b ). Presently, crosses are being made to improve resistance to CMV Resistance to Striga and Alectra Cowpea suffers considerable damage due to Striga gesnerioides in West and Central Africa and to Alectra vogelii in West and Central Africa as well as in eastern and southern Africa. Good progress has been made in breeding improved cowpea varieties with com- bined resistance to Striga andAlectra (Atokple et al. 1995, Berner et al. 1995, Singh and Emechebe 1997, Singh et al. 1997, Singh 2000). The most promising new cowpea varieties are IT93K-693-2, IT95K-l090-l2, IT97K-499-39, IT97K-497-2, and IT97K-8l9-l54 with combined resistance to Striga and Alectra and major diseases. The details of breeding for Striga and Alectra resistance are presented in this volume by B.B. Singh. I nsect resistance Insect pests are a major constraint in cowpea production. Considerable progress has been made in the last four years in developing cowpea varieties resistant to several insects. Pandey et al. (1995) reported TVu 908 to be resistant to leaf beetles. Singh et al. (1996) reported several improved cowpea varieties with combined resistance to aphid, thrips, and bruchid. Of these, IT90K-76, IT90K-59, and IT90K 277-2 are already popular varieties in several countries. Among the new varieties IT97K-207-l5, IT95K-398-l4, and 98K- 506-1 have a high level of bruchid resistance (Singh 1999c). Nkansah and Hodgeson (1995) confirmed resistance of TVu 801 and TVu 3000 to the Nigerian aphid strain but found that the two lines were susceptible to aphids from the Philippines. Similar differential reactions to aphids has been observed in the USA (A.E. Hall, personal communication) indicating the existence of different aphid strains. Shade et al. (1999) also reported a virulent strain of bruchid (Callosobruchus maculatus) which was able to cause severe damage to TVu 2027, which is otherwise resistant to the bruchid strain in Nigeria. Yunes et al. (1998) observed that the 7 s-storage protein, "vicillin" is responsible for bruchid resistance in cowpea lines related to TVu 2027. Only low levels of resistance have been observed for Maruca pod borer and pod bugs, which cause severe damage and yield reduc- tion in cowpea. Jagginavan et al. (1995) observed cowpea lines P120 and Cll to be least damaged by Maruca and Veeranna and Hussain (1997) found TVx 7 to be most resistant to Maruca and has a high density of trichomes (2l.4l1mm2). Veerappa (1998) screened 45 cowpea lines for resistance to Maruca pod borer and observed that the tolerant lines 28 Digitized by Google Recent progress in cowpea breeding had higher phenol and tannin contents compared to susceptible lines. This is in line with the general observation that cowpea varieties with pigmented calyx, petioles, pods, and pod tips suffer less damage due to Maruca. As indicated earlier, a distant wild relative of cowpea Vigna vexillata has shown high levels of resistance to Maruca pod borer and bruchid but all the efforts made at I1TA to transfer Maruca resistance genes from Vigna vexillata to cowpea have not been successful (Fatokun in this volume). Gomathinayagam et al. (1998) reported a successful susceptible cross between Vigna vexillata and cowpea and also made a backcross in F 2 generation but the resulting seeds looked like the wild parent (personal communications). This work is not being followed further raising the question whether the original cross and the backcross seeds were true hybrids. Over the last 10 years concerted efforts were made by I1TA in collaboration with advanced laboratories in the USA and Italy to transform cowpea with the Bt gene for Maruca resistance. However, no success has been achieved as yet. While the wide crosses and transformation of cowpea with the Bt gene have not been successful, considerable progress has been made in pyramiding minor genes for field resistance to Maruca pod borer and pod bugs through conventional breeding. Singh (1999a) screened new improved cowpea breeding lines for field resistance to major insect pests without insecticide sprays and he observed several cowpea lines with grain yield of 500 kg/ha to 856 kg/ha without any chemical protection. The local variety yielded 0 to 48 kg/ha in the same trials. The most promising varieties are IT90K-277-2, IT93K-452-l, IT94K-437-l, IT97K-569-9, IT95K-222-3, IT97K-837, andIT97K-499-38. These lines are resistant to major foliar diseases, aphid, thrips, and bruchid with pods at a wide angle and sufferless damage due to Maruca. IT94K-437-l and IT97K-499-38 also have combined resistance to Striga and Alectra. Developed through conventional breeding approaches, the new field resistant lines require only one or two sprays of insecticide for a normal yield of 1.5 to 2.5 t compared to four to six sprays needed for the susceptible varieties. Drought, heat, and cold tolerance Since cowpea is grown in varied environments it encounters different types of stresses including drought, heat, and cold. Good progress has been made at I1TA on breeding for enhanced drought and heat tolerance, and at the University of California, Riverside on water use efficiency, heat tolerance, and chilling tolerance (Okosun et al. 1998a,1998b, Singh et al. 1 999a, 1999b; Mai-Kodomi et al. 1 999a, 1999b; Hall et al. 1997; Ismail and Hall 1998; Singh 1 99ge). Simple, cheap, and nondestructive screening methods for drought tolerance have been developed and used to identify and breed for drought toler- ant cowpea varieties. Heat tolerant lines have been developed and heat tolerance is now better understood in cowpea than any other crop (Singh 1999b, Ismail and Hall 1998). Recently the effective- ness of heat tolerance has been quantified using pairs of genetically related and unrelated lines with and without heat tolerance genes (Ismail and Hall 1998). This work is reviewed in detail in this volume by Hall et al. Singh (1 999b) grew 102 cowpea breeding lines atIlTA Kano Station from March to May when the temperatures ranged from 24 to 27°C in the night and from 38 to 42 °C during the day. Most of the lines showed severe flower abortion with little or no pods and these were rated as heat susceptible. The most susceptible lines, IT97K -461-2 and IT97K -461-4, showed complete sterility with no development of pollen beyond the micro spore stage. These lines are otherwise normal and very high yielding in 29 Digitized by Google Cowpea genetics and breeding the regular crop season (July-October) when day temperatures are below 35°C and night temperatures below 24 °C. In contrast to the heat susceptible lines, the heat tolerant lines had normal pollen, good pod set, and normal grain yield. The best heat tolerant lines were IT97K-472-l2, IT97K-472-25, IT97K-8l9-43, and IT97K-499-38. The details of work on chilling tolerance are reviewed in this volume by Hall et al. A dehydrin gene involved in chilling tolerance during seedling stage has been identified (Ismail et al. 1997, 1999) and mapped using recombinant inbred lines (Menendez et al. 1997). The role of the dehydrin in chilling tolerance has been confirmed using near-isogenic lines (Ismail et al. 2000) and efforts are underway to understand the mechanism involved in the control of its expression. Enhanced N-fixation and efficient use of phosphorus Significant variation in cowpea rhizobium strains has been observed for nodulation in cowpea (MandaI et al. 1999) but the local rhizobia invariably outpopulate the introduced strains. Therefore, in recent years, major efforts have concentrated on exploiting genetic variability in cowpea as a host for effective nodulation and nitrogen fixation (Buttery et al. 1992). Graham and Scott (1983) observed major genetic differences for nodulation and dry matter and N accumulation among 12 cowpea varieties. They also observed a significant relationship between total N and seed yield and nodule weight. MandaI et al. (1999) also observed significant varietal differences in cowpea for nodule number and nodule weight as well as for nitrogenase activity indicating a good possibility of breeding improved cowpea varieties with enhanced N-fixation. Sanginga et al. (2000) screened 94 cowpea lines and observed major varietal differences in cowpea for growth, nodulation, and arbuscular mycor- rhizal fungi root infection as well as for performance under low and high phosphorus. The improved cowpea variety IT86D-7l5 showed equally good growth under low as well as high phosphorus levels. It also showed better N-fixation than others. Based on its adaptability to grow in low P soils and overall positive N balance, they recommended cultivation ofIT86D- 715 cowpea variety in soils with low fertility. Kolawale et al. (2000) screened 15 cowpea varieties for tolerance to aluminum and to determine the effect of phosphorus addition on the performance of AI-tolerant lines. The results indicated IT9lK -93-10, IT93K -2046-1, and IT90K -277 -2 cowpea varieties to be tolerant to aluminum and they gave a higher response to phosphorus fertilization when grown in soils with aluminum toxicity problems. Singh et al. (1998) evaluated improved cowpea varieties under low and high fertility and they also observed major varietal differences. They found IT96D-772, IT96D-739, IT96D-740, and IT96D-666 cowpea varieties to be good performers under low as well as high fertility whereas most other varieties were poor in poor fertility and good in good fertility. These studies further indicate a good possibility of developing improved cowpea varieties with enhanced nitrogen fixation and higher yields under low phosphorus as well as in soils with aluminum toxicity. There is a need for closer interactions between cowpea breeders and soil scientists and soil microbiologists. Improved nutritional quality Cowpea is a major source of protein, minerals, and vitamins in the daily diets of the rural and urban masses in the tropics, particularly in West and Central Africa where it comple- ments well with the starchy food prepared from cassava, maize, millet, sorghum, and yam. Systematic efforts have just begun at I1TA and a few other institutions to develop 30 Digitized by Google Recent progress in cowpea breeding improved cowpea varieties with enhanced levels of protein and minerals combined with faster cooking and acceptable taste. Singh (1999d) screened 52 improved and local cowpea varieties to estimate the extent of genetic variability for protein, fat, minerals etc. On a fresh weight basis (about 10% moisture), the protein content ranged from 20 to 26%, fat content from 0.36% to 3.34%, iron content from 56 ppm to 95.8ppm, and manganese content from 5 ppm to 18 ppm. The improved cowpea varieties IT89KD-245, IT89KD-288, and IT97K-499-35 had the highest protein content (26%) whereas the local varieties like Kanannado, Bauchi early, and Bausse local had the lowest protein content (21 to 22%). One of the local varieties, IAR 1696, had high protein content (24.78%) and high fat content (3.28%) as well as high iron content (81.55 ppm). Similarly an improved variety, IT95K-686-2, had high protein (25%), high fat content (3.3%), and high iron content (76.5 ppm). Appropriate crosses have been made to study the inheritance of protein, fat, and iron contents and to initiate a breeding program for improving these quality traits. In another experiment, various physical properties of selected cowpea varieties were determined. The relative density of cowpea seed ranged from 1.01 to 1.09, and hardness (crushing weight) ranged from 3.96kg for IT89KD-288 to 8.4 kg for Aloka local. The seed hard- ness was positively correlated with cooking time. There have been earlier reports on the extent of genetic variability for quality traits in cowpea. Hannah et al. (1976) reported high methionine content in TVu 2093 and Bush Sitao (3.24-3.4 mg/g) dry seeds compared to 2.75-2.88 mg/g seeds of the check variety G-8l-1. Rosario et al. (1980) observed the highest typsin inhibitor activity in winged bean and lima bean and the lowest activity in mung bean and rice bean whereas the trypsin inhibitor values for cowpea were interme- diate. Fashakin and Fasanya (1988) analyzed 10 cowpea varieties and observed a range for protein content from 21.5 to 27% and for iron from 8 to 15 mg/lOOg dry seeds. Nout (1996) evaluated five newly released cowpea varieties used to make popular snack food, koose (also called akara and kosai in Nigeria). They found that akara prepared from high yielding new cowpea varieties Ayiyi (IT83S-728-l3) and Bengpla (IT83S-8l8) were the best. Similarly Singh (1999d) in collaboration with the Women in Agriculture (WIA) section of the Kano Agricultural and Rural Development Authority KNARDA (Nigeria) evaluated three improved cowpea varieties, IT98D-867-11, IT89KD-288, and IT90K- 277 -2 and one local variety Dan Ila for four popular local dishes-kosai, danwake, alale, and dafaduka. These were subjected to an independent taste panel of over 50 persons of different economic status and background. The improved variety IT90K-277-2 was rated as the best and others were as good as the local variety. None of the varieties was rated as unacceptable. IT90K-277-2 has already become very popular in Nigeria and Cameroon as a high yielding variety. These observations indicate that high yield is not negatively correlated with improved nutritional and food quality traits and that sufficient genetic variability exists to improve these traits in cowpea. Development and release of cowpea varieties A large number of cowpea varieties have been released in several countries around the world and the collaborative interactions between the I1TA cowpea breeding program and national program scientists have been very effective. A total of 68 countries have identified and released improved cowpea varieties from I1TA for general cultivation. The countries and the name of breeding lines released are presented in Table 3. The availability of high yielding disease and insect resistant varieties with desired seed and growth types is quietly 31 Digitized by Google Cowpea genetics and breeding Table 3. Countries that have released IITA developed improved cowpea varieties. Country Variety released Country Variety released Angola TVx 3236 Argentina IT820-716 Australia IT82 E-18 (Big Buff) Belize VITA-3, IT820-889, IT82E-l Benin Republic VITA-4, VITA-5, Bolivia IT820-889,IT830-442 IT810-1137, IT84S-2246-4 Botswana ER-7, TVx 3236 Brazil VITA-3, VITA-6, VITA-7, TVx 1836-01) Burkina Faso TVx 3236, VITA-7 Burma VITA-4 (Yezin-l) (KN-l) Cameroon IT81 0-985 (BR1), Central VITA-l, VITA-4, VITA-7, IT810-994, (BR2), African VITA-5, TVx 1948-01 F, TVx 3236, Republic IT81O-1137,IT83S-818, IT880-363 (GLM-92), IT82 E-18, IT81 0-994 IT90K-277-2 (GLM-93) Colombia IT83S-841 Costa Rica VITA-l, VITA-3, Cote IT880-361, IT880-363 VITA-6, VITA-7 d'ivoire Cuba IT840-449 (Titan) Cyprus IT850-3577 IT840-666 (Cubinata-666) IT860-314 (Mulatina-314) IT860-368, (IITA-Precoz) EI Savador TVx 1836-013) IT860-782 (Tropico-782) (Castilla deseda), IT860-792 (Yarey-792) VITA-3 (TECPAN V-3), IT88S-574-3 (OR 574-3) VITA-5 (TECPAN V-5 Oemocratic VITA-6, VITA-7 Equador VITA-3 Rep. of Congo IT89KO-349, IT89KO-349, Ethiopia TVx 1977-010, IT82E-16, IT89KO-389, IT82E-32 IT89KO-355 Equatorial IT870-885 Guinea Fiji VITA-l, VITA-3 Egypt TVu 21, IT820-716 Gambia IT84S-2049, (Sosokoyo) IT820-709,IT820-812, IT83S-728-13 IT82E-16 Ghana IT82 E-l 6 (Asontem) Guinea IT81 0-879, IT830-340-5, IT83S-728-13 (Ayiyi) Conakry IT82 E-16, IT85 F-867-5 IT83S-818 (Bengpla) (pkoku Togboi) TVx 1843-1 C (Boafa) IT85F-2805, IT83S-990, TVx 2724-01 F (Soronko) IT87S-1463, IT84S-2246-4 Guinea Bisau IT82 E-9, IT82 0-889 Guyana ER-7, TVx 2907-020, TVx 66-2H, VITA-3, IT870-611-3 Guatemala VITA-3 Haiti VITA-4, IT870-885 .. .lcontinued 32 Digitized by Google Recent progress in cowpea breeding Table 3 (continued) Country Variety released Country Variety released India VITA-4, TVx 1502, Jamaica VITA-3, ER-7, IT84S-2246-4, IT85 E2 02 0 (Vamban 1) IT82E-124 Lesotho IT82E-889,IT87O-885 Liberia IT82 0-889, TVx 3236, IT82 E-16, IT82 E-32 VITA-5, VITa-4, VITA-7 Malawi IT82 0-889, IT82 E-16 Mali TVx 3236, IT89KO-374 IT82E-25 (Korobalen) IT89KO-245 (Sangaraka) Mozambique IT82 0-812, IT83S-18, IT85F-2020 Mauritius TVx 3236 Namibia IT81 0-985, IT89KO-245-1, Nicaragua VITA-3 Nepal Niger Pakistan Panama Peru Sierra Leone South Yemen South Africa Sri Lanka IT870-453-2 IT820-752 (Aakash) IT82 0-889 (Prakash) IT89KO-374, IT90K-372-1-2 IT90K-82-2, IT89KO-288 VITA-4 VITA-3 VITA-7 TVx 1990-01 E, IT860-721, IT860-719, IT860-1 01 0, IT82E-32, TVx 3236, TVu 1990, VITA-3 VITA-5, VITA-7 IT90K-59, IT82 E-16 (Pannar 311) IT82 0-789 (Wijaya) IT82 0-889 (Waruni) TVx 309-01 EG, VITA-4 TVx 930-01 B, (Lita) Nigeria TVx 3236, IT81 0-994, IT860-719, IT880-867-11, IT89KO-349, IT860-721, IT880-867-11, IT82 E-60, IT89KO-374,IT90K-277-2, Paraguay IT860-1 01 0, IT870-378-4, IT870-697-2, IT87O-2075 Philippines IT82 0-889 Senegal TVx 3236 Somaila TVx 1502, IT82 0-889 IT82E-32 South Korea VITA-5, IT835-852, IT82 0-889 Sudan IT84S-2163 (Oaha EIGoz = Gold from sand) Swaziland IT82 0-889 (Umtilane), IT82 E-18, IT82 E-2 7, IT82 E-71 Thailand VITA-3,IT820-889 Uganda TVx 3236, IT82 E-60 .. .lcontinued 33 Digitized by Google Cowpea genetics and breeding Table 3 (continued) Country Variety released Country Variety released Suriname IT82D-889,IT82-D789 USA IT845-2246-4, IT845-2049, (for nematode resistance) IT89KD-288 Tanzania TKx 9-11 D (Tumaini) Yemen TVx 3236, IT82D-789, TVx 1948-01 F (Fahari) VITA-5 IT82 D-889 (Vu I i-1 ) IT85F-2020 Venezuela VITA-3, IT81 D-795, Togo VITA-5, TVx 3236, IT82 D-504-4 TVx 1850-01 E, IT81 D-985, (VITOCO) Zambia TVx 456-01 F, TVx 309-01 G, IT82 E-16 (Bubebe) Zimbabwe IT82D-889 catalyzing rapid increase in cowpea cultivation including its extension in nontraditional areas. Many countries where new cowpea varieties are making a difference, have given specific names to the new varieties and, in some areas, farmers themselves have given names and facilitated farmer to farmer diffusion of seeds. A few examples are Big Buff in Australia; BR -1 in Cameroon; Titan and Cubinata in Cuba; Asontem and Bengpla in Ghana; Akash (sky) and Prakash (light) in Nepal; Sosokoyo in Gambia; Pkoko Togboi in Guinea Conakry; Korobalen and Sangaraka in Mali; Dan I1TA (son of I1TA) and Dan Bunkure in Nigeria; Pannar 31 in South Mrica; Vuli-l in Tanzania; Dahal Elgoz (gold from the sand) in Sudan; Umtilane in Swaziland; and Bubebe in Zambia. The US Vegetable Laboratory at Charleston, South Carolina, has released several cowpea cultivars in the past five years. These include the "snap" cultivar Bettersnap (F ery and Dukes 1995b), the cream type cultivar Tender Cream (Fery and Dukes 1996), and the persistent-green cultivars Charleston Greenpack, (Fery 1998), Petite-N-Green (Fery 1999), Green Pixie (Fery 2000), and Green Dixie, (USDA 2000). The persistent-green varieties are an important new market class of cowpea for the freezing industry in the US (Ehlers, Fery, Hall in this volume) because they are virtually identical in appearance to fresh -shelled cowpeas after they are imbibed with water, but the harvesting costs are much lower because persistent-green grains may be harvested dry with fast, efficient combines, and cleaned and stored dry. With the appearance of a freshly harvested vegetable product, low product cost, and ease of storage and handing, the persistent-green cowpea is attractive to vegetable processors for use in new products or blends with other vegetables. This could help increase cowpea consumption in the US and elsewhere. California Blackeye No. 27 (CB27) is a new blackeye cowpea cultivar for producing dry grain that was released by the University of California, Riverside in 1999. CB27 has high yield, heat tolerance, strong, broad-based resistance to root-knot nematodes, resistance to two races of Fusarium wilt, excellent canning quality, and a brighter white seed, compared to the standard blackeye variety in California, CB46 (Ehlers et al. 2000b). Brazil has released 18 varieties in the last 12 years for the northern region. Two of these, Monteiro (Freire F ilho et al. 1998) and Riso do Ano (Fernandes et al. 1 990a) were obtained through collection and selection in local populations. Sixteen varieties were developed using pedigree breeding methods. Most of these have been mentioned in the 34 Digitized by Google Recent progress in cowpea breeding virus resistance section. Dry grain yields during the rainy season typically range from 1000 to 1200 kg/ha, while the production under irrigation during the dry season is from 1500 to 2000 kg/ha. All these varieties were selected under the rainfed system. Therefore, it is possible that varieties can be developed with much higher yields under irrigation if selection is conducted under these conditions. It is worth noting that even with these low yield levels, positive economic returns are realized. To overcome local constraints, variet- ies are needed with resistance to a wide spectrum of diseases and pests. Several other varieties have been released in different countries such as Charodi-l (Sreekumar et al. 1993) and Vamban 1 (IT85F -2020) (Viswanathan et al. 1997) in India; Big Buff (IT82E-18 Imrie, 1995) and Ebony PR (ADTA 1996) in Australia; IT83S-852 and IT82D-889 (Lee et al. 1996) in South Korea; Melakh and Mouride (Cisse et al. 1997) in Senegal; IT87D-611-3 (Singh et al. 1994) in Guyana; Cream 7 (Hassan 1996) in Egypt; IT90K-76, IT90K-277-2, IT90K-82-2 in Nigeria; Sangaraka (IT89KD-374-57) and Korobalen (IT89KD-245) in Mali; INIFAT 93 (Diaz et al. 1997) in Cuba; and GLM 93 (IT90K-277-2) in Cameroon. This is not an exhaustive list as the information from all countries is not available. References Adu-Dapaah, H.K., B.B. Singh, and C. Fatokun. 1999. A fascinated mutant in cowpea (Vigna unguiculata [L.] Walp.). Acta Agronomica Hungarica 47: 371-376. Atokple, I.D.K., B.B. Singh, and A.M. Emechebe. 1995. Genetics of resistance to Striga and Alec- tra in cowpea. Journal Heredity 86: 45-49. Australia Division of Tropical Agriculture (ADTA). 1996. Variety: "EbonyPR" sny Line 4 A. Application No:96/159. Plant Varieties Journal 9(4): 25. Berner, D.K., lG. Kling, and B.B. Singh. 1995. Striga research and control-a perspective from Africa. Plant Disease 79: 652-660. Bashir, M., Z. Ahmed, R. Zafar, and B.A. Malik. 1995. Sources of immunity in cowpea against blackeye cowpea mosaic potyvirus. Pakistan Journal of Phytopathology. 7(2): 94-97. Buttery, B.R., S.l Partk, and D.J. Hume. 1992. Potential for increasing nitrogen fixation in grain legumes. Canadian Journal of Plant Science (Canada). V 72 (2): 323-349. Barreto, D.P.D.,A.A. dos Santos, M.A.W. Quindere, J.C. Vidal, lP.P. Araujo, E.E. Walt, G.P. Rios, e, B.P. Neves. Epace-IO: Nova Cultivar DE Caupi PARA 0 CEARA. Fortaleza: EPACE, 1988. Folder. Cardoso, M.J., F.R. Freire Filho, e, C. Athayde Sobrinho. BR 14-Mulato: nova cultivar de feijao macassar para 0 estado do Piaui. Teresina: Embrapa-Uepae de Teresina, 1990.4 pp. (Embrapa- Uepae de Teresina. Comunicado Tecnico 48.) Cardoso, M.l, A.S.A. dos Santos, F.R. Freire Filho, e, A.B. Frota. "BR 12-Caninde": cultivar de feijao macassar precoce com resistencia multipla a virus. Teresina: Embrapa-Uepae de Teresina, 1988.3 pp. (Embrapa-Uepae de Teresina. Comunicado Tecnico 39.) Cisse, N., M. Ndiaye, S. Thiaw, andA.E. Hall. 1997. Registration of "Melakh" cowpea. Crop Sci- ence 37(6): 1978. Diaz, M., T. Shagarodsky, N. Lastres, F. Canet, and G. Puldon. 1997. INFAT 93, a new variety of cowpea (Vigna unguiculata). Agrotecnia-de-Cuba. 27(1): 148. Ehlers, J.D., W.C. Matthews,A.E. Hall, and P.A. Roberts. 2000a. Inheritance of a broad-based form of nematode resistance in cowpea. Crop Science 40: 611-618. Ehlers, lD., A.E. Hall, P.N. Patel, P.A. Roberts, and W.C Matthews. 2000b. Registration of 'Cali- fornia Blackeye 27' Cowpea. Crop Science 40: 854-855. Embrapa. Centro Nacional de Pesquisa de Arroz e Feijao (Goiania, GO). 1986. Cultivares de arroz, feijao caupi lan.,:adas em coopera.,:ao com 0 Centro Nacional de Pesquisa de Arroz e Feijao. Goiania, EMBRAPA-CNPAF Documentos 15: 43-68. 35 Digitized by Google Cowpea genetics and breeding Fashakin, lB. and J.1. Fasanya. 1988. Chemical composition and nutritive changes of some improved varieties of cowpea (Vigna unguiculata). I. Some selected varieties from the International Institute of Tropical Agriculture, Ibadan, Nigeria. Tropical Science (UK) 28 (2): I I I-I I 8. Fernandes, lB., IS. de Holanda,AA. Simplicio, F. Bezerra Neto, J. Torres, e, I Rego Neto. 1990a. Comportamento ambiental e estabilidade produtiva de cultivares de caupi no Rio Grande do Norte. PesquisaAgropecm'tria Brasileira 25(11): 1555-1560. Fernandes, J.B., NA. de Sousa, e, J.S. de Holanda. 1990b. BR 16-Chapeu-de-couro: nova cultivar de feijao macassar para 0 sertiio do Rio Grande do Norte. Natal: EMPARN. Folder. F ery, RL. and P.D. Dukes. 1995a. Registration ofUS-566, US-567, and US-568 root-knot nematode resistant cowpea germplasm lines. Crop Science 35: 1722. Fery, RL. and P.D. Dukes. 1995b. 'Bettersnap' southernpea. HortScience 30: 1318-1319. Fery, RL. and P.D. Dukes. 1996. 'Tender Cream' southernpea. HortScience 31: 1250-1251. Fery, RL. 1998. 'Charleston Greenpack', a pinkeye-type southernpea with a green cotyledon phe- notype. HortScience 33: 907-908. Fery, RL. 1999. 'Petite-N-Green', a small-seeded, full-season, green cotyledon, pinkeye-type southernpea. HortScience 34: 938-939. Fery, RL. 2000. 'Green Pixie', a small-seeded, green cotyledon, cream-type southernpea. Hort- Science 35: (in press). Freire Filho, F.R., VQ. Ribeiro,P'H.S. da Silva,e,PA.C. Carvalho 1998. Monteiro: cultivardecaupi de tegumento branco para cultivo irrigado, Teresina: Embrapa-Cpamn. Embrapa-Cparnn, Comicado Tecnico 85: 1-3. Freire Filho, F.R,A.A. dos, Santos,A.G. deAraujo, M.J. Cardoso, P.H.S. da Silva, e, VQ. Ribeiro. 1994. BR 17-Gurgueia: nova cultivar de caupi com resistencia a virus para 0 Piaui. Teresina: Embrapa-Cpamn, 6 pp. Embrapa-Cpamn. Comunicado Tecnico 61. Gomathinayagam, P., S.G. Ram, R Rathnaswamy, and N.M. Ramaswamy. 1998. Interspecific hybridisation between Vigna unguiculata (L.) Walp. and V vexillata (L.)A. Rich. through in vitro embryo culture. Euphytica 102(2): 203-209. Graham, RA. and T.W. Scott. 1983. Varietal characteristics and nitrogen fixation in cowpea. Tropical Agriculture (Trinidad and Tobago) 60(4): 269-271. Gunasekaran, M., U. Selvaraj, and T.S. Raveemdram. 1998. Induced polygenic mutations in cowpea (Vigna unguiculata L. Walp). South-Indian Horticulture 46: 1-2,13-17. Hall, A.E., B.B. Singh, and ID. Ehlers. 1997. Cowpea breeding. Plant Breeding Reviews 15: 215-274. Hannah, L.C., J. Ferrero, and D.W. Dessauer. 1976. High methionine lines of cowpea. Tropical Grain Legume Bulletin (Nigeria) 4: 9. Hassan, H.M. 1996. New selected strains ofthe cowpea cultivar "Cream 7". Alexandria Journal of Agricultural Research 41(2): 399-406. Ismail, A.M., A.E. Hall, and T.J. Close. 1997. Chilling tolerance during emergence of cowpea associated with a dehydrin and slow electrolyte leakage. Crop Science 37: 1270-1277. IPA (Recife, PE). Caupi-IPA-206: nova cultivar de feijao macassar (Vigna unguiculata [L] Walp.) tipo moita para Pernambuco. Recife, 1989. Ismail,A.M. andA.E. Hall. 1998. Positive and potential negative effects of heat-tolerance genes in cowpea. Crop Science 38: 381-390. Ismail, A.M., A.E. Hall, and T.J. Close. 1999. Allelic variation of a dehydrin gene co-segregates with chilling tolerance during seedling emergence. Proceedings ofthe Natural Academy of Sci- ence 96: 13566-13570. Ismail, A.M., A.E. Hall, and J.D. Ehlers. 2000. Delayed-leaf-senescence and heat tolerance traits mainly are independently expressed in cowpea. Crop Science 40: 1049-1055. Imrie, B.C. 1995. Register of Australian grain legume cultivars. Vigna unguiculata L. (cowpea) cv. Big Buff. Australian Journal of Experimental Agriculture 35(5): 678. Jagginavan, S.B., KA. Kulkarni, and S. Lingappa. 1995. Reaction of cowpea genotypes to the damage of pod borer complex. Karnataka Journal of Agricultural Science 89(1): 90-93. 36 Digitized by Google Recent progress in cowpea breeding John, SA. 1999. Mutation frequency and chlorophyll mutations in parents and hybrid of cowpea following gamma irradiation. Indian Journal of Genetics and Plant Breeding 59(3): 357-361. Knyast, RG., W.E. Odland, Rl Okagaki, O. Riera-Lizarazu, S.M. Maguieira, CD. Russell, H.W. Rines, and RL. Phillips. 2000. Complete set of maize individual dual-chromosome additions in oat. Agronomy Abstracts 2000: 188. Kolawale, G.O., G. Tian, and B.B. Singh. 2000. Differential response of cowpea varieties to alu- minum and phosphorus application. Journal of Plant Nutrition 23: 731-740. Latunde-Dada, A.O., RJ. O'Connell, P. Bowyer, and JA. Lucas. 1999. Cultivar resistance to anthracnose disease of cowpea caused by Colletotrichum destructivum O'Gara. European Jour- nal of Plant Pathology 105: 445-450. Lee, S.M., K.J. Yun, J.B. Tae, S.M. Lee, J.Y. Koo, and B.T. Jeon. 1996. Studies on the growth characteristics and yield of cowpea cultivars for silage. Journal ofthe Korean Society of Grass- land Science 16(2): 105-112. Lima, JA. de A., RC.A. Lima, M.F.B. Gon9alves, and I.M. Sittolin. 1998. Biological and sero- logical characteristics of a genetically different cowpea severe mosaic virus strain. Virus Reviews and Research 3(12): 57--65. Lima, JA. deA. and M.R. Nelson. 1977. Etiology and epidemiology of mosaic of cowpea in Ceara Brasil. Plant Disease Report 61(10): 864-867. Lima, JA. deA., CD.G. Santos, e, L.F.S. Silveira. 1986. Comportamento de gen6tipos de caupi em rela9ao aos dois principais virus que ocorrem no Ceara. Fitopatologia Brasileira 11: 151- 161, Lima, JA. de A. e, AA. Santos. 1988. Virus que infestam 0 caupi no Brasil. Pages 507-545 in lP.P. de Araujo and E.E. Watt. org 0 Caupi no Brasil. Goiania: EMBRAPA-CNPAFlIbadan, I1TA. Lin, M.C., H.P. Chen, Y.F. Wang, W.Z., Zhang, M.C., Lin, H.P., Chen, and Y.F. Wang. 1995. Evaluation of cowpea varieties resistant to cowpea leaf mould (Cercospora cruenta Sacc.). Crop Genetic Resources 4: 36-37. Mai-Kodomi, Y., B.B. Singh, O. Myers Jr., lH. Yopp, P.J. Gibson, and T. Terao. 1999a. Two mechanisms of drought tolerance in cowpea. Indian Journal of Genetics 59: 309-316. Mai-Kodomi, Y., B.B. Singh, T. Terao, O. Myers Jr., J.H. Yopp, and P.J. Gibson. 1999b. Inheritance of drought tolerance in cowpea. Indian Journal of Genetics 59: 317-232. Mandai, l,A. Chattopadhyay, P. Hazra, T. Dasgupta, and M.G. Som. 1999. Genetic variability for three biological nitrogen fixation components in cowpea (Vigna unguiculata [L.] Walp.) cultivars. Crop Research (Hisar) 18: 222-225. Menendez, C.M., A.E. Hall, and P. Gepts. 1997. A genetic linkage map of cowpea (Vigna unguicu- lata) developed from a cross between two inbred, domesticated lines. Theoretical and Applied Genetics 95: 1210-1217. Mehta, D.R and P.P. Zaveri. 1997. Single seed versus single plant selection in cowpea. Legume Research 20(2): 130-132. Nakawuka, C.K. and E. Adipala. 1997. Identification of sources and inheritance of resistance to Sphaceloma scab in cowpea. Plant Disease 81(12): 1395-1399. Nkansah, P.K. and C.l Hodgson. 1995. Interaction between aphid resistant cowpea cultivars and three clones of cowpea aphid and the effect of two light intensity regimes in this interaction. International Journal of Pest Management 41: 161-165. Nout, M.J.R. 1996. Suitability of high yielding cowpea cultivars for koose, a traditional fried paste of Ghana. Tropical Science (UK) 36(4): 229-236. Odeigah, P.G.C.,A.O. Osanyinpeji, and G.O. Myers. 1996. Induced male sterility in cowpea (Vigna unguiculata L. Walp.). Journal of Genetics and Breeding 50(2): 171-175. Okosun, L.A., M.E. Aken'ova, and B.B. Singh. 1998a. Screening for drought tolerance at seedling stage in cowpea (Vigna unguiculata [L.] Walp. I.) The significance ofthe trait permanent wilting percentage. Journal of Arid Agriculture 8: 1-10. 37 Digitized by Google Cowpea genetics and breeding Okosun, L.A., M.E. Aken' ova, and B.B. Singh. 1998b. Screening for drought tolerance at seedling stage in cowpea (Vigna unguiculata [L.] Walp. II.) Selecting for root length and recovery ability traits. Journal of Arid Agriculture 8: 11-20. Paiva, lB., E.M. Te6filo, J.H.R Santos, Lima JAA. dos, M.F.B. Gon9alves, e, L. de F. S. Silveira. 1988. "Setentao": nova cultivar de feijao-de-corda para 0 estado do Ceara. Fortaleza: UFC, Folder. Pandey, K.C., N. Hasan, RB. Bhaskar, S.T. Ahmed, and K.S. Kohli. 1995. Genetic evaluation of cowpea (Vigna unguiculata [L.] Walp.) lines for multiple pest resistance. Indian Journal of Genet- ics and Plant Breeding 55(2): 198-203. Ponmariammal, land VL.D. Das. 1996. Diallel analysis for fodder yield and its components in cowpea. Madras Agricultural Journal 83(11): 699-701. Rios, G.P. e B.P. das Neves. 1982. Resistencia de linhagens e cultivares de caupi (Vigna unguicu- lata [L.] Walp) ao virus do mosaico severo (VMSC). Fitopatologia Brasileira 7: 175-184. Rios, G.P. E.E. Watt, J.P.P. deAraujo, e, B.P. das Neves. 1982. Cultivar CNC 0434 imune ao mosaico severo do caupi. Pages 113-115 in Reuniao nacional de pesquisa de caupi, 1. Goiania, Resumos. Goiania: EMBRAPA-CNPAF. Roberts, P.A., W.C. Matthews, and lD. Ehlers. 1996. New resistance to virulent root-knot nematodes linked to the Rk locus in cowpea. Crop Science 36: 889-894. Roberts, PA., lD. Ehlers,A.E. Hall, and W.C. Matthews. 1997. Characterization of new resistance to root-knot nematodes in cowpea. Pages 207-214 in Advances in cowpea research, edited by B.B. Singh D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (J1RCAS), I1TA, Ibadan. Nigeria. Rodriguez, I., M.G. Rodriguez, L. Sanchez, and A. Iglesias. 1996. Expression of resistance to Meloidogyne incognita in cowpea cultivars (Vigna unguiculata). Revista de Protection Vegetal 11(1): 63--65. Rodriguez, VlL.B., M. Menezes, RS.B. Coelho, and P. Miranda. 1997. Identification of resistance sources on genotypes of cowpea (Vigna unguiculata [L.] Walpers) a Macrophomina phaseolina (Tass.) Goid., em condicoes de casa-de-vegetacao. Summa Phytopathologica 23(2): 170-172. Rosario, RR del, Y. Lozano, S. Pamorasamit, and M.G. Noel. 1980. The trypsin inhibitor activity oflegume seeds. Philippine Agriculturist 6(4): 339-334. Rocha, M.M., lAA Lima, F.R Freire Filho, C.lS. Rosal, e, VC.V Lima. 1996. Resistencia de caupi de tegumento branco a algumas estirpes de comovirus, potyvirus e cucumovirus. Pages 100-101 in Reuniao nacional de pesquisa de caupi, 4. Teresina. Resumos. Teresina: EMBRAPA- CPAMN. (EMBRAPA-CPAMN, Documentos 18). Santos,AA. dos, e F.R Freire Filho. 1986. Gen6tipos de caupi (Vigna unguiculata [L.] Walp.) com resistencia de campo ao virus do mosaico dourado do caupi. Pages 191-203 in Semim'trio de pesquisa agropecm'tria do piaui, 4. Teresina,Anais. Teresina: EMBRAPA-Uepae de Teresina. Santos,A.A. dos, F.R. Freire Filho, e, M.J. Cardoso. 1987. "BR-lO Piaui", cultivar de feijao macas- sar (Vigna unguiculata [L.] Walp.) com resistencia multipla a virus. Fitopatologia Brasileira 12(4): 402. Saber, M.M. and M.H. Hussein. 1998. Induced mutations for resistance to rust disease in cowpea (Vigna sinensis). Bulletin of Faculty of Agriculture, University of Cairo 49(1): 47-48. Sanginga, N., O. Lyasse, and B.B. Singh. 2000. Phosphorus use efficiency and nitrogen balance of cowpea breeding lines in a low P soil ofthe derived savanna zone in West Africa. Plant and Soil 220: 119-128. Santos, A.A., MA.W. Quindere, and M.B. Melo. 1997. Evaluation of cowpea genotypes for resis- tance to cowpea smut (En tyloma vignae). Fitopatologia Brasileira 22(1): 77-78. Shade, RE., L.L. Murdock, and L.w. Kitch. 1999. Interactions between cowpea weevil (Coleoptera: Bruchidae) populations and Vigna (Leguminosae) species. Journal of Economic Entomology 92(3): 740-745. 38 Digitized by Google Recent progress in cowpea breeding Singh,B.B., S.K.Asante, L.E.N. Jackai, and J.d'Hughes. 1996. Screening for resistance to parasitic plants, virus, aphid and bruchid. I1TAAnnual Report 1996 project 11. Page 24. Singh, B.B., S.K. Asante, D. Fiorini, L.E.N. Jackai, C. Fatokun, and K. Wydra. 1997. Breeding for multiple disease and insect resistance. I1TAAnnual Report. 1997. Project 11. Page 22. Singh, B.B. and H.K. Adu-Dapaah.1998. A partial male sterile mutant in cowpea. African Crop Science Journal 6: 97-101. Singh, B.B. 1998. Sources of resistance to septoria, scab, bacterial blight and Cercospora leaf shot. I1TAAnnual Report 1998. Project 11. Pages 24-27. Singh,B.B. and J. d'Hughes. 1998. Sources of multiple virus resistance. I1TAAnnual Report 1998. Project 11. Pages 24-27. Singh, B.B. and H. Ajeigbe. 1998. Evaluation of improved cowpea varieties in the Sudan savanna. I1TAAnnual Report 1998. Project 11. Pages 14-15. Singh, B.B. and J. d' Hughes 1999. Sources of multiples virus resistance. I1TAAnnual Report 1999. Project 11. Page 30. Singh, B.B. 1999a. Evaluation of new improved breeding lines without insecticide sprays. I1TA Annual Report 1999. Project 11. Page 26. Singh, B.B. 1999b. Screening for heat tolerance. I1TAAnnual Report 1999. Project 11. Page 40. Singh, B.B. 1999c. Improved breeding lines with resistance to bruchid. I1TAAnnual Report 1999. Project 11. Pages 29-30. Singh, B.B. 1999d. Breeding for improved quality. IlIA Annual Report 1999. Project 11. Pages 31-32. Singh, B.B., Y Mai-Kodomi, and T. Terao. 1999a. A simple screening method for drought tolerance in cowpea. Indian Journal of Genetics 59: 211-220. Singh, B.B., Y Mai-Kodomi, and T. Terao. 1999b. Relative droughttolerance ofmajor rainfed crops ofthe semi-arid tropics. Indian Journal of Genetics 59: 1-8. Singh, B.B., O.L. Chambliss, and B. Sharma. 1997. Recent advances in cowpea breeding. Pages 30-49 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS), IlIA, Ibadan. Nige- na. Singh, B.B. 2000. Breeding cowpea varieties for wide adaptation by minimizing genotype x envi- ronment interactions. Pages 173-181 in Genotype x environment interactions analysis of I1TA mandate crops in sub-Saharan Africa, edited by I.J. Ekanayake and R Ortiz. International Insti- tute of Tropical Agriculture, Ibadan, Nigeria. Singh, W.M., H. Adams, R Beveney, and T. Motiram.1994. Evaluation ofthe agronomic performance of cowpea (Vigna unguiculata [L.] Walp.) varieties in the intermediate savannahs of Guyana. Pages 118-121 in Annual review conference proceedings, 20-23 October 1992, National Agri- cultural Research Institute, Caribbean Agricultural Research and Development Institute, Guyana. Sreekumar, S.G., VG. Nair,andRB.Asan. 1993. Gujarath cowpea 2 (Chharodi 1): an ideal cowpea variety for intercropping in coconut garden. Journal of Tropical Agriculture 31(1): 9-12. Tyagi, D.K. and H.S. Chawla. 1999. Effects of seasons and hormones on crossability barriers and in vitro hybrid development between "Vigna radiata and "Vigna unguiculata. Acta Agronimica Hungarica 47: 147-154. Tumwegamire, S., P.R Rubaihayo, and E. Adipala. 1998. Genetics of resistance to Sphaceloma scab of cowpea. African Crop Science Journal 6(3): 227-240. Uday, B. and S. Lodha. 1996. Macrophomina phaseolina induced changes in plant water relations of resistant and susceptible cowpea genotypes. Indian Phytopathology 49(3): 254-259. United States Department of Agriculture (USDA). 2000. Notice of release of Green Dixie Blackeye, a green cotyledon blackeye-type southern pea. USDA, ARS, Washington, DC 20250, USA, 28 April 2000. 39 Digitized by Google Cowpea genetics and breeding Veeranna, Rand MA. Hussain. 1997. Trichomes as physical barriers for cowpea pod borer Maruca testulalis (Geyer) (Lepidoptera: Pyralidae). Insect Environment 3(1): 15. Veerappa, R1998. Phenol and tannin reduce the damage of cowpea pod borer Maruca testulalis. Insect environment 4: 5-6. Van Boxtel, 1, B.B. Singh, G. Thottappilly, andA.J. Maule. 2000. Resistance of (Vigna unguiculata [L.] Walp.) breeding lines to blackeye cowpea mosaic and cowpea aphid borne mosaic potyvirus isolates under experimental conditions. Journal of Plant Disease and Protection 107: 197-204. Vale, C.C. do, e lA. de A. Lima. 1995. Heran9a da imunidade da cultivar Macaibo de Vigna unguiculata ao virus do mosaico severo de caupi. Fitopatologia Brasileira, 20(1): 30-32. Wydra, K. and B.B. Singh 1998. Breeding for resistance to multiple strains of cowpea bacterial blight. IITAAnnual Report 1998. Project 11. Page 27. Viswanathan, P.L. and N. Nadarajan. 1996. Genotype x environment interaction for grain yield in cowpea. Madras Agricultural Journal. 83(11): 707-708. Viswanathan, P.L., S. Murugesan, N. Ramamoorty, P. Veerabadran, K.S. Jehangir, N. Natarajan, C.V Dhanakodi, and P.P.RC. Vamban. 1997. Vamban 1, a new cowpea variety for Tamil Nadu. MadrasAgricultural Journal 84(5): 271-272. Wang, P.Z., Q.P. Feng, lL. Yan, and FX. Han. 1995. Integrated evaluation of elite cowpea germ- plasm resources. Crop Genetic Resources 3: 14-16. Wydra, K. and B.B. Singh 1998. Breeding for resistance to multiple strains of cowpea bacterial blight. IITAAnnual Report 1998. Project 11: Pages 25-27. Yunes,AA., M.T. de Sales, M.P. Andrade, RA. Morais, K.VS. Fernandes, VM. Gomes, 1 Xavier- Filho, and M.T. de Andrade. 1998. Legume seed cicilins (7S storage proteins) interfere with the development of the cowpea weevil (Callosobruchus maculates [F]). Journal of the Science of Food and Agriculture 76(1): 111-116. Zohri, AA. 1993. Studies on some cowpea cultivars. II. Suitability for Aspergillusflavus growth and aflatoxin production. Qatar University Science Journal 13(1): 57-62. 40 Digitized by Google 1.4 Breeding and evaluation of cowpeas with high levels of broad-based resistance to root-knot nematodes J.D. Ehlersl, w.e. Matthews2, A.E. Halll, and P.A. Roberts2 Abstract Host-plant resistance to root-knot nematodes (Meloidogyne spp.) is often the most practical solution for control ofthis pest in cowpea. Resistance in current cultivars is conferred by gene Rk. This gene has been used extensively by breeders and it provides protection to many isolates ofM incognita but only moderate resistance to M javanica. Recently, isolates of M. incognita have been identified at multiple sites in California that are virulent to gene Rk. Development of cultivars with broad-based resistance would increase the effectiveness of host-plant resistance to root-knot nematodes and simplity nematode management. California Blackeye No. 27 was released in 1999 and has broad-based resistance expressed at a high level due to the additive effect of genes Rk plus rk3. Improved cultivars are being devel- oped that carry the broad-based resistance gene f?k2. Twelve sources of additional resistance to root-knot nematodes have been identified recently. Genetic studies on the uniqueness of each source and the potential to pyramid resistance genes are being pursued as part of a comprehensive breeding effort. Introduction Root-knot nematodes (Meloidogyne spp.) are distributed widely in warm temperate, sub- tropical, and tropical regions around the world (Sasser 1980). Nearly all major agronomic, vegetable, and fruit crops, including cowpea, cotton, and tomato, are suitable hosts for one or more root-knot nematode species. At very low population levels, they do little damage to crops. Intensification of cropping with susceptible varieties, particularly on sandy soils, can lead to rapid increases of nematode populations and substantial damage to crops. Soil fumigants are effective in controlling root -knot nematodes and are often used prior to plant- ing high-value orchard or horticultural crops. For most crops, however, genetic resistance and cultural practices, such as periodic fallows and rotation to nonhost crops, are the only practical means of managing these pests. The intensification of agriculture that is occurring in many developing countries will exacerbate root-knot nematode problems. Irrigated land in the tropics that is cropped continuously is especially vulnerable to the buildup of root- knot nematodes to devastating levels. In these situations, cultivation of resistant cowpeas may result in a substantial decrease in root-knot nematode populations. For example, in California in a cover-crop study, resistant cowpea breeding line IT84S-2049 (carrying gene RIC) and the resistant cowpea cultivar Iron Clay (carrying gene Rk) reduced soil 1. Department of Botany and Plant Sciences, University of California, Riverside, CA 92521-0124, USA. 2. Department of Nematology, University of California, Riverside, CA 92521-0124, USA. 41 Digitized by Google Cowpea genetics and breeding population densities of M. incognita significantly, compared to fallow and susceptible cowpea treatments (Matthews et al. 1998). The nematode-resistant cowpea cover crops also produced significantly more biomass, compared to the susceptible check, and the reduced nematode populations resulted in significantly higher yields of a susceptible tomato cultivar, compared to yields obtained when planted after susceptible cowpea. All of the known resistance to root-knot nematodes in cowpea is due to a single gene or locus designated Rk by F ery and Dukes (1980), with alleles rk, ric!, Rk, and Rk2 (F ery and Dukes 1982; Roberts et al. 1996), except for gene rk3 whose effectiveness and inheritance are discussed below. Breeders in the USA and elsewhere have incorporated gene Rk into many cowpea cultivars developed for dry bean and fresh southernpea production (Table 1). In southeastern USA, gene Rk effectively controls M. incognita, M. javanica, M. hap la, and M. arenaria populations (F ery and Dukes 1980). In California, gene Rk confers strong resistance to some biotypes of M incognita, but is only partially effective against aggressive isolates ofM.javanica (Roberts et al. 1997). Rk-virulent populations of M. incognita were identified at two geographically distinct sites in California in the early 1990s (Roberts and Matthews 1995). Since then, additional sites with Rk-virulent field populations of M. incognita and Rk-aggressive populations of M javanica have been iden- tified (Roberts, personal communication), suggesting that the problem is widespread. The emergence of root-knot nematode populations that can be damaging to cowpea carrying the Rk gene suggests new sources of resistance are needed to ensure the con- tinued effectiveness of resistance as a management tool. In addition, new, broad-based resistance is needed to simplify management decisions, because it is difficult to quantify the virulence profile of root-knot nematodes causing damage in a particular field without conducting expensive and lengthy bioassays. These considerations have prompted efforts to identify new sources of broad-based resistance, to understand its genetic basis, and to incorporate this resistance into cultivars. Root-knot nematode resistance classes in cowpea It is convenient to classify the known resistance to root-knot nematodes in cowpea into three types: (1) resistance conferred by gene Rk, (2) broad-based resistance conferred by gene Rk2 and other alleles at the Rk locus distinguishable from the resistance conferred by geneRk, and (3) broad-based resistance found in blackeye cultivar CB27 and breeding line H8-8R, which has been shown to be the result of an additive effect of genes Rk and rk3. Due to the limitations and potential problems in relying on the Rk gene as the sole resis- tance factor in cowpea, we initiated a search for new resistance sources in the early 1 990s. Through extensive screening of more than 600 cowpea accessions from the germplasm collection maintained at the University of California (DC) Riverside, we found that I1TA breeding lines IT84S-2049 and IT84S-2246 had strong resistance to at least several root- knot nematodes, including isolates of Rk-virulent M incognita and Rk-aggressive M. javanica (Table 1) (Roberts et al. 1992, 1994). In subsequenttests, IT84S-2049 had slightly greater resistance to the marker isolate of Rk-virulent M incognita than did IT84S-2246. Therefore, IT84S-2049 was used in a comprehensive genetic study that showed that the resistance is conferred by a dominant allele at the Rk locus, or by another tightly linked gene within 0.17 map units of geneRk (Roberts et al. 1996). This allele was designatedRk2. Accessions PI 441917, PI 441920, and PI 468104 also represent new sources and possess higher levels of resistance to M incognita than Rk-cultivar Mississippi Silver (F ery et al. 42 Digitized by Google Table 1. Summary of level of cowpea resistance to three Me/oidogyne isolates, genotype (hypothesized genotype in parenthesis), and reference for selected cowpea varieties and accessions. Level of resistance to:' M. incognita O:l Cu Itivar/ Accession Source Avirulent Virulent M.javanica Genotype Reference ~ <1> Calif. Blackeye 27 UC Riverside Res Res Res RkRk rk3rk3 Ehlers et al. 2000a ~ IT84S-2049 IITA Res Res Res Rk'Rk' Rk3Rk3 Roberts et al. 1996 Otl II> IT84S-2246 IITA Res Res Res (Rk'Rk' Rk3Rk3) Roberts, unpublished dat :J PI441917 Brazil Res(Res) Res (Sus) Res (Sus) Fery et al. 1994; (Roberts, unpubl. data) 0... <1> PI 441920 Brazil Res(Res) Res (Res) Res (Res) Fery et al. 1994; (Roberts, unpubl. data) Q:; Calif. Blackeye 5 UC Davis Res Sus Mod RkRk Rk3Rk3 Roberts et al. 1997 ~ Calif. Blackeye 46 UC Davis Res Sus Mod RkRk Rk3Rk3 Roberts et al. 1997 II> g. Iron/Clay USA Res NT Mod RkRk Rk3Rk3 Roberts, unpublished data :J Iron USA Res NT Mod (RkRk Rk3Rk3) Mackie 1946 0 Mississippi Silver U Miss Res Sus Mod RkRk Rk3Rk3 Roberts et al. 1995 ...... 8 Holstein Australia Res NT Mod RkRk Rk3Rk3 Roberts, unpublished data ~ Aloomba Australia Res NT NT Roberts, unpublished data .I:>- Christando Australia Res NT NT Roberts, unpublished data <1> w II> Victor USA Res NT Mod Roberts, unpublished data '" ~ Carolina Crowder USDA Res NT NT RkRk Rk3Rk3 Fery and Dukes 1992 §: Bettergro Blackeye USDA Res NT NT RkRk Rk3Rk3 Fery and Dukes 1993 :J- Bettersnap USDA Res NT NT RkRk Rk3Rk3 Fery and Dukes 1995a ciQ. TenderCream USDA Res NT NT RkRk Rk3Rk3 Fery and Dukes 1996 :J- Chinese Red-WR USA Res NT Mod Roberts, unpublished data iil 0 N'dout Senegal Res NT NT Roberts, unpublished data (§ o· <;;- "" Chinese Red USA Sus Sus Sus rkrk Rk3Rk3 Roberts, unpublished data N· 0 CD TVu 4552 Nigeria Mod Sus Sus rkrk rk3rk3 Ehlers et al. 2000b ...... a. ~ 0- Calif. Blackeye 3 UC Davis Sus Sus Sus rkrk Rk3Rk3 Roberts et al. 1997 '< '" C"') Charleston Greenpack USDA Sus NT Sus rkrk Rk3Rk3 Roberts, unpublished data tn· Q; Melakh Senegal Sus NT Sus (rkrk Rk3Rk3) Roberts, unpublished data :J 0 Mouride Senegal Sus NT Sus (rkrk Rk3Rk3) Roberts, unpublished data hl ~ Cabbage Pea USA Sus NT NT (rkrk Rk3Rk3) Roberts, unpublished data 6" Red Ripper USA Sus NT NT (rkrk Rk3Rk3) Roberts, unpublished data :J ......... Tiny Lady USA Sus NT NT (rkrk Rk3Rk3) Roberts, unpublished data <1> (i) :3 'Res = resistant; Mod = moderately resistant; Sus = susceptible to California isolates for work of Roberts or Ehlers et al. and to southeastern US isolates for work of Fery II> 6" and Dukes. ?5-NT = not tested. '" Cowpea genetics and breeding 1994). Genetic analysis of these PI accessions indicated that this heightened resistance also is conferred by a single dominant allele at the Rk locus. Though more effective than gene Rk, these new allelic sources of broad-based resistance are poorly adapted to com- mercial production in the United States. They have low yields, nonmarketable seed types, and other undesirable traits; therefore, a substantial breeding effort was required to utilize their resistance traits. The resistance in IT84S-2049 has been transferred to large-seeded blackeye cowpea breeding lines adapted to California (Ehlers et al. 1999). While these alleles may be useful for broadening the genetic base of resistance to root-knot nematodes, additional nonallelic resistance is desirable to enhance the durability and perhaps the level of nematode resistance in cowpea. University of California Riverside blackeye cowpea breeding line H8-8R, originally selected for heat tolerance, was discovered to also have greater levels of resistance to root-knot nematodes than cultivars possessing gene Rk, such as California Blackeye S (CBS), California Blackeye 46 (CB46), and California Blackeye 88 (CB88) (Roberts et al. 1994). Reproduction and galling on H8-8R caused by M javanica (Rk-aggressive) and Rk-virulent M incognita were about half that observed on CBS and CB46 (Ehlers et al. 1996; Roberts et al. 1997). In two years of field testing on ground infested with an Rk-virulent population of M incognita and races 3 and 4 of Fusarium oxysporum f. sp. tracheiphilum (Fusarium wilt), H8-8-27 (a subline ofH8-8R) exhibited low galling and had higher yields than entries possessing the Rk gene (Ehlers et al. 1995, 1996). H8-8- 27 was released as California Blackeye No. 27 (CB27) in 1999 by the California Crop Improvement Association (Ehlers et al. 2000a). Some of the yield difference observed in these field tests could be attributed to the fact that H8-8-27 also carries a gene that confers resistance to an additional race of Fusarium wilt. The broad-based nematode resistance in H8-8-27 does appear to confer some yield benefit, because it yielded higher than Rk-genotypes that also carried the dual resistance to Fusarium wilt. The presence of strong broad-based nematode resistance in H8-8-27 was also indicated by the comparatively low galling scores (in relation to entries carrying gene Rk) and low numbers of second-stage root-knot juveniles (extracted from soil at harvest) in a field trial conducted on ground infested withMjavanica (Ehlers et al. 1996). The broad-based nematode resistance found in H8-8R is due to the additive effects of the dominant gene Rk and recessive gene rk3 (Ehlers et al. 2000b). These conclusions were drawn from an allelism test to determine the presence of gene Rk in H8-8R and genetic analysis of F l' F 2' and F 2 - derived F 3 (F 2:3) generations of crosses between H8-8R and genotypes with gene Rk (CB88 and CB46). Resistance assays were conducted with either greenhouse-grown potted plants or a modified growth-pouch technique described by Omwega et al. (1988) in a controlled environment chamber. The growth-pouch technique employed a commercial seed germination test pouch (16 x 17 cm) that consisted of a paper wick between two sealed sheets of clear plastic. This allowed full view of the developing root system in two dimensions. Ten- to four- teen-day-old plants in pouches were inoculated with second-stage, root-knot juveniles. The nematodes were allowed to develop and complete one generation (indicated when egg masses were observed on the root surface). This took approximately 30 days at 26.7 °C. A nematode egg mass selective dye was used to stain the egg masses, which were then counted under a lOx magnifying lens to quantify nematode reproduction. Root systems from some pouch tests were processed by extracting nematode eggs with NaOCl 44 Digitized by Google Breeding and evaluation of cowpeas with high levels of resistance to nematodes (Hussey and Barker 1973), which were then counted and expressed as eggs per root system and eggs per gram of root. This permitted further discrimination of resistance reactions in the pouch tests. Resistance in greenhouse tests was determined either by a direct assessment of nematode reproduction on the cowpea plants by counting nematode eggs following their extraction from tomato root systems (Hussey and Barker 1973) or indirectly by using a tomato bioassay technique that consisted of growing susceptible tomatoes in the same soil that hosted the cowpea, and visually rating the tomato root systems for extent of galling. ARk-avirulentM. incognita isolate was used in the allelism test (conducted in growth pouches) to detect the presence of susceptible recombinants in a large F 2 population of a cross between H8-8R and CB88 (a cultivarwith gene Rk). AnM.javanica isolate and aRk-virulent M. incognita isolate were used to distinguish the heightened resistance phenotype from the phenotype conferred by gene Rk in other plant populations (F 1 s, F 2S, and F 3S) developed through crosses between H8-8R and genotypes with gene Rk. In the allelism test, a lack of susceptible recombinants indicated that H8-8R, like CB88, is homozygous for gene Rk or a similar allele that confers resistance equivalent to that of Rk (Ehlers et al. 2000b). This result suggested that H8-8R possesses a unique resistance factor responsible for the enhanced resistance. CBS is predominant in the pedigree of both CB88 and H8-8R and it is the likely donor of the Rk allele presumed present in both lines. Several tests were conducted with F IS obtained from crosses between H8-8-R and CB88 or CB46. In each test, the F 1 S were not distinguishable from the Rk parents, indi- cating that the additional resistance in H8-8R to both M. javanica and Rk-virulent M. incognita is recessive. No differences were found among reciprocal F IS, indicating an absence of maternal effects (Ehlers et al. 2000b). A bimodal F 2 distribution of egg masses per plant was obtained from a preliminary pouch test of a population from the cross CB88 x H8-8R evaluated for resistance to the Rk-virulentM. incognita isolate (Ehlers et al. 2000b). This indicated probable segrega- tion of a single gene conferring a higher level of resistance to this isolate. Determination of inheritance was constrained by the difficulty of designating the class of individual plants that fell between the range of the two parents. It was concluded that F 3 family data would provide clearer genotypic separation than data from single, pouch-grown F 2 plants. Therefore, random F 2:3 consisting of ten plants per family were evaluated for resistance to the Rk-virulentM. incognita isolate in growth pouches. The results of three separate tests with F 2:3 families were consistent with the segregation of a single recessive gene for a higher level of resistance to this isolate (Table 2) (Ehlers et al. 2000b). A similar conclusion was reached for the inheritance of resistance to the M. javanica isolate using an F 2 population of 100 plants derived from the cross CB88 x H8-8R. This F 2 pot test indicated the resistance in H8-8R to M. javanica is also controlled by a single recessive gene (Table 2). A greenhouse pot test of F 3 families derived from 20 random plants from the F 2 was conducted to further test the hypothesis. The F 3 families could be placed into three classes: (1) five non segregating families with all individuals per family expressing a phenotype equivalent to CB88, (2) five nonsegregating families with all individuals per family expressing a phenotype equivalent to H8-8R, and (3) ten families with individual plants within families segregating for the two phenotypes 45 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) .j:>. 0" Table 2. Reaction of parental lines and F2 plants (CB88 x H8-8R) to Me/oidogyne javanica (Rk-aggressive) in a greenhouse pot test and reaction of parental lines and randomly selected F 2,3 families (CB88 x H8-8R) to M. incognita (Rk-virulent) in growth pouches. Number of plants or families Generation Tested with Me/oidogyne javanica (pot test): Parents CB88 H8-8R F2 (CB88 x H8-8R) Equivalent to CB88§ Equivalent to H8-8R§ U nclassified+ Tested with Me/oidogyne incognita (pouch test): Parents CB88 H8-8R F203 (CB88 x H8-8R) Equivalent to CB88§ Equivalent to H8-8R§ Eggs per g of root 36714 20547 41395 18464 25381 28803 14868 37449 20338 Observed 10 10 76 22 2 24 30 63 23 'Expected numbers of plants in F, and numbers of F',3 families for a single recessive gene model (1 :3). 'Determined using Yates correction for continuity. §Classification of F, plants and F',3 families based on log,o transformed data. +Plants not in the range (mean +/- SO) of the parental classes. Source: Ehlers et al. (2000b). Expected' 74 24 64 22 Chi-square* Pvalue 0.217 0.50-0.75 0.062 0.75-0.90 bl ~ ffi ~ :J ::a a" II> :J Q.. 0- til tb ~ ClQ Breeding and evaluation of cowpeas with high levels of resistance to nematodes (Ehlers et al. 2000b). The 20 F 3 families segregated 1:2: 1, thereby confirming that the additional resistance to M. javanica in H8-8R is controlled by a single recessive gene unlinked to gene Rk. To test whether the single recessive gene identified by the Rk-virulent M. incognita isolate was the same gene conferring resistance to theMjavanica isolate, an identical set of 28 F 3 families was evaluated for resistance to both nematodes. Five of the seven F 3 families that were identified as homozygous for resistance to M incognita (resistance equivalent to H8-8R) were also identified as being homozygous for resistance toM javanica (Ehlers et al. 2000b). Although this was not an absolute confirmation, it was strong evidence that the same gene operates to confer the additional resistance to both nematodes. The single recessive gene identified in this study could be viewed as a modifier gene that operates either as a recessive enhancer of gene Rk or as a dominant suppressor of Rk (Roberts et al. 1997). An alternative hypothesis is that this recessive gene confers resistance that is independent of the resistance controlled by the Rk gene, and that the two genes combine to give the higher level of resistance observed in H8-8R. To test the second hypothesis, examination of the pedigree ofH8-8R (UCR breeding line 336 x UCR breeding line 1393) suggests that UCR 1393, and not UCR 336, donated the recessive resistance gene. UCR 336 was the result of a cross between a susceptible parent (CB3) and a parent known to carry only the Rk gene (CB5). Among the parental lines used to develop UCR 1393, the most likely donor(s) of a resistance gene unlinked to theRk gene would be Prima and TVu4552. Therefore, we hypothesized that the Rk allele most likely came from UCR 336, and the recessive resistance allele came from UCR 1393. An assessment of TVu4552 and Prima in growth pouches and inoculated with an Rk-avirulent M. incognita isolate was conducted to determine the probable source of the recessive allele and the nature of its resistance. In this test, the susceptible check (CB3) had a mean egg mass count of 89 (range 37 to 152), and the nematode was controlled by the resistant (gene Rk) check (CB46) (mean of 1; range 0 to 3). While Prima was clearly susceptible (mean of 86; range 27 to 155), TVu4552 had a mean of 35 (range 14 to 50) egg masses per root system. Analysis of variance of these data showed that TVu4552 was significantly more resistant than CB3 but significantly more susceptible than CB46. Prima was not different from CB3 (Ehlers et al. 2000b). These results suggest that TVu4552 carries some resistance, possibly the recessive gene, but not Rk. TVu4552 was crossed reciprocally to susceptible CB3 to test whether the resistance of TVu4552 is recessive and thus the probable donor of the gene identified in H8-8R. The parents, reciprocal F 1 s, and a known Rk genotype, CB46, were screened in growth pouches using the same Rk-avirulentM. incognita isolate as before. The results indicated the moderate resistance observed in TVu4552 is recessive (Ehlers et al. 2000b). The symbol rk3 was proposed for this recessive resistance gene and the probable genotype of TVu4552 is rkrkrk3rk3 (Table 1). The high level of resistance observed in H8-8R appears to be the result of an additive effect of the moderate resistance in TVu4552 conferred by a single recessive gene, rk3, and dominant gene Rk. The occurrence of a root -knot nematode resistance gene in cowpea not linked to the Rk locus is an important finding in that when it is combined with Rk, strong, broadened resistance results that is effective againstMeloidogyne spp. isolates that have become virulent to the Rk gene. Identification of an independent resistance locus opens up the possibility for new gene combinations that may provide resistance that is 47 Digitized by Google Cowpea genetics and breeding more effective than resistance based on Rk. For example, it is possible that combining rk3 with the Rk2 resistance discovered by Roberts et al. (1996), which confers a higher level of resistance than Rk to Rk-virulent M. incognita and M javanica, could lead to an even higher level of resistance that approaches immunity to virulent isolates. Breeding Rk2 resistance into large-seeded blackeyes The broad-based resistance (Rk2) present in I1TA breeding line IT84S-2049 has been transferred to large-seeded blackeye breeding lines that are well adapted to California. Two crossing cycles back to adapted large-seeded blackeye cowpea cultivars was sufficient to obtain elite large-seeded lines with nematode resistance equivalent to IT84S-2049. In the crossing program, blackeyes fully susceptible to root-knot nematode were used in crosses to IT84S-2049. In this way, resistant Rk2 segregants could be identified easily without the potential confounding effects due to the presence of both Rk and Rk2 phenotypes in the populations. Several of the lines with RF resistance are expected to be included in multilocation yield trials in California in 2001. New gene combinations and sources of resistance The existing array of root-knot nematode resistance genes consists of two alleles at one locus (Rk, Rk2), and rk3. The genotype of CB27 (derived from H8-8R) is RkRkrk3rk3, while IT84S-2049 and the derived advanced breeding lines are probably RFRk2Rk3Rk3 (Table 1). It is possible that genotype Rk2Rk2rk3rk3 would have even greater resistance due to the additive effect of rk3 and RF To determine this possibility, F? progeny of crosses between H8-8R and IT84S-2049 are being screened in growth-pouch tests with M. javanica to detect any individual plants expressing the heightened resistance expected from combining rk3 and Rk2. Through recent extensive screening in both field and growth pouch tests, 11 new sources of resistance to root-knot nematodes have been identified in breeding lines from I1TA and in accessions from Australia, Botswana, Kenya, and Niger (Table 3). In addition, breed- ing line 96-11-27, developed at DC Riverside, is a potentially unique source of resistance because it has wild cowpea parentage (from Vigna unguiculata ssp. pubescens). Genetic studies are underway to characterize the inheritance of these resistance sources and their relationship to Rk and Rk2, and the potential for combining resistance factors to obtain even more effective resistance. Some of these new sources of resistance have unique phenotypes (Table 3) or F 1 domi- nance relationships (data not shown) that distinguish them from other known resistance sources, suggesting that they may have unique resistance genes. Highly resistant breeding line IT84S-2049 and most of the other accessions that have equivalent effective resis- tance to M. javanica also express very high resistance to Rk-avirulent M. incognita. An exception is TVu47 65. While highly resistant to M. javanica, both in terms of suppressing root galling and nematode reproduction, TVu4765 is more susceptible to M. incognita (Rk-avirulent) than any of the other accessions when compared to the susceptible check. TVu4765 supported nematode reproduction (as measured by number of egg masses) at a level approximately 40% of that seen with the susceptible check, compared to less than 5% for the other accessions tested (Table 3). This difference may be due to either a new gene that is highly resistant to M. javanica but less effective against avirulent isolates of M. incognita than gene Rk, or perhaps an allele on the Rk locus with a root-knot isolate specificity clearly discernable from that of the known alleles (rk, Rk, Rk2). Allelism tests of crosses between TVu47 65 and genotypes with Rk and Rk2 should help to elucidate the genetic basis for this interesting and potentially useful resistance source. 48 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) .j::o. <.C Table 3. Egg mass production in growth pouches and root galling reaction in a field test to Me/oidogyne javanica (Rk-aggressive> and egg mass production in growth pouches to an Rk-avirulent isolate of M. incognita on cowpea genotypes possessing different resistance factors. M. javanica M. incognita Resistance Egg masses Egg masses per Genotype Origint factor* per root system§ Galling+ root system§ CB3 UCD rk 384 b (2.28) 69.1 CB46 UCD Rk 295 c 2.13 0.8 IT93K-503-1 IITA (Nigeria) ?? 53 e 0.1 IT84S-2049 IITA (Nigeria) Re 54 e 1.10 IT89KD-288 IITA (Nigeria) ?? 66 de 0.50 0.1 TVu 4765 Niger ?? 73 de 0.96 27.5 IT82E-18 IITA (Nigeria) ?? 79 de 1.40 0.5 IT92KD-370 IITA (Nigeria) ?? 80 de 0.09 Bots 249C Botswana ?? 06 de 0.67 0.0 Bots 514B Botswana ?? 106 de 0.07 0.0 96-11-27 UCR wild x cultivated ?? 116 de (1.90) KVu 515 Kenya ?? 118 de 1.00 2.0 Bots 444 Botswana ?? 122 de 0.07 0.5 Bots Bl Botswana ?? 126 d 0.00 0.0 TVu 1015 Nigeria ?? BOd 0.00 0.0 1C2899 India ?? 211 - 0.51 0.0 TVu 8016 Nigeria ?? 456 a 1.20 H8-9 UCR rk 3.45 LSD (P = 0.05) 59 tUCD-University of California, Davis; UCR-University of California, Riverside; IITA-Internationallnstitute ofTropical Agriculture. *rk = susceptible allele and Rk' = broad-based resistance allele found in IT845-2049. §Means of five replicates. Values in column followed by the same letter are not significantly different according to Duncan's multiple range test (P :J 0... (J) Q:; t g. :J o ...... 8 ~ ~ '" ~ §: ~ ciQ. ~ iil (§ <;;- o ...... ffi tn· Q; :J hl 6" :J ~ II> 6" ?5- '" Cowpea genetics and breeding In response to inoculation by M javanica, accessions Bots 444, Bots B 1, and TVu 1015 supported nematode reproduction that was more than twice as high as in IT84S-2049, but they exhibited virtually no root galling (Table 3). Again, these differences in resistance phenotypes could be accounted for by either the existence of new resistance alleles or loci, or genetic background effects. Further genetic studies are in progress to clarify the nature and effectiveness of these genotypes. The phenotypes displayed by accessions IC2899 and TVu80l6 are of particular inter- est (Table 3). IC2899 supports virtually no reproduction of M. incognita, but supports high levels of reproduction with very low root galling in response to inoculation with M. javanica. TVu80 16 supports extremely high reproduction of M javanica and yet its galling score is in the range of most of the accessions with relatively high levels of resistance to this nematode, including IT84S-2049. Although they are not likely candidates as strong resistance sources, genetic investigations of IC2899 and TVu80 16 may shed light on the relationship between nematode reproduction and galling in cowpea. These lines may pos- sess a gene(s) for controlling the galling reaction, but lack any genes resisting nematode reproduction. Separate genetic control of reproduction and galling in response to root-knot nematodes has been observed in lima bean (Phaseolus lunatus), where analyses of F l' F 2' and F 3 generations of a cross between a resistance source (PI 256874) and a susceptible commercial lima bean identified two independently inherited genes (Matthews et al. 2000). One of the genes was found to confer resistance to nematode reproduction. The second gene was found to confer resistance to the root galling reaction, but had no effect on reproduction. It is reasonable to expect this also may occur in cowpea. Observations of some phenotypes in cowpea lines and cultivars that were inoculated with an Rk-avirulent isolate of M. incognita support the possibility of independent genetic control of these responses to root-knot nematodes. For example, we have observed that cultivar CB3 supports high levels of reproduction with little or no galling, whereas cultivar Chinese Red supports high levels of both reproduction and galling. It may be that while both of these cultivars lack genes resisting reproduction, CB3 may carry one or more genes that control the galling reaction to this nematode. Crosses between CB3 and Chinese Red have been made and will be analyzed to investigate this phenomenon. Acknowledgements This research was supported in part by grants from the Blackeye Council of the California Dry Bean ResearchAdvisory Board and the Bean/Cowpea Collaborative Research Support Program, USAID Grant no. DAN-G-SS-86-00008-00. The opinions and recommendations are those of the authors and not necessarily those of USAID. References Ehlers, J.D.,A.E. Hall, P.A. Roberts, W.C. Matthews, A.M. Ismail, andA.N. Eckard. 1995. Black- eye varietal improvement. Pages 41-55 in University of California Dry Bean Research: 1995 Progress Report. California Dry Bean Advisory Board, Dinuba, CA, USA. Ehlers, J.D., A.E. Hall, PA. Roberts, W.C. Matthews, A.M. Ismail, B.L. Sanden, CA. Frate, and A.N. Eckard. 1996. Blackeye varietal improvement. Pages 51-54 in University of California Dry Bean Research: 1996 Progress Report. California Dry Bean Advisory Board, Dinuba, CA, USA. Ehlers, J.D.,A.E. Hall, A.M. Ismail, PA. Roberts, W.C. Matthews, B.L. Sanden, CA. Frate, and S. Mueller. 1999. Blackeye varietal improvement. Pages 47-61 in University of California Dry Bean Research: 1996 Progress Report. California Dry Bean Advisory Board, Dinuba, CA, USA. Ehlers, J.D., A.E. Hall, P.N. Patel, P.A. Roberts, and W.C. Matthews. 2000a. Registration of Cali- fornia Blackeye 27 Cowpea. Crop Science 40: 854-855. 50 Digitized by Google Breeding and evaluation of cowpeas with high levels of resistance to nematodes Ehlers, J.D., W.C. Matthews,A.E. Hall, and PA. Roberts. 2000b. Inheritance of a broad-based form of root-knot nematode resistance in cowpea. Crop Science 40: 611-618. F ery, RL. and P.D. Dukes. 1980. Inheritance of root-knot nematode resistance in the cowpea (Vigna unguiculata [L.] Walp.). Journal of American Society of Horticultural Science 105: 671-674. F ery, RL. and P.D. Dukes. 1982. Inheritance and assessment of a second root-knot resistance factor in southernpea (Vigna unguiculata [L.] Walp.). HortScience 17: 152 (Abstract). Fery, RL. and P.D. Dukes. 1992. Carolina Crowder southernpea. HortScience 27: 1335-1337. Fery, RL. and P.D. Dukes. 1993. Bettergro Blackeye southernpea. HortScience 28: 62-63. F ery, RL. and P.D. Dukes. 1995a. Registration ofUS-566, US-567, and US-568 root-knot nematode resistant cowpea germplasm lines. Crop Science 35: 1722. Fery, RL. and P.D. Dukes. 1995b. Bettersnap southernpea. HortScience 30: 1318-1319. Fery, RL. and P.D. Dukes. 1996. Tender Cream southernpea. HortScience 31: 1250-1251. F ery, RL., P.D. Dukes, and J.D. Thies. 1994. Characterization of new sources of resistance in cowpea to the southern root-knot nematode. HortScience 29: 678-679. Hussey, RS. and K.R Barker. 1973. A comparison of methods of collecting inocula for Meloidogyne spp., including a new technique. Plant Disease Reports 57: 1025-1028. Mackie, w.w. 1946. Blackeye beans in California. University of California Agricultural Experi- mental Bulletin 696. Matthews, W.C., J.D. Ehlers, W. Graves, PA. Roberts, and J.V Samons. 1998. Use of resistant cover-crop cowpeas in crop rotations to reduce levels of root-knot nematodes. Page 114 in 1998 Annual Meeting Abstracts, American Society of Agronomy, 90th Annual Meeting, 18-22 Octo- ber 1998, Baltimore, Maryland, USA. ASA, CSSA, SSSA, Madison, Wisconsin, USA. Matthews, W.C., D.M. Helms, and PA. Roberts. 2000. Evidence for independent genetic control of reproduction and galling to Meloidogyne javanica in lima bean. Page 49 in Program and Abstracts of the 37th Annual Meeting, Society of Nematologists, 24-28 June 2000, Quebec, Canada. SON, Lawrence, Kansas, USA. Omwega, C.O., I.J. Thomason, and P.A. Roberts. 1988. A nondestructive technique for screening bean germplasm for resistance to Meloidogyne incognita. Plant Disease 72: 970-972. Roberts, P.A., W.C. Matthews,A.E. Hall, J.D. Ehlers, S.R Temple, and D.M. Helms. 1992. Black- eye bean tolerance and resistance to root-knot nematodes. Pages 43-49 in University of Calif or- nia Dry Bean Research: 1992 Progress Report. California Dry Bean Advisory Board, Dinuba, CA,USA. Roberts, P.A., W.C. Matthews,A.E. Hall, J.D. Ehlers, S.R Temple, and D.M. Helms. 1994. Black- eye bean tolerance and resistance to root-knot nematodes. Pages 57-59 in University of Calif or- nia Dry Bean Research: 1994 Progress Report. California Dry Bean Advisory Board, Dinuba, CA,USA. Roberts, PA. and W.C. Matthews. 1995. Virulence in Meloidogyne spp. to resistance in cowpea. Nematologica 41(3): 336 (abstract) Roberts, PA., C.A. Frate, W.C. Matthews, and P.P. Osterli. 1995. Interactions of virulent Meloido- gyne incognita and Fusarium wilt on resistant cowpea genotypes. Phytopathology 85: 1288- 1295. Roberts, P.A., W.C. Matthews, and J.D. Ehlers. 1996. New resistance to virulent root-knot nematodes linked to the Rk locus of cowpea. Crop Science 36: 889-894. Roberts, PA., J.D. Ehlers,A.E. Hall, and W.C. Matthews. 1997. Characterization of new resistance to root-knot nematodes in cowpea. Pages 207-214 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (ilTA) and Japan International Research Center for Agricultural Sciences (J1RCAS). ilTA, Ibadan, Nigeria. Sasser, J.N. 1980. Root-knot nematodes: a global menace to agriculture. Plant Disease 64: 36-41. 51 Digitized by Google 1.5 Breeding cowpea for resistance to insect pests: attempted crosses between cowpea and Vigna vexillata c.A. Fatokun1 Abstract Cowpea is grown mainly for its protein-rich grains, which is consumed in various forms in sub-SaharanAfrica.Average grain yield in farmers' fields is generally low due to a number of biotic and abiotic stresses. The most important of the biotic stress factors causing extensive grain yield losses in cowpea are postflowering insect pests such as the legume pod borer and pod sucking bugs. Availability of varieties with resistance to these pests will be attractive to cowpea farmers as the crop could then be grown with less dependence on expensive, often adulterated chemicals that are not particularly environmentally friendly. To be able to develop such varieties, it is necessary that genes conferring resistance to these pests are available in the cowpea genome. Genes conferring resistance to these pests were found to exist in the genomes of some wild Vigna species such as V vexillata and V oblongifolia and efforts were made to transfer these genes from the wild Vigna sp. to cowpea. Pods were retained for up to seven days after pollination when V vexillata lines served as female parents with cowpea. Embryos in pods resulting from these crosses did not develop beyond the globular stage. Several procedures aimed at overcoming this incompatibility were adopted without success. Among the techniques used to overcome incompatibility were in vitro culture of interspe- cific hybrid embryos, hormonal treatments offlower buds prior to pollination, and polyploidization. No interspecific hybrids were obtained following the several attempts made, thus suggesting that very strong cross-incompatibility exists between cowpea and V vexillata. Introduction Cowpea (Vigna unguiculata [L.] Walp.) is grown mainly for its grain, which contains between 22 and 32% protein on a dry weight basis. The grain is one of the cheapest sources of protein in the diets of peoples of West and Central Africa where cowpea is also an important crop. The dried grain is consumed after being processed into different food forms while the haulms from dried and shelled pods as well as fodder, are a good source of quality feed for livestock. Farmers in the dry savanna areas of West and Central Africa derive some income from selling cowpea fodder to livestock owners, particularly during the dry season. Every stage in the life cycle of cowpea has at least one major insect pest that could cause serious damage and impact yield negatively. When postflowering insect pests infest 1. International Institute of Tropical Agriculture, PMB 5320, Oyo Road, Ibadan, Nigeria. 52 Digitized by Google Breeding for resistance to insect pests: crosses between cowpea and Vigna vexi lIata cowpea fields and cause heavy damage to grain yield, farmers, especially those in the dry savanna area, resort to harvesting the fodder in order to get some income. There is, however, no doubt that farmers get more financial benefit from cowpea grain than from fodder. In order for farmers to obtain high grain yield from their cowpea fields it is nec- essary for them to spray the cowpea plants with insecticides a number of times. Until recently, cowpea required being sprayed with insecticides up to five times or even more if high grain yield was to be obtained. Relatively high grain yield can now be obtained with two or three insecticide sprays. The high grain yield which can now be obtained with fewer insecticide spray regimes can be attributed to progress that has been made through genetic improvement whereby genes for resistance to some diseases and preflowering insect pests such as aphids have been incorporated into new cowpea varieties. Also, there are cowpea lines that combine these resistance genes with low levels of resistance to the flower bud thrips (Megalurothrips sjostedti). Even some traditional farmer varieties have also been improved by introgressing into them these generally simply inherited resistance genes. Furthermore, the population dynamics for most of the insects have been studied and information obtained has been found useful to target the time of intervention by farmers. No cowpea line has been identified as possessing the desired levels of resistance to the legume pod borer (Maruca vitrata) and pod sucking bugs (Clavigralla tomentosicollis, Anoplocnemis curvipes, and Riptortus dentipes) all of which are postflowering pests. The legume pod borer and pod sucking bugs can cause tremendous grain-yield losses in cowpea if appropriate control measures are not taken. The most economical and environmentally friendly way of controlling these insect pests would be through host-plant resistance. Introgressing genes for resistance to the insect pests into cowpea should result in the availability of varieties which can be grown by farmers in sub-Saharan Africa with minimal use of chemicals. This will lead to a reduction in the cost of cowpea production, thereby increasing the profit margin for farmers. Essentially cowpea production will become more attractive to the generally resource-poor farmers in the savanna zones of Africa. In addition, the farmers would be healthier as they no longer need to handle toxic chemicals while at the same time pollution of their environment would be immensely reduced. Because of the poten- tially immense benefits of growing insect-resistant cowpea varieties, no efforts should be spared in the search for and transfer of the desired genes from landraces and wild Vigna species to cultivated cowpea. To this end, a wide range of accessions from the cowpea germplasm collection as well as those of its wild and weedy cross-compatible relatives was screened in order to identify those with genes for resistance to the pests that wreak havoc on cowpea production. None of the tested cultivated cowpea lines and their cross-compatible wild relatives showed the desired high level of resistance to these pests. Several accessions of some Vigna species, such as those belonging to ~~ vexillata, ~~ davyi, ~~ oblongifolia, and ~~ luteola, were also screened, among others, for resistance to insect pests of cowpea. The results showed that some accessions of ~~ vexillata and ~~ oblongifolia have good levels of resistance to the insect pests that devastate cowpea. The Vigna species whose accessions showed resistance to the major postflowering insect pests of cowpea do not belong to same primary or secondary gene pool as cowpea and this could constitute a major constraint to moving the desirable genes into cultivated cowpea varieties. A phylogenetic study that was carried out involving various Vigna species and based on RFLP markers indicated that among species showing high levels of resistance to the 53 Digitized by Google Cowpea genetics and breeding insect pests, ~~ vexillata is the closest to cowpea (~~ unguiculata) (Fatokun et al. 1993). In this same study, a wild and cross-compatible relative of cowpea v~ unguiculata ssp. dekindtiana var. pubescens was linked to ~~ vexillata when the various accessions of Vigna species tested were displayed on a minimum spanning tree. The various accessions of ~~ vexillata showed high levels of resistance to pod sucking bugs, flower thrips, Maruca vitrata, bruchid, and Striga gesnerioides among others. Pos- session of these traits makes interspecies crosses between it and cowpea very attractive and worth pursuing. Hence, crosses were initiated between cowpea and ~~ vexillata with the aim of transferring the genes conferring resistance to insect pests from the latter to cowpea. While making these crosses (~~ vexillata x cowpea) it was observed that some pods were retained for up to seven days or even more when cowpea is the pollen parent. However, in the reciprocal crosses pods were not retained as emasculated flowers drop within one day following cross pollination. On the other hand, the pods that were retained by pollinating ~~ vexillata with cowpea developed slowly as the seeds contained therein. By the time these pods attained their maximum size, they only approximated the size attained by four-day-old pods resulting from selfing. In all the crosses, flowers were emasculated and pollinated a day before anthesis. This was to ensure that pollen tubes reached the ovule in order to release the male (sperm) nuclei in time for fertilization to take place. No viable interspecific hybrid seed was obtained from any of the several hundreds of crosses made, thus suggesting a strong cross incompatibility between the two species, ~~ vexillata and ~~ unguiculata. Overcoming interspecies incompatibility There are a number of procedures that have been used by breeders to overcome barriers that prevent gene exchange between distantly related plant species. These have been used to successfully effect interspecies hybridization in several crops. Among the procedures commonly used are making reciprocal crosses (Thomas and Waines 1982), crossing between different accessions of both species (Harlan and de Wet 1977), polyploidization followed by crossing, polyploidization of the F 1 interspecies hy brid (where the F 1 is sterile), embryo rescue (przywara et al. 1989), bridging crosses (Hermsen and Ramanna 1973), and hormonal treatment of flower buds prior to or after pollination (Larter and Chaubey 1965; Sitch and Snape 1987), among others. Some of the methods used to overcome constraints to gene exchange through wide crossing in several crops were tried in the attempted cross between cowpea and ~~ vexillata and these are reported in the following sections. Crossing several accessions of both species: Reports of previous wide crossing activities in some crops have shown that hybrids between certain accessions of a species are more productive than others. This is because certain accessions of a species are better combin- ers with some other individuals of another species. In tobacco (Nicotiana tabacum) Pit- tarelli and Stavely (1975) observed that when three different cultivars were crossed to N. repanda only one combination gave F 1 hybrid plants. Harlan and de Wet (1977) also tested a number of Tripsacum dactyloides in combination with com and found that only one of the T. dactyloides accessions was effective in transferring genetic material to maize. In the attempted cross between cowpea and ~~ vexillata, some pods are retained on ~~ vexillata when emasculated flowers are pollinated with cowpea but none were retained in reciprocal crosses. It is conceivable that not all1ines of ~~ vexillata will respond in the same way as, for example, in the frequency of pod retention when flowers are pollinated with different 54 Digitized by Google Breeding for resistance to insect pests: crosses between cowpea and Vigna vexi lIata cowpea accessions. It is also possible that in some specific combinations the embryos may develop beyond the globular stage. Perhaps some V vexillata lines might even support the development of pods with well-formed seeds to maturity while others do not. Hence several accessions of ~~ vexillata were selected for crossing with cowpea. Pollen from four wild cowpea relatives belonging to ~ ~ unguiculata ssp. dekindtiana and ten cultivated cowpea (~~ unguiculata) lines were used to pollinate emasculated flowers of 64 different accessions of ~~ vexillata. There were differences among the accessions of ~~ vexillata used in making these crosses in the frequency of pods they retained fol- lowing pollination with cowpea lines or wild cowpea relatives. While accession TVNu 73 retained up to 70% pods following the interspecies pollination, only a few pods were retained by some other accessions such as TVNu 1359 (Table 1). It should be noted that the retained pods were on the plants for no longer than eight to ten days after pollination. They shriveled and fell off the plant prematurely. Pods resulting from selfing on ~~ vexil- lata remain on the parent plants until they dry and are harvested. No appreciable differ- ences were observed in the frequency of pod retention on the basis of which cowpea or dekindtiana line was used as pollen parent. Also, there were no observed differences in embryo development when random samples of ovules in retained pods were dissected. Essentially none of the ovules from the interspecific hybridization had an embryo that developed beyond the globular stage. Use of mixed pollen: Cowpea pollen grains do produce tubes albeit at low frequencies when placed on the stigma of V vexillata. Also, some of the developed pollen tubes are malformed and are therefore unable to penetrate the style fast enough to reach the ovule in order to effect fertilization (Barone and Ng 1990). A few pollen grains of the female W vexillata) plants were deliberately placed on the stigma along with some of cowpea. Pods developed on the vexillata plants when the mixed pollen grains were used. The number of normal sized seeds in each pod was few but none of the seeds resulted from interspecific hybridization. Payan and Martin (1975) used the mixed pollen technique to successfully effect interspecies cross in the genus Passiflora (passion fruit). Application of growth hormones: Growth promoting hormones have been used to facili- tate interspecies crosses in many crops. Generally, hormones are known to prolong the period during which fruits are retained on plants. In P haseolus, Al-Yasiri and Coyne (1964) used growth hormones to prolong the period of pod retention following wide crossing to 30 days as against 15 for untreated pods. Gibberellic acid and NAA are two commonly Table 1. Frequency of pod retention by some accessions of Vigna vexillata following pollination with cowpea. Accession TVNu 73 TVNu 1616 TVNu 719 TVNu 1344 TVNu 64 TVNu 72 TVNu 1544 TVNu 180 TVNu 1359 55 Percentage pod retained 42 41 36 33 33 23 Digitized by Google 8 3 1 Cowpea genetics and breeding used hormones to treat flowers in order to enhance interspecies crosses. Two auxins (2,4-D & NAA) and one cytokinin (kinetin) were applied as sprays at low concentrations (approximately 1.0 mg/l) and in various combinations on flowers of ~~ vexillata before or after pollination with cowpea. In particular, 2,4-D was effective in promoting the reten- tion of ~~ vexillata flowers pollinated with cowpea and subsequently the pods resulting from the cross-pollination. Pods that formed from ~~ vexillata flowers sprayed with 2,4-D and pollinated with cowpea developed on the plants and at maturity were bigger in size than those resulting from selfing with no 2,4-D sprayed (Fig. 1). These pods remained on the peduncles until they dried as for normal pods resulting from self-pollination. When pods resulting from flowers sprayed with 2,4-D had matured and were opened, the ovules contained in them did not develop beyond the size of three-day -old ovules of selfed pods. In addition, there was a mass of white colored loose callus-like structures, which filled the spaces between adjacent ovules (Fig. 2). When emasculated flowers were sprayed with 2,4-D but not pollinated, the flowers remained attached to the peduncle for up to six days before falling off. Pods were, however, not initiated from such nonpollinated flowers even with the hormonal treatment. The retention and development of pods on ~~ vexillata following pollination with cowpea and 2,4-D spray is further evidence that fertilization does occur, leading to embryo initiation. According to Barone and Ng (1990), between 15 and 20% of ovules are fertilized when ~~ vexillata flowers are pollinated with cowpea. However, the embryos in ovules could not go through the normal stages of development for some reasons. These observations show that prolonging the retention and develop- ment of pods resulting from the ~~ vexillata by cowpea crosses on the female parent did not lead to further embryo development. Deakin et al. (1971) made similar observations in interspecies crosses in cucumber. The application ofNAA as spray to flower buds was not as effective as 2,4-D in promot- ing retention of pods on ~~ vexillata following pollination with cowpea. The pods result- ing from flowers sprayed with NAA increased in size and were only slightly bigger than those that were not sprayed. Also NAA did not increase the frequency of pod retention as compared to when cross-pollinated flowers were not treated with a hormone. Embryo rescue: Developments and improvements in tissue and cell culture techniques have contributed immensely to progress made in the exchange of genes between species in many crops. In vitro culture methods have been used to rescue young interspecific hybrid embryos prior to their abortion. This is particularly important in situations where the cause of incompatibility occurs postfertilization such as endosperm abortion and eventual starvation of the embryo. Fatokun and Singh (1987) needed to rescue embryos of the cross between cowpea and a wild relative, ~~ unguiculata ssp. pubescens, otherwise the embryos resulting from the cross collapsed before attaining full development. Barone et al. (1992) reported that the embryo and endosperm resulting from the cross between ~~ vexillata and cowpea collapsed within five and eight days following pollination. The development of an embryo especially during the early stages depends on the existence of a well-formed endosperm, which is the primary source of nourishment for the embryo. Also, it is essential that a harmonious relationship should exist between the embryo and endosperm tissue if the former is to go through the process of development. When excised, the embryos in ovules resulting from pollinating ~~ vexillata with cowpea attained the globular stage of development (F atokun 1991). The successful rescue of interspecific hybrid embryos that are this young (i.e., at the globular stage) has been 56 Digitized by Google Breeding for resistance to insect pests: crosses between cowpea and Vigna vexi lIata Figure 1. Pods of Vigna vexillata; (A) mature selfed pods, (8) mature pods from flowers pollinated with cowpea and sprayed with 2,4-D, and (C) seven-day-old pods from flowers pollinated with cowpea but not treated with 2,4-D. Figure 2. Opened pod of Vigna vexillata from flowers pollinated with cowpea and sprayed with 2,4-D showing ovules (arrows) which did not develop into seeds. 57 Digitized by Google Cowpea genetics and breeding difficult to achieve in many plants. Embryo rescue is more successful as the embryo gets older. Usually the development of embryos into plants occurs more readily once they have passed the globular stage of development and beyond. It is obvious from the observations and reports mentioned above that fertilization does take place when pollen of cowpea are placed on the stigma of ~~ vexillata. Entire ovules resulting from pollinating ~~ vexillata with cowpea were cultured in MS media containing 10% coconut water, 1 % casein hydrolysate, and varying levels and combinations of sucrose and benzyl adenine. In ovulo culture was used because the embryos are small and difficult to dissect and excise. The presence oflow levels (1-3%) of sucrose along with the other organic components added to the media encouraged the development of young selfed embryos (as young as four days) of both species to develop into plants in the culture tube. However, none of the hybrid embryos developed into plants following placement in the culture media. A few of the ovules (both selfed and hybrid) formed calluses especially in the media containing growth hormone, but when subcultured, no plants could be regenerated from them. Polyploidization: Increasing chromosome number of one or both species can enhance crossability between two species. This is particularly so in cases where the two species being crossed differ in their genome number. However, both cowpea and V vexillata have the same number of chromosomes (2N = 22) as all other members of the genus Vigna with the exception of ~~ glabrescens which has 2N = 2X = 44. Vigna glabrescens is the only naturally occurring polyploid in the subtribe Phaseolinae (Marechal et al. 1978). Polyploids were induced in cowpea following treatment of shoot tips of young seedlings with a weak solution of colchicine. Different accessions responded differently to colchicine such that a higher frequency of polyploids were induced in some than in others following similar treatments. The induced polyploids are fertile but produce fewer seeds per pod compared to their diploid counterparts. In addition the plants produced only a few pods each. The polyploid cowpea plants were characterized by thick leaves with large guard cells around the stomates, larger flowers, and pollen grains that were mostly rectangular in shape. Root tip cells obtained from the plants had 2N = 2X = 44 chromosomes. Young seedlings of ~~ vexillata, treated with colchicine were more sensitive to the chemical than cowpea. Further growth and development of seedlings were arrested at the shoot tips fol- lowing application of 0.1 % colchicine for a period of 12 hours. Application of colchicine at 0.1 % to seedling shoot tips for 24 hours was found to be most effective in cowpea. In ~~ vexillata seedlings treated with colchicine new shoots developed from the roots rather than from the shoot tips. In an attempt to promote development of new shoots from treated buds of ~~ vexillata, young shoots were grafted on cowpea plants as stock. Shoot tips and axillary buds of the scion were treated with 0.1 % colchicine for 12 hours. No shoots developed from any of the axillary buds or shoot tips of the scion treated with colchicine. Hence no polyploids could be induced in V. vexillata using the same concentration of colchicine that was effective on cowpea. Pollen of V. vexillata was placed on the stigma of polyploid cowpea flowers but all such flowers dropped within 24 hours of pollination hence no pods developed on the polyploid cowpea plants. Bridge crossing: Successful interspecific crosses have been made in plants through bridg- ing of crosses. Where direct crosses are not feasible between two species their genomes can be brought together by indirect means. For example Nicotiana tabacum does not readily cross with N. repanda. However, both species can cross successfully with N. sylvestris. 58 Digitized by Google Breeding for resistance to insect pests: crosses between cowpea and Vigna vexi lIata The desirable gene for disease resistance present in N. repanda could be transferred to tobacco by first crossing N. repanda to N. sylvestris and the progeny of this cross was then crossed to tobacco (Burk 1967). In order to effect gene transfer from ~~ vexillata to cowpea, crosses were made between the former and a close relative ~~ davyi on the one hand and between cowpea and ~~ unguiculata ssp. dekindtiana on the other. Both ~~ vexillata and ~~ davyi belong to the same section Plectotropis in the genus Vigna and this was the first reported successful cross between V vexillata and any other Vigna species. However, ~~ davyi and ~~ unguiculata ssp. dekindtiana could not be crossed successfully. Cowpea and ~~ unguiculata ssp. dekindtiana belong to the same section Catiang of the genus. The hybrid resulting from the cross between ~~ vexillata and ~~ davyi was partially fertile (Table 2) as that between cowpea and its wild relative. The hybrids W vexillata x ~~ davyi and ~~ unguiculata x ~~ unguiculata ssp. dekindtiana) were crossed to each other and to the four parents but the efforts did not yield the desired products as no seeds were set in the crosses between members from different sections. Use of a parthenocarpic cowpea line: A cowpea line (RI 36) showing parthenocarpy was identified among the progeny of a cross between IT84s-2049 and IT88s-524-B at the University of California, Riverside (1. Ehlers, personal communication). This cowpea line has the capacity to form and retain pods to maturity from emasculated flowers even when not pollinated. This cowpea line therefore served as the female parent and pollinated using a number of ~~ vexillata accessions. Seeds formed in the pods and these appeared to develop normally for the first ten days after which they started to shrivel. There was a mass of cells connecting each seed to the pod wall. Seeds were excised from pods on different days after pollination for placement in the culture media. Embryos could not be readily distinguished in the seed hence all seeds (ovules) were excised and placed in culture media. The only development observed on the cultured seeds was root initiation (Fig. 3) but no shoots were formed. Table 2. Morphological attributes of F, interspecific hybrid between Vigna vexillata and V. davyi and their parents. Character tTVNu 1335 Leaf length (em) 8.3 Petiole length (em) 3.8 Pod length (em) 9.6 Seeds per pod 10.8 Pollen stainability ('Yo) 96.2 Peduncle length (em) 24.6 tv. davyi. tv. vexillata. Conclusion F, 12.7 5.6 9.0 6.3 59.2 25.6 *TVNu 72 15.5 8.8 11.4 15.9 97.9 17.4 Attempts have been made to cross cowpea with ~~ vexillata using various techniques that have successfully been used to effect wide crosses in some other crops. These efforts did not yield the expected results, thus suggesting the existence of a strong cross incompatibility barrier between cowpea and ~~ vexillata. The causes of incompatibility between the two species are both pre- and postfertilization. In the first place, only a few 59 Digitized by Google Cowpea genetics and breeding Figure 3. Root developing from cultured seed of cowpea line RI 36 following pollination with Vigna vexillata. cowpea pollen tubes are able to penetrate the styles of ~~ vexillata and reach the ovule in order to effect fertilization. There is sufficient evidence that fertilization does occur, albeit at relatively low frequency. The fertilization probably gives rise to diploid zygotes, which develop to the globular stage embryo. The nondevelopment of the hybrid embryos beyond the globular stage may be an indication of incomplete fertlization in which the second male nucleus does not fuse with the diploid endosperm nucleus to give the triploid tissue that normally feeds the embryo. Consequently there is no triploid endosperm tissue formed following the cross between cowpea and ~~ vexillata. The absence of the endosperm leads to starvation and subsequent collapse of the embryos. There is in the genome of ~~ vexillata a repertoire of genes that could confer resistance to several of the pests and diseases to which cowpea succumbs. The attempts made so far, using sexual means, to move desirable genes from ~~ vexillata to cowpea have not yielded the desired results. Perhaps other avenues by which these genes could be accessed should be explored. An approach would be the identification and cloning of these genes which eventually could be used to transform cowpea. This is a much longer route to take but it might be worth the efforts because of the potential benefit. References AI-Yasiri, A. and D.P. Coyne. 1964. Effects of growth regulators in delaying pod abscission and embryo abortion in the interspecific cross Phaseolus vulgaris x P. acutifolius. Crop Science 4: 433-435. 60 Digitized by Google Breeding for resistance to insect pests: crosses between cowpea and Vigna vexi lIata Barone, A., A. Del Guidice, and N.Q. Ng. 1992. Barriers to interspecific hybridisation between Vigna unguiculata and V vexillata. Sexual Plant Reproduction 5: 195-200. Barone, A. and N.Q. Ng. 1990. Embryological study of crosses between Vigna unguiculata and V vexillata. Pages 151-160 in Cowpea genetic resources, edited by N.Q. Ng and L.M. Monti. International Institute of Tropical Agriculture, Ibadan, Nigeria. Burk, L.G. 1967. An iterspecific bridge-cross-Nicotiana repanda through N. sylvestris to N. taba- cum. Journal of Heredity 58: 215-218. Deakin, J.R., G.w. Bohn, and T.W. Whitaker. 1971. Interspecific hybridisation in Cucumis. Econonomics of Botany 25: 195-210. Fatokun, CA. 1991. Wide hybridisation in cowpea: problems and prospects. Euphytica 54: 137- 140. Fatokun, C.A., D. Danesh, N.D. Young, and E.L. Stewart. 1993. Molecular taxonomic relationships in the genus Vigna based on RFLP analysis. Theoretical and Applied Genetics 86: 97-104. Fatokun, C.A. and B.B. Singh. 1987. Interspecific hybridisation between Vigna pubescens and V unguiculata (L.) Walp. through embryo rescue. Plant Cell, Tissue, and Organ Culture 9: 229-233. Harlan, J.R. and J.M.J. de Wet. 1977. Pathways of genetic transfer from Tripsacum to Zea mays. Proceedings ofthe National Academy of Science, USA 74: 3494-3497. Hermsen, J.G. and M.S. Ramanna. 1973. Double bridge hybrids of Solanum bulbocastanum and cultivar S. tuberosum. Euphytica 22: 457-466. Larter, E. and C. Chaubey. 1965. Use of exogenous growth substances in promoting pollen tube growth and fertilisation in barley-rye cross. Canadian Journal of Genetic Cytology 7: 511-518. Marechal, R., J.M. Mascherpa, and F. Stainer. 1978. Etude taxonomique d'un groupe complexe d' especes des genresPhaseolus et Vigna (Papillionaceae) sur la base de donnees morphologiques, et polliniques traitees par l'analyse informatique. Boissiera 28: 1-273. Payan, F.R. and F.w. Martin. 1975. Barriers to the hybridisation of Passiflora species. Euphytica 24: 709-716. Pittarelli, G.w. and J.R. Stavely. 1975. Direct hybridisation of Nicotiana rependa x N. tabacum. Journal of Heredity 66: 281-284. Przywara, L., D.W.R. White, P.M. Sanders, and D. Maher. 1989. Interspecific hybridisation of Trifolium repens with T. hybridum using in ovulo embryo culture. Annals of Botany 64: 613- 624. Sitch, L.A. and J.w. Snape. 1987. Factors affecting haploid production in wheat using the Hordeum bulbosum system. 1. Genotype and environmental effects on pollen grain germination, pollen tube growth and the frequency of fertilisation. Euphytica 36: 483-496. Thomas, C.Y. and J.G. Waines. 1982. Interspecific hybrids between Phaseolus vulgaris L. and P. acutifolius: Field trials. Annual Report of Bean Improvement Cooperative 25: 58-59. 61 Digitized by Google 1.6 Cowpea breeding in the USA: new varieties and improved germplasm J.D. Ehlersl, R.L. Fery2, and A.E. HalP Abstract Cowpea is utilized in the USA as both a vegetable crop and a dry bean, and breed- ing efforts are focused on development of cultivars for specific end uses. Blackeye cultivars are developed for production of dry beans for national and international markets. California Blackeye No. 27 (CB27), a cultivar with a combination of high-value traits, was released in 1999. CB27 has high yield potential, superior seed quality, heat tolerance, and broad-based resistance to root-knot nematodes and Fusarium wilt. Breeding programs in southeastern US have traditionally been directed towards the development of various classes of horticultural-type cultivars for the canning, freezing, fresh market, and home garden market sectors. The most interesting recent development in the horticultural arena is the acceptance of green- seeded cultivars by the freezing industry. Charleston Greenpack is a leading source of raw products for the freezing industry. Introduction Cowpea is an important soil-building crop in the rotation of cotton and vegetable crops in the southern half of the USA. Most cowpea is consumed by people in southeastern US, where it has been a traditional crop since the early l800s. Cowpea is grown as a veg- etable crop in all of the southern states, and it is a popular home garden item throughout the region. Canning or freezing companies process much of the commercial crop in this region, but a significant amount is sold as fresh-shell "peas". In the southwest, primarily California and Texas, about 45000 t of dry blackeye type cowpea ("blackeyes") is pro- duced annually on about 20000 ha. It has been estimated that 20-30% of the production is exported internationally, mostly to southern Europe, the Middle East, and Asia. Most of the blackeyes are sold through the dry-package trade. Perhaps 5-10% of the blackeye crop is canned. Immature whole pods are also consumed in the southeast and by Asian communities throughout the US, and specialized cultivars, similar to snap beans (P haseolus vulgaris), have been developed for this purpose. Prior to the Second World War, cowpea was a major forage crop for horses and cattle (hence the name cowpea). Currently, the acreage of cowpea being used as a soil-building cover crop, particularly in organic agriculture, is increasing rapidly. US genetic resources and improvement of cowpea Reasonably comprehensive germplasm collections have been assembled (about 8000 accessions are held by the United States Department of Agriculture [USDA], and 5500 1. Department of Botany and Plant Sciences, University of California, Riverside, CA, 92521-0124, USA. 2. US Vegetable Laboratory, USDA-ARS, Charleston, SC, 29414-5334, USA. 62 Digitized by Google Cowpea breeding in the USA: new varieties and improved germplasm accessions by the University of California [UC] Riverside). Passport and characterization data for accessions in the USDA collection are available from the Germplasm Resources Information Network (GRIN) of USDA through the World Wide Web (address: http// www.ars-grin.gov). Requests for small quantities of seed of these accessions may be made to USDA or to the UC, Riverside. Lack of research funding in the US has hindered unlocking the full potential of these collections. Nevertheless, important and unique traits have been identified in cowpea, such as heat and chilling tolerance (Hall 1992; Ismail et al. 1999), and resistance to pests such as root-knot nematodes, cowpea curculio, and Fusarium wilt (Ehlers and Hall 1997; Hall et al. 1997). Cowpea breeding programs are being conducted in the US by USDA at their vegetable laboratory in South Carolina, by Louisiana State University, the University of California, at their Riverside and Davis campuses, at Texas A&M University, and at the University of Arkansas. The UC breeding programs are developing improved dry-grain blackeye cowpea varieties and complementary management systems that increase profitability through increased yield and grain quality, and decreased production costs. Specific objectives of these programs include development of black eye varieties with high yield, erect plant type, large grain with low seed coat cracking that cans well, heat tolerance, and resistance to Fusarium wilt (races 3 and 4), "early cut-out" disease, root-knot nematodes, cowpea aphid, and lygus bug. New objectives include the development of cover-crop cowpea varieties, and cowpea cultivars with unique grain types such as the persistent green, "sweet," or large, white grained types. The cowpea genetics and breeding program at the US Vegetable Laboratory in Charles- ton, South Carolina, has been in progress for well over three decades. This program has been successful in identifying unique value-added traits and new sources of needed resistance to root-knot nematodes, diseases, and insects (Cuthbert et al. 1974; Cuthbert and Fery 1975,1979; Fery and Cuthbert 1979; Fery and Dukes 1995a; Fery et al. 1975, 1977, 1994; Schalk and Fery 1982, 1986); in determining the mode of inheritance of major economi- cally important traits (Fery and Cuthbert 1975, 1978; Fery and Dukes 1977, 1980; Fery et al. 1976); and in the development of many cultivars with multiple resistance to pests and diseases. USDA has released 13 cowpea cultivars (recent releases in Table 1) and 10 germplasmlbreeding lines in the past 25 years. The green cotyledon gene in cowpea was discovered by this program, and this new gene is the basis of the first commercially suc- cessful cowpea cultivars with a "persistent green" seed phenotype (F err 1998, 1999, 2000; Fery and Dukes 1994; Fery et al. 1993; USDA 2000). New dry grain blackeye variety released California Blackeye No. 27 (CB27), developed by the UC Riverside breeding program, was released as a new variety by the California Crop Improvement Association in 1999 (Ehlers et al. 2000a). This is the first new blackeye variety available to California growers in about ten years. Plant Variety Protection (PVP) is being sought. Small quantities of seed are available from UC Riverside for research purposes. CB27 is an erect, compact blackeye-type cowpea with heat tolerance and high yields and a number of other desirable features, including brighter white seed coat and broader based resistance to Fusarium wilt and root-knot nematodes than the currently available 63 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) ~ Table 1. Recently released cowpea cultivars, specific trait improvement, and reference. Cultivar Santee Early Pinkeye Carolina Crowder Bettergro Blackeye Bettergreen Bettersnap Tender Cream Charleston Greenpack Petite-N-Green California Blackeye 27 Green Pixie Green Dixie Blackeye Improved trait(s) Early maturing processing cultivar for use to extend season or replace lost/delayed plantings Resistant to curculio, blackeye cowpea mosaic virus (BICMV), root-knot nematodes Resistant to curculio, root-knot nematodes, rust, and powdery mildew First cowpea cultivar with green cotyledon phenotype (cream-type) Snap-type resistant to root-knot nematodes, BICMV, and southern bean mosaic virus Excellent culinary quality and multiple insect, nematode, and disease resistance First pinkeye-type with green cotyledon trait; no hard seed; resistant to BICMV Small-seeded, full-season, green cotyledon, pinkeye-type developed for home gardeners Heat tolerant and broad-based resistance to nematodes, Fusarium wilt Green cotyledon, cream-type with seed size and shape characteristics preferred by industry First blackeye-type cultivar with the green cotyledon phenotype to be released Reference Fery and Dukes 1990 Fery and Dukes 1992 Fery and Dukes 1993 Fery et al. 1993 Fery and Dukes 1995(b) Fery and Dukes 1996 Fery 1998 Fery 1999 Ehlers et al. 2000 Fery 2000 USDA 2000 bl ~ ~ ~ :J ~ a" I\) :J Q.. 0- til rl) ~ 01:i Cowpea breeding in the USA: new varieties and improved germplasm varieties, California Blackeye No. 46 (CB46) and California Blackeye No. S (CBS) (Table 2). CB27 begins flowering at about S2 days and matures its first flush of pods about 9S days from sowing under typical conditions in California. The average individual seed weight has been 224 mg in California and the grain has excellent canning quality. CB46 and CBS carry the nematode resistance gene Rk that confers strong resistance to common strains ofMeloidogyne incognita root-knot nematode. CB27 carries geneRk and another recessive gene, rk3 (Ehlers et al. 2000b) that act together in an additive fashion to provide greater protection against Rk-virulent forms of M. incognita and M. javanica root-knot nematodes. Reproduction and root galling on CB27 caused by Rk-virulent M. incognita andM.javanica are about half those observed on CB46 and CBS (Roberts et al. 1997). Several new fields with root-knot nematodes causing galling on CB46 were identi- fied in 1999, indicating the Rk-virulent strains of root-knot nematodes may be widespread in California (Ehlers et al. 1999). CB27 has resistance to both race 3 and race 4 of Fusarium wilt, while CB46 only has resistance to race 3 of this disease organism and CBS is susceptible to both races. Race 3 is the predominant race of Fusarium wilt in California, but additional fields with race 4 were identified in 1997, 1998, and 1999, suggesting that this race may be widespread (Ehlers et al. 1999). New horticultural cowpea varieties for southeastern US Persistent-green cow peas The development of cowpea cultivars with a persistent-green seed color has been the sub- ject of much interest among both food processors, especially freezers, and plant breeders because seeds of such cultivars can potentially be harvested at the near-dry or dry seed stage of maturity without loss of their green color. The retention of the green color is important because the choice of harvesting method is often a compromise between cost and product quality. Harvesting dry cowpeas can be done efficiently and with minimal losses compared to mechanically harvesting mature-green cowpeas. Also, the crop does not need to be processed immediately and may be stored until it is convenient to freeze. Several hours prior to freezing, the precise amount of grain needed would be soaked in water. This product would have low production and storage costs similar to other dry -grain crops yet the product would be used in high-priced, vegetable-type applications. Compared to other frozen vegetable products that are harvested fresh and have high harvesting and storage costs, the relatively low costs of dehydrated cowpeas and relatively high potential profit margin should encourage processors to use this ingredient in frozen vegetable mixes and other applications, so the market potential may be significant and help increase demand for cowpeas in the USA. Chambliss (1974) reported that the green testa gene (gt) conditions a green seed coat that persists in the dry seed, and this trait results in a processed product with improved consumer appeal. Although a cultivar homozygous for the gt gene was released (Cham- bliss 1979), the green testa trait was never accepted by the processing industry. Fery et al. (1993) discovered a green cotyledon mutant in the cream-type cultivar Carolina Cream, and released a green cotyledon selection as Bettergreen. The new trait is similar to the green cotyledon trait reported in lima bean by Magruder and Wester (1941). The green coty ledon trait revolutionized the lima bean industry, and the quick acceptance of the green cotyledon 65 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) 0' 0' Table 2. Comparative features of California Blackeye No. 27 (CB27), California Blackeye No.5 (CB5), and California Blackeye No. 46 (CB46) blackeye cowpeas. Fusarium wi It Resistance to root-knot nematodes M. incognita Seed weight Entry Race 3 Race 4 avirul. virulent M.javanica Heat Plant habit (gl1 00 seeds) CB27 Yes Yes Yes Yes Yes Yes Compact 22 CB46 Yes No Yes No Partial No Moderately compact 22 CBS No No Yes No Partial No Vine 25 avirul. = avirulent, effectively controlled by gene Rk; virulent = not effectively controlled by gene Rk alone. bl ~ ~ ~ :J ~ a" I\) :J Q.. 0- til rl) ~ 01:i Cowpea breeding in the USA: new varieties and improved germplasm cultivars Bettergreen and Charleston Greenpack by the freezing industry indicates that the trait will have a similar impact on the cowpea industry as most processing horticultural cultivars are now harvested at the near-dry or dry seed stage of maturity. Fery and Dukes (1994) concluded that the green cotyledon trait in southernpea is conditioned by a single recessive gene, symbolized gc, and that this gene is neither allelic to nor linked with the gt gene. Apparently, these genes prevent the normal breakdown of chlorophyll that occurs as seeds reach maturity. Dry persistent-green grain stored in sacks retains its green color for many months, however, the green color of the grain can be bleached by exposure to sunlight for several weeks, giving the grain its background color, e.g., white. Apparently there are no negative pleiotropic effects on yield or other agronomic characters (Freire Filh, unpublished report). Several important new persistent-green cowpea varieties have recently been developed, including Charleston Greenpack (Fery 1998), Petite-N-Green (Fery 1999), Green Pixie (Fery 2000), and Green Dixie (USDA 2000). Bettergreen was the first southernpea cultivar to be developed that exhibits the green cotyledon trait (Fery et al. 1993). It was derived from a single mutant plant harvested from a 1986 field planting of Carolina Cream. The mutant plant was homozygous for a newly discovered gene (gc) conditioning a unique green cotyledon trait. Bettergreen has a medium, bushy plant habit. A typical pod is slightly curved, 15 cm long, and contains 12 to 14 peas. Pods are green when immature, green with a distinct purple shading at green-shell maturity, and pale tan or straw when dry. The fresh peas are small and ovate- reniform in shape. The dry peas have a smooth seed coat and can be harvested at dry seed maturity without loss of the seeds fresh green color. Bettergreen has resistance to the cowpea curculio, Cercospora leaf spot, southern blight, rust, and powdery mildew. The cultivar exhibits tolerance to seedling diseases. Bettergreen is used in the US as a commercial cultivar by the frozen food industry. Charleston Greenpack was the first pinkeye-type southernpea to be developed that exhibits the green coty ledon trait (F ery 1998). Charleston Greenpack originated as a bulk of an F 8 [Kiawah x (Kiawah x Bettergreen)] population grown in 1994. Except for the green seed color and a tendency for a slightly smaller pea size, the phenotype of Charleston Greenpack is quite similar to those of the leading US pinkeye-type processing cultivars Coronet and Pinkeye Purple Hull-BVR. The plant habit is low, bushy, and somewhat more compact than that of either Coronet or Pinkeye Purple Hull-BVR. A typical Charleston Greenpack pod is moderately curved, 17 cm long, and contains 14 peas. Pod color is green when immature and dark purple when ready for mature-green or dry harvest. The fresh peas are kidney-shaped and have a pink eye. The dry peas have a smooth seed coat, and are slightly smaller than those of Coronet and Pinkeye Purple Hull-BVR. Charleston Greenpack has excellent field resistance to blackeye cowpea mosaic virus (BlCMV), a major pathogen of southernpea in the USA. In the brief period since it was first released in 1997, Charleston Greenpack has already become a leading pinkeye-type cultivar for the US frozen food industry. Charleston Greenpack peas produce an attractive frozen pack. Protection for Charleston Greenpack is being sought under the US Plant Variety Protection Act. Petite-N-Green is a small-seeded, full-season, green cotyledon, pinkeye-type south- ernpea that was released in 1998 (Fery 1999). Petite-N-Green originated as a bulk of an F 9 (Coronet x Bettergreen) population grown in 1994. Petite-N-Green has a low, bushy 67 Digitized by Google Cowpea genetics and breeding plant habit similar to that of Coronet. It has a more procumbent vine than does Charleston Greenpack. Petite-N-Green produces dry pods at Charleston, South Carolina, in 70 to 76 days, 4 to 7 days later than Charleston Greenpack and 2 to 9 days later than Coronet and Pinkeye Purple Hull-BVR. A typical Petite-N-Green pod is moderately curved, 14 cm long, and contains 14 peas. Pod color is green when immature, and dark purple when ready for mature-green harvest or when dry. Fresh peas are ovate to kidney-shaped and have a pink eye that is quite similar to fresh Charleston Greenpack, Coronet, and Pinkeye Purple Hull-BVR peas. Dry peas are small and have a smooth seed coat. Petite-N-Green peas are 12-20% smaller than Charleston Greenpack peas, 11-25% smaller than Coronet peas, and 12-24% smaller than Pinkeye Purple Hull-BVR peas. Petite-N-Green yields are compa- rable to those of Charleston Greenpack, Coronet, and Pinkeye Purple Hull-BVR. Petite- N-Green is recommended particularly for use as a home-garden cultivar in southeastern USA. The peas can be harvested not only fresh for immediate consumption or storage in home freezers, but also when fully dry for storage as an attractive dry pack. The dry peas can be removed from storage and soaked to restore a near-fresh green color. Protection for Petite-N-Green is being sought under the US Plant Variety Protection Act. Green Pixie is a small-seeded, green cotyledon, cream-type southernpea that was released in 1999 (Fery 2000). Green Pixie originated as a bulk of an F9 (Bettergreen x White Acre) population grown in 1994. Green Pixie has a high, bushy plant habit similar to that of White Acre. Green Pixie produces dry pods at Charleston, South Carolina, in about 76 days, 5 days later than Bettergreen, and 5 days earlier than White Acre. A typical Green Pixie pod is slightly curved, about 15 cm long, and contains about 16 peas. Pod color is light green when immature, purple when ready for mature-green harvest, and light straw color when dry. Green Pixie peas are rhomboid-kidney in shape, similar to the shape of fresh White Acre peas, but very different from the ovate to reniform shape characteristic of fresh Bettergreen peas. Dry peas are small and have a smooth seed coat. Green Pixie peas are similar in size to White Acre peas, but much smaller than Bettergreen peas. Green Pixie was developed for use by the frozen food industry, either as a replacement for the popular White Acre or as a substitute for Bettergreen when grown to produce the raw product for a blended pack of Bettergreen and White Acre. Green Dixie Blackeye is the first blackeye-type southernpea to be released that exhib- its the green cotyledon phenotype (USDA 2000). Green Dixie Blackeye originated as a bulk of an F 9 (Bettergreen x Bettergro Blackeye) population grown in 1994. Green Dixie Blackeye has a high bushy plant habit. A typical Green Dixie Blackeye pod is slightly curved, 21 cm long, and contains 14 peas. Pod color is light green when immature, light green with a tendency for slight pigmentation (purple) on the tip when ready for mature- green harvest, and light straw color when dry. Green Dixie Blackeye peas have an oblong shape. The dry peas have a smooth seed coat and small, black -colored, hilar eyes. Results of replicated tests conducted at Charleston, South Carolina, indicate that Green Dixie Blackeye has a much greater yield potential than Bettergro Blackeye. Green Dixie Black- eye is recommended for use by home gardeners and the dry -pack bean industry. The peas can be harvested not only fresh for immediate consumption or storage in home freezers, but also when fully dry for storage or sale as an attractive dry pack. The dry peas can be soaked to restore a near-fresh green color. The UC Riverside program is also developing persistent-green California blackeye- type varieties for potential use in frozen products. Like Green Dixie Blackeye, the grain 68 Digitized by Google Cowpea breeding in the USA: new varieties and improved germplasm of these varieties resembles fresh-shell blackeyes after being soaked in water for several hours. Because the green color bleaches to white after prolonged exposure to sunlight in the field, "double-flush" production practices, wherein growers accumulate two flushes of pods over 120-140 days, will not be possible. In other aspects, however, production, harvesting, and storage practices would be as for traditional blackeye cowpeas that are now produced in California. Fresh green pods Certain southernpea cultivars are grown for their immature fresh pods or snaps, and some processors have traditionally included a small portion of snaps in the processed product (F ery 1990). Loa and Halsey (1964) noted that "the snap ingredient has always consisted of immature pods of standard shell-pea varieties." The cultivar Snapea was developed specifically for its attractive, long, low fiber pods (Loa and Halsey 1964). Patel and Hall (1986) evaluated five vegetable cowpea (snap-type) breeding lines and a snap bean cultivar in a summer field test at Riverside, California. They concluded that vegetable cowpea lines have a potential for producing large yields of pods in environments in which snap bean produces only small yields due to hot weather. Fery (1981) observed earlier that southernpeas are tolerant to drought and hot weather, and can be grown quite success- fully under conditions that are totally unsuitable for table legumes such as the common bean and the lima bean. Fery and Dukes (1995b) released the edible podded cultivar Bettersnap in January 1994. Bettersnap, which is resistant to root-knot nematodes and blackeye cowpea mosaic and southern bean mosaic viruses, quickly replaced Snapea as the snap-type cultivar of choice for commercial food processors. Development of new specialty cowpea grain types One major problem facing the cowpea industry in the USA is stagnant demand. Per capita consumption of dry cowpeas has steadily decreased in the USA, in part because this product takes more time to prepare than other foods, and little time is available for meal preparation with most families having both spouses working full-time. Also, more and more meals are eaten outside the home, and cowpeas have not been frequently offered on restaurant menus or used in convenience foods. Increased awareness of the health benefits of consuming grain legumes has helped increase demand in some markets, but new types of cowpeas and quick-to-prepare cowpea food products are needed to stimulate cowpea consumption in the USA. Development of all-white cowpeas for value-added products Cowpea is processed into many traditional West African foods, such as akara, that are delicious yet virtually unknown outside West Africa. Such foods could find wide accep- tance in US markets as processed convenience foods or "fast-foods". Akara is traditionally prepared from cowpeas that are soaked, dehulled, and milled wet. If the milled product is not used immediately, expensive or laborious drying or refrigera- tion is necessary for its preservation. Dry milling of whole grain cowpea would be much more efficient than wet milling and produce an easily storable product. This would make possible the development of "ready to cook" cowpea flour mixes for akara production or for use in other products. Unlike pigmented cowpea cultivars, an all-white cowpea would produce an all-white flour that would be preferred for most products. Cowpea flour can 69 Digitized by Google Cowpea genetics and breeding be substituted for wheat flour up to 30% in the preparation of yeast breads without loss in quality (K.H. McWatters, personal communication). High yielding breeding lines have been developed at UC Riverside with large all-white grains that are adapted to the US by crossing California blackeye varieties with the all- white cultivars Bambey 21 from Senegal and Montiero from Brazil. In crosses between Bambey 21 and California blackeyes, a ratio of 3 blackeye: 1 all-white type seed coat was observed in the F 2 generation (1.D. Ehlers, unpublished data), indicating a single recessive gene confers this trait. Plants having all-white seed may be recognized in the vegetative stage because they lack any red pigmentation on the stem or branch nodes or on other plant parts. Presumably, Bambey 21 carries a gene-blocking formation of pigments. Complex segregation is observed in the F 2 generation of crosses between blackeyes and Montiero. In Montiero, the capacity for pigment to be produced is retained but pigment is restricted to a barely visible ring around the hilum. One all-white linefrom the U C Riverside breeding program, 97 -15-33, developed using Bambey 21 as a source of the all-white character, was compared to four other cowpea lines for use in akara production and found to be as good as the control blackeye variety (McWatters et al. 2000). ""Sweet" cowpea Breeding line 24-l25B is a sweet-tasting cowpea developed by the breeding program of the Research Institute for Agricultural Development (lRAD)lPurdue University Bean-Cowpea Collaborative Research Program (CRSP) that is based in Maroua, Cameroon. Line 24- l25B was developed from a single cross of two I1TA lines, IT86D-364 and IT8ID-1138 (L.w. Kitch, personal communication, 1999) neither of which is considered "sweet". In 1993, a single plant selection was made from an F 4 family at Maroua. The following year, seed of the resulting F 5 family was bulked and used for yield trials. Cameroonian farm- ers who had been brought to the IRADlPurdue University CRSP project plots as part of a farmer-assisted selection process noted that this line tasted "sweet" or "good". For over three years, Cameroonian farmers consistently chose this line as one of their favorites (Kitch et al. 1998). Subsequent analysis of the sugar content of dry seeds of this line by Purdue University researchers revealed that it has a sugar content of about 6% compared to sugar content of about 2% for "normal" cowpea varieties (L. Murdock, unpublished data). Purdue researchers also conducted a triangular taste panel test comparing cooked samples of 24-l25B with its nonsweet sister line 24-l25A. In this test, two samples of the nonsweet line 24-l25A and one sample of the sweet line were placed before a panel. The tasters, who were generally unfamiliar with cowpeas, were able to correctly differentiate the sweet cowpea from its nonsweet sister line about 83% of the time. The discovery of the sweet trait opens up the possibility of developing new products and markets for cowpea in the US and elsewhere. One possibility is the development of "sweet" versions of existing market classes. Another possibility is the development of new market classes. One type might resemble garden peas (Pisum sativum) having grain that are sweet, round-shaped, and persistent-green in color. The sweet trait is being rapidly bred into cultivars targeted to the US, Senegal, and Ghana. Line 24-l25B has been crossed to CB27, to CB46, to the Senegal variety Melakh, and to the Ghananian variety Sul-5l8 for development oflocally adapted "sweet" varieties and for genetic analysis of the trait. F 1 data from several crosses indicate that sweetness is 70 Digitized by Google Cowpea breeding in the USA: new varieties and improved germplasm completely recessive. F 2 seed of selected crosses was sent to Ghana and Senegal in June 2000, and is being grown in California. F 3 seed from F 2 plants of these crosses will be analyzed for sugar content and inheritance of "sweetness" trait determined. If the trait is simply inherited, a backcross procedure would be appropriate to introduce the trait into adapted cultivars suited to many regions. Progress in breeding for pest resistance Improved nematode resistance in blackeye cow peas Resistance to root-knot nematodes in US cultivars and probably most other cultivars in the world is based on the Rk allele. Rk provides very strong protection from most iso- lates of Meloidogyne incognita but only moderate resistance to M. javanica (Roberts et al. 1997). Also, gene Rk-virulent strains of M incognita have been identified at several locations in California. Therefore, cowpea cultivars with effective broad-based resistance to root-knot nematodes are needed. At the last world cowpea conference, Roberts et al. (1997) reported that I1TA breeding line IT84S-2049 had much more effective resistance to root-knot nematodes (M. incognita and M. javanica) than cultivars possessing the Rk resistance gene, and that this resistance is due to an allele at the Rk locus, designated Rk2 (Roberts et al. 1996). Unfortunately, IT84S-2049 is poorly adapted to the US and has poor quality seeds from the standpoint of small size (about 0.13 g/seed), and high frequency of seed coat splitting. Therefore, a limited backcrossing program was used to develop high yielding and large-seeded blackeye breeding lines that possess the IT84S-2049 resistance (Ehlers et al. 1999). Identification and improvement of insect resistance Insect -resistant cowpea varieties may become very important in the near future to maintain high bean quality and yield levels in the USA. Restrictions on the use of currently avail- able pesticides are likely to increase, while their effectiveness in some cases is decreas- ing due to the development of insecticide-resistant insect biotypes. Also, due to the high cost of pesticide registration, few new insecticides for minor crops such as cowpea may be available. A major goal of the breeding programs at the USDA vegetable laboratory and the UC Riverside is the development of pest-resistant cultivars that require minimal applications of pesticides. Lygus bug is the most devastating pest of cowpea in California. Early season infestations of lygus bugs reduce the yield of cowpea by feeding on reproductive buds causing them to abort. The extent of yield loss depends on the timing of the infestation, the phenologi- cal stage of the crop, and the intensity and duration of the attack, and a cultivar's ability to recover. Late infestations of lygus bugs damage pods and developing seeds, causing seed or pod abortion, seed pitting or malformation, and superficial scarring to the seed coat. Even superficial scarring of the seed coat is important because it lowers the value of the grain. Hundreds of cowpea accessions have been screened for resistance to lygus at UC Riverside and several promising lines have been identified. In 1999, the grain yields and lygus-induced seed damage of CB46, three exotic African cowpea lines, and three lines developed at UC Riverside from crosses between wild and cultivated cowpeas were evaluated under both lygus bug-protected and unprotected conditions at Riverside. The six lines were chosen based on their performance in similar trials or from unprotected lygus 71 Digitized by Google Cowpea genetics and breeding screening nurseries conducted in 1998. Lygus bug-induced grain yield losses in CB46 were 29% and 14% of the seed of this variety had lygus damage (Table 3). Five of the six entries had significantly less grain yield loss due to lygus than CB46, and all six entries had lower seed damage (Table 3). These data indicate that progress is being made in identify- ing lygus-resistant germplasm. IT92KD-370 and IT96-ll-27 also had significantly less lygus-induced seed damage than CB46 in similar trials conducted in 1998. Wild cowpeas are a potential source of insect-resistance genes, but are themselves difficult to evaluate for resistance to lygus because of their photoperiod sensitivity, slow early growth and development, and morphological and grain characteristics that differ substantially from cultivated cowpeas. UC Riverside has developed many breeding lines derived from three-way crosses between cultivated and wild cowpeas [(wild cowpea x cowpea) x cowpea] that are similar to cultivated cowpeas. Lines 96-11-27 and 96-11- 38, developed from crosses to wild cowpea accession TVNu 597 (ssp. pubescens), have exhibited lygus resistance in terms ofless grain yield and seed quality reductions than the standard cultivar CB46 (Table 3). Over the last five years, more than 1000 accessions and wild cowpea x cultivated and exotic x adapted breeding lines have been screened in the field for resistance to lygus by the UC Riverside program. From this work, I1TA breeding lines IT93K-2046, IT93K- 273-2-1, IT92KD-370, and IT86D-7l6 appear to have moderate resistance to lygus bud blasting or lygus-induced seed damage. This resistance is being bred into lines adapted to California. The resistance appears to have high heritability since it was possible to visually identify resistant F 3 families developed from crosses between California blackeyes and I1TA lines IT93K-2046, IT93K-273-2-l, IT92KD-370, and IT86D-7l6 (Ehlers et al. 1999). Observations in the field suggest that IT93K-2046 also has resistance to the California biotype(s) of cowpea aphid and this resistance has been transferred to breeding lines after crosses with susceptible California cultivars. Strong resistance to cowpea aphid in the US has been difficult to identify. Unfortunately, the aphid resistance that is effective in Africa is not effective against biotypes of this pest in the US. In 1999, I1TAline IT93K-2046 and several breeding lines (99CV-564-2, 99CV- 564-4, 99CV-565-3, and 99CV576-4) developed from crosses of this line and California blackeye cultivars, were initially attacked, but rapidly recovered and produced pods in both replicates of a screening nursery in which all other lines were completely destroyed by aphids. Several selections were made in each of the resistant F 6 lines and are being further tested for resistance to aphids. Cowpea cover crops-progress in breeding Cowpea is increasingly finding favor as a warm-season, nitrogen-fixing cover crop, particularly in organic vegetable production systems in the US. These systems need low- cost sources of organic nitrogen, and cowpea can have high levels of biological nitrogen fixation. Cowpea cover-crop varieties in the US are needed that produce abundant biomass, have strong nematode resistance, photoperiod sensitivity, resistance to Fusarium wilt, reduced pod shattering, vigorous plant growth, and high yields of small seeds. Photoperiod sensi- tive cowpea cover-crop varieties will fix much more nitrogen and produce much more biomass than present-day, neutral cowpea cultivars grown in the US because, under long day lengths, photoperiod sensitivity prevents the early transition to reproductive growth 72 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) ~ Table 3. Grain yield of protected and unprotected, yield loss due to Iygus, and percentage damaged seed under unprotected conditions of check cultivar CB46 and exotic (E) or wild x cultivated (W x C) cowpea lines at Riverside in 1999. ~ Grain yield Yield Protected Unprotected loss Line Type kglha (%) 96-11-38 Wx C 2014 1826 9 IT93K-452 E 1661 1500 10 96-11-27 WxC 2017 1723 14 IT93K-370 E 1766 1475 15 96-11-29 Wx C 1719 1463 15 96-11-111 WxC 1710 1355 21 CB46 Check 2520 1753 29 Mean 1996 1651 16 LSD(0.05) 330 311 9 CV(%) 11 13 42 tB-W/holstein = black pigment covers about 75% of surface area on white background. Lygus damaged seed (%) 5 6 6 8 3 7 14 7 4 41 Size gil 00 seeds 18.0 15.6 13.6 16.7 12.8 11.6 2004 15.5 1.0 4.1 Seed Color/pattern+ B-W/holstein Blackeye B-W/holstein Blackeye B-W/holstein Solid black Blackeye ~ ffi W ~ Otl S· S- rt> c: ~ :J ~ Q'i ~ rf IlJ :J a.. i" rB a.. ~ ~ -0 1iJ '" :3 Cowpea genetics and breeding that leads to sharp decreases in biomass production and nitrogen fixation. In the US, seed of photoperiod sensitive varieties can only be produced reliably in warm fall regions such as the low-elevation deserts of California and southern Florida. At UC Riverside, complementary parental lines have been identified and crossed to develop a variety with the desired traits listed above. UCR 779, a nematode-suscep- tible landrace from Botswana that has a very aggressive spreading plant habit and high biomass production (in the absence of root-knot nematodes), has been hybridized with IT89KD-288 and IT84S-2049 breeding lines from I1TA that have high biomass and very strong resistance to root-knot nematodes (Roberts et al. 1997) and other desirable traits (Aguiar et al. 1998). Nematode-resistant F 5 breeding lines with photoperiod sensitivity, nonshattering pods, small seed size, and high biomass production have been identified from these crosses thus far. Future possibilities Successful genetic transformation of cowpea, which appears imminent at this time, will open up new possibilities for improvement, particularly in the area of insect resistance. Candidate genes that show insecticidal activities with a number of African cowpea pests have been identified (Machuka 2000). These genes code for Bacillus thuringiensis (Bt) endotoxin crystal proteins, plant lectins, protease and alpha-amylase inhibitor pro- teins (Shade et al. 1994), chitinases and ribosomal inactivating proteins. Little is known, however, about the effectiveness of these genes in controlling major US cowpea pests such as the lygus bug and cowpea curculio. Recent improvements in the cowpea genetic map (Fatokun et al. 2000; Ogundiwin et al. 2000) could potentially make applied breeding programs more efficient through marker-assisted selection. Wild cowpea relative Vigna vexillata is highly resistant to many insect pests, but until now it has not been possible to hybridize this species with cowpea (Fatokun 2000). The recent report of the successful hybridization of this species with cowpea (Gomathinayagam and Muthiah 2000) offers the possibility of obtaining insect resistance that is not available from the primary genepool of the species. Acknowledgements This research was supported in part by grants from the Blackeye Council of the California Dry Bean ResearchAdvisory Board and the Bean/Cowpea Collaborative Research Support Program, USAID Grant no. DAN-G-SS-86-00008-00. The opinions and recommendations are those of the authors and not necessarily those of USAID. References Aguiar, J., J.D. Ehlers, and W. Graves. 1998. Desert cover crop varieties identified. California Vegetable Journal, December. Pages 5, 6, and 21. Chambliss,O.L. 1974. Green seed coat: a mutant in southernpea of value to the processing industry. HortScience 9: 126. Chambliss,O.L. 1979. Freezegreen southernpea. HortScience 14: 193. Cuthbert, F.P. Jr., R.L. Fery, and O.L. Chambliss. 1974. Breeding for resistance to the cowpea curculio in southern peas. HortScience 9: 69-70. Cuthbert, F.P. Jr. and R.L. Fery. 1975. CR 17-1-13, CR 18-13-1, and CR 22-2-21, cowpea curculio resistant southernpea germplasm. HortScience 10: 628. 74 Digitized by Google Cowpea breeding in the USA: new varieties and improved germplasm Cuthbert, F.P., Jr. and RL. F ery. 1979. Value of plant resistance for reducing cowpea curculio damage to the southernpea (Vigna unguiculata [L.] Walp.). Journal oftheAmerican Society of Horticultural Science 104: 199-201. Ehlers,JD. andA.E. Hall. 1997. Cowpea (VIgna unguiculata [L.]Walp.). 1997. Field Crops Research 53: 187-204. Ehlers, JD., A.E. Hall, A.M. Ismail, P.A. Roberts, W.C. Matthews, B.L. Sanden, CA. Frate, and S. Mueller. 1999. Blackeye varietal improvement. Pages 47--61 in University of California Dry Bean Research 1999 Progress Report, California Dry Bean Advisory Board, Dinuba, CA, USA. Ehlers, JD., A.E. Hall, P.N. Patel, PA. Roberts, and W.C Matthews. 2000a. Registration of Cali- fornia Blackeye 27 Cowpea. Crop Science 40: 854-855. Ehlers, J.D., W.C. Matthews,A.E. Hall, and P.A. Roberts. 2000b. Inheritance of a broad-based form of nematode resistance in cowpea. Crop Science 40: 611--618. Fatokun, CA. 2000. Breeding cowpea for resistance to insect pests: attempted crosses between cowpea and VIgna vexillata. Proceedings ofthe Third World Cowpea Conference, 4-7 Septem- ber 2000. Ibadan, Nigeria. Fatokun, CA., B. Ubi, T.H.N. Ellis, C. Li, and G. Scoles. 2000. A genetic linkage map of cowpea based on DNA markers. Proceedings of the Third World Cowpea Conference, 4-7 September 2000. Ibadan, Nigeria. (Abstract.) Fery, RL. 1981. Cowpea production in the United States. HortScience 16: 473-474. Fery, R L. 1990. The cowpea: production, utilization, and research in the United States. Horticul- tural Review 12: 197-222. Fery, RL. 1998. Charleston Greenpack, a pinkeye-type southernpea with a green cotyledon phe- notype. HortScience 33: 907-908. Fery, RL. 1999. Petite-N-Green, a small-seeded, full-season, green cotyledon, pinkeye-type south- ernpea. HortScience 34: 938-939. F ery, RL. 2000. Green Pixie, a small-seeded, green cotyledon, cream-type southernpea. HortScience 35(5): 954-955. F ery, RL. and F.P. Cuthbert, Jr. 1975. Inheritance of pod resistance to the cowpea curculio infesta- tion in southernpeas. Journal of Heredity 66: 43-44. Fery, RL. and F.P. Cuthbert, Jr. 1978. Inheritance and selection of non preference resistance to the cowpea curculio in the southernpea (VIgna unguiculata [L.] Walp.). Journal of the American Society of Horticultural Science 103: 370-372. Fery, RL. and F.P. Cuthbert, Jr. 1979. Measurement of pod-wall resistance to the cowpea curculio in the southernpea (Vigna unguiculata [L.] Walp.). HortScience 14: 29-30. Fery, RL., PD. Dukes, and F.P. Cuthbert, Jr. 1975. CR 17-1-34, Cercospora leaf spot-resistant southernpea germplasm. HortScience 10: 627. Fery, RL. and B.B. Singh. 1997. Cowpea genetics: a review of recent literature. Pages 13-29 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan Inter- national Research Center for Agricultural Sciences (J1RCAS). I1TA, Ibadan Nigeria. Fery, RL., PD. Dukes, and F.P. Cuthbert, Jr. 1976. The inheritance ofCercospora leaf spot resis- tance in the southernpea (VIgna unguiculata [L.] Walp.). Journal of the American Society of Horticultural Science 101: 148-149. Fery, RL. and P.D. Dukes. 1977. An assessment of two genes for Cercospora leaf spot resistance in the southernpea (Vigna unguiculata [L.] Walp.). HortScience 12: 454-456. F ery, RL., P.D. Dukes, and F.P. Cuthbert, Jr. 1977. Yield-loss of southernpea (VIgna unguiculata) caused by Cercospora leaf spot. Plant Disease Report 61: 741-743. Fery, RL. and PD. Dukes. 1980. Inheritance of root-knot nematode resistance in the cowpea (VIgna unguiculata [L.] Walp.). Journal of the American Society of Horticultural Science 105: 671--674. Fery, RL. and P.D. Dukes. 1990. Santee Early Pinkeye southernpea. HortScience 25: 990-991. Fery, RL. and P.D. Dukes. 1992. Carolina Crowder southernpea. HortScience 27: 1335-1337. Fery, RL. and P.D. Dukes. 1993. Bettergro Blackeye southernpea. HortScience 28: 62--63. 75 Digitized by Google Cowpea genetics and breeding F ery, RL. and P.D. Dukes. 1994. Genetic analysis ofthe green cotyledon trait in southernpea (Vigna unguiculata [L.] Walp.). Journal of the American Society of Horticultural Science 119: 1054- 1056. Fery, RL. and PD. Dukes. 1995a. Registration ofUS-566, US-567, and US-568 root-knot nematode resistant cowpea germplasm lines. Crop Science 35: 1722. Fery, RL. and PD. Dukes. 1995b. Bettersnap southernpea. HortScience 30: 1318-1319. Fery, RL. and PD. Dukes. 1996. Tender Cream southernpea. HortScience 31: 1250-1251. Fery, RL. PD. Dukes, and F.P. Maguire. 1993. Bettergreen southernpea. HortScience 28: 856. Fery, RL., PD. Dukes, and J.A. Thies. 1994. Characterization of new sources of resistance in cowpea to the southern root-knot nematode. HortScience 29: 678-779. Freire Filho, F.R, O.L. Chambliss, and A.G. Hunter. 1996. Genetic analysis of crosses to produce persistent green seeds in southernpeas using gt and gc genes. Auburn University Research Report, USA. Gomathinayagam, P. and A.R Muthiah. 2000. Study on backcrossed embryo rescued crossed regenerants (Vigna vexillata [L.] A. Rich x Vigna unguiculata [L.] Walp.) with Vigna unguic- ulata [L.] Walp.) Proceedings ofthe Third World Cowpea Conference, 4-7 September 2000. Ibadan, Nigeria. Hall,A.E. 1992. Breeding for heat tolerance. Plant Breeding Reviews 10: 129-168. Hall, A.E., B.B. Singh, and JD. Ehlers. 1997. Cowpea breeding. Plant Breeding Reviews 15: 215-274. Ismail, A.M., A.E. Hall, and T.J. Close. 1997. Chilling tolerance during emergence of cowpea associated with a dehydrin and slow electrolyte leakage. Crop Science 37: 1270-1277. Ismail,A.M.,A.E. Hall, and T.J. Close. 1999. Allelic variation ofa dehydrin gene co-segregates with chilling tolerance during seedling emergence. Proceedings of the Natural Academy of Science 96: 13566-13570. Kitch, L.w., O. Boukar, C. Endondo, and L.L. Murdock. 1998. Farmer acceptability criteria in breeding cowpea. Experimental Agriculture 34: 475-486. Lorz, A.P. and L.H. Halsey. 1964. Snapea, a new cream type southern pea variety for snap pods use. University of Florida Agricultural Experimental Station Circular S-160, Florida, USA. Machuka, J. 2000. Potential role oftransgenic approaches in the control of cowpea insect pests. Proceedings of the Third World Cowpea Conference, 4-7 September 2000. Ibadan, Nigeria. Magruder, Rand RE. Wester. 1941. Green cotyledon, a new character in the mature lima bean (Phaseolus lunatus L.). Proceedings of the American Society of Horticultural Science 398: 581-584. McWatters, K.H., C.-y'T. Hung, Y.-C. Hung, M.S. Chinnan, and RD. Phillips. 2001. Akara making characteristics of five US varieties of cow peas (Vigna unguiculata). Journal of Food Quality 24(1): 53-66. Ogundiwin, E.A., C.A. Fatokun, G. Thottappilly, M.E. Aken'Ova, and M. Pillay. 2000. Genetic linkage map of Vigna vexillata based on DNA markers and its potential usefulness in cowpea improvement. Proceedings of the Third World Cowpea Conference, 4-7 September 2000. Ibadan, Nigeria. (Abstract). Patel, P.N. and A.E. Hall. 1986. Registration of snap-cowpea germplasm. Crop Science 26: 207-208. Roberts, P.A., W.C. Matthews, and J.D. Ehlers. 1996. New resistance to virulent root-knot nematodes linked to the Rk locus in cowpea. Crop Science 36: 889-894. Roberts, P.A., J.D. Ehlers,A.E. Hall, and W.C. Matthews. 1997. Characterization of new resis- tance to root-knot nematodes in cowpea. Pages 207-214 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInter- national Institute of Tropical Agriculture (lITA) and Japan International Research Center for Agricultural Sciences (nRCAS). I1TA, Ibadan Nigeria. 76 Digitized by Google Cowpea breeding in the USA: new varieties and improved germplasm Schalk, J.M. and RL. F ery. 1982. Southern green stink bug and leaffooted bug: Effect on cowpea production. Journal of Economic Entomology 75: 72-75. Schalk, J.M. and RL. Fery. 1986. Resistance in cowpea to the southern green stink bug. Hort- Science 21: 1189-1190. Shade, RE., H.E. Schroeder, J.J. Pueyo, L.M. Tabe, L.L. Murdock, T.I V Higgins, and M.J. Crisp- eels. 1994. Transgenic pea seeds expressing the alpha-amylase inhibitor of the common bean are resistant to bruchid beetles. Biocontrol Science and Technology 12: 793-796. US Department of Agriculture. 2000. Notice of release of Green Dixie Blackeye, a green cotyledon, blackeye-type southernpea. USDA,ARS, Washington, DC 20250. 28 April 2000. 77 Digitized by Google Digitized by Google Section II Cowpea integrated pest management Digitized by Google Digitized by Google 2.1 The importance of alternative host plants for the biological control of two key cowpea insect pests, the pod borer Maruca vitrata (Fabricius) and the flower thrips Megalurothrips sjostedti (Trybom) M. Tama', D.Y. Arodokoun>, N. Zenz3, M. Tindo4, C. Agboton ', R. Adeoti ' Abstract The interactions between naturally occurring and cultivated host plants, and bio- logical control, are first evaluated for the major lepidopteran pest attacking cowpea in West Africa, the pod borer Maruca vitrata. Significantly higher larval mortality due to parasitism by the ovolarval parasitoidPhanerotoma leucobasis was observed on wild alternative host plants in perennial habitats (e.g., Pterocarpus santalinoi- des) than in agroecosystems such as the cowpea field. Experimental assessment of the impact ofthe only egg parasitoid recorded fromU vitrata, Trichogrammatoi- dea ? eldanae, indicated that it is present in a variety of agroecosystems. A second important legume pest, the flower thrips Megalurothrips sjostedti, is attacked by the larval parasitoid Ceranisus menes, but overall parasitism rates are low, depend- ing on host plant and season. However, another parasitoid of the same genus, C. femoratus, recently discovered in Cameroon, has showed higher efficiency in parasitizingM. sjostedti on most of the important host plants, including cowpea. The potential of this new parasitoid as a biocontrol candidate in West Africa is being assessed through experimental releases in Benin and Ghana. Introduction In West Mrica, cowpea (Vigna unguiculata Walp.) is cultivated mainly as a rainfed crop from April to November, depending on the location. In the moist savanna with a bimodal rainfall pattern, where cowpea can produce two crops, the first rainy season lasts from April to July, and the second from mid-September to November, with a short dry spell from August to early September. In the regions of monomodal rainfall, the beginning and length of the rainy season usually depend on the latitude. In the areas considered in this review (see below), the monomodal rainy season normally lasts from May to November. During the long dry season from December to March, cowpea is cultivated on residual moisture in small isolated areas only (e.g., the Oueme valley in southern Benin, or the fadamas in northern Nigeria) (Arodokoun 1996; Bottenberg et al. 1997). As a consequence, insect pests attacking cowpea would need either to find alternative hosts to survive during this 1. I1TA Benin Research Station, 08 BP 0932 Tri Postal, Cotonou, Benin (m.tamo@cgiar.org). 2. INRAB, Cotonou, Benin. 3. ICIPE, Nairobi, Kenya. 4. I1TA Humid Forest Center, Yaounde, Cameroon. 81 Digitized by Google Cowpea integrated pest management period or to diapause. For the two pests considered in this paper, the pod borer Maruca vitrata Fabricius (Lep., Pyralidae) and the flower thrips Megalurothrips sjostedti Trybom (Thys., Thripidae), Arodokoun et al. (2001) and Tamo et al. (1993b) have demonstrated that neither species goes through diapause during the dry season, both of them being capable of feeding and reproducing on a wide range of alternative host plants in the absence of cowpea. Parallel studies (Arodokoun 1996; Tamo et al. 1997; Zenz 1999) have indicated that natural enemies ofbothM vitrata andM. sjostedti, and particularly parasitic Hyme- noptera, also survive on the same alternative host plant habitat. This paper summarizes the status of knowledge on the interactions between these two cowpea pests, their most important natural enemies, and the alternative wild, host plant habitat. Case study I: the legume pod borer, Maruca vitrata Fabricius (Lepidoptera, Pyralidae> The host plants The host range of M vitrata was studied in Nigeria and Benin by Taylor (1978), Atachi and Djihou (1994), Zenz (1999), and Arodokoun et al. (2001). However, only the latter two studies provide information concerning year-long monitoring of host plants across ecological regions (from the coast to the southern Guinea savanna). The larvae of other Pyralidae occurring in West Africa (e.g., Mussidia nigrivenella Ragonot [Lep., Pyralidae], see Setamou et al. 2000), particularly their early instars, are difficult to distinguish from M. vitrata, and could be mistaken for the latter. For this reason, the larvae sampled by Arodokoun et al. (2001) and suspected to be M. vi trata, were reared until the emergence of the adult moth. The studies by Zenz (1999) and Arodokoun et al. (2001) used quantitative sampling procedures which permitted an assessment of the seasonal abundance ofM. vitrata on the different host plants. This information was used to establish a list of the most important noncrop host plants organized by seasons and habitats (Table 1). Apart from cowpea, other cultivated plants such as pigeonpea (Cajanus cajan) and common bean (Phaseolus vulgaris) are also attacked by larvae of M. vitrata, but were not considered in this study. Without exception, all plants listed in Table 1, and the less important plants cited by Zenz (1999) and Arodokoun et al. (2001), belong to the family of the F abaceae, which lets us conclude that M. vitrata is a stenophagous insect. Although all these host plants occur naturally in the wild, some of them, e.g., the herbaceous legumes Centrosema pubescens and Pueraria phaseoloides, have been introduced as cover crops during the first half of last century, mainly from tropical America and Asia, but are now part of the spontaneous vegetation. Another interesting feature concerning the association of M. vitrata with these host plants is its feeding habit. None of the larval instars feed on growing pods, as is the case for cultivated legumes (cowpea and common beans). Instead, they either feed inside single flowers or spin a web around the whole inflorescence, feeding on several flowers. The most important outcome of these studies is the certainty that, in the area under study, M vitrata does not need cowpea, nor any other cultivated legume, as an obligate host plant in order to complete its annual cycle. This is particularly important during the main dry season, when cowpea cultivation is restricted to moister areas, and cowpea possibly offers a less favorable microhabitat for M vitrata larval development than trees such as 82 Digitized by Google The importance of host plants for biological control of two cowpea pests Pterocarpus santalinoides (Leumann 1994). During the intermediate period from August to November (short dry spell and subsequent short rainy season), cowpea is planted with the return of the rain in mid-September, and thus becomes available for M vitrata larvae approximately one month later, depending on the variety planted. Since the cowpea crop from the first rainy season would have been harvested by late July learly August, this leaves a two-month gap without cowpea. This gap is filled by several Tephrosia and Sesbania species; the most important of them are given in Table 1. The interactions with its natural enemies Earlier studies of the natural enemies complex of M. vitrata (e.g., Usua and Singh 1978) focused on organisms associated withM. vitrata larvae on cowpea. Away from that crop- centered approach for studying natural enemies of a pest, Arodokoun (1996) was the first to look for M vitrata natural enemies, particularly parasitic Hymenoptera, on alternative wild host plants, and to compare their occurrence and parasitism levels with those found on cowpea. Percentage parasitism for each parasitoid was calculated after Bellows et al. (1992) and van Driesche et al. (1991), and is summarized in Figure 1 for each of the important host plants (Arodokoun 1996). Table 1. Flowering season and habitat of major alternative host plants for Maruca vitrata in southern and central Benin (all belonging to the family Fabaceae) (adapted from Zenz [1999] and Arodokoun et al. [2001 n. Host plant Flowering during the main dry season (December-March) Centrosema pubescens Lonchocarpus sericeus Milletia thonningii Pterocarpus erinaceus Pterocarpus santalinoides Pueraria phaseoloides Flowering during the main rainy season (April-July) Afromosia laxiflora Andira inernis Canavalia virosa Centrosema pubescens Dolichos africanus Lonchocarpus cyanescens Lonchocarpus sericeus Pterocarpus santalinoides Pueraria phaseoloides Flowering during the intermediate period (August-November) Sesbania pachycarpa Tephrosia candida Tephrosia humilis Tephrosia platycarpa Vigna racemosa 83 Habitat Ubiquitous Wetland, river banks (coast) Firmland (savanna) Firmland (savanna) Wetland, river banks (savanna) Ubiquitous Firmland (savanna) Firmland (savanna) Firmland (savanna) Wetland (savanna) Firmland (savanna) Firmland (coast, savanna) Firmland (coast, savanna) Firmland (savanna) Wetland (savanna) Firmland (savanna) Firmland (savanna) Firmland (savanna) Firmland (savanna) Firmland (savanna) Digitized by Google Cowpea integrated pest management Compared to cowpea, aggregate parasitism levels were generally higher on wild alterna- tive host plants, with the exception of Lonchocarpus sericeus. Highest overall rates were observed on P. santalinoides during the main dry season, followed by L. cyanescens, while on the herbaceous legumes parasitism rates were lower than 15%. The difference between woody plant species (trees and shrubs such as P. santalinoides, L. sericeus, and L. cyanescens) and herbaceous legumes lies not only in the apparently higher parasitism rates, but even more in the composition of the parasitoid community. AlthoughArodokoun (1996) observed a total of eight different species oflarval parasitoids, and one unidentified parasitic nematode, only three of them seemed to be important and are therefore used in our comparison below. The dominant parasitoid recovered from M. vitrata larvae collected in the flowers of the woody plant species (Fig. 1) was Phanerotoma leucobasis Kriechbaumer (Hym., Braconidae). Parasitoids of the genus Phanerotoma were already observed by Taylor (1967), U sua (1975), and U sua and Singh (1978), but were not identified to species level. P. leucobasis was also found on the herbaceous legumes P. phaseoloides, T. platycarpa, and cowpea (Fig. 1), but with much lower parasitism rates, and in a lower proportion compared with other parasitoid species. Zenz (1999) recovered P. leucobasis mainly from Dolichos africanus and Tephrosia spp., while only one specimen was reared from a total of three larvae found on L. cyanescens. The main reason for the discrepancy between the 40 35 30 25 E ~ .~ 20 99%. Trap catches were counted daily and trapped moths discarded at that time. Insecticides were not sprayed in the fields. In the first experiment, four pheromone blends were evaluated. These were EE 10, 12-16: AId alone or in combination with one or both of the two minor components, EE 1 0, 12-16: OH and E1O-l6:Ald, both of which were present at a level of 5% relative to the EE1O,12- l6:Ald. These synthetic blends were presented in polyethylene vial and rubber septa dispensers as lures in white, sticky, delta traps (Agrisense-BCS, Pontypridd, UK). For each of these, two doses, 0.01 mg or 0.1 mg, were compared, making 16 treatment combinations. These were compared with traps containing two virgin females confined to small wire-mesh cages and with unbaited controls. Females were two days old when placed in traps and were replaced every two days. Sticky card inserts in delta traps were replaced on a weekly basis. The experiment consisted of three cowpea fields, forming replicate blocks; in each, traps were positioned in a grid formation with 10-m spacing. 125 Digitized by Google Cowpea integrated pest management Four subsequent experiments included a comparison oflure age and shielding, two of different trap designs and one of trap height. The lures used in each of these were O.l-mg polyethylene vials. Sticky, delta traps (Fig. la) were used in the lure age and shielding experiment; green plastic funnel traps (Fig. lb) (Agrisense-BCS, Pontypridd, UK), with DDVP insecticide strips to kill trapped moths, were used in one trap design experiment and the trap height comparison. Four water-trap designs (Figs. lc-lf), each constructed from cheap, locally obtained materials, were also evaluated in the trap design comparisons. A water-pan (Fig. lc) trap was made from a green plastic bowl (20-cm diameter) and plate held 5 cm apart with steel wire. Others were made from a 1.5-liter clear plastic bottle (Fig. ld) and 2-1 and 5-1 white plastic jerry cans (Figs. Ie, If) in which four windows had been cut from the sides. Lures were suspended within the center of each trap. A little soap powder was added to the water within each trap to reduce surface tension, and vegetable oil to reduce evaporation. In the lure age and shielding experiment, shielded and unshielded lures were pre-aged for two or four weeks before use by exposing them in sticky, delta traps. Each experiment was carried out to a randomized complete-block design with five replications. Traps within a replicate block were set out in lines or rectangular formations, the exact layout depending on the number of treatments being compared. Individual traps were positioned 20 m apart. Blocks were at least 50 m apart and were usually situated in separate fields. During the trapping experiments, it is possible that there were some interactions between traps within replicate blocks. This may have occurred as individual pheromone plumes overlapped and moths, initially attracted by the plume of one trap, passed on to the plumes of other traps. This would have acted to blur treatment differences. However, the random positioning of treatments within blocks and night-to-night variation in wind direction would have meant that no systematic biases occurred. F or statistical analysis the total catches by each trap over the respective trapping periods were used. With the blend experiment and the lure age/shielding experiment, analysis involved the raw data, since these met the normality and constant variance assumptions. However, it proved necessary to transform data of the trap height (square root) and trap design (loglO [x + 1]) experiments. Analysis of variance was carried out using Genstat 5 for Windows (release 4.1). Where this indicated statistically significant effects, treatment means were separated using the least significant difference (LSD) at the 5% level. Observations relating pheromone trap catches to light trap catches and larval infestations I1TAoperates a light trap (500 W mercury-vapor bulb) at its Cotonou station. During the relevant period this was situated several hundred meters from any experimental cowpea fields. Catches of M vitrata were recorded on a daily basis and compared to those in pheromone traps forming part of the trap and lure optimization experiments. Weekly inspections for larvae were carried out in the fields containing traps from the optimization experiments; all individuals on four randomly selected plants per field were counted. On 22 September 1998, before the second cropping season began, a ring of 20 sticky, delta pheromone traps was established around the perimeter of the I1TA station. These traps were baited with polyethylene vial lures containing 0.1 mg of the 3-component blend (100:5: 5 ratio). They were placed 150 m apart and at least 80 m from the nearest cowpea field. Data from these traps were also compared to the light trap and other data. 126 Digitized by Google Development of sex pheromone traps for monitoring legume podborer, Maruca vitrata (F.) Figure 1 a. Sticky delta trap. tI-sl'j£ p;;' III ...... Figure 1 c. Water-pan trap. " . , - I.... _" -" .~~ -'" ... ~. ~ ..... -... ~' • I ~ ..,.: ...,: . '", .. . ...,. ... .., .... ........ ' . Figure 1 b. Agrisense-BCS funnel trap. Figure 1 d. Plastic bottle water trap (1.5 liters). Figure 1 e. Two-liter bottle water trap. 127 Digitized by Google Cowpea integrated pest management Figure 1 f. Five-liter jerry can water trap. In a separate set of observations, conducted from mid-January to mid-March 2000, two pheromone traps (5-ljerry can design-see trap design experiments) were monitored three times/week in each of 10 on-farm plots. Plots were situated in the villages of Agonguey, Agbonou, and Wosounme in the Oueme valley in southeast Benin. The nearby river allows cowpea to be grown here in what is the off-season elsewhere in Benin. The size of plots varied from 400 to 600 m2. Plots were not treated with insecticides but were situated within larger blocks that were treated. The cowpea variety in each case was Chawe Daho, which has a growing season of 90 days. Sowing dates were 30 November-IS December 1999, flowering commenced 9-25 January 2000, and harvest was 25 February-5 March 2000. At weekly intervals from 19 January to 21 February, 20 flowers per plot were inspected for the presence of larvae. Results Trap and lure optimization experiments In the pheromone blend experiment, traps baited with lures containing all three of the proposed components, EE1O,12-16:Ald, EE1O,12-16:0H, and E1O-16:A1d, caught significantly more male M. vitrata moths than those baited with one or two component blends or live females (P < 0.05), all of which attracted similar numbers of male moths. No males were captured in unbaited control traps (Table 1). Although the polyethylene vial dispensers loaded with 0.1 mg pheromone attracted slightly more males than other combinations of dose or dispenser (Table 2), there was no significant overall effect of dispenser type or dose (P> 0.05 LSD). About 20% of total catches in traps baited with synthetic lures were female moths although almost no females were attracted to live females or unbaited controls. The trends in respect of different blends, doses, and dispensers were similar to those for males (Tables 1 and 2). When the attractiveness of lures of different ages was compared, separate analyses of variance indicated highly significant effects in respect of captures of both sexes (P < 0.01) (Table 3). Four to six-week-old lures were significantly less attractive to males than 0-2 and 2-4-week-old lures (P < 0.05); in respect of females, 0-2-week-old lures were significantly more attractive than both older sets oflures (P < 0.05). Analyses of variance showed that male captures were not influenced by shielding of the lures (P = 0.75), but female captures with shielded lures were significantly greater than with unshielded lures (P < 0.01). Captures of female moths made up 14% of the total in this experiment. 128 Digitized by Google Development of sex pheromone traps for monitoring legume podborer, Maruca vitrata (F.) Table 1. Mean catches/trap of M. vitrata in the blend experiment at IITA, Cotonou, Benin, June-August 1998 (catches for synthetic blends averaged across dose and dis- penser type). Males* lure or ratio of componentst Mean SE 100:0:0 7.0 bc 100:5:0 5.3 c 100:0:5 8.9 b 100:5:5 33.1 a 2 x virgin females 5.8 c Blank, control 0.0 d t(E,E)-1 0,12-16:Ald: (E,E)-10,12-16:0H: (E)-10-16:Ald. 1.4 0.9 1.1 2.4 0.8 0.0 Females* Mean SE 1.3 cd 1.8 bc 2.9 b 5.3 a 0.2 d 0.0 d 0.4 0.4 0.6 0.9 0.1 0.0 *Means within a column followed by a common letter were not significantly different (P > 0.05, LSD following ANOVA). Table 2. Mean catches per trap of M. vitrata with synthetic dispensers in the blend experiment at IITA, Cotonou, Benin, June-August 1998 (catches averaged across different blends). Malest Femalest lure dose/dispenser type Mean SE Mean SE 0.01 mg vials 12.3 3.3 2.5 0.5 0.1 mg vials 16.8 4.2 2.7 0.6 0.01 mg septa 13.3 3.4 3.2 0.9 0.1 mg septa 12.0 3.8 3.0 1.0 t There was no significant effect of dose or dispenser type on male or female catches (P > 0.05, F-ratio ANOVA). Table 3. Mean catches/trap of M. vitrata with lures of different ages, shielded or not shielded from sunlight, at IITA, Cotonou, Benin, August-November 1999. Shielding Males Females lure age (weeks) Yes/No Mean SE Mean SE 0-2 Yes 11.8 a 1.0 3.6a 0.8 1/ No 12.0 a 2.0 1.8 b 0.6 2-4 Yes 11.4 a 1.5 1.6 b 0.7 1/ No 9.8 a 1.9 0.8b 0.2 4-6 Yes 5.0 b 0.8 1.4 b 0.4 1/ No 7.6ab 1.4 0.4 b 0.2 129 Digitized by Google Cowpea integrated pest management In both trap design experiments, significant treatment effects were observed (P < 0.05). In the first comparison, the sticky, delta trap attracted the fewest moths of both sexes (Table 4). Three to four times more males were captured by the Agrisense-BCS funnel trap than the delta trap, but the locally constructed water-pan trap was most effective in capturing females (three times more than the delta trap). The sticky, delta trap was also less effective than two other locally constructed water traps in the second experiment. In this experiment, for both sexes, the 5-1 and 2-1 jerry can designs proved superior to both the delta trap and the 1.5-1 bottle design (Table 5). Overall percentage captures of females in the two experiments were 46% and 35%. The trap height experiment indicated that 120 cm was optimal in respect of catches of males (Table 6). Mean catches of males at this height were significantly greater than at 20 and 170 cm (P < 0.05), though not at 70 cm. Overall catches of females were around 11 % of the total, and there were no significant differences in respect of trap height. Table 4. Mean catches/trap of M. vitrata in the first trap design experiment at IITA, Cotonou, Benin, October-December 1998. Trap design Sticky, delta Water-pan Funnel Males Mean SE 3.0 b 7.6ab 11.0 a 1.6 3.4 4.0 Females Mean SE 3.2 b 9.0a 6.4 ab 1.5 3.5 2.7 Table 5. Mean catches/trap of M. vitrata in the second trap design experiment at IITA, Cotonou, Benin, September-November 1999. Males Females Trap design Mean SE Mean 5-1 jerry 13.0 a 1.8 7.4 a 2-1 jerry 10.8a 2.0 6.0ab Sticky, delta 4.0 b 0.8 1.4 e 1.5-1 bottle 5.0 b 1.1 2.8 be Table 6. Mean catches/trap of M. vitrata at different heights aboveground at IITA, Cotonou, Benin, July-October 1999. Males Females Height (em) Mean SE Mean 20 5.6 be 1.2 0.2 a 70 6.8ab 0.6 1.4 a 120 10.4 a 1.4 0.6 a 170 3.4 e 1.3 1.2 a 130 SE 1.3 1.7 0.5 0.6 SE 0.2 0.4 0.4 1.0 Digitized by Google Development of sex pheromone traps for monitoring legume podborer, Maruca vitrata (F.) Observations relating pheromone trap catches to light trap catches and larval infestations Catches in the light trap were always much greater than in individual pheromone traps, and while males predominated in pheromone trap catches, females tended to form the majority in the light traps. Within each type of trap, temporal patterns of catches of each sex were similar, so that catches of one sex accurately reflected the presence of the other. During two seasons of on-station trials at IITA-Cotonou, a general observation was made that the timing of catches in the light trap and those in pheromone traps within fields did not correspond closely. However, there was a better temporal correspondence between the light trap catches and those in perimeter traps (Fig. 2). This was notable at the start of the second season of 1998. Following several weeks of zero catches in the light trap and the perimeter pheromone traps, the latter detected the first small peak of moths at exactly the same time (29 October) as the light trap, although there appeared to be little subsequent quantitative correlation. During this period, the first appearance of moths in traps in cowpea fields was at least 12 days after catches were first noted in the light trap and perimeter pheromone traps. These initial within-field catches were 33-50 days after the fields were sown. The first crop inspection, eight days after the initial catches in perimeter traps, showed that larvae were already present in each of three fields sampled at that time; but this was several days before within-field catches in two of the fields and simultaneous with the first catches in a third. Representative data for one field are shown in Figure 3 and can be compared to Figure 2. Trap catches within fields in the second season were confined to periods of 8-12 days. Results from the on-farm observations in the Oueme valley are summarized in Figure 4. Although overall catches were relatively low-rarely exceeding an average of 0.5 moths per trap per count-the timing of their onset across all plots was consistent. In eight of the 10 plots, the first catches were noted on 28 January, while first catches occurred in the remaining plots on the subsequent count three days later. Catches were evenly distributed across alII 0 plots and three village sites until the end ofF ebruary, when they began to decline. Males and females were trapped in approximately equal numbers. Larvae were only found on two dates: 9 and 14 February. On the first occasion they were noted in four plots, on the second they were observed in seven plots. Since some of the larvae were late instars it is probable that eggs were laid five to ten days after the first adults were trapped. Discussion From the results of the trap and lure optimization experiments an effective and practical trapping system for M. vitrata has now been developed for the first time. The best phero- mone blend is amixtureofEE1O,12-16:Ald, EE1O,12-16:0H, andE1O-16:Aldin the ratio 100:5:5. Although no significant differences were evident in respect of dose or dispenser, the O.I-mg polyethy lene vials would be expected to show the greatest longevity of the lures tested on the basis of dose and release rate characteristics. Our results indicate no loss of attractiveness for up to four weeks under field conditions. Therefore, these lures have now been adopted as standard for use in further work. The best trap height is 120 cm and the most effective traps are those produced from locally available plastic jerry cans. Not only are these relatively much cheaper than imported, commercial designs (US$0.30-O.80 as 131 Digitized by Google Cowpea integrated pest management 90 80 70 60 ., OJ .c 1; 50 " .... ~ 1: 40 ~ >-~ 30 .e!> Z 20 10 12 til 10 .... ~ "0 E ~ 8 .c .... ~ .S! OJ E 6 ·t .... Q N E 4 til OJ .c ~ IU " >- 2 ~ .e!> z n 22Sep 07 Oct 22 Oct ~ ~ 06 Nov 21 Nov 06 Dec •••••• Females _Males 21 Dec 05 Jan 20 Jan 04 Feb 19 Feb I _Males I _Females 06 Nov 21 Nov 06 Dec 21 Dec 05 Jan 20 Jan 04 Feb 19 Feb Figure 2. M. vitrata catches in a light trap (top) and in 20 pheromone traps in non crop areas around the perimeter of the IITA station (bottom) during and after the second crop- ping season in 1998. 132 Digitized by Google Development of sex pheromone traps for monitoring legume podborer, Maruca vitrata (F.) 50 40 10 o ___ Females --Males • larvae present o larvae not found o o o O++++++++++++++++++++++++++++++++++~~++++++++++++++++rH 21 Oct 26 Oct 31 Oct 5 Nov 10 Nov 15 Nov 20 Nov 25 Nov 30 Nov 5 Dec 10 Dec 15 Dec Duration of trapping Figure 3. Total catches of M. vitrata by seven pheromone traps in field e3, forming one block of a pheromone blend experiment during the second cropping season in 1998. 8 2 o 0 • Larvae present D Larvae not found Females Males O+-__ ~ __ ~ __ ~ __ ~-L~ __ ,-__ ,-__ ,-__ ~ __ ~~~~~ __ ~ __ ~ __ 5 Jan 10 Jan 15 Jan 20 Jan 25 Jan 30 Jan 4 Feb 9 Feb 14 Feb 19 Feb 24 Feb 29 Feb 5 Mar 10 Mar 15 Mar Duration of trapping Figure 4. Total catches of M. vitrata by 20 pheromone traps in lOon-farm plots in the Oueme valley, Benin, 2000. 133 Digitized by Google Cowpea integrated pest management against approximately US$3. 00 for sticky, delta traps and more for plastic funnel traps), they are easy to construct and robust in use. Further work on trap and lure optimization is required. Experiments are planned to determine the effect of trap color and frequency of checking on trap catches. An experiment concerning isomeric purity of the pheromone blend components is currently underway and the results will be of particular relevance. If a lower level of purity can be used without a marked loss of attraction, it will be possible to reduce the cost of lure production to around US$0.50 per lure. This will be important in helping to ensure the economic viability of pheromone trap monitoring of M. vitrata. The consistent capture of significant numbers of female moths with synthetic sex pheromone lures is, to the best of our knowledge, unprecedented. It could suggest an incomplete pheromone blend, but extensive analytical work with the natural pheromone, to be reported elsewhere, has failed to find any evidence for further blend components (M.C.A. Downham and D.R. Hall, unpublished data). Furthermore, incomplete phero- mone blends generally produce lower catches of males, rather than co-attraction of both sexes. Thus a better explanation may lie in some unknown aspect of the species' mating behavior or ecology, and further work to explain the phenomenon would be very help- ful. Regardless of explanation, catches of females may actually improve the predictive power of traps, since they should more accurately reflect local population events than males alone. With regard to the practical use of traps for monitoring purposes, results to date indicate real potential. In the on-station trials at Cotonou, catches in pheromone traps outside cowpea fields preceded larval infestations in fields; in contrast, catches in traps within the fields occurred only after larvae appeared. Thus it seemed that traps near but outside fields might be better able to predict pest attacks. However, a subsequent trial at a different time of year in farmers' fields suggested that within-field traps could give early warning of larval infestations. Resolving the question of the best positioning of traps to detect immigrating moths will be the task of further work currently in progress at several on-farm sites around Benin, and soon to be extended to Ghana. This work is being carried out in association with the West African regional PRONAF (projet de Niebe pour I' Afrique) project, a partnership between I1TA and various NARS that aims to promote the transfer and implementation of research on cowpea to subsistence farmers. It will be necessary to determine the relationship between larval infestations and catches by pheromone traps with confidence. Whether any good quantitative correlation between catches and damage exists remains to be seen, but it seems likely that pheromone traps could be used as the basis for timing the application of control measures. Afun et al. (1991) found that by using action thresholds based on larval/flower infes- tation rates to time insecticide applications on cowpea, the number of sprays could be reduced, relative to a calendar-based approach, with no loss of control and at reduced cost. Ultimately it is hoped that pheromone trap-based action thresholds could be used in conjunction with other promising sustainable control methods being developed through PRONAF, e. g., neem-based insecticides CW. Hammond, pers. comm.; see also Bottenberg and Singh 1996). Furthermore, Bottenberg (1995) found that many farmers are unable to link the adult stage of M. vitrata with the highly destructive larval stage. Pheromone traps could also serve a training role by assisting in making up this gap in knowledge. 134 Digitized by Google Development of sex pheromone traps for monitoring legume podborer, Maruca vitrata (F.) References Abate, T. and J.K.O.Ampofo. 1996. Insect pests of beans inAfrica: their ecology and management. Annual Review of Entomology 41: 45-73. Adati, T. and S. Tatsuki. 1999. Identification ofthe female sex pheromone ofthe legume pod borer, M vitrata and antagonistic effects of geometrical isomers. Journal of Chemical Ecology 25: 105-115. Afun, lVK., L.E.N. Jackai, and C.l Hodgson. 1991. Calendar and monitored insecticide applica- tion for the control of cowpea pests. Crop Protection 10: 363-370. Bottenberg, H. 1995. F arrners' perceptions of crop pests and pest control practices in rainfed cowpea in Kano, Nigeria. International Journal of Pest Management 41: 195-200. Bottenberg, H. and B.B. Singh. 1996. Effect of neem leaf extract applied using the 'broom' method, on cowpea pests and yield. International Journal of Pest Management 42: 207-209. Bottenberg, H., M. Tamo, D.Arodokoun, L.E.N. Jackai, B.B. Singh, and O. Youm. 1997. Population dynamics and migration of cowpea pests in northern Nigeria: implications for integrated pest management. Pages 271-284 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agri- culture (I1TA) and Japan International Center for Agricultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Dreyer, H., 1 Baumgartner, and M. Tamo. 1994. Seed damaging field pests of cowpea (Vigna unguiculata L. Walp.) in Benin: occurrence and pest status. International Journal of Pest Manage- ment 40: 252-260. Jackai, L.E.N. 1995. Integrated pest management of borers of cowpea and beans. Insect Science and its Application 16: 237-250. Jackai, L.E.N., RS. Ochieng, and J.R. Raulston. 1990. Mating and oviposition behavior in the legume pod borer, Maruca testulalis. Entomologia Experimentalis et Applicata 56: 179-186. Onyango, F.O. and J.P.R Ochieng-Odero. 1993. Laboratory rearing ofthe legume pod borer, Maruca testulalis Geyer (Lepidoptera: Pyralidae) on a semi-synthetic diet. Insect Science and its Appli- cation 14: 719-722. Shanower, T.G., J. Romeis, and E.M. Minj a. 1999. Insect pests of pigeonpea and their management. Annual Review of Entomology 44: 77-96. Singh, S.R., L.E.N. Jackai, J.H.R Dos Santos, and C.B. Adalia. 1990. Insect pests of cowpeas. Pages 43-90 in Insect pests of tropical legumes, edited by S.R Singh. John Wiley and Sons, Chichester, UK. Taylor, TA. 1967. The bionomics of Maruca testulalis Gey. (Lepidoptera: Pyralidae), a major pest of cowpeas in Nigeria. Journal of the West African Science Association 12: 111-129. 135 Digitized by Google 2.4 Evaluation of a novel technique for screening cowpea varieties for resistance to the seed beetle Callosobruchus maculatus A.D. Devereau1, L.E.N. JackaP, I.B. Olesegun2, and A.N.J. Asiwe2 Abstract A novel method for screening cowpea varieties for resistance to the postharvest insect pest Callosobruchus maculatus (F.) (Coleoptera: Bruchidae) was compared to conventional screening techniques at the International Institute for Tropical Agriculture (ITTA) laboratory in Ibadan, Nigeria. The new technique assesses seed resistance by measuring larval feeding activity via electronic sensors. Initial small- scale trials demonstrated that the method could be successfully applied in the laboratory with a potential saving in time and effort. The results were used to design a larger scale device that was able to screen 32 varieties of cowpea in 19 days using a methodology designed to identity only the most resistant cowpea varieties. Resistant varieties were identified but some problems were encountered with analysis of the results. Practical application ofthe technique for large-scale resis- tance screening is discussed. Introduction The "biomonitor" technique, which uses ultrasonic transducers to detect sounds made by insect larvae feeding within seeds, was first described by Shade, Fergason, and Murdock (1990). The sounds were counted automatically and used as a measure of insect feeding activity. Work at the Natural Resources Institute (NRI) (Devereau et al. 1999) has inves- tigated the use of this technique for screening varieties of cowpea, Vigna unguiculata (L.) Walp., for resistance to the bean weevil Callosobruchus maculatus (F.) (Coleoptera: Bruchidae). A device was developed which could simultaneously monitor eight cowpeas, each containing one insect, and a methodology was developed that detected significant differences between susceptible and resistant cowpea varieties by comparing feeding activity between 14 and 15 days after oviposition. This paper describes two sets of trials undertaken at the International Institute of Tropi- cal Agriculture (I1TA) in Ibadan, Nigeria, to further develop the technique. The first set of trials tested the ability of the technique to differentiate between four cowpea cultivars of different but known susceptibility. The time and effort required in comparison to the conventional screening method were also measured. The results were used to develop a 1. Natural Resources Institute, Central Avenue, Chatham Maritime, Kent, ME4 4TB, UK. Tel.: +44 1634883796, Fax.: +44 1634883567, email: A.Devereau@gre.ac.uk. 2. International Institute of Tropical Agriculture, c/o Lamboum (UK) Ltd., Carolyn House, 26 Dingwall Road, Croydon, CR9 3EE, UK. Tel.: +23422412626, Fax.: +23422412221. 136 Digitized by Google Evaluation of novel technique for screening cowpea larger scale device designed to be both faster than conventional screening and to require less staff time. This new device was tested in the second set of trials. Materials and methods Biomonitor system The system used in the first set of trials, shown schematically in Figure 1, was essentially the same as that used by Shade et al. (1990). It was housed in a small metal cabinet to help screen it from electrical interference. Each of the eight transducers was monitored for five minutes in turn, with the number of counts recorded automatically by a datalogger (Delta-T Devices, UK). Each sensor was therefore monitored for five out of every forty minutes, resulting in 36 readings being recorded from each sensor during the 24-hour monitoring period. The system was placed in the culture room at I1TA in which insects were being maintained for the trial. The temperature and relative humidity (rh) were 26 ± 2 °C and 70-80%. For the second trial a larger scale biomonitor device was designed in an attempt to approach a practical level of throughput. The number of sensors was increased to 32 and the monitoring time was reduced to one minute per channel, giving a reading every 32 minutes from each sensor, i.e., 45 readings per 24 hours. Insect population The laboratory population of C. maculatus that has been maintained at I1TA since 1973 with regular crossing using local (lbadan market) insects was used in the study. The insects were cultured on dry seeds of a susceptible cowpea cultivar, either Ife Brown, IT84D-7l5, a b ,.,., .. ,.,.".,.".,., .. ,.,.".,.".,., .. ,.,.,h..".,.,""""""'r"""""""""'[ ! ' g Figure 1. Block diagram of the biomonitor. a = infested cowpea; b = ultrasonic transducer; c = pre-amplifier; d = electronic switch; e = filter/amplifier; f = trigger; g = datalogger; h = clock signal (for electronic switch operation). 137 Digitized by Google Cowpea integrated pest management or IT82E-889. They were maintained under laboratory conditions until adults started to emerge after about 21-26 days. During the trials cultures were initiated every five days to provide a regular supply of 0-6-day-old adults. Cowpea varieties The four cowpea varieties used in the first trial were Ife Brown, IT87D-1827, TVu 2027, and IT84S-2246-4. Ife Brown is used at I1TA as a susceptible reference, and TVu2027 was established by Singh and lackai (1985) as the only accession in I1TA's germplasm collection to show resistance. Ofuya and Credland (1995) measured the relative suscepti- bilities of cowpea varieties including Ife Brown, TVu2027, and IT84S-2246-4. The latter two had significantly longer (P ::::: 0.05) development periods than Ife Brown, but only TVu2027 showed lower percentage adult emergence. Recent work at I1TA (L. lackai, personal communication.) showed that IT87D-1827 was a susceptible variety and also suggested that TVu2027 showed less resistance to the strain of C. maculatus used at I1TA than to other strains of the species. Thirty-three cowpea varieties were used in the second trial. Conventional screening (Singh and lackai 1985) was used to establish the relative susceptibility of 24 of these varieties, which are listed with the results in Table 4. The 32 varieties screened using the biomonitor included all those screened conventionally with the exception of variety Ngouya Local, as well as varieties IT-534, IT8ID-994, IT433-1, IT82E-25, TVu1509, TVu6867, Vicaml, TVullOll, and TVu3000. Infestation method Cowpeas were brought out of cold storage and allowed to return to room temperature in the laboratory 24 hours before infestation. Clean, unbroken cowpeas were placed in a single layer inside labelled plastic CORNING® 35 mm tissue culture dishes (35 x 10 mm) with one cowpea variety per dish. The dishes were then introduced into a culture j ar containing hundreds of 0--6-day-old adults for 1 hour between 9am and lOam for oviposition (Ofuya and Credland 1995). Dishes were removed from the culture after 55 minutes and the insects removed using a vacuum generator during the remaining 5 minutes. Excess eggs were removed from the exposed cowpeas after 24 hours to leave one egg on the cheek of each seed. Cowpeas with no eggs or with eggs laid in the wrong area were discarded. The remaining cowpeas were left in the laboratory for 14 days, after which they were ready for evaluation on the biomonitor. At this time the seeds were inspected to ensure that the eggs had hatched and that the larvae had penetrated the seed. Seeds not showing penetration were discarded. Experimental design A randomized block design was used for the first trial. Cowpeas of all four varieties were exposed simultaneously to oviposition on consecutive days, using fresh cowpeas each day, to form a series of infested cowpea sets. Exactly 14 days after oviposition on the first cowpea set, two replicates of each variety were selected, placed at random on one of the eight biomonitor sensors and monitored for 24 hours. After 24 hours they were removed, the data collected, and the process repeated using the next set of cowpeas. This was repeated for eight days. The data were analyzed after five days (10 replicates of each treatment) and eight days (16 replicates of each treatment) with each day treated as a block to account for variations in, for example, parent insect age, laboratory conditions, etc. 138 Digitized by Google Evaluation of novel technique for screening cowpea A randomized design was used for the second set of trials using the 32-channel device. This device was designed to detect significant differences between eight cowpea varieties during one 24-hour monitoring period using only four replicates of each variety, so no blocking was required. The susceptible and resistant references, Ife Brown and TVu2027, were included in each set of eight cowpeas. All eight varieties were exposed to oviposition and monitored as before, with the cowpeas placed at random on the 32 sensors. Data analysis The mean and peak activity over 24 hours, as counts per five minutes or counts per minute for the 32-channel device, were calculated for each replicate. The statistical package SPSS for Windows was used to analyze this data graphically and using ANOVA, for which a 10gIO transformation was used. Contrasts between each cowpea variety and the reference varieties were then made. Time and effort requirement The time and staff effort required for screening using both conventional and biomonitor techniques were estimated by I1TA staff as the trials were conducted. Results First trials Screening results Figures 2 and 3 show the mean and peak feeding activity for each of the four cowpea variet- ies after five days, i.e., with 10 replicates per treatment. The susceptible varieties showed much larger ranges of feeding activity than the resistant varieties. ANO VA performed on these data showed significant differences (F(3,32) = 25.26, P < 0.001, and F(3,32) = 22.49, P < 0.001) due to cowpea variety for mean and peak feeding activity, respectively. :rJ '5 c ·e Q/ :! 400 .. Q/ Q. 300 II> C :::I 8 200 100 Ife Brown IT87D-1827 Key to box plot (Marsh 1988) Outliners * Upper quartile Median Lower quartile TVu2027 I Main body of data IT845-2246-4 Figure 2. Mean activity (counts per five minutes> after five days for the four cowpea varieties in the first trials. 139 Digitized by Google Cowpea integrated pest management Ife Brown IT87D-1827 TVu2027 IT845-2246-4 Figure 3. Mean peak activity (counts per five minutes) after five days (key as in Figure 2). Table 1 records the mean and peak feeding activity after five days, and shows that varieties IT84S-2246-4 and TVu2027 exhibited significantly lower mean feeding activ- ity (t32 = 8.43, P < 0.00001; and t32 = 3.38, P = 0.002, respectively) when compared with Ife Brown, while IT87D-1827 showed a difference which was less significant (t32 = 2.32, P = 0.026). For peak feeding activity, varieties IT84S-2246-4 and TVu2027 again showed significantly lower activity (t32 = 7.67,P<0.00001; and t32 = 2.78,P =0.009, respectively), but IT87D-1827 did not differ from Ife Brown. The larva in one replicate of IT87D-1827 was inactive during the whole monitoring period, in contrast to all other individuals in cowpeas of that variety. This had a large effect on the results, reducing the mean activity feeding values for variety IT87D-1827 by 13 counts per five minutes and peak activity feeding values by 46.4 counts. Table 2 shows the mean and peak feeding activity after eight days, i.e., from 16 repli- cates per treatment. The results are very similar to those after five days. Time and effort requirements Both methods required an initial 28 days for preparation of insect cultures. The conven- tional method then required a further 64 days to complete, while the biomonitor took 19 days. The conventional bioassay and the biomonitor technique needed almost identical amounts of staff effort, 21.1 and 21.2 hours, respectively, to reach a conclusion. This included prepa- ration of insect cultures, oviposition, and collection and analysis of results. Second trials Screening results The screening methodology used was designed to identify only the most resistant variet- ies, allowing susceptible varieties to be rapidly eliminated and resistant varieties to be subjected to further, more detailed analysis. The number of replicates needed to show significant differences between varieties was estimated using a formula given by Sokal and Rohlf (1981). This suggested that four 140 Digitized by Google Evaluation of novel technique for screening cowpea Table 1. Mean and peak activity (counts per five minutes) and loglo transformation of mean and peak activity for four cowpea varieties tested at IITA. Standard error of the difference between transformed mean and mean peak activity = 0.198 and 0.18, respec- tively. Transformed Transformed Cowpea variety Mean activity mean activity Peak activity peak activity Ife Brown 248.2 2.21 445.5 2.56 IT87D-1827 119.7 1.75 425.4 2.34 TVu2027 37.6 1.54 123.4 2.06 IT845-2246-4 4.97 0.54 21.5 1.18 Table 2. Mean activity (counts per five minutes) and loglo transformation of mean and peak activity for each cowpea variety after eight days. Standard error of the difference between transformed means = 0.198 and 0.18, respectively. Mean activity Mean peak Transformed (counts per Transformed activity (counts mean peak Cowpea five minutes) mean activity perfive minutes) activity Ife Brown 234.2 2.12 412.6 2.49 IT87D-1827 99.3 1.66 329.1 2.19 TVu2027 45.6 1.59 182.8 2.13 IT845-2246-4 26.9 0.70 54.9 1.25 replicates would be required to show a significant difference at the 5% level between the mean activities of Ife Brown and the most resistant variety, IT84S-2246, in 90% of experiments, and this was selected as the criterion or threshold for indicating resistance in the second trial. By increasing the number of biomonitor sensors to 32 and using four replicates, eight cowpea varieties could be monitored per day. Figures 4a-4e show the boxplots of mean feeding activity from the five sets of cowpeas monitored during the second trial. There were clearly contrasting levels of activity, with many varieties showing very low activity, suggesting resistance. Peak feeding activity showed a very similar pattern for most varieties with the exception of IT82E-25 (Figure 4d) which showed a relatively higher range of peak activity than mean activity. Table 3 shows the mean and peak feeding activity for the set of cowpeas shown in Figure 4a. ANOVAfor this data showed significant differences due to cowpea variety (F (7,24) = 6.727, P < 0.001 and F(7,24) = 6.73, P < 0.001, respectively) for the 10gIO transformed mean and peak data. Only variety IT82E-7l6 showed significantly lower feeding activity at the 5% level than Ife Brown, and was therefore identified as the only resistant variety in the batch. The other sets of cowpeas, i.e., those shown in Figures 4b-4e, were similarly analyzed. Varieties IT89KD-245, IT-534, IT84S-2246, and IT8ID-994 all showed signifi- cant lower feeding activity than Ife Brown at the 5% level and were therefore identified as resistant. Table 4 shows the results of the conventional screening trials. Not all of the variet- ies that were identified as resistant by the biomonitor method, i.e., those that showed a significantly lower level of activity than Ife Brown, were screened conventionally. Those that were included IT89KD-245, IT84S-2246, and IT82E-7l6, and these varieties were all 141 Digitized by Google Cowpea integrated pest management Table 3. Mean and peak activity for four replicates of eight cowpea varieties between 14 and 15 days after oviposition. S.e.d. between transformed means = 0.41 and 0.38, respec- tively. Mean activity Peak activity (counts per Tranformed (counts per Transformed Cowpea variety minute) mean activity minute) peak activity MRx17-85S 250.9 2.32 577.5 2.70 TVu 9525 112.4 2.01 266.0 2.38 Ife Brown 107.7 1.19 185.0 1.69 IT86D-498 104.4 1.75 176.0 2.07 Moussa Local 84.4 1.92 162.8 2.21 TVu 801 52.2 1.59 114.8 1.97 TVu 2027 12.1 0.79 48.3 1.26 IT82E-716 1.08 3.4 x 10-2 4.0 0.46 Table 4. Relative susceptibility of cowpea varieties determined by conventional bioas- say at IITA. Total development time (TDn = mean development time per insect. Growth index (GI) = ([in % adult emergence]/TDn. Susceptibility index = (GI test material/GI Ife Brown) x 100. Adult Total emergence development Growth Susceptibility Cowpea variety (%) time index index Ife Brown (SC) 100.0 23.1 0.20 100.0 Moussa L. 100.0 23.7 0.19 97.4 IT89KD-457 100.0 24.0 0.19 96.3 TVu 13731 100.0 24.1 0.19 95.7 MRx17-85S 100.0 24.3 0.19 95.1 IT91K-180 100.0 24.4 0.19 94.7 TVu 9525 95.0 24.1 0.19 94.6 IT86D-888 100.0 24.8 0.19 93.3 Ngouya L. 100.0 25.0 0.18 92.5 MRx10-85S 79.2 23.9 0.18 91.8 IT86D-400 87.5 24.6 0.18 90.7 TVu 12151 87.5 24.9 0.18 89.7 TVu 801 83.3 24.7 0.18 89.2 Zonkwa L. 91.7 25.5 0.18 88.8 IT87D-1827 81.3 27.0 0.16 82.2 I T89KD-245 100.0 30.9 0.15 74.6 IT86D-364 62.5 29.0 0.14 71.0 IT82E-716 83.8 31.8 0.14 70.4 Maiduguri B 59.8 30.2 0.14 69.1 IT87D-697-2 90.0 33.3 0.13 67.6 IT84S-2246 81.7 33.4 0.13 66.1 TVu 2027 68.8 36.2 0.11 57.5 IT86D-498 23.8 32.0 0.09 45.8 IT87S-1393 26.7 39.0 0.09 43.8 in the lower half of Table 4, i.e., they were less susceptible, though they showed higher susceptibility than variety TVu2027. Other varieties with low susceptibility in Table 4 were not shown to be resistant by the biomonitor method however. Figure 4 shows that most of the varieties in the lower half of Table 4 showed low levels of feeding activity in 142 Digitized by Google Evaluation of novel technique for screening cowpea 500'1 30QI 2001' 100+ IT875- Zonkwa lie IT86D TVu IT86KD- IT86[)' TVu -ass 801 498 9525 2027 1393 local Brown 364 12151 457 888 2027 (a) (b) -I -I fl i i.:: u '']____ iii II _ '- T I - . ---,----1T89KD IT- Ife MRx17 TVu IT86D- TVu TVu IT82E- Ife TVu IT87D IT81 0 IT845- IT91 K 11433 -245 534 Browr, -ass 801 498 9525 2027 25 Brown 2027 1827 -994 2246 -ISO (e) (d) 400 300 200 100 -1509 6"7 11011 3000 2027 Bro>.lln (e) Figure 4. Boxplots of mean feeding activity (counts per minute) for five sets of eight cowpea varieties monitored between 14 and 15 days after oviposition (key as in Figure 2). 143 Digitized by Google Cowpea integrated pest management comparison to insects in Ife Brown, but they were not found to be significantly different due to one or two replicates of Ife Brown being inactive during monitoring of each set of cowpeas. The other main disagreements were variety IT86D-498, which was the second most resistant variety according to the conventional screening but showed high levels of feeding activity on the biomonitor (Figure 4a), and varieties IT86D-888 and IT9lK-180 which were of high susceptibility according to conventional screening but showed low levels of feeding activity on the biomonitor (Figures 4b and 4d, respectively). Time and effort requirements The 24 cowpea varieties screened by the conventional technique required approximately 95 hours of effort and took 64 days to complete. The biomonitor by contrast required 36 hours of effort and took 19 days to reach a conclusion for 32 varieties. Discussion The first trial showed clear differences between feeding activity in the cowpea varieties using the biomonitor method. These corresponded to the known susceptible status of these varieties. It also confirmed that variety TVu2027, when tested using I1TA insects, was not the most resistant variety. There was however the problem caused by variability of the insects' development rates. The monitoring period of 24 hours after 14 days' development was designed to correspond to the highly active fourth instar larvae in susceptible varieties and provide a significant contrast to the low activity of larvae in resistant varieties. However, for variety IT87D-1827, the fourth instar started or stopped during the monitoring period in some cases and caused some of the moult period before the fourth instar or the pupal period after the fourth instar to be monitored. As no feeding activity occurs during these periods of development this caused the mean activity to be reduced. Including the peak activity in the analysis helped to identify when this had happened. One replicate of IT87D-1827 showed no larval feeding activity throughout monitor- ing, and this had a relatively large effect on the mean feeding activity levels. This could have been caused by the fourth instar finishing before monitoring began or starting after it had finished for this replicate, or by the larva dying between penetration and monitoring for reasons which may have been related or unrelated to susceptibility. Because of this uncertainty, it is difficult to justify removing the point from the analysis, especially as a similar situation in a resistant variety would have been easily overlooked or attributed to resistance. A similar amount of effort was required to conduct both the biomonitor and conven- tional techniques during the first trial. The biomonitor was much faster though, reaching a conclusion in less than a third of the time of the conventional method. This is a distinct advantage as cowpea breeders need to know the resistant status of new varieties as soon as possible after harvest. Another advantage of the new method was that data is collected automatically and downloaded directly to a computer for analysis, which is less laborious and less prone to error than manually recorded results. The methodology for the second trial was designed to be as fast as the first trial but to require less effort. This was achieved by increasing the capacity of the biomonitor to 32 channels and by reducing the number of replicates to four, a level that would only identify the most resistant cowpea varieties. The varieties identified as resistant in the second trial 144 Digitized by Google Evaluation of novel technique for screening cowpea were among those identified as having lower susceptibility by conventional screening, but other varieties shown to be resistant by conventional screening were not identified by the biomonitor technique. The problem in most cases was that, despite some varieties showing consistently low levels of feeding activity typical of resistance, one or two replicates of the susceptible variety Ife Brown to which they were being compared showed no feeding activity during monitoring despite their know susceptibility. Removing these inactive replicates from the analysis would allow more varieties to be identified as resistant, but, as discussed above, this is hazardous as the cause of the inactivity was not determined and could have been due to a number of reasons. When such low feeding activity occurs in the susceptible reference it would be sensible to repeat the trial. Three cowpea varieties, IT86D-498, IT86D-888, and IT9IK-180, showed complete disagreement between the biomonitor results and conventional screening. It was possible that the results from either method were incorrect-given the number of tests being made and the natural variability in susceptibility there will always be some errors or anomalies in screening results. Relatively susceptible varieties being identified in error as resistant is not a serious problem provided it does not occur too often, as the more detailed screening that should follow the initial rapid screen will identify these varieties. Failing to detect a resistant variety is more serious however. The methodology needs to be designed to minimize the risk of this happening. It is also possible that resistance or susceptibility was not manifest in these varieties in the same way as in other varieties. Further, detailed investigation will be needed to establish whether this is the case or not. In conclusion, although the biomonitor technique was successful on a small scale with clear advantages of reduced time and effort over conventional screening, further work is necessary to ensure its reliability for large-scale screening. The degree of replication used was too small given the amount of variability which occurred, so this should be increased and the criterion used for identifying resistance revised. It may also be necessary to modify the monitoring period to take into account variability in larval development time. To achieve these modifications while retaining a practical level of throughput will require a device with a larger capacity, and this is being addressed by current work at NRI. Acknowledgements This publication is an output from a research project funded by the United Kingdom Department for International Development (DFID) for the benefit of developing countries. The views expressed are not necessarily those ofDFID. DFID Project code R6508, Crop Post-Harvest Programme. References Devereau,AD., P.F. Credland, J. Appleby, and L. Jackai. 1999. Rapid screening of grain for insect resistance. Pages 18-26 in Proceedings ofthe 7th International Working Conference on Stored- Product Protection, edited by J. Zuxun, L. Quan, L. Yongsheng, T. Xianchang and G. Lianghua, 14-19 October 1998, Beijing. Sichuan Publishing House of Science and Technology, Chengdu, P.R. China. Marsh C. 1988. Exploring data. Polity Press, Cambridge, Massachusetts, USA. 385 pp. Ofuya, T.I. and P.F. Credland. 1995. Responses of three populations of the seed beetle, Callosobruchus maculatus (p.) (Coleoptera: Bruchidae), to seed resistance in selected varieties of cowpea, Vigna unguiculata (L.) Walp. Journal of Stored Product Research 31: 17-27. 145 Digitized by Google Cowpea integrated pest management Sokal, RR and F.I Rohlf. 1981. Biometry. 2nd edition. W.H. Freeman and Company, New York, USA. 859pp. Shade, RE., E.S. Fergason, and L.L. Murdock. 1990. Detection of hidden insect infestations by feeding generated ultrasonic signals. American Entomologist 36(3): 231-234. Singh, S.R. and L.E.N. Jackai. 1985. Insect pests of cowpeas in Africa: their life cycle, economic importance and potential for control. Pages 217-231 in Cowpea research production and utiliza- tion, edited by S.R. Singh and K.O. Richie. John Wiley and Sons, New York, USA. 146 Digitized by Google 2.5 Detection of fumonisin 81 in cowpea seeds Q. Kritzingerl, T.A.S. Aveling2, W.F.O. Marasas3, G.S. Shephard3, and N. Leggott3 Abstract Cowpeas (Vigna unguiculata [L.] Walp) are important nutritious legume crops for many subsistence farmers and rural communities. In tropical and subtropical Africa, cowpeas are often stored at high relative humidities and high ambient temperatures and are susceptible to fungal contamination. Some of these fungi produce myco- toxins, which can have adverse effects on the health of both farm animals and humans. Eight cowpea seed samples from four different cultivars were analyzed for the Fusarium mycotoxins, fumonisins Bl' B2, and Br Samples were extracted with methanol/water (70:30) and cleaned up on strong anion exchange solid phase extraction cartridges. High-performance liquid chromatography with precolumn derivatization using o-phthaldialdehyde was used for the detection and quantifica- tion offumonisins Bl' B2, and Br The analyses revealed that all eight samples were contaminated with fumonisin Bl at levels ranging between 81 and 1002 ng/g, whereas fumonisins B2 and B3 were not detected. It is believed that this is the first report of the natural occurrence offumonisin Bl in cowpea seeds. Since none of the known fumonisin-producing fungi were isolated from the cowpea seeds, it is necessary to identify which species are responsible for toxin production. Introduction Cowpeas (Vigna unguiculata [L.] Walp.) are regarded as popular and important indigenous African legume crops by many rural communities living in less developed countries of tropical and subtropical Africa. They are grown as a pulse, vegetable, fodder, and as a cover crop (Ushamalini et al. 1998). Cowpeas are mainly consumed as a favorite food- stuff in the form of dried seeds, either as flour or split (Johnson and Raymond 1964; van Wyk and Gericke 2000). They are a good source of carbohydrates, vitamins, and protein, providing more than half of plant protein in human diets in some areas of the semiarid tropics (Singh et al. 1997; Tuan and Phillips 1992). It is well known, however, that cowpea seeds are susceptible to fungal contamination when poorly stored at high relative humidities and high ambient temperatures (Esuruoso 1975; Hitokoto et al. 1981; Seenappa et al. 1983). It is also under these conditions that certain fungi may produce toxic secondary metabolites, namely mycotoxins (van Warmelo 1967). The ingestion of mycotoxins in contaminated agricultural products can lead to detrimental health problems for humans and farm animals (Desjardins and Hohn 1997; Moss 1996). Mycotoxins exhibit properties of acute, subacute, and chronic toxicity, leading to interference with the functioning of various body systems (Coker 1994; 1. Department of Botany, University of Pretoria, Pretoria, 0002 SouthMrica. 2. Department of Microbiology and Plant Pathology, Forestry and Agricultural Biotechnology Institute, University of Pretoria, Pretoria, 0002 South Africa. 3. Programme on Mycotoxins and Experimental Carcinogenesis (PROMEC), Medical Research Council, PO Box 19070, Tygerberg, 7505 South Africa. 147 Digitized by Google Cowpea integrated pest management Saber et al. 1998). Furthermore, they are capable of causing mutations and deformities in developing embryos (Saber et al. 1998). Fumonisins are the most recently characterized mycotoxins that have major significance in human health (Moss 1996). They are primarily produced by Fusarium verticil/ioides (Sacc.) Nirenberg, Fusarium proliferatum (Matsushima) Nirenberg, and Fusarium nyg- amai Burgess and Trimboli (Coker 1994; Marasas 1994). Fumonisins are acutely toxic to the liver and kidneys (Desjardins and Hohn 1997). They are amino poly alcohols that inhibit the activity of sphingosine N-acetyltransferase that leads to the accumulation of toxic sphingoid bases (Desjardins and Hohn 1997). Various fumonisins have been isolated and characterized (Musser 1996), of which fumonisin BI (FBI)' fumonisin B2 (FB2), and fumonisin B3 (FB3) are the most important analogs found in contaminated maize (Sheph- ard et al. 1996). FBI and FB2 are known to be toxicologically significant. FBI has been known to cause leukoencephalomacia (LEM), a fatal brain disease in horses (Coker 1994; Desjardins and Hohn 1997; Marasas 1996) and pulmonary edema syndrome (PES) in pigs (Marasas 1996). FBI is also toxic to the central nervous system, liver, pancreas, kidneys, and lungs in numerous animal species (Coker 1994). Furthermore, it is a cancer promoter and initiator in rat liver, hepatotoxic to horses, pigs, rats, and vervet monkeys, and phy- totoxic to several plants (Marasas 1995; 1996). Lastly, FBI has been statistically linked to the incidence of human esophageal cancer rates in Transkei, South Africa, and China (Marasas 1996). FBI has been classified as a group 2B carcinogen by the International Agency for Research on Cancer (lARC) which considers it to be possibly carcinogenic to humans (Vainio et al. 1993). There are various reports concerning mycotoxins associated with legume seeds, includ- ing chickpea (Cicer arietinum L.) (Ahmad and Singh 1991), lupine (Lupinus spp. L.) (Abdel-Hafez 1984), pea (Saber et al. 1998), and various types of beans (El-Kady et al. 1991; Saber 1992; Tseng and Tu 1997). There is, however, little literature regarding cowpea seeds and mycotoxins. Seenappa et al. (1983) found cowpea samples to be susceptible to Aspergillus parasitic us Speare infection, and in subsequent aflatoxin contamination. There is no report, however, concerning the presence of fumonisins in cowpea seeds. This paper deals with the detection and quantification of the Fusarium toxins, specifi- cally FBi' FB2, and FB3 in cowpea seeds. Materials and methods Apparatus • Liquid chromatography-Waters 6000 A pump (Waters Corp., Milford, MA 01757, USA) and Rheodyne injector. • Fluorescence detecto-Waters Fluorescence 474 set at 335 nm (excitation) and 440 nm (emission) (Waters Corp., Milford, MA 01757, USA). • Column-Phenomenex Ultracarb 5 ODS (20) (150 x 4.6 mm id.). • Integrator-Borwin Chromatography Software 1.22 (JMBS Developments, France). • Solid-phase extraction (SPE) columns-Chromabond® Strong anion exchange (SAX) cartridges, 6 ml capacity, containing 500 mg SiOH (Machery-Nagel, Duren, D-523l3, Germany). • SPE manifold-12-place vacuum manifold (Lida). 148 Digitized by Google Detection of fumonisin B 1 in cowpea seeds • Reacti-ThermTM Heating module (pierce, Rockford, IL 61105, USA). • Reacti-VapTM Evaporator (pierce, Rockford, IL 61105, USA). Reagents Fumonisin Bl' B2, and B3 standards were obtained from PROMEC, Medical Research Council, Tygerberg, South Africa. All other reagents and solvents were obtained from Merck (Darmstadt, D-6427 1 , Germany). Seed samples Four cultivars (Bechwana White, Glenda, Iron Grey, and Rhino) were used. Approximately 100 g each of cowpea seeds were received from A. Haasbroek from the Agricultural Research Council (ARC), Grain Crops Institute, Potchefstroom, South Africa. The seeds were harvested from experimental fields at the institute and were kept in cold storage (approximately 5°C) for four months prior to the analyses. Determination of seedborne fungi One hundred seeds were randomly chosen from each sample. Prior to plating, 50 seeds from each sample were surface sterilized in 1 % sodium hypochlorite for 1 min. The remaining 50 seeds from each sample were not surface sterilized. The seeds were plated on malt extract agar (MEA) consisting of 15 g malt extract (diastase free), 17 g Bacto agar, 1000 ml distilled H20, and 0.125 g novobiocin. The plates were incubated at 25°C for two to seven days. The fungi were isolated, identified with the aid of various references (Samson et al. 1981; Nelson et al. 1983; Watanabe 1994), and recorded. The Fusarium spp. were identified by Dr J.F. Rheeder ofPROMEC, Medical Research Council, Tyger- berg, South Africa. Sample preparation, extraction, and clean up Cowpea seeds of the four different cultivars were used as samples. The sample extraction and clean up were based on the method described by Sydenham et al. (1992) and were carried out at the Department of Botany, University of Pretoria, Pretoria, South Africa. Approximately 50 g of seeds from each sample were ground using a coffee grinder and 20 g of the ground seeds weighed. After adding 100 ml 70% (v/v) methanol, the ground samples were homogenized for 3 min at 5000 rpm using a hand-held mixer. The samples were then centrifuged for 10 min at 4000 rpm and filtered through Whatman No.4 filter paper. The pH of the filtrate was measured and adjusted with 0.1 M NaOH to between pH 5.8 and 6.5. Clean up and extraction of the filtrate were carried out on strong anion exchange (SAX) cartridges attached to a solid phase extraction (SPE) manifold. Prior to adding 10 ml of the filtrate, the SAX cartridges were conditioned by washing successively with 5 ml 100% methanol followed by 5 ml 70% (v/v) methanol, whilst maintaining a flow rate of 1 mUmin. Cartridges were then washed with 5 III 70% (v/v) methanol and 3 mllOO% methanol. This was followed by elution with 10 1111 % (v/v) methanolic acetic acid at a flow rate of 1 mUmin and the eluate collected in vials. Eluates then were evaporated to dryness in vials on a Reacti -Therm heating module and evaporator at 50°C under a slight stream of nitrogen (AFROX). The collection vials were washed with methanol and the additional methanol was evaporated until a dry residue formed. The dry residues were maintained at 4°C until used for high performance liquid chromatography. 149 Digitized by Google Cowpea integrated pest management High Performance Liquid Chromatography (HPLC) The HPLC analyses were undertaken at PROMEC, Medical Research Council, Tygerberg, South Africa. A derivatization agent, o-phthaldialdehyde (OPA), was added to both the standards and samples prior to HPLC. This is necessary since fumonisins are unable to absorb either UV or visible light and are unable to fluoresce. OPA derives the fluorescent products from the fumonisins (Sydenham et al. 1992). OPA (225 Ill) was added to 25 III of the standard and 10 III was injected into the HPLC, whilst 150 ml OPA was added to 100 III of the sample (which had been redissolved in 200 III CH3CN:HP) and 50 III was injected into the HPLC (Sydenham et al. 1992). Results and discussion The percentage of fungi isolated from each sample was higher in the untreated seeds than in the surface-sterilized seeds (Table 1). The most fungi was isolated from Iron Grey (98% infection) followed by Rhino with 94% infection, Bechwana White with 92% infection, and Glenda with 88% infection. In the surface-sterilized seeds, the most fungi were iso- lated from Rhino (68% infection) followed by Iron Grey (52% infection). Glenda and Bechwana White had low counts of fungal colonies (8 and 4% infection, respectively). The most common fungi found included members of the genera, Aspergillus and Phoma, present in both surface-sterilized and untreated seeds in all four samples. Aspergillus glaucus Link ex. Gray was the predominant species, present in three samples, followed by both Aspergillus flavus Link ex. Fries and Aspergillus niger van Tieghem. Seenappa et al. (1983) reported that all cowpea samples analyzed were susceptible to Aspergillus infection and subsequent aflatoxin production. Table 1. Percentages of fungi isolated from four cultivars of cowpea seeds. Cultivar Bechwana Fungi Glenda White Rhino Iron Grey +a _b + + + Aspergillus {favus 4 10 - 26 2 A. gfaucus 4 8 8 40 68 A. niger 18 - 14 4 2 Chaetomium sp. 2 2 2 2 Cladosporium sp. 18 - 14 2 Dipfodia sp. 4 Fusarium chfamydosporum - 2 F. equiseti 2 2 10 F. graminearum 2 F. sambucinum 2 F. scirpi 6 F. subgfutinans 2 Penicillium sp. 4 - 32 16 Phoma sp. 2 14 4 28 52 36 2 Trichothecium roseum 2 2 2 Other 10 4 4 2 6 Total % infection 8 88 4 92 68 94 52 98 'surface-sterilized seeds, buntreated seeds. 150 Digitized by Google Detection of fumonisin B 1 in cowpea seeds Six Fusarium species were isolated; Fusarium equiseti (Corda) Sacco appeared to be dominant. Four of these Fusarium species were present in the Rhino seeds, two in the Bechwana White sample, and one in the Glenda sample. An interesting occurrence can be noted here. While the most important fumonisin-producing species are F. verticil- lioides and F. proliferatum (Coker 1994; Marasas 1994), neither of these two species were isolated from the samples. However, Esuruoso (1995) recorded F. verticil/ioides on nearly all cowpea samples (81) examined. Other Fusarium species known to produce high concentrations of other mycotoxins but not fumonisins, including F. equiseti, F. sambucinum Fuckel, and F. subglutinans (Wollenw. and Reink.) Nelson, Toussoun, and Marasas were isolated. Further research is required to identify the fungal species present on cowpea seeds responsible for the fumonisin production. Other fungal genera isolated from the samples included Chaetomium, Cladosporium, Penicillium, and Trichothecium spp. Penicillium spp. are also known to produce mycotoxins including ochratoxins (Moss 1996) and citrinin (pitt 1998). From the eight samples analyzed for Fusarium toxins, specifically FBi' FB2, and FB3, FBI was found to be present in all the samples (Table 2), while FB2 and FB3 were not detected. The highest concentration ofFB I was found in the Rhino A cultivar (1002 ng/g), followed by Rhino B (213 ng/g), Bechwana White A (178 ng/g), Glenda A and B (161 ng/g), and Iron Grey A (127 ng/g). Levels below 100 ng/g were detected in Bechwana White B and Iron Grey B. This is the first report of the natural occurrence of FBI on cowpea seeds. Since large quantities of cowpea seeds are produced and consumed in tropical and subtropical countries (Seenappa et al. 1983) and in the light of the various toxicological consequences as a result of fungal mycotoxin contamination, a potential health risk exists for both humans and animals. It is thus essential that care be taken when seeds are stored such that fungal infestation and subsequent mycotoxin production can be effectively controlled and prevented. There are various reports concerning the antifungal activity of essential plant oils (Adegoke and Odesola 1996; National Research Council 1992) which can be used as an alternative approach to controlling and preventing fungal contamination of cowpea seeds. Table 2. Fumonisin concentrations in cowpea seed cultivars. Cultivar Fumonisin concentration (nglg) FBI FB2 FB3 Bechwana White A 178 0 0 Bechwana White B 81 0 0 Glenda A 161 0 0 Glenda B 161 0 0 Iron Grey A 127 0 0 Iron Grey B 99 0 0 Rhino A 1002 0 0 Rhino B 213 0 0 151 Digitized by Google Cowpea integrated pest management References Abdel-Hafez, S.I.1. 1984. Mycoflora of bean, broad bean, lentil, lupine and pea seeds in Saudi Arabia. Mycopathologia 88: 45-49. Ahmad, S.K. and P.L. Singh. 1991. Mycofloral changes and aflatoxin contamination in stored chickpea seeds. Food Additives and Contaminants 8: 723-730. Adegoke, G.O. and B.A. Odesola. 1996. Storage of maize and cowpea and inhibition of microbial agents of biodeterioration using the powder and essential oil of lemon grass (Cymbopogon citratus). International Biodeterioration and Biodegradation 37: 81-84. Coker, N.R.1. 1994. Biodeterioration of grain and the risk of mycotoxins. Pages 27-38 in Grain storage techniques: evolution and trends in developing countries, edited by D.L. Proctor. FAO Agricultural Services Bulletin 109, Rome, Italy. Desjardins,A.E. and T.M. Hohn. 1997. Mycotoxins in plant pathogenesis. Molecular Plant- Microbe Interactions 10: 147-152. El-Kady, LA., S.S.M. El-Maraghy, and AA. Zohri. 1991. Mycotoxin production on different cul- tivars and lines of broad bean (Viciafaba L.) seeds in Egypt. Mycopathologia 113: 165-169. Esuruoso,O.F. 1975. Seed-borne fungi of cowpea (Vigna unguiculata) in Western Nigeria. Nigerian Journal of Plant Produce 2: 87-90. Hitokoko, H., S. Morozumi, T. Wauke, S. Sakai, and H. Kurata. 1981. Fungal contamination and mycotoxin-producing potential of dried beans. Mycopathologia 73: 33-38. Johnson, RM. and W.D. Raymond. 1964. The chemical composition of some tropical food plants II. Pigeon peas and cowpeas. Tropical Science 6: 68-73. Marasas, W.F.O. 1994. Fusarium. Pages 522-530 in Food-borne disease handbook: diseases caused by viruses, parasites and fungi, edited by YH. Hui, J.R. Gorham, KD. Murrell, and D.O. Cliver. Marcel Dekker Inc, New York, USA. Marasas, W.F.O. 1995. Fumonisins: their implication for human and animal health. Natural Toxins 3: 193-198. Marasas, W.F.O. 1996. Fumonisins: history, worldwide occurrence and impact. Page 3 in Fumoni- sins in food, edited by L. Jackson. Plenum Press, New York, USA. Moss,O.M. 1996. Mycotoxins. Mycological Research 100: 513-523. Musser, S.M. 1996. Quantification and identification offumonisins by liquid chromatography/mass spectrometry. Pages 65-74 in Fumonisins in food, edited by L. Jackson. Plenum Press, New York, USA. National Research Council. 1992. Neem: a tree for solving global problems. National Academy Press, Washington DC, USA. Pages 53-55. Nelson, P.E., TA. Tousson, and W.F.O. Marasas. 1983. Fusarium species: an illustrated manual for identification. Pennsylvania State University Press, University Park, Pennsylvania, USA. 89 pp. Pitt, J.1. 1998. Toxigenic aspergillus and penicillium species. In Mycotoxin prevention and control in food grains edited by RL. Semple, A.S. Frio, P.A. Hicks, J. V Lozare. Food and Agriculture Organization of the United Nations (FAO) and the Information Network on Post-Harvest Operations (INPhO). Bangkok, Thailand. http://www.fao.org/inpho/vlibrary/x0036e/ x0036eOO .html Saber, S.M. 1992. Fungal contamination, natural occurrence of mycotoxins and resistance for aflatoxin accumulation of some broad bean CViciafaba L.) cultivars. Journal of Basic Microbiol- ogy 32: 249-258. Saber, M.S., M.B. Aboul-Nasr, and O.M.O. El-Maghraby. 1998. Contamination of pea (Pisum sativum L.) seeds by fungi and mycotoxins. African Journal of Mycology and Biotechnology 6: 53--64. Samson, RA., E.S. Hoekstra, and C.A.N. Van Oorschot. 1981. Introduction to food-borne fungi. Centraalbureau voor Schimelcultures, Netherlands. 190 pp. 152 Digitized by Google Detection of fumonisin B 1 in cowpea seeds Seenappa, M., C.L. Keswani, and T.M. Kundya. 1983. Aspergillus infection and aflatoxin produc- tion in some cowpea (Vigna unguiculata [L.] Walp) lines in Tanzania. Mycopathologia 83: 103-106. Shephard, G.S., P.G. Thiel, S. Stockenstrom, and E.W. Sydenham. 1996. Worldwide survey of fumonisin contamination of com and com-based products. Journal AOAC International 79: 671-687. Singh, B.B., D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. 1997. Pages x-xii in Advances in cowpea research. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (J1RCAS). IlIA, Ibadan, Nige- na. Sydenham, E.W., G.S. Shephard, and P.G. Thiel. 1992. Liquid chromatographic determination of fumonisins Bl' B2, and B3 in foods and feeds. Journal AOAC International. 75: 313-317. Tseng, T.C. and IC. Tu. 1997. Mycoflora and mycotoxins in adzuki and mung beans produced in Ontario, Canada. Microbios Letters 90: 87-95. Tuan, YH. and R.D. Phillips. 1992. Nutritional quality of hard-to-cook and processed cowpea. Journal of Food Science 68: 1371-1374. Ushamalini, C., K. Rajappan, and K. Gangadharan. 1998. Seed-borne mycoflora of cowpea (Vigna unguiculata [L.] Walp.) and their effect on seed germination under different storage conditions. Acta Phytopathologica et Entomologica Hungarica 33: 285-290. Vainio, H., E. Heseltine, and J. Wilburn. 1993. Report on an IARC working group meeting on some naturally occurring substances. International Journal of Cancer 53: 535-537. Van Warmelo, K.T. 1967. The fungus flora of stock feeds in South Africa. Onderstepoort Journal of Veterinary Research 34: 439-450. Van Wyk, B-E. and N. Gericke. 2000. People's plants: a guide to useful plants of southern Africa. Briza Publications, Pretoria, South Africa. 30 pp. Watanabe, T. 1994. Pictorial atlas of soil and seed fungi. CRC Press Inc., Boca Raton, FA, USA. Pages 159-399. 153 Digitized by Google 2.6 Breeding cowpea varieties for resistance to Striga gesnerioides and Alectra vogelii B.B. Singh 1 Abstract Two parasitic flowering plants, Striga gesnerioides (Wild.) Vatke and Alectra vogelii (Benth.), cause substantial yield reduction in cowpea in the dry savannas of sub-Saharan Africa. Alectra is more prevalent in the northern Guinea savanna and southern Sudan savanna of West Africa, as well as in East and southernAfrica whereas Striga is mostly found in West and Central Africa. However, both are fast spreading beyond these limits. Collaborative studies with national and regional programs have revealed the presence of five strains of S. gesnerioides of which strain 1 is presently found in Burkina Faso, strain 2 in Mali, strain 3 in Nigeria and Niger, strain 4 in Benin Republic, and strain 5 in Cameroon. A locallandrace, B 301 from Botswana, confers complete resistance to Striga and Alectra in Burkina Faso, Cameroon, Mali, Niger, and Nigeria. However, it has moderate levels of resistance to the strain from Benin Republic. Other lines such as IT81D-994, IT89KD-288, 58-57, and Gorom local confer complete resistance to strains from Benin Republic and Burkina F aso. Therefore, crosses were made among the selected complementary parents and a number of new varieties have been developed with combined resistance to Alectra as well as all the five strains of Striga. Most ofthese lines also serve as a false host for S. hermonthica reducing its seed bank in the soil when grown as an intercrop or in rotation with cereals. Introduction Cowpea is the most important food legume in West and Central Africa and this region represents over 66% of the 12.5 million ha grown worldwide. It contains about 25% pro- tein and so is a cheap source of protein in the daily diet of rural and urban populations. Its haulms are also an important source of nutritious fodder for the livestock in the dry savannas (Bressani 1985; Singh et al. 1997; Tarawali et al. 1997). However, the average yield of cowpea is very low due to numerous biotic and abiotic constraints. Of these, two parasitic flowering plant species, Striga gesnerioides and A lectra vogelii, cause consider- able yield reduction in cowpea (Emechebe et al. 1991). Striga causes severe damage to cowpeas in the Sudano-Sahelian belt whereas Alectra is more prevalent in the Guinea savanna and Sudan savanna covering most parts of West and Central Africa. Alectra is also widespread in East and southern Africa. The Striga infection in cowpea (Fig. 1) is more devastating in areas with sandy soils, low fertility, and low rainfall. Both parasites are difficult to control because they produce large numbers of seed and up to 75% of the crop damage is done before they emerge from the ground. Therefore, concerted efforts are being made to develop improved cowpea varieties with combined resistance to both parasites. 1. International Institute of Tropical Agriculture, Kano Station, PMB 3112, Kano, Nigeria. 154 Digitized by Google Breeding cowpea varieties for resistance to Striga gesnerioides and Alectra vogelii Sources of resistance and strain variation in cowpea Striga Varietal differences with respect to Striga infection in cowpea were first noticed in 1981 in Burkina Faso, and two lines, Suvita-2 and 58-57, were found to be completely resis- tant (I1TA 1982, 1983). However, the results of subsequent regional trials revealed that these lines were not resistant to Striga in Niger and Nigeria indicating strain variation in cowpea Striga (Aggarwal 1985). Further screening of new lines at several locations in West and Central Africa showed that IT82D-849 (breeding line from I1TA) and B 301 (a landrace from Botswana) were completely resistant to Striga (Fig. 2) in Burkina Faso, Cameroon, Mali, Nigeria, and Niger. B 301 had earlier been identified as being resistant to Alectra in Botswana (Riches 1989) and it was found to be resistant to Alectra in Nigeria also. However, IT82D-849 and Suvita-2 were found to be highly susceptible to Alectra. Subsequently, several other lines were identified which had moderate to high levels of resistance to both Striga andAlectra. These included IT86D-534, IT8ID-994, IT86D-371 , IT84D-666 (Singh and Emechebe 1991), and Tvu 9238, TVu 11788, TVu 12415, TVu 12432, and TVu 12470 (Singh 1994). The Striga seeds germinate and the radicles attach to the roots of resistant and susceptible plants (Fig. 3) but the resistant roots do not permit haustorium development (Fig. 4.). The Striga seedling dies leaving the resistant plants completely healthy and productive. On the other hand, there is a normal development of haustorium on roots of susceptible varieties (Fig. 5) permitting Striga to parasitize cowpea plants and cause up to 100% yield reduction (Fig. 6). Several of these resistant lines were tested at Zakpota in the coastal savanna of Benin Republic where severe Striga infestation had been reported. All the TVu lines as well as IT86D-534, IT86D-371 , and IIT84D-666 were susceptible to the Zakpota strain, whereas B 301 and IT82D-849 showed moderate levels of resistance such that about 10% to 30% plants of these varieties show susceptibility. However, Suvita-2, 58-57 and IT8lD-994 were completely resistant indicating that the Zakpota strain was different from strains from Burkina Faso and Nigeria. Systematic collection of Striga seed from different parts of West and Central Africa and testing against selected cowpea varieties revealed the presence of 5 strains (Lane et al. 1994; Lane et al. 1997). Of these, strain 1 is presently found in Burkina F aso, strain 2 in Mali, strain 3 in Nigeria and Niger, strain 4 in Benin Republic, and strain 5 in Cameroon. The host differentials for different strains are presented in Table 1. As evident from Table 1, B 301 and IT82D-849 are resistant to strains from Burkina Faso, Camer- oon, Mali, and Nigeria, but moderately resistant to the Benin strain, which causes 10% to 30% susceptibility in these lines. IT8lD-994 is resistant to all the strains except for the Nigerian strain. Suvita-2 is only resistant to Burkina Faso and Benin Republic strains, and 58-57 is resistant to strains from Benin, Burkina Faso, and Cameroon, but susceptible to strains from Mali and Nigeria. Thus, a combination of B 301 or IT82D-849 on one hand and IT8lD-994 or Suvita-2 and 58-57 on the other will provide resistance to all the existing strains of Striga. These data also indicate that lines resistant to the Nigerian strain (strain 3) confer resistance to all the strains except the Zakpota strain. Therefore, testing of new lines in Nigeria and Benin Republic will be adequate for identification of lines with combined resistance to all five strains. 155 Digitized by Google Cowpea integrated pest management ~ . Figure 1. Cowpea Striga in the field. .. . .. -. " .: ~ .:' -:;"~~~\' ~,' , < - • . '.. , .. Figure 3. Striga attaching to cowpea root. Figure 5. Striga parasitizing susceptible root. 156 Figure 2. Striga resistant and susceptible plants . Figure 4. Resistant cowpea root kills Striga. Figure 6. Striga resistant and susceptible lines in the field. Digitized by Google Breeding cowpea varieties far resistance ta Striga gesnerioides and Alectra vogelii Table 1. Host differentials for different strains of cowpea Striga gesnerioides. Cowpea Reaction to s. gesneriaides strain+ variety 2 3 4 5 Blackeye 5 5 5 5 5 TVx 3236 5 5 5 5 5 58-57 R 5 5 R R Suvita-2 R 5 5 R R IT81 D-994 R R 5 R 5 B 301 R R R MR R IT82D-849 R R R MR R +Strain 1 occurs in Burkina Faso, 2 in Mali, 3 in Nigeria/Niger, 4 in Benin Republic, and 5 in Cameroon. R = 100% plants resistant. S = 100% plants susceptible. Effect of Striga infection on growth characters in susceptible and resistant cowpea varieties Cowpea varieties with complete resistance to Striga stimulate germination and permit attachment of Striga radic1es to their roots but the haustorium development is inhibited. The question has been raised whether the initial attachment of germinating Striga to resistant cowpea roots causes any shock to the plants and reduces plant growth even though further development of Striga is checked. Therefore, a set of 23 resistant and nine susceptible cowpea varieties were planted in pots infested with Striga gesneriaides seeds as well as in pots without Striga seeds and notes were taken on various parameters. As expected, the resistant varieties did not show any Striga emergence in both infested and noninfested pots and the susceptible varieties showed severe infestation (Table 2). The data on plant height, and shoot and root dry -matter of resistant varieties in infested pots did not differ significantly from the noninfested pots whereas the susceptible varieties suffered signifi- cant reduction in plant height as well as in dry matter in the infested pots. These results suggest that Striga attachment in resistant plants does not affect plant growth. Genetics of resistance to Striga gesnerioides and Alectra vogelii Using the sources of resistance mentioned in the earlier section, systematic genetic studies were conducted to elucidate the nature of inheritance of resistance to Striga and Alectra. Singh and Emechebe (1990) reported that a single dominant gene, designated Rsg1 conditions resistance to S. gesneriaides in cowpea variety B30l. Singh et al. (1993) found that duplicate dominant genes, designatedRav 1 and Rav 2 (resistant to Alectra vage Iii) control resistance to Alectra in cowpea variety B 301. Atokple et al. (1993) demonstrated that the genes conditioning the resistance to Striga and Alectra in B30 1 are neither allelic nor linked. Atokple et al. (1995) reported the results of extensive allelism tests among cowpea lines resistant to Striga and Alectra. This work revealed that different genes are responsible for the Striga resistance exhibited by B 301, IT82D-849, and Suvita-2. Atokple et al. (1995) also reported that the single dominant gene conditioningAlectra resistance in IT8lD-994 is not one of the two duplicate dominant genes conditioning resistance in B 301. They proposed the symbols RsgI' Rsg2' and Rsg3 for the genes conditioning resistance to 157 Digitized by Google Cowpea integrated pest management Table 2. Effect of Striga infection on growth characters in susceptible and resistant cowpea varieties. Growth character/plant No. of Striga plants emerged No. of Striga plants attached Plant height at flow. (cm ) Plant height at mat. (cm) Shoot dry matter at mat. g Root dry matter at mat. g Susceptible varieties Infected Not infected 8** 13** 29* 36** 5** 1.8 NS o o 37 61 7.7 2.3 Resistant varieties Infected Not infected o NS o NS 49 NS 59 NS 8 NS 2.5 NS o o 50 50 8.2 2.9 ** = Significantly different (0.01); NS = not significantly different, flow. = flowering, mat. = maturity. Striga gesnerioides in B 301, IT82D-849, and Suvita-2, respectively. They also proposed the symbols Rav] and Rav2 for the genes conditioning resistance to Alectra vogelii in B 301, and the symbol Rav Jor the gene conditioning resistance to Alectra in IT8lD994. The fact that the resistance in B 301 is due to a single dominant gene indicated that this gene confers resistance to four strains and the genes for resistance in Suvita-2 and IT8lD-994 confer resistance to two to three strains including the Zakpota strain. Therefore, B 301 derived Striga resistant lines and Suvita-2 or other lines showing resistance to Zakpota strain can be used as complementary parents for breeding cowpea varieties resistant to all the 5 strains. Recently Ouedraogo et al. (2001) have confirmed monogenic inheritance of Striga resistance in Suvita-2 cowpea and they have also identified AFLP (amplified fragment length polymorphism) markers tightly linked to genes conditioning resistance to Striga. This will permit marker-assisted selection for Striga resistance and the eventual cloning and characterization of the genes conferring resistance to Striga in cowpea. Breeding cowpea for resistance to Striga and Alectra A systematic breeding program for resistance to Striga and Alectra using B 301 as a resistance source, was undertaken in 1987. This line was crossed to a susceptible variety, IT84S-2246-4, which is otherwise a high yielding variety with resistance to aphid, bru- chid, and several diseases. The F 1 was backcrossed to IT84S-2246-4. From the resistant BC1-F 1 plants, F 2' F 3' F 4' F 5 and F 6 progenies were developed and selected under suitable disease, insect, and StrigalAlectra pressures. This led to the selection of a number of F 6 breeding lines, which are very similar to IT84S-2246 and have combined resistance to aphid, bruchid, thrips, Striga,A lectra, and several diseases. These were evaluated for yield and other characters in a replicated trial in 1991 and promising lines distributed to various national programs in Africa. Based on their performance in Striga infested fields, IT89KD- 374-57 (Sangaraka) and IT89KD-245 (Korobalen) have been released in Mali; IT90K-76 and IT90K -82-2 have been released for general cultivation in Nigeria, and IT90K-59-2 in South Africa. The last four varieties have combined resistance to aphid, bruchid, thrips, Striga, andAlectra. These have been used as parents in the breeding program and a large number of Striga-resistant cowpea varieties have been developed. Performance of a few promising Striga-resistant varieties in the Striga-infested area of the Sahelian region is presented in Table 3. The Striga resistant varieties yielded about 50 to 100% more than the susceptible checks. 158 Digitized by Google Breeding cowpea varieties for resistance to Striga gesnerioides and Alectra vogelii Table 3. Perfomance of Striga-resistant cowpea varieties in the Sahel (Toumnia, Niger). Yield (kg/ha) No. of Strigalplot Cowpea variety Grain Fodder (6 m2) Early maturing varieties IT97K-499-8 1467 891 0 IT97K-499-39 1274 891 0 IT97K-497-2 1166 807 0 IT97-819-180 1105 1420 0 Dan lIa (susceptible) 529 1280 26 SED 199 214 2.3 Early semideterminate varieties IT97K-819-45 1319 1447 0 IT98K-205-29 1150 1002 0 IT94K-437-1 1092 1030 0 IT97K-499-38 1029 1948 1 IAR 48 (susceptible) 726 1058 23 Dan lIa (susceptible) 546 1336 21 SED 160 183 5.0 Breeding for resistance to multiple strains of S. gesnerioides After the discovery of the Zakpota strain of Striga in Benin Republic, a large number of crosses were made between IT8lD-994 and 58-57 with B 301 derived lines like IT90K-59 and IT90K-76 which are similar to B 301 with respect to Striga and Alectra resistance, but with higher yield and better seed quality. The segregating F 2 populations were first screened at Kano and then part of the seeds from resistant F 3 progenies were tested at Zakpota. The remnant seeds of resistant F 3 progenies at Zakpota were then planted at Kano and F 4 plants were selected. This was continued until the F 6 generation. This procedure had to be adopted because seeds of the Benin strain of Striga cannot be brought to Kano. The advanced breeding lines derived from this program were tested at several locations and selected lines distributed to national programs. The most promising lines were IT93KZ-4-3 -1-7, IT93KZ-8-2-2-3 -6, and IT93KZ-4-5-6-1-5 with over two t/ha grain yield with two sprays of insecticide and 100 kg/ha fertilizer (NPK 15-15-15). Subsequently, the breeding procedure was simplified to minimize record keeping and save costs. The crosses are made between complementary parents and the segregation populations screened for resistance to Striga with artificial infestation at Kano (Nigeria) and at Babura (Nigeria) with natural infestation. The lines are also subjected to disease and insect pressure while advancing the generations. The selected F 6 lines are then tested at several locations including Zakpota where selection for resistance to the Zakpota strain is made. These lines are also tested at Samaru (Nigeria) for resistance to Alectra. This strategy has been very effective and a number of new breeding lines have been selected with combined resistance to all the strains of Striga and Alectra, as well as resistance to aphid, bruchid, thrips, viruses, and several diseases. The yield performance and level of resistance of the newly developed breeding lines at Striga-free (Minjibir) and Striga- infested (Babura) locations are indicated in Table 4. The results indicated that in a good environment (Minjibir), the yield potential of most of the lines is between 1500 and 2500 kg/ ha. However, significant varietal differences were observed at a poor environment 159 Digitized by Google Cowpea integrated pest management Table 4. Performance of selected improved cowpea varieties at Minjibir (less Striga) and Babura (severe Striga). Yield (kg/ha) Grain Fodder Strigalplot (6m2) Variety Min. Bab. Min. Bab. Min. Bab. Zak. IT97K-400-3 2761 207 1420 459 0 112 19 IT97K-351-5 2559 0 2004 109 2 60 10 IT97K-825-8 2362 855 2881 752 0 0 0 IT97K-817-178 2311 642 2422 551 0 0 0 IT97K-825-21 2310 931 3006 1461 0 0 0 IT97K-499-35 2297 805 1587 676 0 3 1 IT90K-277-2 2579 99 1378 134 1 122 10 IT86D-719 2171 184 1879 159 8 167 2 IT97K-826-86 855 1653 2088 626 0 2 0 IT97K-819-154 1787 1552 1253 793 0 0 0 IT97K-819-132 1519 1284 1420 1670 0 5 2 Dan lIa 1415 223 1975 710 1 57 2 SED 365 318 103 385 1.6 3 2 Min. = Minjibir, Bab. = Babura, Zak. = Zakpota. (Babura) where soils are sandy, less fertile, and heavily infested with Striga. The Striga- resistant lines yielded between 642 and 1653 kg/ha but the Striga susceptible lines yielded from nothing to 223 kg/ha. The difference in the performance of Striga-resistant lines at Minjibir and Babura is due to low fertility at Babura and not due to Striga. It is interest- ing to note that a few Striga resistant lines such as IT97K-826-86, IT97K-8l9-l54, and IT97K-8l9-l32 yielded between 1384 and 1653 kg/ha grain with 2 sprays of insecticide even at Babura indicating their adaptability to poor soils and their ability to make efficient use of limited soil nutrients. A number of these lines have been multiplied and distributed to various national programs in cowpea international trials. A Striga-resistant cowpea variety, IT 97K-499-38, was tested at eight farmers' fields in Benin Republic along with the respective local varieties grown by the farmers. The resistant variety, IT97K -4 99-38, performed as well as or better than local varieties at Striga- free locations but was much superior at Striga-infested locations (Table 5). The number of Striga in plots of local varieties (48m2) ranged from 1000-2000 and their grain yield ranged from 5 kg/ha to 220kg/ha whereas the number of Striga in the resistant variety ranged from 23-456 and the grain yield from 457-678 kg/ha. Thefactthat IT97K-499-38 showed some level of Striga infestation indicates that it is not immune to the Striga strain present in Benin Republic. Breeding for combined resistance to Striga and Alectra Through planned crosses among complementary parents and screening of the derived breeding lines at Minjibir, Babura, and Zakpota for Striga and at Samaru for Alectra, a number of cowpea varieties have been developed which have combined resistance to all the strains of Striga as well as Alectra. Based on their resistance and yield performance in different trials, IT94K-437-l, IT94K-440-3, IT96D-748, IT97K-499-39, and IT97K- 819-154 appear to be very promising (Table 6). 160 Digitized by Google Breeding cowpea varieties for resistance to Striga gesnerioides and Alectra vogelii Table 5. Performance of Striga-resistant varieties in the coastal savanna under low fertil- ity without insecticide spray. Cowpea variety Location in IT97K-499-38 Local varie!}' Benin Republic No. of Striga' Yield (kg/ha) No. of Striga Yield (kg/ha) Mlinkpin 0 787 47 587 Adjoko 0 662 29 650 Maikpin 0 775 23 500 Oukombe 0 587 1461 202 Kodota 57 312 589 50 Aligodon 70 609 1053 300 Some 437 300 1526 262 Zakpota 360 225 2724 5 , No. of Striga in 40 m' plots (single replicate). Table 6. Reaction of improved cowpea breeding lines to Striga gesnerioides and A/ectra vogelii. Emerged Strigalplott Emerged Alectra/plot Breeding line Samaru Babura Zakpota Samaru IT93K-596 0.0 1.8 4.5 0.0 IT93 K-693-2 0.0 0.0 0.0 0.0 IT94K-437-1 0.0 0.0 0.0 0.0 IT94-440-3 2.0 0.0 3.0 0.0 IT95K-1090-12 0.0 0.3 0.0 0.0 IT95K-1091-3 0.0 1.7 0.0 0.0 IT96D-748 0.0 0.0 0.0 0.0 IT97K-499-39 0.0 0.0 1.0 0.0 IT97K-819-154 0.0 0.0 0.0 0.0 Tvx 3236 (check) 12 10 35 28 tplot = 6 m' Screening cowpea, sorghum, and millet varieties as false hosts for Striga Spp. In view of the fact that cowpea is mostly planted as an intercrop with pearl millet and sorghum, it would be ideal to select cowpea varieties that can stimulate suicidal germina- tion of S. hermonthica on one hand, and millet and sorghum varieties that can stimulate suicidal germination of S. gesnerioides on the other hand, thereby reducing the seed bank of both types of Striga. Therefore, a range of cowpea, millet, and sorghum varieties were tested from 1993 to 1995 for their ability to cause suicidal germination of Striga spp. Most of the cowpea varieties were able to cause from 65% to 80% suicidal germination of S. hermonthica (Table 7). Of these, IT90K-76, IT8ID-994, and Suvita-2 are resistant to several strains of cowpea Striga. From a total of 55 sorghum varieties tested, only 6 could stimulate the germination of Striga gesnerioides and of these, varieties Yalan and BES were the best with about 60% germination which was close to that of cowpea variety, TVx 3236. Of the 50 millet varieties tested, none was able to cause significant germina- tion of S. gesnerioides. 161 Digitized by Google Cowpea integrated pest management Table 7. Percentage suicidal germination of Striga hermonthica by cowpea varieties and Striga gesnerioides by sorghum and millet varieties. Host crop/variety Cowpea IT90K-277-2 IT81D-994 Suvita-2 IAR48 IAR 1696 Sorghum Yalang BES ICSV 1007 47 others Pearl millet ICMV-1589201 ICMV-1 5 94110 48 other lines Suicidal germination (%) s. hermonthica 82-3 80.9 80.9 71.4 65.4 S. gesnerioides 63.5 60.0 48.0 0.0 S. gesnerioides 3.3 0.0 0.0 These results indicate that most of the cowpea varieties can cause suicidal germination of S. hermonthica. The new Striga-resistant cowpea breeding lines have been tested and they cause similar germination indicating that these varieties would be ideal for inter- cropping or as a rotation crop with millet and sorghum. Although some sorghum variet- ies have shown ability to cause suicidal germination of S. gesnerioides, it is desirable to screen more sorghum and millet varieties to identify lines that can cause higher levels of suicidal germination of S. gesnerioides and use these lines in the breeding program. Thus, there is a need for cowpea breeders and millet and sorghum breeders to work together to identify complementary combinations of millet-cowpea and sorghum-cowpea intercrops to minimize Striga infestation on both crops in the dry savannas. Conclusion Cowpea suffers considerable damage due to Striga and Alectra and the yield reduction can be up to 100% in severe cases. Current annual losses due to these parasitic plants are estimated to be over US$ 200 million in West and Central Africa where over 8 million ha of cowpea are grown mostly by smallholder farmers who cannot afford to control these parasites by chemical means. Development of cowpea varieties with combined resistance to both parasites is the cheapest and best method of reducing the losses due to these parasitic weeds. A great deal of progress has been made and a number of improved StrigaiAlectra resistant cowpea varieties have been developed, which are fast becoming popular with the farmers. References Aggarwal, VD. 1985. Cowpea-Striga gesnerioides research. Pages 335-340 in Cowpea research, production, and utilization, edited by S.R. Singh and K.O. Rachie. John Wiley and Sons, Chichester, UK. Atokple, ID.K., B.B. Singh, and A.M. Emechebe.1993. Independent inheritance of Striga and Alectra resistance in cowpea genotype B301. Crop Science 33: 714-715. 162 Digitized by Google Breeding cowpea varieties for resistance to Striga gesnerioides and Alectra vogelii Atokple, I.D.K., B.B. Singh, and A.M. Emechebe. 1995. Genetics of resistance to Striga and Alectra in cowpea. Journal of Heredity 86: 45-49. Bressani, R. 1985. Nutritive value of cowpea. Pages 353-360 in Cowpea research, production, and utilization, edited by S.R Singh and K.O. Rachie. John Wiley and Sons, Chichester, UK. Emechebe, A.M., B.B. Singh, O.I. Le1eji, I.D.K. Atokple, and J.K. Adu. 1991. Cowpea Striga problems and research in Nigeria. Pages 18-28 in Combating Striga in Africa, edited by S.K. Kim. I1TA, Ibadan, Nigeria. I1TA (International Institute of Tropical Agriculture). 1982. Screening for resistance to Striga gesnerioides. Page 148 in Annual Report 1981. I1TA, Ibadan, Nigeria. I1TA (International Institute of Tropical Agriculture). 1983. Striga gesnerioides resistance. Pages 69-70 in Annual Report 1982. I1TA, Ibadan, Nigeria. Lane, lA., T.H.M. Moore, D. V Child, K.E. Cardwell, B.B. Singh, and lA. Bailey. 1994. Virulence characteristics of a new race of the parasitic angiosperm Striga gesnerioides from southern Benin on cowpea. Euphytica 72: 183-188. Lane, lA., T.H.M. Moore, D.V Child, and JA. Bailey. 1997. Variation in virulence of Striga gesnerioides on cowpea: new sources of crop resistance. Pages 225-230 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublica- tion ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Ouedraogo, J.T., V Maheshwari, D.K. Berner, C.A. St-Pierre, and P. Timko. 2001. Identification of AFLP markers linked to resistance of cowpea to parasitism by Striga gesnerioides. Theo- retical and Applied Genetics 102: 1029-1036. Riches, C.R. 1989. The biology and control of Alectra vogelii in Botswana. PhD Thesis, Univer- sity of Reading, UK. Singh, B.B. 1994. Collection and utilization of germplasm of cowpea resistant to Striga and Alectra. Pages 135-144 in Plant genetic resource management in the tropics. llRCAS Inter- national Symposium 2. Japan International Research Center for Agricultural Sciences, Tsukuba, Japan. Singh, B.B. and A.M. Emechebe. 1990. Inheritance of Striga gesnerioides: resistance in cowpea genotype B 301. Crop Science 30: 879-881. Singh, B.B. and A.M. Emechebe. 1991. Breeding for resistance to Striga and Alectra in cowpea. Pages 303-305 in Proceedings of 5th International Symposium on Parasitic Weeds, edited by lK. Ransom, L.J. Musse1man,AD. Worsham, and C. Parker, 24-30 June 1991, Nairobi, Kenya. The International Maize and Wheat Improvement Center (CIMMYT), Mexico, D.F. Mexico. Singh, B.B.,A.M. Emechebe, and I.D.K. Atokple. 1993. Inheritance ofAlectra vogelii resistance in cowpea, genotype B 301. Crop Science 33: 70-72. Singh, B.B., O.L. Chamblis, and B. Sharma. 1997. Recent advances in cowpea breeding. Pages 30-49 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (llRCAS). I1TA, Ibadan, Nige- na. Tarawali, S.A., B.B. Singh, M. Peters, and S.F. Blade. 1997. Cowpea haulms as fodder. Pages 313-325 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashi- ell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (llRCAS). I1TA, Ibadan, Nigeria. 163 Digitized by Google Digitized by Google Section III Biotechnology for cowpea Digitized by Google Digitized by Google 3.1 Isolation, sequencing, and mapping of resistance gene analogs from cowpea (Vigna unguiculata L.) B.S. Gowda" J.L. Miller2, 5.5. Rubin" D.R. Sharma3, and M.P. Timko' Abstract Cowpea (f/igna unguiculata L.) is a staple crop of significant economic importance worldwide and for many people in emerging areas ofthe globe, it is a major source of protein necessary for proper human nutrition. Degenerate oligonucleotide prim- ers designed to recognize conserved regions within the nucleotide binding site (NBS) ofknownNBS-LRR-type resistance genes from various plant species were used in PCR amplification reactions to identify resistance gene analogs (RGAs) from the Striga gesnerioides-resistant cowpea cultivar Suvita-2. The PCR reaction products consisted of a group of related fragments approximately 500 bp in length which migrated as a single band during agarose gel electrophoresis. The nucleotide sequences of 50 different fragments were determined and their predicted protein sequences compared to each other and to the proteins encoded by known resistance genes and RGAs from other plant species. A total of eight different classes ofRGAs were found in cowpea. Gel blot analysis revealed that each class recognized a dif- ferent subset of genes in the cowpea genome. Several ofthe RGAs were associated with restriction fragment length polymorphisms, which allowed them to be placed on the cowpea genomic map. The potential for using these sequences to isolate their corresponding genes and the subsequent direct manipulation of disease and pest resistance through genetic engineering is discussed. Introduction Cowpea (Vigna unguiculata L.) is a staple food crop of significant economic importance worldwide. In the semiarid and humid tropical regions of Africa, cowpea is a major source of protein and of considerable importance for human nutrition. It is estimated that cowpea is now cultivated on at least 12.5 million hectares, with an annual production of over 3 million tonnes worldwide (Singh et al. 1997). While cowpea is grown on some 80000 hectares in the USA (F ery 1990, Ehlers and Hall 1997), Central and West Africa account for more than half of the cultivated area, followed by South America, Asia, East and South Africa (Singh et al. 1997). Cowpea production is limited by numerous insects, microbial and fungal diseases, and other pests including the parasitic angiosperms Striga gesnerioides and A lectra volgetii (Bashir and Haptom 1996; Singh and Emechebe 1997). Because of its widespread use, numerous initiatives have been undertaken to improve various agronomic and nutritional traits of cowpea. These initiatives include selective 1. Department of Biology, University of Virginia, Charlottesville, Virginia 22903 USA. 2. Cell and Molecular Biology Program, Yale University, New Naven, CT 06511. 3. Department of Biotechnology, Dr Y.S. Parmar University of Horticulture and Forestry, Nauni, Solan, 1732309 Himachal Pradesh, India. 167 Digitized by Google Biotechnology for cowpea breeding programs aimed at screening wild and cultivated germplasm for sources of disease and pest resistance, improved plant cell culture and cell transformation methods, and gene isolation and characterization analysis for the direct manipulation of the cowpea genome through genetic engineering. Genes conferring resistance to the major classes of plant pathogens, including bacteria, virus, fungi, and nematodes, have been isolated from a variety of plant species, includ- ing almost all of the agronomically important grasses and legumes (Baker et al. 1997; Gebhardt 1997; Hammond-Kosack and Jones 1997). The products of the resistance (R) genes have been suggested to act as receptors that specifically bind ligands encoded by the corresponding pathogen avirulence factors in a gene-for-gene recognition process (Baker et al. 1997; Hammond-Kosack and Jones 1997). The R-gene productlavirulence factor complex is thought to initiate a series of signaling cascades within the cell leading to disease resistance. Among the downstream cellular events that characterize the resistant state are rapid oxidative bursts, cell wall strengthening, the induction of defense gene expression, and rapid cell death at the site of infection (Morel and Dangl 1997). From comparisons of the predicted protein sequences of cloned disease and pest resistance genes from various plants, researchers were able to identify common motifs in R-gene products from plants of diverse evolutionary origin working against a broad array of pathogens. The structural motifs present in the various R-gene products, when compared to other prokaryotic and eukaryotic proteins of known function, also suggest possible roles and cellular locations for the R-gene encoded proteins. Based on shared molecular features, the products of R -genes from various plants have now been grouped into several maj or classes (parker and Coleman 1997; Hammond -Kosack and Jones 1997; Van der Beizen and Jones 1998; Pan et al. 2000). The majority of plant-resistance genes encode cytoplasmic receptor-like proteins that contain a leucine-rich repeat (LRR) domain and a nucleotide triphosphate binding site (NBS). Included in this class ofR -genes are the N gene from tobacco conferring resistance to tobacco mosaic virus (TMV)(Whitham et al. 1994), Prj, I2CI, and Mi from tomato (Milligan et al. 1998; Ori et al. 1997; Salmeron et al. 1996), RPMl, RPS2, RPP5, RPS5, RP P l, andRP P8 from Arabidopsis (Bent et al. 1994; Botella et al. 1997; Grant et al. 1995; McDowell et al. 1998; Mindrinos et al. 1994; Parker et al. 1997; Warren et al. 1998), the rust resistance genes M and L6 from flax (Anderson et al. 1997; Lawrence et al. 1995), RGC2 from lettuce (Meyers et al. 1998), Xal from rice (Yoshimura et al. 1998), and the nematode resistance locus of wheat, Cre3 (Lagudah et al. 1997). Some members of this group contain domains near their amino terminus which have significant similarity to the Drosophila Toll or human interleukin receptor-like (TIR) region (Hammond-Kosack and Jones 1997; Whitham et al. 1994). In others, the amino-terminus of the protein contains coiled-coil (CC) motifs (pan et al. 2000). Interestingly, TIR-NBS-LRR type resistance proteins appear to be found only in dicotyledonous plants, whereas CC-NBS-LRR type resistance genes are found in both monocots and dicots (Meyers et al. 1999; Pan et al. 2000). The second subfamily includes the tomato Cf-2, Cf-4, Cf-5, and Cf-9 genes which confer resistance to different races of the fungus Cladosporium fulvum. (Jones et al. 1994; Dixon et al. 1996, 1998; Thomas et al. 1997) and the sugar beet nematode resis- tance gene HSpro-l (Cai et al. 1997; Hammond-Kosack and Jones 1997). These R-genes encode putative transmembrane molecules with extracellular LRR domains. The riceXa2l 168 Digitized by Google Resistance gene analogs from cowpea gene encodes a third class of R proteins (Song et al. 1995), which have a transmembrane segment, an extracellular LRR domain and an intracellular serine-threonine kinase. The bacterial blight resistance gene Xal from rice also contains both the NBS and LRR but differs significantly from theXa21 protein (Yoshimura et al. 1998). Pto, which confers resistance to the bacterial pathogen P. syringae pv. tomato (Martin et al. 1993) constitutes yet another class ofR gene. Pto contains a serine-threonine kinase domain but lacks both the LRR and NBS and requires Prjfor function. As a result of the high degree of sequence conservation among R-genes encoding NBS- LRR type proteins, various investigators have designed degenerate oligonucleotides for use in polymerase chain reaction (PCR) amplification reactions to clone resistance genes and resistance gene analogs (RGAs) from the genomes of diverse plants species, including soybean (Kanazin et al. 1996; Yu et al. 1996), common bean (Rivkin et al. 1999),Arabi- dopsis (Speulman et al. 1998, Aarts et al. 1998), and numerous other dicot and monocots (see Meyers et al. 1999; Pan et al. 2000). The term RGA is used throughout the text to denote cloned R-gene sequences for which no function has yet been assigned in the plant species. In some cases, it has been possible to map RGAs within a plant genome and show that they are linked to a known disease or pest resistance locus (Yu et al. 1996; Chen et al. 1998; Collins et al. 1998; Seah et al. 1998; Shen et al. 1998; Mago et al. 1999; Tada 1999). RGAs have not only been found that are linked to known single dominant R-loci, but also to quantitative trait loci (QTL) (pflieger et al. 1999). Yields of edible cowpea seed are severely reduced by infection of the roots by the para- sitic angiosperm Striga gesnerioides (Aggarwal and Ouedraogo 1989; Aggarwal 1991). Attempts to control the parasite by altering cultural practices have not been effective and the use of chemical treatments have been economically impractical for most local farmers (Aggarwal and Ouedraogo 1989; Muleba et al. 1996, 1997; Singh and Emechebe 1997). The identification of local varieties with natural resistance and their incorporation into breeding programs has been the most successful strategy used to date for controlling the parasite (Singh and Emechebe 1997). The identification and cloning of resistance genes to this and other disease pathogens would contribute significantly to the future improvement of cowpea germplasm. Given the success of these previous investigators, we have used primers based on the conserved motifs of previously isolated disease resistance genes to amplify similar regions from the Striga gesnerioides-resistant variety Suvita-2 (Atokple et al. 1995; Toure et al. 1997, 1998). The different RGAs were subsequently cloned, sequenced, and several of them mapped onto the cowpea genetic map. Materials and methods Plant growth and materials Seeds of the cowpea accession Suvita-2 were obtained from the USDA-ARS, Plant Genetic Resources Conservation Unit, University of Georgia (Griffin, Georgia). Breed- ing lines IT84S-2049, 524B, and 96 recombinant inbred lines (RILs) derived from the cross between these two lines were from Paul Gepts (University of California, Davis, California). IT84S-2049 is a breeding line developed at the International Institute for Tropical Agriculture (I1TA) Ibadan, Nigeria, and is reported to have multiple disease and pest resistance (Menendez et al. 1997). Cultivar 524B is a California black-eye type developed from a cross between cultivars CB5 and CB3, which encompass the genetic variability in cowpea in California. Plants used for DNA isolation were grown in pots 169 Digitized by Google Biotechnology for cowpea in the greenhouse, the leaves were harvested, frozen immediately in liquid nitrogen, and stored at -70°C until use. DNA isolation, peR amplification, and RGA cloning Genomic DNA was isolated following the protocol ofVaradarajan and Prakash (1991). Conserved regions within the NBS and hydrophobic regions of the tobacco N gene, RPS2 of Arabidopsis, and L6 of flax were used to design degenerate oligonucleotide primers for amplification of RGAs from cowpea (Kanazin et al. 1996). Two primers were made for these studies, Primer l-5'-GGIGGIGTIGGIAAIACIAC-3' and Primer 2-5' A(A/G)IGCTA(A/G)IGGIA(A/G)ICC-3' (purchased from Gibco-BRL, Life Sci- ences). PCR amplification reactions were carried out with 200 ng of Suvita-2 genomic DNA in 100 III reactions containing 10 mM Tris-HCl, pH 8.3, 1.5 mM MgC12, 50 mM KCl, lmM of each primer, 100 11M dNTPs, and 2.5 units of Taq DNA polymerase. The initial step of the amplification reaction was denaturation at 94°C for 5 min, followed by 35 cycles of 94 °C for 1 minute, 45°C for 1 minute, 72 °C for 2 minutes, and a final extension at 72 °C for 10 minutes. The PCR products were resolved on the 0.8% aga- rose gel, the DNA bands were purified using GeneClean (BlOlOl, Vista, California), and ligated into Bluescript SK (Stratagene, La Jolla, California) which was linearized by digestion with EcoRV and T-tailed (Marchuk et al. 1991) before use. The resulting plasmids were transformed into E. coli DH5u. A total of 100 colonies were randomly selected, liquid cultures grown from each, and plasmid DNA isolated by the alkaline lysis method (Sambrook et al. 1989). Nucleotide sequencing and analysis Nucleotide sequencing was carried out manually using the Sequenase Version 2.0 DNA sequencing kit according to the manufacturer's protocol (United States Biochemical, Cleveland, Ohio) or using the BigDye fluorescence labeling method and anABI Prizm 310 automated sequencer (pEApplied Biosystems, Forest City, California). The open reading frames for the nucleic acid sequences were obtained by using DNASTAR program (DNA- STAR Inc., Madison, Wisconsin) and the nucleotide and predicted amino acid sequences of the various cDNAs were analyzed using BLAST and BLASTX sequence analysis programs (Altschul et al. 1990, 1997; Gish and States 1993). Protein sequence align- ments were carried out using the PILEUP program (Genetics Computer Group Sequence Analysis Package, Version 9.0, Madison, Wisconsin) and the various R-gene sequences available in the NCB! (National Center for Biotechnology Information, Bethesda, MD) nucleotide and protein sequence database. Construction of dendrograms was done using PaupSearch and PaupDisplay programs (Genetics Computer Group Sequence Analysis Package, Version 9.0, Madison, Wisconsin). Manual adjustment of the sequence align- ments was carried out as necessary. Gel blot analysis of genomic DNA Gel blot analysis of genomic DNA was carried out as described by Gowda et al. (1996, 1999) using 10llg aliquots of genomic DNA digested with EcoRI, EcoRV or HindIII. Restriction digestion products were separated on 0.8% agarose gels in TAE buffer and then transferred to NytranPlus membranes (Schliecher and Schuell, Keene, New Hampshire) by alkaline capillary transfer (Sambrook et al. 1989). The blots were hybridized with [u32p]-dCTP-labeled hybridization probes prepared from the inserts of the various RGA 170 Digitized by Google Resistance gene analogs from cowpea clones (e.g., 432, 434, 436, 438, 445, 468, and 490). Pre-hybridization, hybridization, and washing of the membranes were done according to Gowda et al. (1996). Segregation and linkage analysis In order to place the polymorphisms recognized on this study on the existing map of the cowpea genome, segregation of polymorphic fragments was carried out with 96 RILs derived from a cross between IT84S-2049 and 524B (Menendez et al. 1997). Segregation of individual markers was analysed by chi -square test for goodness of fit to a 1 : 2 : 1 or 1 : 3 ratio. Linkage analysis was performed using MAPMAKER 3.0 program (Lander et al. 1987). The "group" command was used to determine linkage groups, pair-wise comparisons, and group markers. An LOD score of 3.0 or above and a maximum recom- bination frequency of 30% were specified. To determine the most likely order within a linkage group, the "compare" command was used and the best order was accepted based on a log-likelihood difference of two or more. The Kosambi mapping function (Kosambi 1944) was used to convert recombination frequency into map distances in centimorgans (cM). Results Cloning and nucleotide sequence analysis of cowpea RGAs PCR amplification of cowpea DNA prepared from the Striga-resistant line Suvita-2 using degenerate oligonucleotide primers recognizing conserved sequences corresponding to the NBS and hydrophobic domains of NBS-LRR type R-genes yielded a heterogeneous mixture of fragments migrating on agarose gels as a single band approximately 500 bp in length. The PCR products were recovered from the gel, cloned into pBluescript plasmids, and the nucleotide sequence of 50 independently derived clones determined. Based upon their nucleotide and predicted amino acid sequences, the various clones were categorized into eight major groups of RGAs. The predicted protein sequence of a representative member of each RGA group is shown in Figure 1. Pairwise comparisons of the nucleotide and predicted protein sequences of the various RGAs showed that they have between 35.6-98.2% and 20.4-97.0% identity, respectively (Table 1). Comparison of the cowpea RGAs to sequences available in the various databases available through NCBI, revealed that at the amino acid levels, the cowpea RGAs were most similar to R-genes and RGAs from other leguminous species (e.g., wild cowpea, alfalfa, soybean) (Table 2). The great- est degree of identity was found with RGAs isolated from Vigna vexillata (wild cowpea) and Vigna unguiculata variety IT94K-2053 (81-86%), followed by RGAs from soybean (39-73%). Comparisons of the cowpea RGAs to those of other leguminous crops, such as Medicago spp., chickpea, and pigeon pea, showed significantly higher levels of iden- tity than RGAs isolated from species more evolutionarily diverged (e.g., Pinus radiata, Sorghum hicolor, Hordeum vulgare) (Fig. 2). The various RGAs from cowpea contain sequences resembling the consensus G-X-X-G-X-G-K-T-T motif (Motif!) present in almost all NBS-LRR type R genes (Meyers et al. 1999; Pan et al. 2000). This motif, referred to as the P-loop or kinase-l domain and is thought to be necessary for properly orienting the nucleotide phosphate group of the bound ATP or GTP (Saraste et al. 1990; Traut 1994; Mago et al. 1999). The presence of two absolutely conserved phenylalanine residues separated by four amino acids (Motif II) and the almost absolutely conserved W-F-G-X-G-S-R kinase-2 171 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) ..... tj Motil: I Mottl: II Motil: III ]. 50 ~OO 4.38 ~:~ •••••• alDGlS~I~SRI.BrJLBB;J:QS1~t.n1LGI •• DlDlNIASBBBI::!II~~_.~ 4510 ~ •••••• ClIDGlStl!i'I~ISRI.Br!~SlLt:.,BlLLG •• DmINL1SBBEJ!IISVIp'il'Qr01i'~.~ 468 ~LILIL!'nDG •••••• CIlDGlS-QIlaNVPlilXSm: .mp:mmgSlL~D ILG •• BIDIILlSIBE.aISIII,gPJU.g:a;o:vIa.LlLmrrD .:tc:;Iil;tQ 4.31 ~~~""lD _ .. -.. lQlBALCIl.tBNVRHAS SI .s&LLJqSTriI,.!,E'l'VG •• B •• IILTSV:O: arS~~UI!l2bP .;lBRQUQ 4.3 2 g~~ •••••• t;iiBGlLOi.t'BNVP;H,AS SI .BDLLJ:IIQSTz:,t."$ETVG •• B •• IILTSV:O:,I:iIIS~,~'Q:P.:V,t.tlL,~ .:IJ$CiLQ 445 QQVR'l!~n;o •••••• aiEG1Ii~DI]VRIil .• AVJ: .BDLvm,;Dlr.LSE'l'VG. _ B •• IILTSVO: aISn~~:t.:ft,:tmv.D .Dm.Q'l4 436 ~~P.D ...... ~GIL~DI] •• AVX .BBL~DIL~I,EIVG •• BIDII1aSVSl~IS~~;V,~~:mm.:Jl.isrn 4.3oi~:9 •••••• ~lS~ISXEsrrlRlRIQ~r,.$1: .LA •• lJ:G1LB1RDLDE'~LPl;D.~V~.DT~1l • ~_IJ'DTLt'G1R)(DSSYQ~GlA~lllil •••• J::A.a:B:S~AJ;,ti~ELLR •• HI. ANYNNBBDI3ImQIASltt.R~;DmJJNIDHlIiE L6 Gai ~f:rm'IINIISS •••••• CllDCCmr:rDNlRlilTQBI .r:UvvvI9I1LV8EILRIDSI]SVI]PNNDsg'_TJ:xmiVs~vv.r.I1D11imJ:.p,; XPE Mot! l: IV HyarophObic !2!!!!n ~o~ ~50 192 4.3 8 TL1dISPI$m~~!'l!RU'l ~l;I •• Ase:t;N.lTTnVXTLNmDALQLL~EJl\IDP:RYVEVINDVVI:f.I.~L 4510 '1'L1I.USmfli'e~Ji9l:t~:et.li •• ASBQVJ:Tr'nVXTLNIDDALQLL'l'WDlTrEJl\lDP:RYVEVumv.Yl;.DIi&I:.Pt.J,h 468 ALAGsmo~!rrlij:n'@L •• ASBl£Y.IBrrtWTLNmDASRLL'l'WDroTD'JYDASYY.BVINBA:~n;BD~ 43 1 ALA;~~QtIi •• 1: SBGYERT'f.iVXBLNBElrALBLL'l'WU.RPlNPDPSYmVINLoAYrUSQLlItaA ... 4.3 2 ALA~~$.CiH~~L •• 1: SBGYERrY.iVXELmmN1lLBLL~lNPDPSYmVINLoA~8Br.Pr,;i.: ... 445 ALA'eRPQfli'IIQIIJlIJ.II!l."l'DIlIt;:lJ.Ili •• 1: SBGYERrnVXBLNBElrALBLL'l'WD.Pn'lNPDPSYmVINLoA'V!l!\!18IJ~h 4.3 6 ATVilvPNt&_*ir:r,t.IiIi:a:ei:.ii .. Q •• CMXmD IEDLNBEBALBLL~SYmI~AiUAIitfi.Pr.l'~ 4.3 oi CL~~:PGslwil:!'r:4u'l QVli •• 1 SBGIlmn:1INLLNSDBSLQLL~mmYLBLSXAV m~h • ¥L1I.~_~I: •• ',BINDI •• $VTALPDBBSIQLP~EVPNBN'PBILSLBV'\I:Nf~~ L6 DE.. risPXD:ii'iSQln~,ri:rtSISD.V:OOTLRBN";lc:n.:'f.lVG'SKSB:PRSLBLPSXBI.l'i:nITPP SYYBTLINDVllDTTA,Dm .. Figure 1. Comparison of the protein sequences for various cowpea RGAs with closely related R-gene products from other plants. Shown is the predicted amino acid sequence of representative cowpea RGAs compared with the relevant portions of the tobacco N (Accession No. 2020282A) and flax L6 (Accession No. T18546) proteins. Amino acid sequence alignment was performed using the PILEUP program. Con- served residues are shaded. The location of various conserved motifs as defined in the text are indicated. The nucleotide sequences for RGAs 431,432,434,436,438,445,468, and 490 reported in this paper appear in Genebank with the Accession numbers AF255460-AF225467, respectively. O:l 0' [ ::J o 0- ~ 0' ... 8 ~ ffi ..... ~ 0 0" "" N" CD D- o- '< C"') 0 ~ ......... (i) Table 1. Percentage identity among cowpea RGAs in nucleotide and predicted protein sequence. RGA 490 468 431 42.9/25.7 42.7/26.3 432 42.9/27.5 42.5/26.9 434 54.2/47.3 52.2/46.2 436 58.4/47.3 60.0/47.9 438 98.2/97.0 90.2/84.0 445 41.2/25.1 41.9/25.7 468 89.6/82.2 100/100 490 100/100 445 438 436 (Percentage identity-nucleotide sequence/protein sequence) 94.4/92.8 95.4/94.6 35.6/20.4 66.0/55.1 41.6/25.7 100/100 43.1/26.3 42.7/26.9 55.2/47.9 59.6/49.1 100/100 60.8/47.9 61.8/49.7 43.7132.3 100/100 434 36.0/19.2 36.2/20.4 100/100 432 97.8/97.0 100/100 431 100/100 ~ tn· ~ ::; Ml ~ ::; rt> I\> ::; I\> 0- 0Jl "" o :3 8 ~ ffi ..... ~ 0 0" "" N" CD a. 0- '< C"') 0 ~ ......... (i) Table 2. Distribution of homo logs to cowpea RGAs among different plant species. Plant species Glycine max Vigna unguiculata Vigna vexillata Medicago truncata Medicago sativa Gcer arietinum Cajanus cajan Lactuca sativa Cucumis melo Elaeis guinensis Arabidopsis thaliana Solanum tuberosum Lysopersicon esculentum Capsicum annuum Helianthus annuus Triticum aestivum Hordeum vulgare Zea mays Sorghum bielor Oryza sativa Common name Soybean Cowpea Wild cowpea Alfalfa Chick pea Pigeon pea Lettuce Muskmelon Oil palm Mouse-ear cress Potato Tomato Pepper Sunflower Wheat Barley Maize Sorghum Rice Genebank accession numbers GMU 55803-55811, 2224379A-2224379H, 2224379), T08819-T08821, T088124, T088131-T088135, T088150 AB020483, AB020484, AB020486, AB020489, AF1410121, AF1410131 AF141014,AF141015 AW559555,AW256570 AF230816,AF230823,AF230825,AF230827,AF230835,AF230838-AF230840,AF230842, AF230843,AF230848,AF230850,AF230851,AF230853,AF230854,AF230856,AF230857, AF230859,AF230862 AF186624-AF186627, AF186629 AF186633-AF186639 AFOl77541 and AF07177511 CME2519691-CME2518711 AF197921-AF197924 AC0080174, AC0080176, AC0080179, AC00801710, AC00801715, AC 0071971, T08196, T081916, ATF26013, ATU97217-ATU97219, ATU97221, ATU97224- ATU97226,C71436, E71436, H71436, A71437, B71437, D71437, G71437 2303415G, 2303415H, 2303415)-2303415M, T07766, T07767, T07769, T07770, T07772, T07774 1168481 AF1214351, AF1214371 966421 AF08751821 AF032679-AF032682,AF03684-AF03687,T04389,T04392-T04394,AF146274 AF0561531,AF0561551 AF186644 AF074886-AF074892, AF074894-AF074899, AF1462701, AF146275, AF032688-AF032690, AF032692, AF032693, AF032697, AF032698, AF032700, AF032702, OSH85583, OSH8558121 O:l 0" [ ::J o 0- ~ 0' ... 8 ~ ffi Resistance gene analogs from cowpea ,------------------Vul '-------------------- Vu2 ,-_______________ RGA438 L-_______________ RGA490 L-_______________ V vex L------------------RGA468 ,------------------RGA431 L-_______________ RGA432 '-----_______________ RGA445 '----------------------RGA436 L-__________________ RGA434 L-___________________ N '-----__________________ L6 Figure 2. Dendrogram showing phylogenetic relationship of cowpea RGAs and related resistance gene products from other plants. A representative sequence from each of the eight cowpea RGAs was compared with corresponding regions from RGAs isolated from Vigna vexillata (V vex; Accession No. AF1410151) and Vigna unguiculata variety IT94K- 2053 Vu1, Accession No. AF141 0121 and Vu2 AF141 0131, the N protein of tobacco (Accession No. 2020282A), and L6 protein of flax (Accession No. T18546). domain (Motif IV) (Mago et al. 1999) are characteristic features of NBS-LLR proteins that fall into the Group I category characterized by Pan et al. (2000). Group I NBS-LRR proteins are generally associated with TIR domains at their amino-terminus and are found only in dicotyledonous species. Another notable feature of the cowpea RGA sequences is Motif III, a short stretch of hydrophobic residues followed by two/three aspartate residues which are conserved in almost all of the sequences. This region has been suggested to be involved in stabilizing nucleotide binding with magnesium (pan et al. 2000). Finally, all of the cowpea RGAs contain a short hydrophobic domain with a consensus amino-acid sequence G-L-P-L adjacent to the NBS. Interestingly, several of the peR fragments recovered (clones 447 and 494) contained sequences matching the other RGAs within the regions adj acent to the primer sites, but did not contain complete open reading frames. These fragments showed homology to various retrotransposon-like elements present in the genebank databases. Retrotransposon-like sequences have also been reported in the noncoding regions oftheXa21 gene from rice (Song et al. 1998). It is possible that clones 447 and 494 represent remnants ofR-genes, which have either lost their function due to disruption/rearrangement during evolution as a result of viral insertion. Genomic complexity and RFLP mapping of RGAs In order to determine the relative complexity of the various gene families which encode the RGAs characterized above, gel blot analysis was carried out using DNA isolated from the cowpea lines IT84S-2049 and 527B. IT84S-2049 and 527B are parental lines used to generate the recombinant inbred F8 population used for mapping of the cowpea genome by Menendez et al. (1997). One representative of each of the eight different classes of 175 Digitized by Google Biotechnology for cowpea RGA was used as hybridization probes against total genomic DNA digested with either EcoRI, EcoRV, or Hind III. The results of this analysis are shown in Figure 3. When used as hybridization probes, clones 434 and 436 hybridized to a single frag- ment from both parental lines indicating that these are likely single copy genes. Clones 431 and 432 hybridized to 2-4 different sized fragments suggesting a small family of related sequences, whereas clones 438, 445, 468, and 490 identified multiple fragments, depending upon the enzymes used in the digest. These clones likely represent members of large multi gene families. Clones 438, 468, and 490 identified similar patterns within the genomic digests, although the relative intensities of the hybridization to the individual bands differed. Similarly, identical hybridization patterns were detected with clones 432 and 445. The similar pattern of hybridization observed was consistent with their position on the dendrogram shown in Figure 3. These data suggest that clones 438, 468, and 490 ........ - 6.M-? _ .... , - INM_ 41131 ....!L --!!.......!L L J t :II L :II -, ... _ .. • . ... 41138 .32 ...!!L --.!!!. ...!L • :t ~ 31 I 31 r- ~-'11 ,- ... -- -- 'III! J -~ • ....!!...... .....!.!......!!....- I ~ • 1 I :. .3" • 431 --.!L -.!.!... ..!!.- ...!L --.!!!. ....!L L 31 • I L :II J I ~ 31 I I: Figure 3. Complexity of the nuclear gene families encoding various cowpea RGAs. Genomic DNA from cowpea cultivars 5348 (lane 1) and IT-84S-2049 (lane 2) were digested with fcoRI, fcoRV or Hind III , the restriction digest products separated byaga- rose gel electrophoresis, blotted to nylon membranes, and hybridized with 32P-labeled probes prepared from the various RGAs (indicated above each blot). The blots were washed under high stringency conditions and visualized by autoradiography. The approxi- mate size of the hybridizing fragments are shown to the left in base pairs (kb). 176 Digitized by Google Resistance gene analogs from cowpea and clones 432 and 445 likely constitute diverged members of a large family of related R-genes in the cowpea genome. The various RGAs from each of the eight different classes were also used as hybridiza- tion probes in order to map their location onto the existing cowpea genomic map and begin determining whether polymorphisms detected by any of them cosegregate with previously mapped disease or pest resistance genes. RFLPs were detected between the two parental lines, IT84S-2049 and 527B, when gel blots were hybridized with probes prepared from clones 434, 438, 468, and 490. Hybridization probes prepared from clones 438, 468, and 490 detected RFLPs in DNA digested with either EcoRI, EcoRV, or Hind III, whereas hybridization probes prepared from clone 434 only detected RFLP in EcoRI and EcoRV digested DNA. Segregation analysis of the RFLP markers in the recombinant inbred F8 population derived from crossing IT84S-2049 x 527B was used to map the location of the RGAs within the cowpea genome. RGA 434 mapped to the end oflinkage group 2 (LG-2), approximately 8.9 cM from D 1289 (Menendez et al. 1997). RGAs 438, 468, and 490 also mapped to LG-2, but in a different location from RGA 434. These loci clustered near the top of the linkage group between markers M 185 (68.3 cM) and OC 1 (52.4 cM). The large distances are due in part to the lack of marker data in this area of the genome. Interest- ingly, loci for cowpea severe mosaic virus (CPSMv) and Fusarium oxysporum (FusR) resistance also map to this general region of LG-2, suggesting that R-genes against a number of different pathogens may be clustered nearby. All the RFLPs segregated in a dominant manner except the EcoRI and EcoRV fragments of 434. None of the RGAs mapped thus far cosegregate with known disease or pest resistance loci. Discussion We have identified at least eight separate classes of RGAs from cowpea based on the presence of a conserved NBS within their predicted protein coding region. Eight classes ofRGAs were also reported to be present in common bean (Phaseolus vulgaris) (Rivkin et al. 1999) whereas nine classes ofRGAs are found in soybean (Glycine max) (Kanazin et al. 1996). At least one class ofRGA in soybean, RGA9, contained multiple stop codons and frame-shift mutations, and was thought to represent a psuedogene. Using slightly different experimental conditions and a different set of degenerate oligonucleotide prim- ers from those reported by Kanazin et al. (1996), Yu et al. (1996) amplified 11 different classes ofRGAs from soybean. As might be expected, significant similarity was found among the various RGAs in the different classes identified in the two studies. Although further analysis is necessary, it is clear that the cowpea RGAs character- ized in this investigation are homologs of the NBS-LLR type R-genes isolated from other plants including rice (Li and Chen 1999; Mago et al. 1999; Tada 1999), Brassica (Joyeux et al. 1999), maize (Collins et al. 1998), common bean (Rivkin et al. 1999), Arabidopsis (Speulman et al. 1998, Aarts et al. 1998), soybean (Kanazin et al.1996; Yu et al. 1996), potato (Leister et al. 1996, 1998), wheat (Seah et al. 1998), barley (Seah et al. 1998), and lettuce (Shen et al.1998). The RGAs identified from cowpea appear to fall into the Group I category of NBS- LRR R-genes characterized by Pan et al. (2000) based upon the nature of conserved residues within various signature motifs within the NBS. This may reflect a bias during the amplification process for a subset of sequences, which could be recognized by the degenerate primers used during PCR. Similar observations were made by other 177 Digitized by Google Biotechnology for cowpea investigators ( Kanazin et al. 1996; Leister et al. 1996, 1998; Aarts et al. 1998; Meyers et al. 1999; Pan et al. 2000). Thus, the diversity recognized in the present study likely grossly underestimates the number of classes ofR-gene sequences present in the cowpea genome. In this regard, work is underway to apply additional RGA sequences using prim- ers corresponding to not only conserved motifs within the NBS, but within the LRR, TIR, and serine/threonine kinase domains of known R-genes. Such a strategy coupled with high resolution polyacrylamide gel electrophoresis has been successfully applied in other plants (Chen et al. 1998). The higher resolution of this system will also assist in the direct identification of polymorphisms between parental lines allowing for a greater ability to map the corresponding RGA directly onto to the existing cowpea map. The number of classes ofRGAs amplified from a plant species not only depends on the type of oligonucleotide primers used, but also depends on the variety/cultivar of a particu- lar plant species that was used as a source of genomic DNA. For example, Speulman et al. (1998) observed that in Arabidopsis, RGA sequences obtained with the cultivar Col-O were all identical and fell into one group, whereas seven different RGA sequence classes were identified when DNA from Nd-l was employed. Furthermore, none of the RGAs identified from Nd-l were identical to those isolated from Col-O. Aarts et al. (1998) also reported that in Arabidopsis, for some RGAfragments, the presence of an R-gene locus and a cosegregating RGA locus is often cultivar dependent. Gel blot analysis of cowpea genomic DNA revealed that the RGAs isolated in the pres- ent study were encoded in gene families that ranged in size from one or two members to large multigene assemblies. RGAs belonging to single, or low copy number gene families have also been reported in rice (Tada 1999). In contrast, RGAs from lettuce, Arabidopsis, and wheat all hybridized to multiple fragments of varying intensity (Aarts et al. 1998; Seah et al. 1998; Shen et al. 1998) indicating that they all were members of large families. Interestingly, the size of the gene family recognized by various RGAs (estimated by the number of hybridizing bands) appears to vary depending upon the variety/cultivar of rice and barley used in the analysis (Leister et al. 1998). In some cases, it has been possible to place RGAs on the respective genomic maps of the plant from which they were derived and to show that a particular RGA was linked to the known disease resistance locus. For example, Collins et al. (1998) reported a perfect segregation between RGA loci and rust resistance loci rpl and rp3 in maize and Seah et al. (1998) showed linkage of RGAs from barley to the loci conferring resistance to cereal cyst nematode and corn leaf aphid. Similarly, various RGAs from rice have been shown to segregate with bacterial blight resistance genes Xa3, Xa4, Xa21, and XalO, blast resistance genes Pi-l(t}, Pi-7(t}, and Pi-lan, green leaf hopper resistance locus Glh, rice tungro spherical virus resistance locus RTS~: and the gall midge resistance gene Gm2 (Mago et al. 1999; Tada 1999). In soybean, five out of 11 RGA subfamilies mapped by Yu et al. (1996) were found linked to the known soybean genes conferring resistance to potyvirus (Rsv 1 and Rpv), P hytophthora root rot (Rps 1, Rps2, and Rps3) and powdery mildew (rmd). RGAs from lettuce were found linked to many downy mildew disease resistance loci (Shen et al. 1998) and Chen et al. (1998) observed a link between the RGA markers and stripe rust resistance genes in wheat. Leister et al. (1996) reported absolute linkage of RGAs to the nematode resistance locus Grol and the Phytophthora infestans resistance locus R7 in potato. RGAs were not only found linked to known single dominant resistance gene loci but also to quantitative 178 Digitized by Google Resistance gene analogs from cowpea loci (QTL). In pepper, a QTL conferring partial resistance to cucumber mosaic virus (CMV) with an additive effect was found closely linked or allelic to one NBS-type family (Pflieger et al. 1999). Of the eight different classes of RGAs recognized in cowpea, only four have been placed on the cowpea map. In the cases of the other sequences, no polymorphisms were detected using our present restriction digestion conditions, which would allow us to map them. It is entirely possible that using a different subset of available restriction enzymes, RFLPs may be recognized that allow the remaining RGAs to be mapped as well. Of the RGAs that were mapped, none showed linkage to any of the known R- genes reported in cowpea thus far, including three different loci conferring resistance to race 1 and race 3 of Striga gesnerioides (Ouedraogo et al. 2000). At the present time, only a small number of R-genes have been mapped in the cowpea genome. As more information becomes available, it is possible that some of the RGAs identified here will be shown to segregate with known pest and disease resistance traits. It is also possible that some of the R-genes already mapped do not fall within the NBS-LRR category. Collins et al. (1998) reported that in their studies on maize, none of the cloned RGAs mapped to known disease resistance loci. In addition, dif- ferent lines or cultivars may have different sets of NBS-LRR genes. In Arabidopsis, sequences hybridizing to RPM 1, an NB S-LRR type R -gene, were totally absent from some lines (Grant et al. 1995). A similar situation was also noted in maize (Collins et al. 1998) where certain RGAs appeared to be absent in some maize lines. It has also been suggested that using a number of different mapping populations may lead to the detection of greater numbers of RGA loci than seen with only one line (Sillito et al. 2000). The four RGAs placed on the cowpea map were all located to the end of LG-2, suggesting that some clustering of R genes may occur. In soybean, RGA6 mapped close to Rpsl and N, whereas a cluster of RGAs representing five different classes mapped to the same linkage group and encompassed an R-gene cluster that included Rmd, Rps2, and Rj 2 (Kanazin et al. 1996). Clustering of RGAs has also been reported in rice, soybean, common bean, lettuce, andArabidopsis (Aarts et al. 1998; Shen et al. 1998; Speulman et al. 1998; Yu et al. 1996; Kanazin et al. 1996; Rivkin et al. 1999; and Mago et al. 1999). Many of the well characterized R loci exist either as complex loci containing tandem arrays of closely linked R genes with different specificities (e.g., M locus for flax rust resistance [Pryor and Ellis 1993]) or major resistance com- plexes (MRCs) conferring resistance to different pathogens (Holub 1997). Several mechanisms have been proposed to explain the origin of complex loci and MRCs. These include gene duplication, unequal crossing over, and gene conversion (pryor 1987; Hammond-Kosack and Jones 1997; Richter and Ronald 2000). However, more detailed study is needed to understand the clustering of RGAs in cowpea. The studies described here are a first step to a broader understanding of the structure and organization of R-genes in cowpea, and should assist in the eventual cloning and characterization ofR-genes against major agronomic pests and diseases of this important food crop. 179 Digitized by Google Biotechnology for cowpea Acknowledgements This work was supported by grants from the Rockefeller Foundation and the Interna- tional Institute of Tropical Agriculture (I1TA) awarded to MPT. We thank Tony Hall for providing the recombinant inbred lines used in mapping and Paul Gepts for help with mapping of the RGAs on the existing cowpea linkage map. References Aarts, M.G.M., B.L. Hekkert, E.B. Holub, J.L. Benyon, W.J. Stiekema, andA. Pereira. 1998. Identification ofR-gene homologus DNA fragments genetically linked to disease resistance loci in Arabidopsis thaliana. Molecular Plant-Microbe Interaction 11: 251-258. Aggarwal, V.D. and J.T. Ouedraogo. 1989. Estimation of cowpea yield loss from S. gesneri- oides infestation. Tropical Agriculture 66: 91-92. Aggarwal, V.D. 1991. Research on cowpea-Striga resistance at I1TA. Pages 90-95 in Proceed- ings, International Workshop organized by I1TA, ICRISAT, and IDRC on Combating Striga in Africa, 22-24 August 1988, Ibadan, Nigeria, edited by S.K. Kim. Altschul, S.F., W. Gish, W. Miller, E.W. Myers, and D.J. Lipman. 1990. Basic local alignment search tool. Journal of Molecular Biology 215: 403-410. Altschul, S.F., T.L. Madden, A.A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D.J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search pro- grams. Nucleic Acids Research 25: 3389-3402. Anderson, P.A., G.J. Lawrence, B.C. Morrish, M.A. Ayliffe, J. Finnegan, and J.E. Ellis. 1997. Inactivation ofthe flax rust resistance gene M associated with loss ofa repeated unit within the leucine-rich repeat coding region. Plant Cell 9: 641-651. Atokple, I.D.K., B.B. Singh, and A.M. Emechebe. 1995. Genetics of resistance to Striga and Alectra in cowpea. Journal of Heredity 86: 45-49. Baker, B., P. Zambriski, B. Staskawicz, and S. P. Dinesh-Kumar. 1997. Signaling in plant- microbe interactions. Science 276: 726-733. Bashir, M. and R.O. Haptom. 1996. Sources of genetic resistance in cowpea (Vigna unguicu- lata L. Walp.) to cowpea aphid-borne mosaic virus. European Journal of Plant Pathology 102: 411-419. Bent,A.F., B.N. Kunkel, D. Dhalbeck, K.L. Brown, R. Schmidt, J. Giraudt, J. Leung, and B.J. Staskawi cz. 1994. RP S2 of Arabidopsis thaliana: a I eucine-ri ch repeat class of plant di sease resistance genes. Science 265: 1856-1860. Botella, M.A., M.J. Coleman, D.E. Hughes, M.T. Nishimura, J.D.G. Jones, and C.R. Somer- ville. 1997. Map positions of 47 Arabidopsis sequences with sequence similarity to disease resistance genes. Plant Journal 12: 1197-1211. Cai, D., M. Kleine, S. Kifle, H. Hans-Joachim, N.N. Sandal, K.A. Marcker, R.M. Klein-Lank- horst, E.M.J. Salentijn, W. Lange, W.J. Stiekema, U. Wyss, F.M.W. Grundler, and C. Jung. 1997. Positional cloning of a gene for nematode resistance in sugar beet. Science 275: 832-834. Chen, X.M., R.F. Line, and H. Leung. 1998. Genome scanning for resistance-gene analogs in rice, barley, and wheat by high-resolution electrophoresis. Theoretical and Applied Genet- ics 97: 345-355. Collins, N.C., C.A.Webb, S. Seah, J.G. Ellis, S.H. Hulbert, andA. Pryor. 1998. The isolation and mapping of disease resistance gene analogs in maize. Molecular Plant-Microbe Interac- tion 11: 968-978. Dixon, M.S., D.A. Jones, J.S. Keddie, C.M. Thomas, K. Harrison, and J.D.G. Jones. 1996. The tomato Cf-2 disease resistance locus comprises two functional genes encoding leucine- rich repeat proteins. Cell 84: 451-459. 180 Digitized by Google Resistance gene analogs from cowpea Dixon, M.S., K. Hatzixanthis, D.A. Jones, K. Harrison, and JD.G. Jones. 1998. The tomato Cf- 5 disease resistance gene and six homologs show pronounced al1elic variation in leucine-rich repeat copy number. Plant Cel1lO: 1915-1925. Ehlers, JD. andA.E. Hall. 1997. Cowpea (Vigna unguiculata L. Walp.). Field Crops Research 53: 187-204. F ery, RL. 1990. The cowpea: production, utilization, and research in the United States. Horticultural Reviews 12: 197-222. Gebhardt, C. 1997. Plant genes for pathogen resistance-variation on a theme. Trends in Plant Sci- ence 2: 243-244. Gish, W. and D.I States. 1993. Identification of protein coding regions by database similarity search. Nature Genetics 3: 266-272. Gowda, B., M. Sar, X. Mu, I Cidlowski, and T. Welbourne. 1996. Coordinate modulation of glu- cocorticoid receptor and glutaminase gene expression in LLC-PK1-F + Cel1s. American Journal of Physiology 270: C825-C831. Gowda, B.S., IL. Riopel, and M.P. Timko. 1999. NRSA-1: a resistance gene homolog expressed in roots of nonhost plants fol1owing parasitism by Striga asiatica (witchweed). Plant Journal 20: 217-230. Grant, M.R, L. Godiard, E. Straube, T. Ashfield, J. Lewald, A. Sattler, RW. Innes, and J.L. Dang. 1995. Structure oftheArabidopsis RPM1 gene enabling dual specificity disease resistance. Sci- ence 269: 843-846. Hammond-Kosack, K.E. and JD.G. Jones. 1997. Plant disease resistance genes. Annual Reveiwof Plant Physiology, Plant Molecular Biology 48: 575-607. Holub, E.B. 1997. Organization of resistance genes inArabidopsis. Pages 5-26 in The gene-for-gene relationship in host-parasite interactions, edited by I.R Crute, J.J. Burdon, and E.B. Holub, CAB International, Wal1ingford, UK. Jones, DA., C.M. Thomas, K.E. Hammond-Kosack, P.J. Balint-Kurti, and ID.G. Jones. 1994. Isolation ofthe tomato Cf-9 gene for resistance to Cladosporiumfulvum by transposon tagging. Science 266: 789-793. Joyeux, A., M.G. Fortin, R Mayerhofer, and A.G. Good. 1999. Genetic mapping of plant disease resistance gene homologs using a minimal Brassica napus L. population. Genome 42: 735- 743. Kanazin, V, L.F. Marek, and RC. Shoemaker. 1996. Resistance gene analogs are conserved and clustered in soybean. NationalAcademy of Sciences ofthe United States of America Proceedings 93: 11746-11750. Kosambi, DD. 1944. The estimation of map distances from recombination values. Annual Eugen- ics 12: 172-175. Lagudah, E.S., O. Moul1et, and R Apples. 1997. Map-based cloning of a gene sequence encoding a nucleotide-binding domain and a leucine-rich region at the Cre3 nematode resistance locus of wheat. Genome 40: 659-665. Lander, E.S., P. Green, J. Abrahamson,A. Barlow, M. Daly, S.E. Lincoln, and L. Newberg. 1987. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1: 174-181. Lawrence, G.J., IE. Finnegan,M.A.Ayliffe,and J.G. El1is. 1995. TheL6 gene for flax rust resistance is related to the Arabidopsis bacterial resistance gene RPS2 and the tobacco viral resistance gene N. Plant Cel1 7: 1195-1206. Leister, D.,A. Bal1vora, F. Salamini, and C. Gebhardt. 1996. A PCR-based approach for isolating pathogen resistance genes from potato with potential for wide application in plants. Nature Genetics 14: 421-428. Leister, D., I Kurth, D.A. Laurie, M. Yano, T. Sasaki, K. Devos, A. Graner, and P. Schulze-Lefert. 1998. Rapid reorganization of resistance gene homologs in cereal genomes. National Academy of Sciences ofthe United States of America Proceedings 95: 370-375. 181 Digitized by Google Biotechnology for cowpea Li, Z. Y. and S.Y. Chen. 1999. Molecular cloning, chromosomal mapping and expression analysis of disease resistance gene homologs in rice (Oryza sativa L.). Chinese Science Bulletin. 44: 1202-1207. Mago, R., S. Nair, and M. Mohan. 1999. Resistance gene analogues from rice: cloning, sequencing and mapping. Theoretical and Applied Genetics 99: 50-57. Marchuk, D., M. Drumm, A. Saulino, and F.S. Collins. 1991. Construction of T-vectors, a rapid and general system for direct cloning of unmodified PCR products. Nucleic Acids Research 19: 1154. Martin, G.B., S.H. Brommonschenkel, J. Chunwongse, A. Frary, M.W. Ganal, R. Spivey, T. Wu, E.D. Earle, and S.D. Tanksley. 1993. Map-based cloning of a protein kinase gene conferring disease resistance in tomato. Science 262: 1432-1436. McDowell, J.M., M. Dhandaydham, T.A. Long, M.G.M. Aarts, S. Goff, E.G. Holub, and J.L. Dang!. 1998. Intragenic recombination and diversifying selection contribute to the evolution of downy mildew resistance at the RPP810cus of Arabi do psis. Plant Cell 10: 1861-1874. Menendez, C.M., A.E. Hall, and P. Gepts. 1997. A genetic linkage map of cowpea (Vigna unguiculata) developed from a cross between two inbred domesticated lines. Theoretical and Applied Genetics 95: 1210-1217. Meyers, B.C., K.A. Shen, P. Rohani, B.S. Gaut, and R.W. Michelmore. 1998. Receptor-like genes in the major resistance locus ofletluce are subject to divergent selection. Plant Cell 10: 1833-1846. Meyers, B.C., A.W. Dickerman, R.W. Michelmore, S. Sivaramakrishnan, B.W. Sobral, and N.D. Young. 1999. Plant disease resistance genes encode members of an ancient and diverse protein family within the nucleotide binding family. Plant Journal 20: 317-332. Milligan, S.B., J. Bodeau, J. Yaghoobi, I. Kaloshian, P. Zabel, and V.M. Williamson. 1998. The root-knot nematode resistance gene Mi from tomato is a member ofthe leucine zipper, nucleotide binding, leucine-rich repeat family of plant genes. Plant Cell 10: 1307-1319. Mindrinos, M., F. Katagiri, G.L. Yu, and F.M. Ausube!. 1994. TheA. thaliana disease resistance gene RPS2 encodes a protein containing a nucleotide-binding site and leucine-rich repeats. Cell 78: 1089-1099. Morel, J. and J.L. Dang!. 1997. The hypersensitive response and the induction of cell death in plants. Cell Death and Differentiation 4: 671-683. Muleba, N., J. T. Ouedraogo, and I. Drabo. 1996. Yield stability in relation to Striga resistance in cowpea production in West and Central Africa. African Crop Science Journal 4: 29- 40. Muleba, N., J.T. Ouedraogo, and J.B. Tignegre. 1997. Cowpea yield losses attributed to Striga infestations. Journal of Agricultural Science (Cambridge) 129: 43-48. Ori, N., Y. Eshed, I. Paran, G. Presting, D. Aviv, S. Tanksley, D. Zamir, and R. Fluhr. 1997. The 12C1 family from the wilt disease resistance locus 12 belong to the nucleotide binding, leucine-rich repeat superfamily of plant resistance genes. Plant Cell 9: 521-532. Ouedraogo, J.T., V. Maheshwari, D.K. Berner, C.-A. St-Pierre, F. Belzile, and M.P. Timko. 2000. Identification of AFLP markers linked to resistance of cowpea (Vigna unguiculata L.) to parasitism by Striga gesnerioides. Theoretical and Applied Genetics. In press. Pan, Q., J. Wendel, and R. Fluhr. 2000. Divergent evolution of plant NBS-LRR resistance gene homologs in dicot and cereal genomes. Journal of Molecular Evolution 50: 203- 213. Parker, J.E. and M.J. Coleman. 1997. Molecular intimacy between proteins specifying plant- pathogen recognition. Trends in Biochemical Sciences 22: 291-296. Parker, J.E., M.J. Coleman, V. Szabo, L.N. Frost, R. Schmidt, E.A. van der Biezen, T. Moores, C. Dean, M.J. Daniels, and J.D.G. Jones. 1997. The Arabidopsis downy mildew resistance 182 Digitized by Google Resistance gene analogs from cowpea gene RPP5 shares similarity to the toll and interleukin-l receptor with Nand L6. Plant Cell 9: 879-894. Pflieger, S., V. Lefebvre, C. Caranta, A. Blattes, B. Goffinet, and A. Palloix. 1999. Disease resistance gene analogs as candidates for QTLs involved in pepper-pathogen interactions. Genome 42: 1100-1110. Pryor, T. 1987. The origin and structure of fungal disease resistance genes in plants. Trends in Genetics 3: 157-161. Pryor, T. and J. Ellis. 1993. The genetic complexity offungal resistance genes in plants. Advances in Plant Pathology 10: 281-305. Richter, T.E. and P.C. Ronald. 2000. The evolution of disease resistance genes. Plant Molecular Biology 42: 195-204. Rivkin, M.I., C.E. Vellajos, and P.E. McClean. 1999. Disease-resistance related sequences in common bean. Genome 42: 41-47. Salmeron, J.M., G.E.D. Oldroyd, C.M.T. Rommens, S.R Scofield, H.-S. Kim, D.T. Lavelle, D. Dhalbeck, and B.J. Staskawicz. 1996. Tomato Prfis a member ofthe leucine-rich repeat class of plant disease resistance genes and lies embedded within the Pto kinase gene cluster. Cell 86: 123-133. Sambrook, 1, E.F. Fritsch, and T. Maniatis. 1989. Molecular cloning-a laboratory manual. Second edition. Cold Spring Harbour Lab Press, New York, USA. Saraste, M., P.R. Sibbald, and A. Wittinghofer.1990. The P-loop: A common motif in ATP- and GTP-binding proteins. Trends in Biochemical Sciences 15: 430-434. Seah, S., K. Sivasithamparam,A. Karakousis, and E.S. Lagudah. 1998. Cloning and characterization of a family of disease resistance gene analogs from wheat and barley. Theoretical and Applied Genetics 97: 937-945. Shen, K.A., B.C. Meyers, M. Nurul Islam-Faridi, D.B. Chin, D.M. Stelly, and RW. Michelmore. 1998. Resistance gene candidates identified by PCR with degenerate oligonucleotide primers map to clusters of resistance genes in lettuce. Molecular Plant-Microbe Interaction 11: 815- 823. Sillito, D., I.A.P. Parkin, R Mayerhofer, D.J. Lydiate, andA.G. Good. 2000. Arabidopsis thaliana: A source of candidate disease resistance genes for Brassica napus. Genome 43: 452-460. Singh, B.B., D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai (editors). 1997. Advances in cowpea research. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan Inter- national Reseach Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Singh, B.B. and A.M. Emechebe. 1997. Advances in research on cowpea, Striga and Alectra. Pages 215-224 in Advances in cowpea research. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS), edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. I1TA, Ibadan, Nigeria. Song, w.-Y., G.-L. Wang, L.-L. Chen, H.-S. Kim, L.-Y. Pi, T. Holsten, J. Gardner, B. Wang, w.- X. Zhai, L. -H. Zhu, C. F auquet, and P. C. Ronald. 1995. A receptor kinase-like protein encoded by the rice disease resistance gene, Xa21. Science 270: 1804-1806. Song, W.-Y., L.-Y Pi, T.E. Bureau, and P.C. Ronald. 1998. Identification and characterization of 14 transposon-like elements in the noncoding regions of members oftheXa21 family of disease resistance genes in rice. Molecular and General Genetics 258: 449-456. Speulman, E., D. Bouchez, E.B. Holub, and J.L. Benyon. 1998. Disease resistance gene homologs correlate with disease resistance loci of Arabi do psis thaliana. Plant Journal 14: 467-474. Tada, Y. 1999. PCR-amplified resistance gene analogs link to resistance loci in rice. Breeding Science 49: 267-273. Thomas, C.M., D.A. Jones, M. Parniske, K. Harrison, P.J. Balint-Kurti, K. Hatzixanthis, and JD.G. Jones. 1997. Characterization of the tomato Cf-4 gene for resistance to Cladosporium 183 Digitized by Google Resistance gene analogs from cowpea fulvum identifies sequences that determine recognitional specificity in Cf-4 and Cf-9. Plant Cell 9: 2209-2224. Toure, M.,A. Olivier, B.R. Ntare, J.A. Lane, and C.-A. St-Pierre. 1997. Inheritance of resistance to Striga gesnerioides biotypes from Mali and Niger in cowpea (Vigna unguiculata [L.] Walp.). Euphytica 94: 273-278. Toure, M., A. Olivier, B.R Ntare, J.A. Lane, and C.-A. St-Pierre. 1998. Reaction of cowpea (Vigna unguiculata) cultivars to Striga gesnerioides races from Mali and Niger. Canadian Journal of Plant Science 78: 477-480. Traut, T.W. 1994. The functions and consensus motifs of nine types of peptide segments that form different types of nucleotide-binding sites. European Journal of Biochemistry 222: 9-19. Van der Biezen, EA. and JD.G. Jones. 1998. Plant disease-resistance proteins and the gene-for-gene concept. Trends in Biochemical Sciences 23: 454-456. Varadarajan, G.S. and C.S. Prakash. 1991. A rapid and efficient method for the extraction oftotal DNA from the sweetpotato and its related species. Plant Molecular Biology Reporters 9: 6-12. Warren, RF.,A. Henk, P. Mowery, E. Holub, and RW. Innes. 1998. A mutation within the leucine- rich repeat domain oftheArabidopsis disease resistance gene RPS5 partially suppresses multiple bacterial and downy mildew resistance genes. Plant Cell 10: 1439-1452. Whitham, S., S.P. Dinesh-Kumar, D. Choi, R Hehl, C. Corr, and B. Baker. 1994. The product of the tobacco mosaic virus resistance gene N: Similarity to toll and the interleukin-1 receptor. Cell 78: 1101-1115. Yoshimura, S., u.Yamanouchi, Y Katayose, S. Toki, ZX. Wang, I. Kono, N. Kurata, M. Yano, N. Iwata, and T. Sasaki. 1998. Expression ofXal, a bacterial blight-resistance gene in rice, is induced by bacterial inoculation. National Academy of Science of the USA Proceedings 95: 1663- 1668. Yu, YG., G.R. Buss, and M.A. Sahai-Maroof. 1996. Isolation of a superfamily of candidate disease- resistance genes in soybean based on a conserved nucleotide-binding site. National Academy of Science ofthe USA Proceedings 93: 11751-11756. 184 Digitized by Google 3.2 Regeneration and genetic transformation in cowpea J. Machuka1 , A. Adesoye, and 0.0. Obembe2 Abstract Over the last three decades, sporadic efforts have been made to develop regenera- tion and transformation systems in cowpea (Vigna unguiculata L. Walp). This paper reviews the progress made to date, including highlights of culture media and explants used for regeneration and chimeric gene constructs employed in transfor- mations. Progress has been slow, mainly due to limited resources, since very few laboratories have been involved. There is an urgent need for more focused and consistent efforts to develop genotype, and tissue-culture dependent and indepen- dent approaches for obtaining stable genetic transformation in cowpea. Introduction Cowpea faces several biotic and abiotic stresses for which conventional breeding alone may not provide ultimate solutions. For example, grain yield losses are mainly due to damage caused by insect pests and diseases, as well as abiotic stresses such as heat and drought (Singh et al. 1997). Plant molecular biology and genetic engineering approaches offer alternative ways of overcoming these stresses. In addition to direct transfer of genes of agronomic interest, genetic transformation techniques can be used to answer many basic questions pertaining to cowpea biology such as understanding of gene function and regulation of physiological and developmental processes (Gelvin 1998). These benefits require the development of reliable, efficient, and reproducible methods for cowpea transformation and regeneration. Although legumes are considered "recalcitrant" to regeneration and transformation, routine protocols for obtaining stable transformants are now available for the major grain legumes such as the common bean (Phaseolus vulgaris), soybean (Glycine max), pea (Pisum sativum), peanut (Arachis hypogea), and alfalfa (Medicago sativa), as well as the model legume, barrel medic (Medicago truncata) (Christou 1992; Puonti-Kaerlas et al. 1990; Russell et al. 1993). In contrast, development of tractable gene transfer systems in cowpea has been impeded by several constraints. Cowpea is not of major economic importance to the most technologically advanced countries in North America and Europe. This crop is mainly grown in tropical Africa, Asia, and Latin America where technical expertise and infrastructure for biotechnology research are either lacking or poor. There- fore, comparatively little work has been done to develop and optimize regeneration and transformation procedures, relative to temperate crops that are of economic importance in the North, including recalcitrant cereals (Komari et al. 1998). This paper reviews 1. PO Box 347, Kilifi, Kenya. 2. Biotechnology Laboratory, International Institute of Tropical Agriculture, PMB 5320, Oyo Road, Ibadan, Nigeria. 185 Digitized by Google Biotechnology for cowpea previous work on cowpea cell and tissue culture and transformation. It also highlights future research directions that could hold promise for the establishment of reliable gene transfer systems for a crop that has tremendous potential as a rich source of dietary protein for millions of people in Africa and Asia. Cell and tissue culture The two methods commonly used for regeneration of plants from cell cultures are somatic embryogenesis and organogenesis. Both methods are controlled by plant hor- mones and other factors added to the culture medium. As the name suggests, somatic embryogenesis involves the generation of embryos from somatic tissues, such as roots, cotyledons, leaves, stems, and reproductive organs. The proliferating somatic embryos are either induced in liquid culture or on solid medium. Since embryogenic tissues are very prolific and usually originate from single cells, the embryos are considered excellent targets for transformation (Hansen and Wright 1999). This is why somatic embryogenesis is the method of choice for most genetic transformation protocols for recalcitrant legumes and mono cots such as soybean, maize, and rice, respectively (Komari et al. 1998; Puonti-Kaerlas 1993; Trick et al. 1997). In cowpea, induction of somatic embryos has been reported to occur in suspension cultures of calli derived from seedling leaf explants (Ganapathi and Anand 1998). Embryogenic calli were induced on solid Murashige and Skoog (MS) medium (Murashige and Skoog 1962) supplemented with 1.5 mg/liter (mg/L) of2-, 4-dichlorophenoxy-acetic acid (2-, 4-D). The maximum frequency of somatic embryos was obtained when callus was transferred to liquid MS medium supplemented with 0.5 mg/L 2-, 4-D. This work is repeated in other laboratories, including characterization of the stages and processes of somatic embryo development. Additionally, other explant sources other than young leaves should also be investigated for their ability to produce somatic embryos in solid and liquid suspension cultures. The basal medium developed for embryo development by Pellegrineschi et al. (1997) could form a starting point for formulating media for growth of somatic embryos in vitro. Growth medium supplements that enhanced embryo development included addition of sucrose, casein hydrolysate, and anyone of three commonly used cytokins, namely zeatin, benzyl amino purine (BAP), and kinetin, for enhancing embryo maturation. The establishment and maintenance of embryogenic cultures as well as recovery of plants can be an extremely labor intensive and lengthy process that has the added risk of encountering morphological abnormalities and sterility among regenerants. In contrast, multiple shoot formation via organogenesis is simpler once a suitable explant has been identified. Various laboratories have independently reported success- ful regeneration of cowpea by direct organogenesis from a variety of explants. These include roots, stem pieces, intact immature cotyledons or protoplasts derived from them, leaves, stem apices, stimulated shoot bud formation following gamma irradiation, or germination of mature seeds in the presence of the herbicide thidiazuron (Kartha et al. 1981; Subramaniam et al. 1968). Shoot regeneration has also been reported using axenic cowpea hypocotyls and cotyledons excised from green immature pods of advanced breeding lines and varieties developed at the International Institute of Tropical Agriculture (I1TA) in Ibadan, Nigeria, (pellegrineschi 1997). The apical parts of the embryos were removed and the hypocotyls were transferred to regeneration 186 Digitized by Google Regeneration and genetic transformation in cowpea media modified from a formulation that was previously employed for embryo rescue (Pellegrineschi et al. 1997). Fertile cowpea plants have been regenerated successfully using nodal thin cell layer (TCL) explants. The TCL, approximately eight cells thick, was obtained by cutting twice over each cotyledonary node, followed by regeneration on MS media containing either 1.1 mglL zeatin and 0.05 mglL indole butyric acid (IBA) or 1.1 mglL BAP and 0.05 mglL IBA. Among these explants, direct organogenesis from cotyledons, cotyledonary nodes, epicotyls, and primary leaves cultured on MS containing optimal levels of either ~-ben­ zyladenine (BA) or BAP appear to be reproducible and hold promise for use in transfor- mation (Brar et al. 1999; Muthukumar et al. 1995; Obembe et al. 2000a; Pellegrineschi 1997). At I1TA, organogenesis has been obtained in several genotypes such as 90K-277, 89D-288, 83D-442, 86D-1O 10, 93K-624, Vita 3, and Ife Brown (Fig. 1 a). Shoot meristem regeneration on MS media supplemeted with either the herbicide thidiazuron or BAP has been successfully demonstrated in various genotypes, including CB5, TARS 36, SUV-2, 283, 1137,275, TN88-63, B30l, Tvu 9062, Vita 3, Vita 4, and 58-57 (Kononowicz et al. 1997; Monti et al. 1997). Brar et al. (1999) have recently reported a regeneration system that was applicable to 17 US commercial cowpea cultivars and breeding lines. Cotyle- dons were initiated on 113 MS medium containing 15-35 mglL ofBAfollowed by shoot regeneration on MS containing 1.0 mglL of BA. Depending on the genotype, regenera- tion percentages ranged from 1 to 11, with 4-12 multiple shoots produced per explant. F or rooting of cowpea plantlets, the report of Brar et al. (1999) and our results show that hormone-free MS medium works well. However, addition of 1.0 mglL of indole-3-acetic acid (IAA) or 0.05 mglL of nephthalene acetic acid (NAA) significantly enhances rooting and survival of plantlets in soil during the hardening and acclimatization phase following transfer from tissue culture conditions (Obembe et al. 2000a). A procedure for protoplast isolation from leaf mesophy 11 cells and regeneration leading to production of microcalli has also been described. However, plant regeneration from protoplast-derived calli was not possible, rendering the system inapplicable for heritable gene transfer. A Figure 1 a. In vitro cowpea regeneration from cotyledonary nodes cultured on 0.5 mgtl of benzyl amino purine (RAP). 187 Digitized by Google Biotechnology for cowpea Transformation systems Currently used methods for genetic transformation have been classified into natural and non-natural or in vitro methods (Gelvin 1998). The latter include DNA microinjection (Neuhaus and Spangenberg 1990), direct DNA uptake into protoplasts with or without the use of electroporation (Shillito 1999), use of silicon carbide whiskers (Kaepplar et al. 1990) and biolistic bombardment (Hadi et al. 1996; McCabe et al. 1998; Shillito 1999). Natural methods involve the use of viral vectors that will result in transient but not stable transformation (Choi et al. 2000; Masuta et al. 2000) and Agrobacterium tumefaciens T- DNA-mediated transformation (Zupan et al. 2000). There are two major causes for the delay in the development of methods for the genetic transformation of legumes, in comparison to other dicotyledonous species. First, is the problem of recalcitrancy to regeneration by somatic embryogenesis and organogenesis, as already discussed. Secondly, transformation mediated by the soil bacteriumA. tumefa- ciens was not, initially, readily applicable to legumes. Therefore, attempts at gene transfer initially focused on direct DNA delivery, especially by microprojectile (particle) bom- bardment which is still a popular technique since it is species- and genotype-independent (Christou 1992; McCabe et al. 1998). It has now been demonstrated that A. tumefaciens can efficiently transform legumes such as soybean (Trick et al. 1997). In the following section, we will review the methods and results of previous work that has been done on genetic transformation in cowpea. Agrobacterium-mediated transformation The earliest report on Agrobacterium-mediated cowpea transformation was based on the tobacco leaf disc transformation method (Horsch et al. 1985). Cowpea leaf discs were punched from primary leaves obtained from 6-day old seedlings and co-cultivated withA. tumefaciens strains harboring tumor inducing (Ti)-derived vectors containing two copies of a chimeric kanamycin resistance gene (Garcia et al. 1986a, 1986b). A. tumefaciens strain C58CI harboring the non-oncogenic Ti plasmid pGV3850:: l103neo, or its deriva- tives, strain LBA 1010 containing the octopine type Ti plasmid p TIB6 and strain LBA 958 containing a nopaline type Ti plasmid were all infective on cowpea leaves and stems. For selection of transformed tissues, G4l8 (50 mglL) was initially incorporated into the culture media, but tissues were transferred and selected on kanamycin (100 mglL) during later subcultures. This procedure resulted in stable transformation of callus, but no transgenic plants were regenerated. The full length eDNA of cowpea mosaic virus (CPW) gene under the control of either the cauliflower mosaic virus (CaMV35S) or nopaline synthase (nos) promoter was stably transferred and expressed in cowpea calli (Garcia et al. 1986b). The CaMV35S was also more than ten times stronger than the nos promoter. Moreover, this work showed that 7-day old cowpea plants (stems) are susceptible to Agrobacterium infection, since both oncogenic Agrobacterium strains LBA 1010 and LBA 958 induced crown galls at wounded stem sites. An earlier study by Saedi et al. (1979) showed that cowpea seedlings fail to develop tumors after being inoculated withA. tumefaciens if, at times earlier than one day later, they were inoculated on the primary leaves with a cowpea mosaic virus that systemically infects them. Inoculation with buffer or with a virus that is restricted to a localized infection, or to which the cowpea is immune, did not interfere with the subsequent development of tumors. These observations indicated that systemic virus infection may induce in cowpeas a translocated substance that prevents tumor induction 188 Digitized by Google Regeneration and genetic transformation in cowpea by A. tumefaciens. Therefore, the pathology of cowpea tissues may be an important factor to consider during Agrobacterium-mediated transformation. We have found LBA 4404 (carrying octopine type plasmid pTiA6) to be least virulent on cowpea tissues cultured in vitro, compared to AGL 1, a disarmed, hypervirulent strain harboring mannopine-type Ti plasmid pTiB0542. PGV3850, another disarmed, wide host range hypervirulent strain harboring a nopaline-type Ti plasmid pTiC58, is also very virulent on cowpea (Obembe et al. 2000b). Only a few other reports have appeared in scientific literature concerning Agrobac- terium-mediated transformation of cowpea since the excellent early work of Garcia et al. (1986a, 1986b). Perkins et al. (1987) and Filippone (1990) were able to show stable transformation of callus by co-cultivation of mature embryos, cotyledonary node buds, epi- cotyls, and apical meristems withA. tumefaciens. Cowpea accessions used in Filippone's work were IT8ID-994, Tvu 9062, and cv VlTA4. Transformations utilized the hyper- virulent A. tumefaciens strain 6044 containing plasmid pGA472 carrying the neomycin phosphotransferase (NPTIJ) gene. Selection of transformed calli was carried out on 100 mglL kanamycin or 50 mglL geneticin. When cowpea embryos were used, the parts most amenable to transformation were the collar and epicotyls (Filippone 1990). Penza et al. (1991) reported the production of chimeric beta-glucuronidase (gus) (Jefferson 1989) in transgenic cowpea plants from mature embryos co-cultivated withA. tumefaciens. Using excised, ungerminated embryos was seen as a way of bypassing problems associated with regeneration from callus and differentiated tissues. Co-cultivation of embryos with the disarmedA. tumefaciens strain C58 (pGV2260/p35SGUSINT) carrying a gus intron resulted in chimeric, transformed shoots derived from axillary buds. Transformed cells were mostly located in subepidermal regions of the plant stems where the L2 meristematic layer is positioned (Fletcher and Meyerowitz 2000). Since the L2 layer potentially can contribute to flower buds, it still remains unclear why the transgenes were not transmitted through the germline, despite extensive plant propagation through nodal culture (penza et al. 1991). The ability to regenerate cowpea in planta (Machuka 2000) as well as the use of positive selection systems (Joersbo et al. 1998) may provide avenues for recovery of stable transformed plants. If successful, the mature embryo co-cultivation method would be simple and easy to use for large-seeded legumes such as cowpea. Using excised leaf, epicotyl, and hypocotyl explants, stable callus transformation was obtained after co-cul- tivation of the explants with LBA 4404 carrying the gus-intron plasmid p35SGUSINT. Through co-cultivation of these explants with A. rhizogenes, the same workers demon- strated production of transgenic hairy roots following in vitro selection on kanamycin. Hairy root transformation was also reported earlier (Suzuki et al. 1993). These workers used a soybean cell wall protein gene (SbPRP 1) promoter-GUS construct to show localiza- tion of SbPRP 1 in actively growing roots (apical and elongating regions) during cowpea seedling growth. Publications on stable Agrobacterium-mediated transformation incorporating southern analysis of primary transform ants are available (Muthukumar et al. 1996; Kononowicz et al. 1997; Monti et al. 1997). Muthukumar and co-workers used mature de-embryonated cotyledons excised from 2-3-day old seedlings. The cotyledons were co-cultivated with A. tumefaciens and transformed tissues selected on 25 mglL hygromycin. Our preliminary work on the effect ofhygromycin on in vitro regeneration and rooting of un transformed cowpea has established significant inhibition levels at ~20 mglL (Obembe et al. 2000b). 189 Digitized by Google Biotechnology for cowpea Although Muthukumar et al. (1996) reported that 15-19% of explants produced shoots on hygromycin selection medium, 13 out of 17 putative transformants died. Unfortunately, seeds from the four remaining plants failed to germinate, thus leaving us without reproduc- ible evidence of stable transformation. Research teams at Purdue University (USA) and the University of Naples (Italy) obtained transformed TO plants using the gus reporter gene as well as two useful genes. However, results from further analysis to establish proof of stable transformation and reliability of the protocols have not been forthcoming. Despite this, the work was useful in many respects. For example, tests pertaining to the virulence of Agrobacterium strains revealed that A28l, a hypervirulent oncogenic strain, was most infective, followed by ERA 101, whereas LBA 4404 had the lowest virulence (Kononowicz et al. 1997; Monti et al. 1997). For many plant species, Agrobacterium-mediated transfor- mation is relatively efficient, and a low copy number of intact, nonrearranged transgenes are frequently integrated into the plant genome (Zupan 2000). These observations and the foregoing discussion indicate that Agrobacterium-mediated transformation in cowpea is feasible and may yet be the preferred choice for laboratories that work or plan to begin work on genetic transformation in cowpea. Transformation with naked DNA Microprojectile bombardment can be performed with any tissue of most species; however, the process is relatively inefficient because few cells are stably transformed. When DNA is delivered by this method, the conversion rate from transient expression to stable integra- tion is estimated to be <1 to 9% (Hansen and Wright 1999; Finer et al. 2000). This method of transformation has been used on cowpea cotyledon segments, immature embryos, and shoot meristems (Ikea 1998; Kononowicz et al. 1997; Monti et al. 1997). However, convincing molecular evidence of transformation in T 1 and subsequent progeny was not provided. In the work of Kononowicz et al. (1997) and Monti et al. (1997), some chimeric gene constructs used in transformations contained the phosphoinothricin (bar) resistance, gus and NPTII genes, driven by CaMV 35S or nos promoters. Other constructs contained sequences encoding the common bean a-amylase inhibitor or Bex (2S albumin) protein from Brazil nut, under control of phaseolin (seed-specific) or CaMV 35S (constitutive) promoters. Putative transformed tissues were selected on 50 mglL kanamycin, which is probably not stringent enough to prevent escapes. Plant transformation using protoplasts is laborious and requires a lot of finesse. Once isolated mechanically or using enzymes, the protoplasts can be transformed either by Agrobacterium or by direct DNA uptake methods, facilitated by polyethylene glycol (pEG) treatment, electroporation, or liposomes (Shillito 1999). The method has the advantage that single cells can be targeted for transformation, provided the protoplasts can regenerate into whole plants. Using cowpea leaf me sophy 11 protoplasts, stable, PEG-mediated protoplast co-transformation of two plasmids (pGL2 and pMONGUS) carrying the hygromycin resistance and gus genes were obtained. Stable transgenic microcalli were obtained that could not be regenerated into plants. Electroporation of cells or tissues in the presence of DNA is used for the introduction of transgenes either stably or transiently into bacterial, fungal, animal, and plant cells (Lurquin 1997; 10ersbo and Brunstedt 1991). The method is not often used in plant trans- formation because of its low reproducibility. However, owing to difficulties encountered in regenerating transformed cowpea cells and tissues in vitro, electroporation of intact 190 Digitized by Google Regeneration and genetic transformation in cowpea tissues and organs has been resorted to with promising results. Early work using cowpea seed-derived embryos showed that chimeric transgenes could be expressed in cowpea protoplasts and seedlings after passive or electroporation-mediated naked DNA transfer (Akella and Lurquin 1993; Penza et al. 1992). Electropration-mediated DNA delivery into seedling tissues was also demonstrated by Dillen et al. (1995), not only in cowpea but also in other grain legumes such as the common bean, pea, and soybean. Linearization of plasmid DNA markedly increased transient DNA expression levels in intact hypocotyls and epicoty Is. It is not clear what is the conversion rate from transient expression to stable integration in the plant genome using electro-transformation, but it is likely to be low (Lurquin 1997; Joersbo and Brunstedt 1990). Chowrira et al. (1995) at Washington State University, Pullman, provided evidence of both transient and stable expression of the gus gene after electroporation of auxilIary nodal meristems in planta. The branches that grew out of the nodal meristems were chimeric and expressed the introduced gene up to 20 days after electroporation (Chowrira et al. 1996). Transgenic T 1 pea, lentil, and cowpea plants were recovered from seeds originating on these chimeric branches as shown by Southern blot hybridization and gus expression. Although transgenic T2 soybean and lentil plants were also obtained, no transgenic T2 cow- peas were reported. Segregation ratios in these populations showed a strong bias against transgene presence or expression. This in vivo transformation approach has at least two advantages. First, electroporation equipment is cheap and the protocols are easy to optimize (Lurquin 1997). Secondly, seeds can be obtained without need for in vitro steps, thereby speeding up the process of generating transgenic plants. The occurrence of chimeras may be reduced if selection systems can be developed for cowpea, such as phosphinothricin (Fig. 1 b) and kanamycin painting and chlorophyll fluorescence for phosphinothricin and kanamycin resistance, respectively (Eu et al. 1998; Rasco-Gaunt et al. 1999). Other promising transformation methodologies The recent development of simple and routine de novo floral and seedling dipping and/or infiltration procedures for Agrobacterium-transformation in Arabidopsis andM. truncata (Clough and Bent 1998; Trieu et al. 2000) has sparked new optimism to develop similar techniques for other crops. In comparison with these model plants, cowpea has few flowers that would be the key target for transformation. Furthermore, comparatively few seeds are set. Since electroporation of cowpea nodal tissue has already been reported (Chowrira et al. 1996), work is in progress at I1TA to maximize the number of vegetative and floral buds produced at every node or at the shoot apex through hormonal applications (Machuka 2000). This procedure has potential for coupling to in planta transformation techniques, notably electroporation and dipping of hormone-induced organs inAgrobacterium suspen- sions (Fig. 1 c). Transient gus expression assays indicate that use of Silwet-L 77 in conjunc- tion with acetosyringone enhances expression following vacuum infiltration of excised mature cowpea embyos (Fig. 2). Experiments utilizing these additives in Agrobacterium seedlings and floral dipping and infiltration solutions are in progress at I1TA. Selection of transformed tissue is likely to be the key obstacle for reliable adoption and exploitation of a de novo cowpea regeneration-based transformation system. Natural plant transformation technologies that include the use of viral vectors for transient transformation should also be explored for cowpea (Choi et al. 2000; Masuta et al. 2000). It is already known that full-length cDNA copies of cowpea mosaic virus RNA cloned downstream of the CaMV 191 Digitized by Google Biotechnology for cowpea B c , Figure 1 B, C. In planta cowpea regeneration of decapitated seedlings (B) and three-week old plants (C) treated with 10 mglL BAP. MglL: 500 100 50 10 5 - . 2 o .................... Figure 2. Phosphinothricin (PPT, Duchefa Bichemie, Haarism, Holland) painting of cowpea plants. Numbers represent PPT concentrations. As seen in this photo, survival of seedlings was nil 7 days after spraying with PPT concentrations exceeding 50 mglL. 35S promoter give rise to cowpea mosaic virus-like symptoms when inoculated onto cowpea plants (Dessens and Lomonossoff 1993). More recently, the clover yellow vein virus has been developed as an efficient vector system for stable foreign gene expression in legumes in planta (Masuta et al. 2000). Techniques for DNA delivery using silicone carbide whiskers (potential carcinogens), microinjection, and laser microbeams (Hansen and Wright 1999) require much finesse and may not be easily adapted for use in African and Asian countries which are likely to 192 Digitized by Google Regeneration and genetic transformation in cowpea benefit most from genetic modification in cowpea breeding. However, groups working on cowpea transformation need to experiment with techniques that combine the best attri- butes of Agrobacterium-mediated transformation (high efficiency, low copy number, and intact transgenes) with particle technologies (Gelvin 1998). For example, a novel strategy termed "Agrolistic" transformation could be used on cowpea tissues that are susceptible to transformation by particle bombardment (Ikea 1998; Kononowicz et al. 1997; Monti et al. 1997). This technique has the potential to integrate a low copy number of transgenes without integration of plasmid vector sequences (Hansen and Chilton 1996). Conclusions The powerful combination of conventional and genetic modification breeding has the potential of greatly enhancing the productivity of cowpeas by increasing resistance to pests, diseases, Striga, and abiotic stress, as well as seed quality and other traits that impact on cowpea utilization for fodder and grain. To be of value, genetically modified plants must faithfully transmit their transgenes. From the works surveyed in this review, it is apparent that this has not been achieved in cowpea. Recalcitrance to plant regeneration of transformed tissues, epidermal transformation, and transgene instability are likely causes of failure to achieve stable transformation and transgene transmission. Improvements in existing cell and tissue culture systems to allow regeneration of stable transformed cowpea plants is urgently needed. With so many available advances and new breakthroughs in plant transformation technologies, it is hoped that cowpea's stubborn resistance to genetic engineering will soon be overcome. Acknowledgements The authors thank S. Akinbade, A.O. Odeseye, and B.J. Akinyemi for technical assis- tance in cowpea regeneration and transformation research; S. Adekunle and Wole for screenhouse maintenance of plants; and M.O. Raji for routine laboratory and glassware maintenance. References Akella,V and P.F. Lurquin. 1993. Expression in cowpea seedlings of chimeric transgenes after electroporation into seed-derived embryos. Plant Cell Reports 12: 110-117. Brar, M.S., 1M. Al-Khayri, T.E. Morelock, and E.l Anderson. 1999. Genotypic response of cowpea Vigna unguiculata (L.) to in vitro regeneration from cotyledon explants. In Vitro Cellular Devel- opmental Biology 35: 8-12. Choi, I-R., D.C. Stenger, T.l Morris, and R. French. 2000. A plant virus vector system for expres- sion of foreign genes in cereals. The Plant Journal 23: 547-556. Chowrira, G., V Akella, and P.F. Lurquin. 1993. Transformation of peas and lentils by in vivo electroporation of nodal meristems. Western Society of Crop Science Abstracts. 2p. Chowrira, G., V Akella, and P.F. Lurquin. 1995. Electroporation-mediated gene transfer into intact nodal meristems in planta: Generating transgenic plants without in vitro tissue culture. Molecu- lar Biotechnology 3: 17-23. Chowrira, G., V Akella, P.E. Fuerst, and P.F. Lurquin. 1996. Transgenic grain legumes obtained by in planta electroporation-mediated gene transfer. Molecular Biotechnology 5: 85-96. Christou, P. 1992. Genetic engineering and in vitro culture of crop legumes. Technomic Publishing, Pennsylvania, USA. 307 p. Clough, S.l and A.F. Bent. 1998. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabi do psis thaliana. The Plant Journal 16: 735-743. 193 Digitized by Google Biotechnology for cowpea Dessens, J. T. and G.P. Lomonossoff. 1993. Cauliflower mosaic virus 35S promoter-controlled DNA copies of cowpea mosaic virus RNAs are infectious on plants. Journal of General Virology 74: 889-892. Dillen, w., G. Engler, M.C. Van Montagu, and G. Angenon. 1995. Electroporation-mediated DNA delivery to seedling tissues of Phaseolus vulgaris L. (common bean). Plant Cell Reports 15: 119-124. Eu, YI, M.H. Lee, H.S. Chang, T.H. Rhew, H. Y Lee, and C.H. Lee. 1998. Chlorophyll fluorescence assay for kanamycin resistance screening in transgenic plants. Plant Cell Reports 17: 189-194. Filippone, E. 1990. Genetic transformation of pea and cowpea by co-cultivation of tissues with Agrobacterium tumefaciens carrying binary vectors. Pages 175-181 in Cowpea genetic resources, contributions in cowpea exploration, evaluation and research from Italy and I1TA, edited by N.Q. Ng and L.M. Monti. I1TA, Ibadan, Nigeria. Filippone, E. and R Penza. 1992. Agrobacterium tumefaciens-mediated gene transfer. Pages 197-202 in Biotechnology: enhancing research on tropical crops in Africa, edited by G. Thot- tappilly, L.M. Monti, D.R. Mohan Raj, and A.W. Moore. CTAIIITA copublication. I1TA, Ibadan, Nigeria. Finer, J.J., K.R Finer, and T. Ponappa. 2000. Particle bombardment-mediated transformation in Current topics in microbiology and immunology plant biotechnology: new products and applica- tions, edited by J. Hammond, P.B. McGarvey, and V Yusibov. Springer-Verlag Vol. 240. Fletcher, C. and E.M. Meyerowitz. 2000. Cell signaling within the shoot meristem. Current Opinion in Plant Biology 3: 23-30. Ganapathi, A. and P. Anand. 1998. Somatic embryogenesis from young leaves of cowpea (Vigna unguiculata (L.) Walp. (Abstracts) in Plant Biotechnology and in vitro biology for the 21st Century. IX International Congress on Plant Tissue and Cell Culture, 14-19 June 1998, Jerusa- lem, Israel, Garcia, IA., J. Hillie, and R Goldbach. 1986a. Transformation of cowpea Vigna unguiculata cells with an antibiotic resistance gene using a Ti-plasmid-derived vector. Plant Science 44: 37-46. Garcia, JA, J. Hillie, and R Goldbach. 1986b. Transformation of cowpea Vigna unguiculata cells with a full length DNA copy of cowpea mosaic virus m-RNA. Plant Science 44: 89-98. Gelvin, S.B. 1998. The introduction and expression of transgenes in plants. Current Opinion in Biotechnology 9: 227-232. Hadi, M.Z., MD. McMuller, and J.J. Finer. 1996. Transformation of 12 different plasmids into soybeans via particle bombardment. Plant Cell Reports 15: 500-505. Hansen, G. and MD. Chilton. 1996. "Agrolistic" transformation of plant cells: integration of T- strands generated in planta. Proceedings National Academic Science USA 93: 14978-14983. Hansen, G. and S.M. Wright. 1999. Recent advances in the transformation of plants. Trends in Plant Science 4: 226-231. Horsch, RB., IE. Fry, N. Hoffman, D. Eichlz, S.G. Rogers, and RT. Fraley. 1985. A simple and general method oftransferring genes into plants. Science 227: 1229-1231. Ikea, I 1998. Transformability of cowpea (Vigna unguiculata L. Walp.) by particle bombardment. PhD Thesis, The University ofIbadan, Ibadan, Nigeria. Jefferson, RA. 1989. The gus reporter gene system. Nature 342: 837-838. Joersbo, M. and I Brunstedt. 1990. Quantitative relationship between parameters of electroporation. Journal of Plant Physiology 137: 169-174. Joersbo, M. and J. Brunstedt. 1991. Electroporation: Mechanism and transient expression, stable transformation and biological effects in plant protoplasts. Physiologia Plantarum 81: 256-264. Joersbo, M., I. Donaldson, I Kreiberg, S.G. Petersen, I Brunstedt, and F.T. Okkels. 1998.Analysis of man nose selection used for transformation of sugar beet. Molecular Breeding 4: 111-117. Kaepplar, H.F., W. Gu, DA. Somers, H.W Rines, and A.F. Cockburn. 1990. Silicon carbide fibre- mediated DNA delivery into plant cell. Plant Cell Reports 8: 415-418. 194 Digitized by Google Regeneration and genetic transformation in cowpea Kartha, K.K., K. Pahl, N.L. Leung, and L.A. Mroginski. 1981. Plant regeneration from meristems of grain legumes: soybean, cowpea, peanut, chickpea, and bean. Canadian Journal of Botany 59: 1672-1574. Komari, T., Y Hiei, Y Ishida, T. Kumashiro, and T. Kubo. 1998. Advances in cereal gene transfer. Current Opinion in Plant Biology 1: 161-165. Kononowicz,A.K., K. T. Cheah, M.L. Narasimhan, L.L. Murdock, RE. Shade, M.J. Chrispeels, E. Filippone, L.M. Monti, RA. Bressan, and P.M. Hasegawa. 1997. Developing a transformation system for cowpea (f/igna unguiculata L.Walp.). Pages 361-371 in Advances in Cowpea Research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublica- tion of International Institute of Tropical Agriculture (ilIA) and Japan International Research Center for Agricultural Sciences (J1RCAS). I1TA, Ibadan, Nigeria. Lurquin, P.p. 1997. Gene transfer by electroporation. Molecular Biotechnology 7: 15-35. Machuka, J. 2000. Cowpea regeneration in planta: can it be coupled to transformation? Page 20 in Abstracts, World Cowpea Research Conference III, 4-7 September 2000, Ibadan, Nigeria. Masuta, C., T. Yamana, Y Tacahashi, I. Uyeda, M. Sato, and T. Matsumura. 2000. Development of clover yellow vein virus as an efficient stable gene expression system for legume species. The Plant Journal 23: 547-556. McCabe, D.E., W.F. Swain, B.J. Martinell, and P. Christou. 1998. Stable transformation of soybean (Glycine max) by particle acceleration. Biotechnology 6: 923-926. Monti, L.M., L.L. Murdock, and G. Thottappilly. 1997. Opportunities for biotechnology in cowpea. Pages 341-351 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (ilIA) and Japan International Research Center for Agricultural Sciences (J1RCAS). ilTA, Ibadan, Nigeria. Murashige, T. and F. Skoog. 1962. A revised medium for rapid growth and bioassays and with tobacco tissue culture. Physiologia Plantarum 15: 473-497. Muthukumar, B., M. Mariamma, andA. Gnanam. 1995. Regeneration of plants from primary leaves of cowpea. Plant Cell Tissue and Organ Culture 42: 153-155. Muthukumar, B., M. Mariamma, K. Valuthambi, and A. Gnanam. 1996. Genetic transformation of cotyledon explants of cowpea (f/igna unguiculata L. Walp.) using Agrobacterium tumefaciens. Plant Cell Reports 15: 980-985. Neuhaus, G. and G. Spangenberg. 1990. Plant transformation by microinjection techniques. Phys- iologia Plantarum. 79: 213-217. Neuhaus, G., G. Spangenberg, O. Mittelsten-Scheid, and H.G. Schweizer. 1987. Transgenic rape seed plants obtained by the microinjection of DNA into microspore-derived embryoids. Theo- retical and Applied Genetics 75: 30-36. Obembe, 0.0., M. Kadiri, and J. Machuka. 2000a. Induction of multiple shoots and regeneration from cotyledonary nodes and epicotyls. Page 20 in Abstracts, World Cowpea Research Confer- ence III, 4-7 September 2000, Ibadan, Nigeria. 32 p. Obembe, O.O.,A.Adesoye, M. Kadiri, and J. Machuka. 2000b. The role of antibiotics in the control of Agrobacterium and as potential selective agents in Agrobacterium-mediated transformation of cowpea. Page 19 in Abstracts, World Cowpea Research Conference III, 4-7 September 2000, Ibadan, Nigeria. 32 p. Pellegrineschi,A. 1997. In vitro plant regeneration via organogenesis of cowpea [f/igna unguiculata (L.) Walp.]. Plant Cell Reports 17: 89-95. Pellegrineschi,A., C.A. Fatokun, G. Thottappilly, andA.A.Adepoju. 1997. Cowpea embryo rescue. Influence of culture media composition on plant recovery from isolated immature embryos. Plant Cell Reports 17: 133-138. Penza, R, V Akella, and P.F. Lurquin. 1992. Transient expression and histological localization of a gus chimeric gene after direct transfer to mature cowpea embryos. Biotechniques 13: 576- 580. 195 Digitized by Google Biotechnology for cowpea Penza, R, P.F. Lurquin, and E. Filippone. 1991. Gene transfer by co-cultivation of mature embryos withAgrobacterium tumefaciens: application to cowpea (Vigna unguiculata. L. Walp.). Journal of Plant Physiology 138: 39-43. Perkins, E.J., C.M. Stiff, and P.F. Lurquin. 1987. Use of Alcaligenes eutropus as a source of genes for 2-, 4-D resistance in plants. Weed Science 35: 12-18. Puonti-Kaerlas,I 1993. Methods in grain legume transformation. Grain Legumes 2: 14-15. Puonti -Kaerlas, 1, T. Eriksson, and P. Engstrom. 1990. Production oftransgenic pea (Pisum sativum L.) plants by Agrobacterium tumefaciens-mediated gene transfer. Theoretical and Applied Genet- ics 84: 443-450. Rasco-Gaunt, S., A. Riley, P. Lazzeri, and P. Barcelo. 1999. A facile method for phosphoinothricin (pPT)-resistant wheats. Molecular Breeding 5: 255-262. Russell, D.R., K.M. Wallace, IH. Bathe, B.J. Martinell, and D.E. McCabe. 1993. Stable transforma- tion of Phaseolus vulgaris via electric discharge-mediated acceleration. Plant Cell Reports 12: 165-169. Saedi, D., G. Bruening, C.1. Kado, and IC. Dutra. 1979. Tumor induction by Agrobacterium tume- faciens prevented in Vigna sinensis seedlings systemically infected by ribonucleic acid viruses. Infection and Immunology 23: 2 298-304. Shillito, R 1999. Methods of genetic transformations: electroporation and polyethylene glycol treatment. Pages 9-20 in Molecular improvement of cereal crop, edited by I. Vasil. Kluwer, Dordrecht, The Netherlands. Singh, B.B., O.L. Chambliss, and B. Sharma. 1997. Recent advances in cowpea breeding. Pages 30-50 in Advances in Cowpea Research edited, by B.B. Singh, D.R Mohan Raj, K.E. Dashiell and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (ilTA) and Japan International Research Center for Agricultural Sciences (JIRCAS), Ibadan, Nigeria. Subramaniam, M.K., S. Subramaniam, P.M. Gopinath, K.C. Gupta, and S. Vasantha. 1968. Histo- genesis, organogenesis and morphogenesis in callus cultures ofTrigonellafoenum-graecum and Vigna unguiculata L. Current Science 37: 398-399. Suzuki, H., T. Fowler, and M. Tierney. 1993. Deletion analysis and localization ofSbPRP1, a soy- bean cell wall protein gene, in roots oftransgenic tobacco and cowpea. Plant Molecular Biology 21: 109-119. Trick, N.U., RD. Dinkins, E.R Santarem, RD.V Samoylo, CA. Meurer, D.R Walker, W.A. Parrot, 11 Finer, and G.B. Collins. 1997. Recent advances in soybean transformation. Plant Tissue Culture and Biotechnology 3(1): 9-26. Trieu,A.T., S. H. Burleigh, I.E. Kardailsky, w.K. Maldonado-Mendoza, L.A. Versaw, H. Shin,A. Blaylock, T. Chion, H. Katagi, G.R Dewbare, D. Weigel, and M.J. Harrison. 2000. Transforma- tion of Medicago truncata via infiltration of seedlings or flowering plants with Agrobacterium. The Plant Journal 22 (6): 531-541. Zupan, J., T.R Muth, O. Draper, and P. Zambryski. 2000. The transfer of DNA from Agrobacterium into plants: a feat of fundamental insights. The Plant Journal 23: 11-28. 196 Digitized by Google 3.3 Molecular cloning in cowpea: perspectives on the status of genome characterization and gene isolation for crop improvement M.P. Timko1 Abstract Cowpea (Vigna unguiculata L.) is a grain legume of significant economic impor- tance worldwide and for many people in the semiarid areas of West and Central Africa, it is the major source of dietary protein necessary for human nutrition. As a result of its widespread use and economic importance, numerous programs aimed atthe improvement of various agronomic and nutritional quality traits are underway. Included among these initiatives are selective breeding programs aimed at identity- ing new sources of disease and pest resistance from wild species for introgression into cultivated varieties, gene isolation, and characterization studies aimed at under- standing the factors controlling plant growth and development, as well as plant cell culture, and genetic transformation programs aimed at the direct manipulation of traits through genetic engineering. Recent progress on genome organization and evolution, and gene characterization in cowpea is reviewed and prospects for future improvement of cowpea through biotechnological applications are discussed. Introduction Cowpea (Vigna unguiculata [L.] Walp.) is a food legume of significant economic impor- tance worldwide. It is grown in North and South America, Africa, Europe, and Asia, primarily in the semiarid and humid tropical regions lying between 35 oN and 30 Os of the equator. It is estimated that cowpea is now cultivated on at least 12.5 million hectares with an annual production of over 3 million tonnes of grains worldwide (Singh et al. 1997). Currently, Central and West Africa account for more than 64% of the total area under cowpea cultivation, followed by South America, Asia, East, and South Africa (F ery 1985; Singh et al. 1997). In the United States, cowpea is a crop of minor significance, grown on just over 80000 hectares (Fery 1985; 1990). Part of the popularity of cowpea as a food staple for people in the semiarid and humid tropical regions of Africa stems from the fact that it is relatively drought-tolerant, perform- ing well under conditions where most other food legumes do not. Its unique ability to fix nitrogen even in very poor soils with low organic matter also contributes to its widespread use among farmers (Singh et al. 1997). Like most crop plants, cowpea production is limited by numerous biotic and abiotic factors. Both severe heat and drought limit cowpea pro- ductivity (Marfo and Hall 1992). Cowpea is also attacked by a wide range of insect pests, microbial and fungal diseases, nematodes, and two different parasitic angiosperms (Bashir and Haptom 1996; Ehlers and Hall 1997; Fery and Singh 1997; Singh and Emechebe 1997). 1. Department of Biology, University of Virginia, Gilmer Hall 044, Charlottesville, Virginia 22903. 197 Digitized by Google Biotechnology for cowpea As a result of its widespread use, nwnerous programs directed at the improvement of agronomic and nutritional quality traits are underway, including breeding programs aimed at screening wild and cultivated germplasm for sources of disease and pest resistance genes as well as plant cell culture and genetic transformation programs aimed at direct manipulation of traits through genetic engineering. With few exceptions, the application of biotechnology to cowpea improvement offers the promise of increased productivity by speeding the development of varieties that yield more, are more resistant to biotic and abiotic stresses, and are more economical and efficient to produce. This paper reviews some recent advances in genome characterization, gene isolation, and genetic manipula- tion of cowpea, and offers perspectives on emerging areas of research. Discussions on other related aspects of cowpea research, including genetics, breeding, and cell culture are also in this paper. Phylogeny and genome organization Cowpea (Vigna unguiculata) is one of several important cultivated species which con- stitute the genus Vigna. Other members include mung bean (~~ radiata), azuki bean (~~ angularis), blackgram (~~ mungo), and the bambara groundnut (~~ subterranea). The genus was initially divided into several subgenera by Marechal et al. (1978) based upon morphological characteristics, extent of genetic hybridization/reproductive isolation, and geographic distribution of species. The major groupings consist of the African subgenera Vigna and Haydonia, the Asian subgenus Ceratotropis, and the American subgenera Sig- moidotropis and Lasiopron. Under the scheme proposed by Marechal and his colleagues, cultivated cowpea was placed in the subgenus Vigna, whereas mung bean and blackgram were placed in the Asian subgenus. The development and use of biochemical-based analytical techniques and molecular marker technologies, such as analysis of restriction fragment length polymorphisms (RFLPs), random-amplified polymorphic DNAs (RAPDs) (Williams et al. 1990), ampli- fied fragment length polymorphisms (AFLPs) (Vos et al. 1995), minisatellites (Sonnante et al. 1994), and simple sequence repeats (SSRs) (Akkaya et al. 1992, 1995), have greatly facilitated the analysis of the structure of plant genomes and their evolution and have contributed significantly to our understanding of cowpea genome organization. Using RFLP analysis, Fatokun et al. (1993a) analyzed 18 Vigna species including five of the subgenus Ceratotropis to determine the taxonomic relationship between the subgenus Ceratotropis and other subgenera. These investigators showed that a high level of genetic variation exists within the genus, with a remarkably higher amount of variation associated with Vigna species from Africa relative to those from Asia. Their data supported the taxonomic separation of the Asian and African genera as proposed by Marechal et al. (1978) and underscored the previously held viewpoint that Africa is likely to be the center of diversity for Vigna. Generally, the placement of species and subspecies based upon molecular taxonomic procedures by Fatokun et al. (1993a) substantiated prior clas- sifications based on classical taxonomic criteria, such as morphological and reproductive traits. Genetic variation in the subgenus Ceratotropis was subsequently reinvestigated by Kaga et al. (1996a) using RAPD analysis. Examining 23 accessions of five species within the subgenus Ceratotropis for polymorphisms, these investigators identified approxi- mately 404 amplified fragments capable of providing comparative information. Based 198 Digitized by Google Molecular cloning in cowpea on the degree of polymorphism at these informative loci, these investigators were able to separate the accessions into two main groups differing by approximately 70% at the molecular level. Within each of the main groups, the accessions could be further divided into five subgroups which composition were in complete agreement with their taxonomic species classifications. Sonnante et al. (1996) examined isozyme variation between v~ unguiculata and other species in the subgenus Vigna and showed that v~ unguiculata was more closely related to v~ vexillata, a member of the subgenus Plectotropis, than to any other species belonging to the section Vigna. This is not surprising since v~ vexillata is thought to be the intermediate species between African and Asian Vigna species. Vaillancourt and Weeden (1996) reached a similar conclusion. Based on their analysis of variation in chloroplast DNA structure (Vaillancourt and Weeden 1992), and isozyme polymorphisms (Vaillancourt et al. 1993), these investigators suggested that v~ vexillata and v~ reticulata were the closest relatives of v~ unguiculata. While the close relationship between v~ unguiculata and v~ vexillata proposed by Vaillancourt and Weeden (1996) is consistent with previous observations (Marechal et al. 1978), v~ reticulata was placed in a different cluster based upon RFLP analysis (Fatokun et al. 1993a). Polymorphisms in 21 different enzyme systems were used by Pasquet (1999) to evaluate the relationship among 199 accessions of wild and cultivated cowpea differing in breeding system and growth characteristic (i.e., annual versus perennial growth habit). Based on these allozyme data, perennial subspecies of cowpea (spp. unguiculata var. unguiculata) were shown to form a coherent group closely related to annual forms (ssp. unguiculata var. spontanea). Among the 10 subspecies studied, v~ unguiculata var. spontanea and ssp. pubescens were the closest taxa to be cultivated into cowpea. Most recently, Ajibade et al. (2000) used inter simple sequence repeat (lSSR) DNA polymorphism analysis to study the genetic relationships among 18 v Igna species. The study showed that closely related species within each subgenus clustered together (e.g., v~ umbellata and v~ angularis (subgenus Ceratotropis), v~ adenantha and v~ caracalla (subgenus Sigmoidotropis), and v~ luteola and v~ ambacensis (subgenus Vigna). Cultivated cowpea was grouped closely with the wild subspecies of v~ unguiculata, and the entire species was separated from its most closely allied species v~ triphylla and v~ reticulata. ISSR polymorphism analysis split Vigna into groupings that differed in their composition from previous classifications. For example, the subgenus Vigna was split into three lineages, with v~ unguiculatalreticulatalfriesorum forming one group, v~ luteolalambacensis forming a second, and v~ subterranea being far from the other two. Ceratotropis was split into two sections, with three species (v~ radiata, v~ mungo, and v~ aconitifolia) in one section and two species (v~ angularis and v~ umbellata) in a second section. While such groupings had been suggested previously (Marechal et al. 1978; F atokun et al. 1993a; Vaillancourt and Weeden 1996), it should be noted that ISSR analysis was not as effective at resolving genetic distance relationships at the sub generic level as it was at resolving relationships at the species level and below. Therefore, the authors note that their conclusions regarding subgeneric classifications should be taken with caution. Thus, there is still considerable need to develop appropri- ate strategies and molecular techniques to resolve exact taxonomic relationships among members of this important genus. Repetitive DNA sequences have been shown to represent a substantial fraction of the nuclear genome of all higher plant species and to account for much of the variation in 199 Digitized by Google Biotechnology for cowpea genomic DNA content observed among species (Flavell et al. 1994). Many of the repeat sequences found in plant genomes appear to have originated through the activity of trans- posable elements (transposons), that move either by first forming an RNA intermediate (i.e., retrotransposons [Boeke et al. 1985]) or by direct DNA transposition intermediates (i.e., transposons [Federoff 1989]). To gain insight into the genomic organization and evolution of species within Vigna, Galasso et al. (1997) examined the genomic organiza- tion and distribution of Ty l-copia type retrotransposons in seven different species and subspecies of Vigna and several related leguminous plants. Gel blot analysis of genomic DNA from ~~ unguiculata, v~ luteola, v~ oblongifolia, ~~ ambacensis, and v~ vexillata probed with radioactively-labeled probes to the reverse transcriptase gene amplified from ~~ unguiculata subsp. unguiculata, ~~ unguiculata subsp. dekindtiana, ~~ luteola, and ~~ vexillata, showed variable hybridization patterns and intensities generally correlating with their previously defined taxonomic position. Fluorescent in situ hybridization analysis of the distribution of the Ty l-copia type sequences showed that these elements represented a major fraction of the cowpea genome and were dispersed relatively uniformly over all the chromosomes. Little or no hybridization was found associated with centromeric, subtelomeric, and nucleolar organizing regions of the chromosomes, indicating that these portions of the genome may not be suitable sites for transposition. Comparisons of ret- rotransposon structural similarity between Vigna and other genera of legumes generally supported the subdivision of the tribes Phaseoleae and Vicieae, with greater homology seen between members of the Cicereae and P haseoleae than Cicer species and those from the Vicieae (Galasso et al. 1997). In addition to providing insight into phy logenetic relationships, molecular marker tech- nologies have also been used in the construction of genetic maps for most of the important crop species, including cowpea. The first attempt to generate a comprehensive linkage map for cowpea was by F atokun et al. (1993b) who used polymorphisms detected by 87 random genomic DNA fragments, five cDNAs, and RAPDs to generate a map consist- ing of ten linkage groups spanning 680 cM. Improvement on this initial map was made by Menendez et al. (1997) who were able to develop a linkage map for ~~ unguiculata consisting of 181 loci falling into 12 linkage groups. The resolution of the map was to approximately 6.4 cM between loci. Similarly, Menancio-Hautea et al. (1993a, b) used RFLP analysis to construct a genome map of mung bean (v~ radiata). The map consisted of 172 markers placed into 11 linkage groups and provided 1570 cM coverage with an average distance of 9 cM between loci. It is worth noting that even at these early stages of genome comparison, significant colinearity was recognized between the cowpea and mung bean genomes (Menacio-Hautea et al. 1993b). A total of 132 markers (108 RAPDs, 19 RFLPs, and five morphological markers) have been mapped in azuki bean using an interspecific population generated from a cross of ~~ angularis x ~~ nakashimae (Kaga et al. 1 996b ). Comparison of the linkage map of azuki bean with those of mung bean and cowpea using 20 RFLP markers indicated that, as might be expected, the three genomes have many linkage blocks in common. Among the most recent developments in understanding cowpea genome organization is the report by Li et al. (1999) who used DNA amplification fingerprinting (DAF) and AFLP analysis to identify additional molecular markers segregating in the F 8 recombinant inbred population derived from a cross between IT84S-2049 and 524B (Menendez et al. 1997). These researchers screened 400 randomly generated DAF decamers and 128 200 Digitized by Google Molecular cloning in cowpea AFLP primer combinations, and were able to place 57 DAF and 90 AFLP markers to the existing cowpea genetic map. Studies are underway to further saturate the map with additional markers to increase its utility for future map-based cloning activities in cowpea. Additionally, a map of the wild relative of cowpea v~ vexillata has also been generated (Ogundiwin et al. 2000) adding even greater breadth to our understanding of genomic relationships in Vigna. The considerable progress made in recent years on the development of genomic maps for cowpea and related species is reflected in the ever increasing number of growth, yield, and resistance trait loci that have now been located within the various genomes (F atokun et al. 1992, 1997; Myers et al. 1996; Roberts et al. 1996; Menendez et al. 1997; Ouedraogo et al. 2001; Gowda et al. 2002). Table 1 lists the various agronomic and disease resistance trait loci that have now been placed on the cowpea genetic map. Table 1. Agronomic, growth habit, and disease and pest resistance trait loci currently placed on the cowpea genetic mapt. Linkage group Locus designation LG1 LG2 LG3 LG5 LG6 Resistance LG7 LG8 LG9 LG12 PodL SW C P Er Rac1 (Rac2) Rsg1, Rsg2, Rsg4 PodN (NTF) CPSMV (ims) FusR CPMV RGA-434 RGA-438,468,490 SBMV(sbc-l,2) PodN SW SW Rk (NemR) BECMV Maturity 50% FL SW Dehydrin Height PodN GluC Character or function Pod length Seed weight (100 seed) General flower color factor Pod pigmentation Pod attachment (erect pod) Resistance to Aphis craccivora (aphid) Race-specific resistance to Striga gesnerioides Pod number per plant Nodes to 1st flower (D1301a) Cowpea severe mosaic virus resistance Resistance to Fusarium oxysporum Cowpea mosaic virus Resistance gene analogs (pathogen unknown) Resistance gene analogs (pathogen unknown) Resistance to southern bean mosaic virus Pod number per plant Seed weight (100 seed; OB6a) Seed weight (100 seed) Root-knot nematode (Meloidogyne incognita) Resistance to blackeye cowpea mosaic virus Maturity 50% flowering Seed weight Dehydrin protein Plant height Pod number per plant tData taken from Fatokun et al. (1992, 1993b, 1997), Myers et al. (1996); Roberts et al. (1996); Menendez et al. (1997); Ouedraogo et al. (2001); Gowda et al. (2002). 201 Digitized by Google Biotechnology for cowpea Gene isolation and characterization Developing innovative biotechnologies for cowpea improvement requires not only an understanding of genome organization and complexity, but also of gene structure and function. In the genus Vigna, only limited progress has been made in basic gene discovery and only a modest number of studies have appeared in the literature examining differ- ential gene regulation during growth and development or in response to biotic or abiotic stress. Table 2 summarizes the number of nucleotide and translated protein sequences currently available from cowpea and related species in the genus Vigna in comparison to other legumes and major crops. Mung bean (~~ radiata) and cowpea (v~ unguiculata) lead among Vigna species in the number of gene sequences available to researchers. Basic information available for these species is between 2- and 5-fold lower than that available for pea (Pisum sativum), French bean (Phaseolus vulgaris), and alfalfa (Medicago sati- vum), and over 500-fold lower than currently available for soybean (Glycine max). A large proportion of the nucleic acid sequences present in the databases for cowpea and mung bean are either ribosomal RNA coding and spacer regions or nuclear genomic sequences of unidentified function developed for RFLP mapping, further accentuating Table 2. Number of nucleotide and deduced amino acid sequences available for various Vigna species and related plantst. Vigna species Vigna aconitifolia -11 (23) Vigna adenantha - 3 (0) Vigna angularis (azuki bean) - 39 (29) Vigna caracalla - 1 (0) Vigna gentryi - 1 (0) Vigna glabrescens - 2 (0) Vigna hosei - 2 (0) Vigna kirkii - 2 (0) Vigna lasiocarpa - 2 (0) Vigna linearis - 1 (0) Vigna lobatifolia - 2 (0) Vigna longifolia - 2 (0) Vigna luteola - 2 (0) Vigna membranacea - 2 (0) Vigna minima - 2 (0) Vigna multinervis - 2 (0) Vigna mungo (blackgram) -17 (20) Vigna mungo subsp. sylvestris - 2 (0) Other plant species Arabidopsis thaliana -180,056 (40,933) Arachis hypogaea (peanut) - 46 (121) Cajanus cajan (pigeon pea) - 9 (11) Canavalia gladiata (sword bean) -10 (16) Cicer arietinum (chick pea) - 229 (218) Glycine max (soybean) - 123,492 (1758) Lathyrus sativus (chickling vetch) -1 (4) Vigna oblongifolia - 2 (0) Vigna parviflora - 1 (0) Vigna parvifolia - 1 (0) Vigna peduncularis - 2 (0) Vigna populnea - 1 (0) Vigna racemosa - 2 (0) Vigna radiata (mung bean) -192 (167) Vigna radiata subsp. sublobata - 2 (0) Vigna reticulata - 2 (0) Vigna sinensis (cowpea) -5 (0) Vigna speciosa - 3 (0) Vigna subterranea (ground-bean) -1 (1) Vigna trilobata - 2 (0) Vigna triphylla - 2 (0) Vigna umbellata - 3 (1) Vigna unguiculata (cowpea) -148 (128) Vigna vexillata - 8 (1) Lens culinaris (lentil) - 20 (34) Medicago sativa (alfalfa) - 1024 (604) Phaseolus vulgaris (French bean) - 462 (490) Pisum sativum (pea) -1081 (1662) Vicia faba (fava bean) - 247 (380) Zea mays -77,295 (3337) tData taken from genebank sequence database, National Center for Biotechnology Information, 1 September 2000. 202 Digitized by Google Molecular cloning in cowpea the paucity of genetic information. Thus, there is a need for more research in basic gene discovery for cowpea. The gene products characterized from cowpea thus far fall into one of several categories based on their confirmed or predicted function (Table 3). Among the largest number of nucleotide sequences available are those encoding rRNA (either of nuclear or plastid origin) and their associated intergenic spacer regions, randomly generated nuclear frag- ments used for RFLP analysis, and cDNAs generated from differential display analysis of root-hairmRNAs collected 24 hours after inoculation with Rhizobium sp. NGR234. These are followed by sequences encoded by genes turned on in response to pathogen attack (e.g., acidic and basic chitinase, pathogenesis-related proteins, and various resistance gene analogs) or in response to abiotic stress such as drought and low temperature (e.g., dehydrin, acid phosphatases, and phospholipases, seed associated proteins (trypsin inhibi- tors, a-amylase), and general metabolic enzymes. Table 3. Summary of gene products characterized from cowpeat • Gene designation Seed associated proteins Accession number(s) a-amylase (AJ225087) Asp-protease (U61396) Arcelin 9 (AF147793) cpi (Z21954) tpi-f IV (X51617) tpi-f IV (X51618) Stress response CpABA1 (AB030295) CPRD8 (D83970) CPRD12 (D88121) CPRD14 (D83971) CPRD22 (D83972) CPRD46 (D88122) CPRD65 (AB030293) CPRD86 (AB030294) Dhn1 (AF159804) papB (AF171230) papA (AF165891) PplC (U85250) PplD (U92656) Identity or function a-amylase Aspartic proteinase Homolog of phaseolus arcelin gene Cysteine proteinase inhibitor Bowman-Birk type trypsin inhibitor (f IV) Bowman-Birk type trypsin inhibitor (f IV) Water stress-inducible genes in the highly drought-tolerant cowpea Water stress-inducible genes in the highly drought-tolerant cowpea Water stress-inducible genes in the highly drought-tolerant cowpea Water stress-inducible genes in the highly drought-tolerant cowpea Water stress-inducible genes in the highly drought-tolerant cowpea Water stress-inducible gene for neoxan thin cleavage enzyme involved in abscisic- acid biosynthesis under water stress Water stress-inducible genes in the highly drought-tolerant cowpea Water stress-inducible genes in the highly drought-tolerant cowpea Chilling tolerance induced dehydrin Phosphatidic acid phosphatase B Phosphatidic acid phosphatase B Phosphoinositide-specific phospholipase C Water stress-induced phospholipase D ... continued 203 Digitized by Google Biotechnology for cowpea (Table 3 continued) Gene designation Accession number(s) Nodulation-nitrogen fixation AKC59 (X79604) flbr lbl lbll Lbll sod E5T1-27 Agronomic traits Vu-Yld (AF181 096) (U33206) (U33207) (2033205) (AF077224) (AI 759142- AI 759148) (AI 755286- AI 755305) (AB028025) Identity or function Nodulation associated lipid transfer protein. Ferric leghemoglobin reductase leghemoglobin I leghemoglobin II leghemoglobin II Iron-superoxide dismutase precursor from root nodules cDNA sequences from display analysis of mRNA collected 24 hr following inocula- tion with Rhizobium sp. NGR234 Regulatory protein for wall yield threshold (yielding) Resistance gene products and pathogen-induced genes chi1 (X88800) Chitinase class 1 chi3 (X88802) Acidic chitinase class 3 chi3B (X88801) Basic chitinase class 3 chi4 (X88803) Chitinase class 4CH5 (X74821) Chalcone synthase CRGA1-8 (AB020483- Nucleotide-binding site sequence A B020490) containing resistance KIND11,12 KINE12 loc431-490 PAL POX PR3 PR4.2 51-1 51-3 (AF141011, AF141012) (AF141 013) (AF255460- AF255467) (AF165998) (U61379) (AB027154) (X98608) (AB038691) (AB038692) MitochondriaVplastid localized functions atpA (AF141143) cox2 (AF211254) cpF2 cpF3 (AF052058) (AF052057) gene analogs; CRGAS is linked to Cry1 (CMV strain Y) resistance locus Resistance gene protein homolog Resistance gene protein homolog Nucleotide-binding site sequence containing resistance gene analogs Phenylalanine ammonia-lyase Ascorbate peroxidase Pathogenesis-related protein PR3 Pathogenesis-related protein PR4.2 Cucumber mosaic virus infection induced mRNA Cucumber mosaic virus infection induced mRNA Chloroplast ATP synthase B subunit Mitochondtial cytochrome c oxidase subunit 2 Chloroplast associated ferritin subunit precursor, nuclear gene 2 Chloroplast associated ferritin subunit precursor, nuclear gene 3 ... continued 204 Digitized by Google (Table 3 continued) Gene designation psbA rbcL rp116, rpl14 Accession number(s) (X80932) (Z95543) (M80799) Molecular cloning in cowpea Identity or function Photosystem II D1 protein Large subunit, ribulose bisphosphate carboxylase Chloroplast ribosomal proteins L 16 and L 14 General cellular and metabolic functions A3 (X90487) Unknown protein (A3 gene) ARF (AF022389) ADP-ribosylation factor (ARF) cdc2 (X89400) Protein kinase (cdc2) homolog cp-wap11 (AF005278) Type lila Golgi-associated membrane cp-wap13 ext127 ext3 ext26 ext26G GRP pfe1,pfe2, pfeS pur1 pur2 pur3 Vupur3 purS SSIII SSV Ted2 AG81-1 Centromeric DNA Ty1-copia-1i ke Unknown (AF005279) (X86028) (X86029) (X86030) (X91836) (X87948) (X67754- X67757) (AF071862) (U30896) (AF160196) (U30875) (U30895) (AJ225088) (AJOO6752) (Y088624) (AF062782) (Z49817) (Y12763, Y12764) (AZ254216- AZ254227) protein Type lila Golgi-associated membrane protein Extensin-like protein127 Root-hair-specific extensin-like protein Root-hair-specific extensin-like protein Extensin 26G gene Glycin-rich protein Ferritin gene exons 1 and 2 Phosphoribosylpyrophosphate amido- transferase Glycinamide ribonucleotide (GAR) synthetase Glycinamide ribonucleotide transformylase Glycinamide ribonucleotide transformylase Aminoimidazole ribonucleotide (AIRS) synthe- tase Starch synthase isoform III; ADP-glucose-starch gl ucosyl transferase Starch synthase isoform V; ADP-glucose-starch gl ucosyl transferase Ted2 protein homolog to marker gene for differentiation Microsatellite AG81-1 repeat region Satellite DNA (centromeric region) Ty1-copia-like retrotransposable element repeat region RFLP sequences of cowpea Vigna unguiculata genomic DNA tData taken from genebank sequence database, National Center for Biotechnology Information, 1 September 2000. 205 Digitized by Google Biotechnology for cowpea Plant cell transformation and cell culture Progress in cowpea improvement over the past several decades relied largely upon tra- ditional selection and breeding strategies for the introduction of new traits into existing cultivars. Excellent discussions of previous and current cowpea breeding activities can be found elsewhere in this volume. With the advent of molecular techniques for gene isolation and gene transfer among species, plant breeders now have at their disposal the ability to rapidly move single gene characteristics among agronomically preferred cultivars. More importantly, the ability to introduce genes into plant cells from distant genera and even other kingdom (e.g., genes of nonplant origin) allow researchers to bypass interspecific barriers which often stymied efforts to introduce desirable traits from wild species into preferred cultivars. Transgenic approaches essentially expand the genepool to include all available genetic information, whether naturally occurring or synthetically created. In order to take full advantage of transgenic approaches for crop improvement, it is necessary to ensure efficient and reproducible methods for gene transfer (i.e., plant cell transformation) and the identification and recovery of transgenic plants. Attempts to estab- lish procedures for plant transformation in cowpea have met with mixed success. Garcia et al. (1986, 1987) reported obtaining transgenic calli and chimeric plantlets following Agrobacterium-mediated leaf-disc transformation of v~ unguiculata. Similar findings were reported by Penza et al. (1991) followingAgrobacterium-mediated transformation of axil- lary buds and embryonic tissues. However, the ability to produce mature transgenic plants with these procedures was never confirmed. Several groups (Finer et al. 1992; Penza et al. 1992; Kononowicz et al. 1997) have attempted to introduce foreign DNA into cowpea leaf tissues and embryos by microprojectile bombardment (biolistics). These research- ers obtained high levels of transient expression of the B-glucuronidase (gus) transgene, but were unable to regenerate plantlets from the transformed cells. Similarly, Akella and Lurquin (1993) described the expression of B-glucuronidase (GUS) activity in a variety of tissues following electroporation of embryos with plasmid DNA. Unfortunately, it was not possible to produce mature transgenic plants that stably inherited the transgene. In contrast, Muthukumar et al. (1996) reported the successful transformation of mature de- embryonated cowpea cotyledons by Agrobacterium-mediated transformation. Cotyledon explants inoculated withA. tumefaciens pUCD26l4 carrying plasmid pUCD2340 contain- ing a hygromycin phosphotransferase (hpt) transgene conferring hygromycin-B resistance were cultured on shooting medium and approximately 15-19% of the explants produced shoots which could be rooted in the presence of antibiotics. The presence of the hpt gene in the transgenic plants was confirmed by genomic DNA gel blot hybridization analysis. It should be noted, however, that no information is available on whether the antibiotic resistance trait was transferred to subsequent generations. Among the more recent reports of attempts to overcome the limitations to cowpea transformation, Brar et al. (1999) showed that there were genotype effects on the performance of various v~ unguiculata cultivars during cell culture. They also found that endogenous ethylene levels influenced in vitro regeneration rates. Machuka and colleagues at the International Institute of Tropi- cal Agriculture (I1TA) are attempting to optimize parameters for cowpea transformation through the establishment of antibiotic thresholds for selection of transformed cowpea tissues and development of shoot elongation and rooting media. Details of these studies can be found elsewhere in this volume. 206 Digitized by Google Molecular cloning in cowpea The development of successful genetic transformation protocols for cowpea is essential to realize the potentials of transgenic approaches for germplasm improvement of cowpea. At the present time, a number of candidate genes that could have substantial impact on yield are available for introduction into cowpea. These include a range of genes whose products (e.g., lectins, serine and thiol-protease inhibitors, a-amylase inhibitors, trypsin inhibitors, cysteine proteases, chitinases, and Bacillus thuringiensis toxin) have been shown to be effective in the control of many of the major insect pests that diminish seed yield and quality, including bruchid beetles (Callosobruchus maculatus), pod-sucking bugs (Clavigralla tomentosicollis), and pod borers (Maruca vitrata). Enhanced resis- tance of cowpea to a wide spectrum of disease pathogens can also be achieved through transgenic manipulation by both the introduction of single or multiple gene resistance traits from other species or through metabolic engineering (Hilder and Boulter 1999). The effectiveness and durability of disease and pest resistance are likely to be greater in engineered transgenic plants in which multiple resistance genes are introduced (so-called "resistance gene pyramiding"). Such pyramiding is time consuming and often difficult to achieve through traditional breeding approaches due to interspecific barriers, but readily achievable through transgenic approaches. Beyond disease and pest resistance, the ability to transform cowpea opens up the potential for manipulation of numerous other plant characteristics including seed protein composition and nutritional quality (e. g., protein content, amino acid balance, etc. [Chopra et al. 1999]) and abiotic stress tolerances (e.g., drought, heat, and salinity tolerance [Van- demark 1999]). Each of these characteristics is being successfully manipulated in other crop species (Hilder and Boulter 1999; Mazur et al. 1999; Somerville and Somerville 1999) where well established protocols for transformation and regeneration already exist. Perspective and future directions Much of the foundation for the future successful manipulation of cowpea by genetic engineering is now in place. A genetic map of the cowpea genome which provides a rea- sonable degree of coverage to rapidly locate loci of interest has already been established and studies are underway in a number of laboratories to further saturate the map with additional markers in order to improve its utility. Genetic linkage maps are also available for a number of related species, including v~ radiata, v~ angularis, and v~ vexillata. Given the high degree of colinearity and conservation in genome organization between species that have been studied, progress made on the mapping of genes in one species should be useful in all species. Numerous single gene and quantitative trait loci have already been placed on the cowpea map. As the use of molecular-marker analysis for gene mapping becomes more widespread in the cowpea community, the number and variety of traits placed on the map will increase. A large number of populations segregating for disease and pest resistance, drought tolerance, growth and yield parameters, and other characteristics which can be used in mapping activities have already been developed through the effort of breeders. Coordination of efforts between laboratories to exploit these resources is important to ensure rapid future progress. In addition to efforts aimed at refining the genetic map, the development of physical maps linking genetically defined markers with DNA fragments is essential for the future map-based cloning of genes in cowpea. Techniques are now available for the construction 207 Digitized by Google Biotechnology for cowpea of ordered libraries of large DNA fragments using either Yeast Artificial Chromosomes (YACs) (Burke et al. 1987; Coulson et al. 1988) or Bacterial Artificial Chromosomes (BACs) (Shizuya et al. 1992; Woo et al. 1994), with the latter being the method of choice in recent years due to higher cloning efficiency, ease of handling, and greater stability of the recombinant clones. The generation of large insert DNA libraries for cowpea and the establishment of a physical map based on an assembly of overlapping contigs should be one of the priorities over the next few years. To complement work on the physical and genetic mapping of the cowpea genome, there should be increased research activity centered on basic gene discovery and studies on gene regulation. The past several years have seen major advances in DNA chip technologies for the rapid measurement of differential gene expression in plants (Lemieux et al. 1998). The use of oligonucleotide and cDNA microarray technologies (Lockhart et al. 1996; DeRisi et al. 1997; Heller et al. 1997) offer researchers an efficient, inexpensive, and rapid means to measure transcript levels for thousands of genes simultaneously, allowing the identification of genes participating in common metabolic activities or activated/repressed in response to changes in any number of selected internal or external cues (e.g., changes in developmental age, challenge by disease or pests, and alterations in physical environ- ment). Information provided from such analysis in combination with data on inheritance of desirable agronomic (growth, yield, and resistance) traits should give researchers the ability to pinpoint changes necessary to achieve rapid improvements in germplasm through marker-assisted breeding or genetic engineering (Somerville and Somerville 1999). Finally, it is essential that open and rapid exchange of information occurs between researchers working within various disciplines whether at the molecular, genetic, cell culture, or agricultural extension level. Working groups and fora established by electronic communication have greatly facilitated progress on other crop plants and efforts to enhance and extend current activities should be a priority within the cowpea community. Integrated with a clear understanding of the needs of producers and desires of consumers, current technologies and new biotechnology-based strategies under development should have significant impact on expanding the economic importance of cowpea in the coming decades. Acknowledgments This work was supported by grants from the Rockefeller Foundation and the International Institute ofTropicalAgriculture (I1TA). Thanks to the laboratory staff for their suggestions and comments on this manuscript. References Ajibade, S.R, N.F. Weedden, and S.M. Chite. 2000. Inter simple sequence repeat analysis of genetic relationships in the genus Vigna. Euphytica 111: 47-55. Akella, V. and P.F. Lurquin. 1993. Expression in cowpea seedlings of chimeric transgenes after electroporation into seed-derived embryos. Plant Cell Reports 12: 110-117. Akkaya, M.S., A.A. Bhagwat, and P.B. Cregan. 1992. Length polymorphisms of simple sequence repeat DNA in soybean. Genetics 132: 1131-1139. Akkaya, M.S., RC. Shoemaker, J.E. Specht, A.A. Bhagwat, and P.B. Cregan. 1995. Integration of simple sequence repeat DNA markers into a soybean linkage map. Crop Science 35: 1439-1445. Bashir, M. and RO. Haptom. 1996. Sources of genetic resistance in cowpea (Vigna unguiculata L. Walp.) to cowpea aphid-borne mosaic virus. European Journal of Plant Pathology 102: 411-419. 208 Digitized by Google Molecular cloning in cowpea Boeke, J.D., D.I Garfinkel, CA. Styles, and G.R. Fink. 1985. Ty elements transpose through an RNA intermediate. Cell 40: 491-500. Brar, M.S., 1M. Al-Khayri, T.E. Morelock, and IE. Anderson. 1999. Genotypic response of cowpea Vigna unguiculata (L.) to in vitro regeneration from cotyledon explants. Cellular and Develop- mental Biology 35: 8-12. Burke, D. T., G.F. Carle, and M. V Olsen. 1987. Cloning oflarge fragments of exogenous DNA into yeast by means of artificial chromosome vectors. Science 236: 806-812. Chopra, VL., VS. Malik, and S.R. Bhat (editors). 1999. Applied plant biotechnology. Science Publishers Inc., Enfield, New Hampshire, USA. 384 p. Coulson, A., R Waterson, J. Riff, 1 Sulston, and Y. Kohara. 1988. Genome linking with yeast artificial chromosomes. Nature 335: 184-186. DeRisi, J.L., VR Iyer, and P.O. Brown. 1997. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278: 680--686. Ehlers, J.D. andA.E. Hall. 1997. Cowpea (Vigna unguiculata L. Walp.). Field Crops Research 53: 187-204. Fatokun, CA., N.D. Young, and G.O. Myers. 1997. Molecular markers and genome mapping in cowpea. Pages 352-360 in Advances in cowpea research. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Reseach Center for Agricultural Sciences (J1RCAS), edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. I1TA, Ibadan, Nigeria. Fatokun, CA., D.1. Menancio-Hautea, D. Danesh, and N.D. Young. 1992. Evidence for orthologous seed weight genes in cowpea and mung bean based on RFLP mapping. Genetics 132: 841- 846. Fatokun, CA., D. Danesh, N.D. Young, and E.L. Stewart. 1993a. Molecular taxonomic relationships in the genus Vigna based on RFLP analysis. Theoretical and Applied Genetics 86: 97-104. Fatokun, C.A., D. Danesh, D.1. Menancio-Hautea, and N.D. Young. 1993b. A linkage map for cowpea (Vigna unguiculata [L.] Walp.) based on DNA markers. Pages 6256--6258 inA compila- tion oflinkage and restriction maps of genetically studied organisms. Genetic maps 1992, edited by IS. O'Brien. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA. Federoff, N.V 1989. About maize transposable elements and development. Cell 56: 181-191. F ery, RL. 1985. The genetics of cowpeas: a review ofthe world literature. Pages 25--62 in Cowpea research, production and utilization, edited by S.R. Singh and K.O. Rachie. John Wiley and Sons, Chichester, UK. F ery, RL. 1990. The cowpea: production, utilization, and research in the United States. Horticulture Review 12: 197-222. Fery, RL. and B.B. Singh. 1997. Cowpea genetics: a review of the recent literature. Pages 13-29 in Advances in cowpea research, Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (J1RCAS), edited by B.B Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. IlIA, Ibadan, Nigeria. Finer, 11, P. Vain, M.W. Jones, and M.D. McMullen. 1992. Development ofthe particle inflow gun for DNA delivery to plant cells. Plant Cell Reports 11: 323-328. Flavell,A.I, S. Pearce, andA. Kumar. 1994. Plant transposable elements and the genome. Current Opinions in Genetics and Development 4: 838-844. Galasso, I., G.E. Harrison, D. Pignone, A. Brandes, and IS. Heslop-Harrison. 1997. The distribu- tion and organization of Tyl-copia-like retrotransposable elements in the genome of Vigna unguiculata (L.) Walp. (cowpea) and its relatives. Analytical Botany 80: 327-333. 209 Digitized by Google Biotechnology for cowpea Garcia, lA., J. Hille, and R Godbach. 1986. Transformation of cowpea Vigna unguiculata cells with an antibiotic resistance gene using a Ti-plasmid-derived vector. Plant Science 44: 37-46. Garcia, JA., J. Hille, P. Vos, and R Godbach. 1987. Transformation of cowpea Vigna unguiculata with a full-length DNA copy of cowpea mosaic virus m-RNA. Plant Science 48: 89-98. Gowda, B.S., lL. Miller, S.S. Rubin, D.R. Sharma, and M.P. Timko. 2002. Isolation, sequencing and linkage mapping of resistance-gene analogs in cowpea (Vigna unguiculata L.). Euphytica. 126:(3) 365-377. Heller, RA., M. Schena, A. Chai, D. Shalon, T. Bedilion, J. Gilmore, D.E. Wooley, and RW. Davis. 1997. Discovery and analysis of inflammatory disease-related genes using cDNA microarrays. Proceedings of National Academy of Science USA 94: 2150-2155. Hilder, VA. and D. Boulter. 1999. Genetic engineering of crop plants for insect resistance-a critical review. Crop Protection 18: 177-191. Kaga,A., N. Tomooka, Y Egawa, K. Hosaka, and O. Kamijima. 1996a. Species relationships in the subgenus Ceratotropis (genus Vigna) as revealed by RAPD analysis. Euphytica 88: 17-24. Kaga,A., M. Ohnishi, T. Ishii, and O. Kamijima. 1996b. A genetic linkage map ofazuki bean con- structed with molecular and morphological markers using an interspecific population (Vigna angularis x V nakashimae). Theoretical and Applied Genetics 93: 658-663. Kononowicz,A.K., K.T. Cheah, M.L. Narasimhan, L.I. Murdock, RE. Shade, M.J. Chrispeels, E. Filippone, L.M. Monti, RA. Bressan, and P.M. Hasegawa. 1997. Developing a transformation system for cowpea (Vigna unguiculata [L.] Walp.). Pages 361-371 in Advances in cowpea research. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan Inter- national Reseach Center for Agricultural Sciences (J1RCAS), edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. I1TA, Ibadan, Nigeria. Lemieux, B., A. Aharoni, and M. Schena. 1998. Overview of DNA chip technology. Molecular Breeding 4: 277-289. Li, J., G. He, P. Gepts, and C.S. Prakash. 1999. Development of a genetic map for cowpea (Vigna unguiculata) using DNA markers. Page 237 in Abstracts, Plant & Animal Genome VII Confer- ence, 17-21 January 1999. San Diego, California, USA. Lockhart, D.l, H. Dong, M.C. Byrne, M.T. Follettie, M.V Gallo, M.S. Chee, M. Mittmann, C. Wang, M. Kobayashi, H. Horton, and E.L. Brown. 1996. Expression monitoring by hybridization to high-density oligonucleotide arrays. Nature Biotechnology 14: 1675-1680. Marechal, R, 1M. Mascherpa, and F. Stainer. 1978. Etude taxonomique d'un groupe complexe d' especes des genresPhaseolus et Vigna (Papillionaceae) sur la base de donnees morphologiques et polliniques traitees par I'analyse informatique. Boissiera 28: 1-273. Marfo, K.O. and A.E. Hall. 1992. Inheritance of heat tolerance during pod set in cowpea. Crop Science 32: 912-918. Mazur, B., E. Krebbers, and S. Tingy. 1999. Gene discovery and product development for grain quality traits. Science 285: 372-375. Menancio-Hautea, D., L. Kumar, D. Danesh, and N.D. Young. 1993a. A genome map for mung bean (Vigna radiata [L.] Wilczek) based on DNA genetic markers (2N = 2x = 22). Pages 6259--6261 in A compilation of linkage and restriction maps of genetically studied organisms. Genetic maps 1992, edited by J. S. O'Brien. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA. Menancio-Hautea, D., CA. Fatokun, L. Kumar, D. Danesh, and N.D. Young. 1993b. Comparative genome analysis of mung bean (Vigna radiata L. Wilczek) and cowpea (V unguiculata L. Walp.) using RFLP mapping data. Theoretical and Applied Genetics 86: 797-810. 210 Digitized by Google Molecular cloning in cowpea Menendez, C.M., A.E. Hall, and P. Gepts. 1997. A genetic linkage map of cowpea (Vigna unguiculata) developed from a cross between two inbred domesticated lines. Theoretical and Applied Genetics 95: 1210-1217. Muthukumar, B., M. Mariamma, K. Veluthambi, andA. Gnanam. 1996. Genetic transformation of cotyledon explants of cowpea (Vigna unguiculata [L.] Wal p.) using Agrobacterium tume- faciens. Plant Cell Reports 15: 980-985. Myers, G.O, C.A. Fatokun, and N.D. Young. 1996. RFLP mapping of an aphid resistance gene in cowpea (Vigna unguiculata [L.] Walp.). Euphytica 91: 181-187. Ogundiwin, E.A., C.A. Fatokun, G. Thottappilly, M.E. Aken'Ova, and M. Pillay. 2000. Genetic linkage map of Vigna vexillata based on DNA markers and its potential usefulness in cowpea improvement. Page 19 in Abstracts, World cowpea research conference III, 4-7 September 2000, Ibadan, Nigeria. Ouedraogo, J.T., V Maheshwari, D.K. Berner, C.-A. St-Pierre, F. Belzile, and M.P. Timko. 2001. Identification of AFLP markers linked to resistance of cowpea (Vigna unguiculata L.) to parasitism by Striga gesnerioides. Theoretical and Applied Genetics 102: 1029-1036. Pasquet, R.S. 1999. Genetic relationships among subspecies of Vigna unguiculata (L.) Walp. Based on allozyme variation. Theoretical and Applied Genetics 98: 1104-1119. Penza, R., P.F. Lurquin, and E. Filippone. 1991. Gene transfer by co-cultivation of mature embryos with Agrobacterium tumefaciens. Application to cowpea (Vigna unguiculata L. Walp.). Journal of Plant Physiology 138: 39-43. Penza, R., V Akella, and P.F. Lurquin. 1992. Transient expression and histological localization of a gus chimeric gene after direct transfer to mature cowpea embryos. Bio Techniques 13: 576-579. Roberts, P.A., W.C. Matthews, and J.D. Ehlers. 1996. New resistance to virulent root-knot nematodes linked to the Rk locus of cowpea. Crop Science 36: 889-894. Shizuya, H., B. Birren, U.J. Kim, V Mancino, T. Slepak, Y. Tachiiri, and M. Simon. 1992. Cloning and stable maintenance of 300-kilobase-pair fragments of human DNA in Esche- richia coli using anF-factor-based vector. Proceedings of National Academy of Science USA 89: 8794-8797. Singh, B.B. and A.M. Emechebe. 1997. Advances in research on cowpea Striga and Alectra. Pages 215-224 in Advances in cowpea research. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Reseach Center for Agricultural Sciences (nRCAS), edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. I1TA, Ibadan, Nigeria. Singh, B.B., D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai (editors). 1997. Advances in cowpea research. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Reseach Center for Agricultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Somerville, C. and S. Somerville. 1999. Plant functional genomics. Science 285: 380-383. Sonnante, G., T. Stockton, R.O. Nodari, VI. Becerra Ve!{lsquez, and P. Gepts. 1994. Evolution of genetic diversity during the domestication of common bean (Phaseolus vulgaris L.). Theoretical and Applied Genetics 89: 629-635. Sonnante, G., A.R. Piergiovanni, Q.N. Ng, and P. Perrino. 1996. Relationships of Vigna unguiculata (L.) Walp., V vexillata (L.) A. Rich., and species of section Vigna based on isozyme variation. Genetic Resources and Crop Evolution 43: 157-165. Vaillancourt, R.E. and N.F. Weeden. 1992. Chloroplast DNA polymorphism suggests a Nigerian center of domestication for the cowpea, Vigna unguiculata (Leguminosae). American Journal of Botany 79: 1194-1199. Vaillancourt, R.E., N.F. Weeden, and J.D. Barnard. 1993. Isozyme diversity in the cowpea species complex. Crop Science 33: 606-613. 211 Digitized by Google Biotechnology for cowpea Vaillancourt, RE. and N.F. Weeden. 1996. Vigna unguiculata and its position within the genus Vigna. Pages 89-93 in Advances in legume systematics 8: legumes of economic importance, edited by B. Pickersgill and J.M. Lock. Royal Botanic Gardens, Kew, UK. Vandemark, G.J. 1999. Transgenic plants for the improvement offield characteristics limiting crop production. Pages 219-273 in Molecular biotechnology for plant food production, edited by O. Paredes-L6pes. Technomic Publishing Co. Inc. Lancaster, UK. Vos, P., R Hogers, M. Bleeker, M. Reijans, T. van de Lee, M. Hornes,A. Fritjers, J. Pot, 1 Peleman, M. Kuiper, and M. Zabeau. 1995.AFLP: a new technique for DNA fingerprinting. NucieicAcids Research 23: 4407-4414. Williams, J.G.K.,A.R Kubelik, J.l Livak, JA. Rafalski, and S.V Tingey. 1990. DNA polymor- phisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18: 6531--6535. Woo, S.S., J. Jiang, B.S. Gill, A.H. Paterson, and RA. Wing. 1994. Construction and character- ization ofa bacterial artificial chromosome library of Sorghum bicolor. Nucleic Acids Research 22: 4922-4931. 212 Digitized by Google 3.4 Potential role of transgenic approaches in the control of cowpea insect pests J. Machuka1 Abstract Crops' incompatibility makes conventional breeding approaches untenable in transferring available insect resistance from wild 'Vigna sp. into cowpea. The alternative recourse is to isolate and transfer alien resistance genes using genetic transformation. This has the added advantage of using useful genes from distantly related organisms to control cowpea pests. Artificial diet bioassays carried out on the Maruca pod borer, pod sucking bugs, and cowpea weevils indicate that these insects can be controlled by Bacillus thuringiensis crystal proteins, plant lectins, protease, a-amylase inhibitors, chitinases, and/or ribosome-inactivating proteins. The challenge now is to express the genes encoding these proteins in transgenic cowpea and hope that what happens in artificial diets will, at least in some cases, be replicated in transgenics. Other candidate genes include enzymes encoding biochemical pathways in secondary metabolism. It can be anticipated that useful information emerging from current global genomics efforts in crop species, includ- ing model legumes, will have a bearing on cowpea improvement through genetic engineering. What cowpea researchers need to do now is develop a comprehensive pest resistance management strategy. Such a strategy must take into account criteria such as transformation of elite cowpea lines that are adapted to each of the major agroecological zones, gene flow between cultivated and wild cowpea, and strategies for dissemination and adoption of biotechnologically improved cowpea lines. This paper reviews previous work on candidate genes, presents some recent results, and makes projections on how research on cowpea breeding through genetic modification for insect resistance may move from the laboratory into farmers' fields, especially in sub-Saharan Africa. Introduction Grain yield losses in cowpea (Vigna unguiculata) are mainly due to biotic stresses, espe- cially insect pests, including aphids, thrips, Maruca pod-borer (MPB [Maruca vitrata]), bruchids, and pod-sucking bugs (PSB). Although modest levels of host plant resistance are available in cowpea germplasm, there is nearly none to MPB. Insect resistance genes are present in wild cowpea relatives (Vigna spp.) as well as other non-Vigna legumes that are infested by MBP such as African yam bean (AYB [Sphenostylis stenocarpa]). How- ever, breeding barriers make conventional breeding approaches untenable in transferring resistance from wild Vigna and other legumes into cowpea. The alternative recourse is to isolate and transfer alien resistance genes using genetic transformation. This has the added advantage of using useful genes from distantly related organisms to control cowpea pests. Artificial diet bioassays carried out on MPB, PSB, and cowpea weevils indicate that these insects can be controlled by Bacillus thuringiensis (Bt) crystal proteins, plant 1. PO Box 347, Kilifi, Kenya. 213 Digitized by Google Biotechnology for cowpea lectins, protease and a -amylase inhibitors, chitinases and/or ribosome-inactivating proteins. The challenge now is to express the genes encoding these proteins in transgenic cowpeas and hope that what happens in artificial diets will, at least in some cases, be replicated in transgenics. Other candidate genes include enzymes encoding biochemical pathways in secondary metabolism. This paper reviews research related to identification of candidate insect resistance genes and makes projections on how cowpea genetic modification breeding for insect resistance may move from the laboratory into farmers' fields, especially in sub-Saharan Africa. Methods for isolation of insect resistance factors The first step in generating insect resistant transgenic crops is to identify insecticidal proteins or compounds that are active against the target pest. The most common way of doing this involves the use of artificial diets or seeds that contain proteins, secondary metabolites, or other compounds that are suspected or known to have anti-insect proper- ties (Duck and Evola 1997). Bacillus thuringiensis crystal proteins were the first to be used to generate transgenic insect resistant crops (reviewed by Krattiger 1997). Proteins from other microorganisms as well as plants have also been used for direct screening for insecticidal activities (Schulera et al. 1998). The common higher plant defense proteins tested to date include lectins, protease, and a-amylase inhibitors (Duck and Evola 1997). In addition to screening known factors, random screening without bias regarding origin or source of protein, chemical, or extract may be performed. The compounds or proteins may even be purchased from commercial sources. For example, Streptomyces choles- terol oxidase, a potent insecticidal enzyme against the cotton boll weevil, was isolated by screening culture filtrates from over 10000 microbial fermentations (purcell 1997). Callus-based insect bioassays from susceptible and resistant crop lines have also been used to investigate insect resistance (Williams et al. 1987). Map-based cloning using techniques such as chromosome walking which utilize molecular probes that map near resistance loci is another approach to isolate genes for deployment in genetic engineering for insect resistance (Gibson and Somerville 1993). Although some work has been done to transfer insect resistance genes from mammals and insects into crops (Schulera et al. 1998), the following discussion focuses mainly on microbial and plant genes that have potential for deployment in transgenic insect resistance in cowpea. Resistance genes from microorganisms Bacillus thuringiensis is a spore-forming soil bacterium that produces insecticidal protein crystals, also called Bt toxins, endotoxins, or crystal (Cry) proteins, within its cells during sporulation. Spores and purified protein crystals of several Bt strains have been used as microbial insecticides since the 1950s and now have an established role in some integrated pest management systems (Fietelson et al. 1992). Different strains of Bt produce different crystal proteins, coded for by Cry genes that are highly toxic to specific insects, nematodes, and other invertebrates. Bt toxins tend to be specific in their activities either to Lepidoptera, Coleoptera or other insects. Their mechanism of action is not quite clear, but it is believed that the proteins damage the membrane of the insect's midgut epithelial cells, causing massive water uptake (Gatehouse et al. 1992). This may in turn lead to the disruption of the electrical K + and pH gradients by creating pores, resulting in irreversible damage to the midgut wall. 214 Digitized by Google Potential role of transgenic approaches in the control of cowpea insect pests To date, several genes encoding different Bt toxins have been engineered into crop plants (Schulera et al. 1998). Research at I1TAhas shown that Cry lAb, Cry 1 C, and Cry I1A proteins are toxic to MPB (Jackai, unpublished). For control of cowpea pests, it is imperative that other different Bt toxins be tested in artificial diets or seeds for their effi- cacy against MPB, bruchids, and PSB for which assay systems are available (Jakai and Raulston 1988; Shade et al. 1986). Moreover, artificial insect resistance assays need to be developed for other problematic pests, particulary thrips, to allow screening of Cry proteins against these pests. Bacillus thuringiensis also produces vegetative insecticidal proteins (ViPs) when it is not sporulating (Estruch et al. 1996; Warren 1997). The ViPs are unrelated to crystal proteins and appear to be active against lepidopteran pests such as fall armyworm, beet armyworm, corn rootworm, and tobacco budworm (Estruch et al. 1997). Other candidate genes for insect protection include Streptomyces cholesterol oxidase (purcell 1997), fungal chitinases (Kramer et al. 1997), the isopenteny I-transferase gene (ipt) fromAgrobacterium tumefaciens (Smigocki et al. 1997) and genes encoding insect viral RNAs (Hanzlik and Gordon1997). Additionally, the bacterium Photorrhabdus luminescens, which lives in entomophagous nematodes has recently been shown to produce insecticidal toxins that may be useful for transgenic insect control (Bowen et al. 1998). Insect resistance genes from higher plants It is important to discover new genes that can be pyramided with Bt genes to enhance resistance levels. Other limitations of Bt genes include possibilities of resistance break- down, limited scope of pests covered by Cry proteins, and public perception issues (Stewart 1999). To overcome some of these limitations, plant-derived genes have been cloned and transferred into several crop species (Schulera et al. 1997, Snow and Palma 1997). Genes for bruchid resistance Coleopteran insects in the family Bruchidae cause serious cowpea grain losses in stor- age. Callosobruchus maculatus is key among these pests. Through conventional breed- ing efforts at I1TA and elsewhere, modest levels of resistance to C. maculatus have been attained (Singh and lackai 1985). To enhance these modest resistance levels, efforts have also been underway to identify plant genes that affect C. maculatus development. The majority of artificial seed bioassays have involved the use of plant lectins (Gatehouse et al. 1984, 1991; Heusing et al. 1991a; Machuka et al. 1999a, 1999b, 2000; Murdock et al. 1990; Omitogun et al. 1999; Pratt et al. 1990). Vicilins (7S seed storage proteins) and protease and a-amylase inhibitors and a-amylase inhibitor-like proteins (AIL), are also insecticidal to bruchids (Hilder et al. 1987; Ishimoto et al. 1999; Pittendrigh et al. 1997; Sales et al. 1996; Yunes et al. 1998; Huesing et al. 1991c). Table 1 summarizes the toxicity mechanism of these proteins. Transgenic pea and azuki seeds containing the bean a-amylase inhibitor are resistant to bruchid beetles (Ishimoto et al. 1996, Shade et al. 1994). Plans are underway to introduce this gene into modestly bruchid resistant I1TA cowpea lines once the transformation system becomes routine. Various compounds are toxic to cowpea beetles. F or example, leaf, fruit, seed, and oil extracts from some Mrican shrubs possess larvicidal and ovicidal activities against C. maculatus (Leonard et al. 1993, Seck et al. 1993). However, these toxins are more applicable in biocontrol than transgenic insect control strategies, at least in the short term. 215 Digitized by Google Biotechnology for cowpea Table 1. Candidate genes for transgenic resistance to bruchids. Protein Cowpea vicilins a-amylase inhibitors Cowpea protease inhibitors e.g. cystein, Bowman-Birk, trypsin, and chymotrypsin inhibitors lectins Possible mechanism(s) of action Bind insect chitin Inhibition in insect a-amylases • Depletion of essential amino acids resulting from hypersecretion of digestive enzymes • Inhibition of insect digestive proteases • Carbohydrate binding to insect midgut epithelium or peritrophic matrix/ membrane • Resistance to proteolysis Genes for resistance to the Maruca pod borer Unlike studies focusing on cowpea weevils, only two studies have been reported that pertain to the biological effects of plant lectins on growth, development, and fecundity of MPB in artificial diet bioassays (Machuka et al. 1999b, 2000). Table 2 shows the list of plant lectins so far tested for their effects against MPB. At least 26 lectins from 15 plant families and representing seven carbohydrate-binding specificity groups have been tested. Results from this screening work indicated that mannose-specific lectins from twayblade (Listera ovata) and snowdrop (Galanthus nivalis) have detrimental effects on MPB larval development at all stages of development. Others, such as wheat germ and jackfruit agglutinins possess latent effects that only manifest at (a) subsequent unique stage(s). A type 1 ribosome-inactivating protein (RIP) from Iris and bean (Phaseolus vulgaris) a-mylase inhibitor are not toxic to MPB larvae although the latter mildly affects pupal development and adult emergence (Machuka et al. 1 999b ). The galactose-specific seed lectin from Nigerian-grown Mrican yam bean (Sphenostylis stenocarpa) does not affect MPB larval development, although it inhibits C. maculatus development (Machuka et al. 2000). Generally, relatively few lectins are toxic to lepidopteran insects, even when they have been found stable to proteolysis by enzymes in the insect gut (Czapla and Lang 1990; Czapla 1997; Gatehouse et al. 1995). Apart from lectins, plant proteinaceous inhibitors (PIs) of insect proteinases (serine, cysteine, aspartic, and metallo proteinases) are considered potential candidates for gene transfer for insect resistance (Ryan 1990). Serine proteases are the dominant class in lepidopteran insects larvae such as MPB, whereas coleopteran species have a wider range of dominant gut proteinases (Gerald et al. 1997). Since serine and cysteine PIs mainly inhibit the growth and development of lepidopteran (and coleopteran) species, it would be useful to screen a wide range of these PIs against MPB in artificial diets. To date, more than 14 different plant PI genes have been introduced into crop plants, with efforts con- centrated on serine PIs from the plant families F abaceae, Solanaceae, and Poaceae (Koiwa et al. 1997, Schulera et al. 1998). So far, the most active PI identified is the cowpea tryp- sin inhibitor (CpTI), isolated from an I1TA bruchid resistant line, TVnu 2027 (Hilder et al. 1987). Serine PI-like proteins have been identified from seeds of Nigerian-grown velvetbeans (Mucuna spp.) (Machuka 2000a). These proteins, as well as affinity purified trypsin and chymotrypsin inhibitors from two wild Vigna species (~~ vexillata and ~~ oblongifolia) andAYB, are not toxic to MPB (Machuka unpublished). The advantage of 216 Digitized by Google Potential role of transgenic approaches in the control of cowpea insect pests Table 2. Plant lectins tested against the Maruca pod borer in artificial diets. Lectin' ASA,Allium sativum (garlic) agglutinin AUA, Allium ursinum (ramson) lectin *GNA, Galanthus nivalis (snowdrop) agglutinin *LOA, Listera ovata (twayblade) agglutinin *NPA, Narcissus pseudonarcissus (daffodil) agglutinin *CSA, Calystegia sepium (hedge bindweed) agglutinin PSL, Pisum sativum (garden pea) lectin SSA, Sphenostylis stenocarpa (African yam bean) agglutin APA,Aegopodium podagraria (ground elder) lectin BDA, Bryonia dioica agglutinin (white bryony) *BPA, Bauhinia purpurea agglutinin (carmel's foot tree) DBA, Dolichos biflorus agglutinin (horse gram) *IRA, Iris hybrid agglutinin (Dutch iris) JCA, Artocarpus integrifolia lectin (jackfruit) SBA, Glycine max agglutinin (soybean) *SNA-II, Sambucus nigra agglutinin (elderberry) DSL, Datura stramonium lectin (jimson weed) *UDA, Urtica dioica agglutinin (stinging nettle) WGA, Triticum aestivum (wheat germ) (Wheat) agglutinin MM, Maackia amurensis (Maackia) agglutinin SNA-I, Sambucus nigra (elderberry) agglutinin CM, Colchicum autumnale (meadow saffron) agglutinin PHA-E, Phaseolus vulgaris (red kidney bean) phytohemagglutinin isoform E *PHA-L, Phaseolus vulgaris (red kidney bean) phytohemagglutinin isoform L TLC-I, Tulipa hybrid (tulip) agglutinin RPA, Robinia pseudoacacia (false/black acacia) agglutinin Plant family Alliaceae Alliaceae Amaryllidaceae Orchidaceaee Convolvulacea Fabaceae Apiaceae Cu rcu rbitaceae Moraceae Fabaceae Iridaceae Caesalpiniaceae Fabaceae Sambucaseae Solanaceae Urticacaceae Gramineae Fabaceae Sambucaceae Liliaceae Fabaceae Fabaceae Liliaceae Fabaceae Lectin specificity group Mannose Mannose/maltose Mannose/glucose Galactosel N-acetyl- galactosamine N-acetyl- glucosamine Sialic acid Complex glycan 'Candidate lectins for transgenic resistance to Maruca pod borer. Detailed references of names and classification of lectins and pod borer bioassays can be found in Van Damme et al. (1998a, b) and Machuka et al. (1999b, 2000). using PIs and other genes from plants, especially edible ones, for enhanced insect resistance is that the nutritional penalty after gene transfer is absent or minimal and there are fewer public perception problems. This has been demonstrated through mammalian toxicity tests, for example, in the case of the cowpea trypsin inhibitor gene (pusztai et al. 1992). Recently, it has been shown that expression of plant proteases rather than protease inhibitors may be a novel insect defence mechanism in plants (pechan et al. 2000). Based on the use of Arginine Sepharose B chromatography for isolation of animal serine prote- ases, novel insecticidal proteins against MPB larvae have been isolated from Mucuna seeds (Machuka 2000b). Although protein database searches revealed that the N-terminus of these proteins is similar to a novel human synovial membrane fluid protein, it is not clear exactly what these proteins are and what their role is in plants. Other candidate genes 217 Digitized by Google Biotechnology for cowpea that may be implicated in MPB resistance may include chitinases and lectin-like proteins (Colucci et al. 1999, Machuka and Okeola 2000). Genes for resistance to pod-sucking bugs PSBs are probably the next most serious pests of cowpea for which conventional breeding approaches have been inadequate. Omitogun et al. (1999) were the first to demonstrate that crude lectin-enriched extracts from AYB affect development of the cowpea coreid bug (Clavigralla tomentosicollis [Stal]) (Hemiptera: Coreidae). Subsequently, the puri- fied seed lectin (SSA) from A YB has been shown to be toxic to C. tomentosicollis in an artificial cowpea seed system (Machuka et al. 1999a, Okeola et al. 2000). Wheat germ agglutinin, the nonprotein amino acid (para -aminopheny lalanine, PAPA) from v~ vexi llata, and a cysteine protease inhibitor (E-64) also inhibit development of C. tomentosicollis nymphs (Jackai, Shade, and Murdock, unpublished). More studies are needed to identify other candidate proteins for resistance to PSB. Some ecological issues related to projected transgenic cowpea release It is clear from the above survey that candidate genes for transgenic insect control in cowpea are available. In order to realise the potential of this approach it is imperative to establish a stable genetic transformation system for this crop. At the same time, it is also crucial for cowpea scientists to begin to discuss the ecological issues associated with release of transgenic cowpeas, particularly in Africa. Although it is true that pest resistance genes identified in wild and cultivated Vigna germplasm have been incorporated into cultivated varieties by farmers and breeders for several years (Singh et al. 1990; Fatokun 1991; Jackai et al. 1996) the use of genetic engineering raises questions related to the transfer of trans genes to compatible wild or weedy Vigna species related to cowpea (Krattiger 1997; Snow and Palma 1997; Stewart 1999). Some of the issues to consider at this point include the possibility that introduced pest resistance may confer added fitness to cowpea, resulting in enhancement of weedy characteristics due to its increased ability to survive and spread outside of cultivation. Secondly, would transgenic cowpeas transfer pest resistance (or other traits) by natural hybridization to produce hybrid progeny that are more aggressive or more difficult to control? Although gene flow to related species is likely to be limited to v~ unguiculata subspecies such as v~ unguiculata ssp. dekindtiana, it is important to carry out field trials to determine rates of gene flow. Such a study is underway at IlIA (Fatokun, personal communication). Obviously, information will be required from many disciplines such as weed science, agronomy, population biology and genetics, entomology, plant breeding, ecology, plant pathology, molecular biology, and from farmers. Conclusion Reliable and efficient bioassay systems need to be continuously developed and refined to aid the discovery of insecticidal proteins for control of key cowpea pests. It can be anticipated that useful information emerging from current global genomics efforts in crop species, including model legumes, will have a bearing on cowpea improvement through genetic engineering. What cowpea researchers need to do now is develop a comprehen- sive pest resistance management strategy that incorporates transgenic approaches. Such 218 Digitized by Google Potential role of transgenic approaches in the control of cowpea insect pests a strategy must take into account criteria such as transformation of elite cowpea lines that are adapted to each of the major agroecological zones, gene flow between cultivated and wild cowpeas, and strategies for dissemination and adoption of biotechnologically- improved cowpea lines. Aknowledgements Thanks to all the technicians in the Cellular and Molecular Technology Laboratory at I1TA for work related to characterization of insecticidal proteins from African legumes. References Bowen, D., TA. Rocheleau, M. Blackburn, O. Andreev, E. Golubeva, R Bhartia, and RH. ffrench- Constant. 1998. Insecticidal toxins from bacterium Photorhabdus luminescens. Science 280: 2129-2132. Colucci, G., J. Machuka, and M.J. Chrispeels. 1999. cDNA cloning of a class III acid chitinase from the African yam bean (Sphenostylis stenocarpa) (Accession no. AF137070). Plant Physiology 120: 633. Czapla, T.H. 1997. Plant lectins as insect control agents in transgenic plant. Pages 123-138 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M. Koziel. Taylor and Francis, London, UK. Czapla, T.H. and BA. Lang. 1990. Effect of plant lectins on the larval development of European corn borer (Lepidoptera: Pyralidae) and southern corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 83: 2480-2485. D'Silva, I., G.G. Poirier, and M.C. Heath. 1998. Activation of cysteine proteases in cowpea plants during the hypersensitive response-a form of programmed cell death. Experimental and Cell Research 245: 2389-2399. Duck, N.B. and S. Evola. 1997. Use of transgenes to increase host plant resistance to insects: opportunities and challenges. Pages 1-18 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M.G. Koziel. Taylor and Francis, London, UK. Estruch, 11, N.B. Carozzi, N.B. Duck, G.w. Warren, and M.G. Koziel. 1997. Transgenic plants: an emerging approach to pest control. Nature Biotechnology 15: 137-141. Estruch, 11, G.w. Warren, M.A. Mullins, G.J. Nye, J.A. Craig, and M.G. Koziel. 1996. ViP3A, a novel Bacillus thuringiensis vegetative insecticidal protein with a wide spectrum of activities against lepidopteran insects. Proceedings National Academy of Sciences ofthe USA 93: 5389- 5394. Fatokun, CA. 1991. Wide hybridization in cowpea: problems and prospects. Euphytica 54: 137- 140. Fietelson 1, J. Payne, and L. Kim. 1992. Bacillus thuringiensis: insects and beyond. Bio/Technol- ogy 10: 271-275. Gatehouse, A.M.R, F.M. Dewey, J. Dove, and KA. Fenton. 1984. Effect of seed lectins from Phaseolus vulgaris on the development of larvae of Callosobruchus maculatus, mechanism of toxicity. Journal ofthe Science of Food and Agriculture 35: 373-380. Gatehouse, A.M.R., D.S. Howe, IE. Flemming, VA. Hilder, and IA. Gatehouse. 1991. Biochem- ical basis of insect resistance in winged bean (Psophocarpus tetragonolobus) seeds. Journal of Science Food and Agriculture 55: 63-74. GatehouseA.M.R, VA. Hilder, and D. Boulter. 1992. Plant genetic manipulation for crop protec- tion. CAB International, Oxford, UK. Gatehouse, A.M.R., K.S. Powell, w.J. Peumans, E.IM. Van Damme, and IA. Gatehouse. 1995. Insecticidal properties of lectins; their potential in plant protection. Pages 35-58 in Lectins: biomedical perspectives, edited by A.J. Pusztai and S. Bardocz. Taylor and Francis, London, UK. 219 Digitized by Google Biotechnology for cowpea Gerald, RR, K.J. Kramer, J.E. Baker, IF. Kanost, and C.A. Behke. 1997. Proteinase inhibitors and resistance oftransgenic plants to insects. Pages 157-183 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M. Koziel. Taylor and Francis, London, UK. Gibson S. and C. Somerville. 1993. Isolating plant genes. Trends in Biotechnology 11: 306-312. Hanzlik T.N. and K.H.J Gordon. 1997. The Tetraviridae. Advances in Virus Research 48: 101- 168. Hilder V. A., A.M.R Gatehouse, S.E. Sheerman, R.F. Barker, and D. Boulter. 1987. A novel mechanism of insect resistance engineered into tobacco. Nature 300: 160-163. Huesing, J.E., L.L. Murdock, and RE. Shade. 1991a. Rice and stinging nettle lectins: insecticidal activity similar to wheat germ agglutinin. Phytochemistry 30: 3565-3568. Huesing, IE., L.L. Murdock, and RE. Shade. 1991b. Effect of wheat germ isolectins on develop- ment of cowpea weevil. Phytochemistry 30: 785-788. Huesing IE., RE. Shade, M.I Chrispeels, and L.L. Murdock. 1991c. a-amylase inhibitor, not phytohemagglutinin, explains resistance of common bean seeds to cowpea weevil. Plant Physi- ology 96: 993-996. Ishimoto M., T. Yamada, and A. Kaga. 1999. Insecticidal activity of an a-amylase inhibitor-like protein resembling a putative precursor of a-amylase inhibitor in the common bean, Phaseolus vulgaris L. Biochimica et BiophysicaActa 1432: 104-112. Ishimoto, M., T. Sato, M.J. Chrispeels, and K. Kitamura. 1996. Bruchid resistance of transgenic azuki bean expressing the seed a-amylase inhibitor of common bean. Entomologia Experimen- talis etApplicata 79: 309-315. Jackai, L.E.N. and J.R Raulston. 1988. Rearing of legume pod borer, Maruca testulalis Geyer (Lepidoptera: Pyralidae) on artificial diet. Tropical Pest Management 34: 168-172. Jackai, L.E.N., S. Padulosi, and Q. Ng. 1996. Resistance to the legume pod borer, Maruca vitrata Fabricius, and the probable modalities involved in wild Vigna. Crop Protection 15: 753-761. Koiwa, H., RA. Bressan, and P.M. Hasegawa. 1997. Regulation of protease inhibitors and plant defense. Trends in Plant Sciences 2: 379-383. Kramer, K.L., S. Muthukrishnan, L. Johnson, and F. White. 1997. Pages 185-193 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M. Koziel. Taylor and Francis, London, UK. Krattiger, A. F. 1997. Insect resistance in crops: a case study of Bacillus thuringiensis (Bt) and its transfer to developing countries. International Service for the Acquisition ofAgri-BiotechAppli- cations, Ithaca, USA. Leonard, v., D. Seck, G. Lognay, C. Gaspqr, and M. Severin. 1993. Biological activity of Cassia occidentalis L. against Callosobruchus maculatus (p.) (Coleoptera: Bruchidae). Journal of Stored Products Research 25: 311-318. Machuka, J. 2000a. Characterization of the seed proteins of velvetbean (Mucuna pruriens) from Nigeria. Food Chemistry 68: 421-427. Machuka, I 2000b. Insecticidal effects of velvet bean seed proteins againstMaruca pod-borer larvae, in Abstracts, Proceedings of World Cowpea Conference III, 4-7 September 2000. I1TA, Ibadan, Nigeria. Machuka J. and O.G. Okeola. 2000. One and two dimensional gel-electrophoresis identification of African yam bean seed proteins. Journal of Agricultural and Food Chemistry 48: 2296-2299. Machuka, IS., O.G. Okeola, E.J.M. Van Damme, M.J. Chrispeels, F.V. Leuven, and W.l Peumans. 1999a. Isolation and partial characterization of galactose-specific lectins fromAfrican yam beans, Sphenostylis stenocarpa Harms. Phytochemistry 51: 721-728. Machuka, J.S., E.J.M. Van Damme, W.l Peumans, and L.E.N. Jackai. 1999b. Effect of plant lectins on larval development ofthe legume pod borer, Maruca vitrata. Entomologia Experimentalis et Applicata 93: 179-187. 220 Digitized by Google Potential role of transgenic approaches in the control of cowpea insect pests Machuka, J.S., O.G. Okeola, M.I Chrispeels, and L.E.N. Jackai. 2000. African yam bean seed lectins affect the development ofthe cowpea weevil but do not affect the development oflarvae oflegume pod borer. Phytochemistry 53: 667-674. Murdock, L.L., J.E. Huesing, S.S. Nielsen, RC. Pratt, and RE. Shade. 1990. Biological effects of plant lectins on the cowpea weevil. Phytochemistry 29: 85-89. Okeola, O.G., J. Machuka, and 1.0. Fasidi. 2000. Insecticidal activities of the African yam bean seed lectin on the development ofbruchid beetle and pod sucking bugs, in Abstracts, Proceedings of World Cowpea Conference III, 14-17 September 2000. I1TA, Ibadan, Nigeria. Omitogun, O.G., L.E.N. Jackai, and G. Thottappilly. 1999. Isolation of insecticidal lectin-enriched extracts from African yam bean, Sphenostylis stenocarpa Harms, and other legume species. Entomologia Experimentalis et Applicata 90: 301-311. Pechan, T., L. Ye, Y Chang, A. Mitra, L. Lin, F.M. Davis, w.P. Williams, and Dawn S. Luthe. 2000. A unique 33-kd cysteine proteinase accumulates in response to larval feeding in maize genotypes resistant to fall armyworm and other Lepidoptera. Plant Cell 12: 1031-1040. Pittendrigh, B.R., J.E. Huesing, RE. Shade, and L.L. Murdock. 1997. Effects of CryiA/CrylB Bt endotoxins, PAPA, protease and a-amylase inhibitors, on the development of the rice weevil, Sitophilus oryzae, using an artificial seed bioassay. Entomologia Experimentalis etApplicata 82: 201-211. Pratt, RC., N.K. Singh, RE. Shade, L.L. Murdock, and RA. Bressan. 1990. Isolation and partial characterization of a seed lectin from Tepary Bean that delays bruchid beetle development. Plant Physiology 93: 1453-1459. Purcell, L. 1997. Cholesterol oxidase for the control of boll weevil. Pages 95-108 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M. Koziel. Taylor and Francis, London, UK. Pusztai A, G. Grant, D.J. Brown, J.C. Stewart, S. Bardocz, S.w. Ewen,A.M.R. Gatehouse, and V Hilder. 1992. Nutritional evaluation of the trypsin (EC 3.4.21.4) inhibitor from cowpea (Vigna unguiculata Walp.). British Journal of Nutrition 68: 783-791. Ryan, C.A. 1990. Protease inhibitors in plants: genes for improving defenses against insects and pathogens. Annual Review of Phytopathology 28: 425-449. Shade RE., H.E. Schroeder, J.J. Pueyo, L.M. Tabe, L.L. Murdock, T.IV Higgins, and M.J. Chrisp- eels. 1994. Transgenic pea seeds expressing the alpha-amylase inhibitor ofthe common bean are resistant to bruchid beetles. Bio/Technology 12: 793-796. Sales M.P., VM. Gomes, K.V Fernandes, and I Xavier-Filho. 1996. Chitin-binding proteins from cowpea (Vigna unguiculata) seeds. Brazillian Journal of Medical and Biological Research 29: 319-326. Shade, RE., L.L. Murdock, D.E. Foard, and MA. Pomeroy. 1986. Artificial seed system for bioas- say of cowpea weevil (Coleoptera: Bruchidae) growth and development. Environmental Ento- mology 15: 1286-1291. Shade, RE., H.E. Schroeder, J.J. Pueyo, L.M. Tabe, L.L. Murdock, T.I V Higgins, and M.I Chrisp- eels. 1994. Transgenic pea seeds expressing the a-amylase inhibitor of the common bean are resistant to bruchid beetles. Biotechnology 12: 793-796. Schulera, T.H., G. Poppya, B.R. Kerrya, and I. Denholmb. 1998. Insect-resistanttransgenic plants. Trends in Biotechnology 16: 168-175. Seck, D., G. Lognay, E. Haubruge, J-P. Wathelet, M. Marler, C. Gaspqr, and M. Severin. 1993. Biological activity of the shrub Boscia senegalensis (Pers.) Lam. Ex Poir. (Capparaceae) on stored grain pests. Journal of Chemical Ecology 19: 377-389. Singh, S.R. and L.E.N. Jackai. 1985. Insect pests of cowpea in Africa: their life cycle, economic importance, and potential for control. Pages 217-231 in Cowpea research, production, and uti- lization, edited by S.R. Singh and K.O. Rachie. John Wiley and Sons, Chichester, UK. 221 Digitized by Google Biotechnology for cowpea Singh S.R., L.E.N. Jackai, J.H.R. Dos Santos, and C.B. Adalia. 1990. Insect pests of cowpea. Pages 43-91 in Insect pests oftropical food legumes. John Wiley and Sons, Chichester, UK. Smigocki,A., S. Heu, I. Mccanna, C. Wozniak, and G. Buta. 1997. Insecticidal compounds induced by regulated overproduction of cytokinins in transgenic plants. Pages 225-2363 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M. Koziel. Taylor and Francis, London, UK. Snow, A.A. and P.M. Palma. 1997. Commercialization of transgenic plants: potential ecological risks. BioScience 47: 86-96. Stewart, C.N. 1999. Insecticidal transgenes into nature: gene flow, ecological effects, relevancy, and monitoring. Pages 179-190 in Gene flow and agriculture, relevance for transgenic crops. Proceedings of a symposium held at the University of Keele, 12-14 April 1999, edited by P.IW. Lutman. British Crop Protection Council, Surrey, UK. Van Damme, E.J.M., W.I Peumans,A. Pusztai, and S. Bardocz. 1998a. A handbook of plant lectins: properties and biomedical applications. John Wiley and Sons, Chichester, UK. Van Damme, E.IM., W.I Peumans,A. Barre, and P. Rouge. 1998b. Plant lectins: a composite of several distinct families of structurally and evolutionary related proteins with diverse biological roles. Critical Reviews in Plant Sciences 17: 575-692. Warren, G.W. 1997. Vegetative insecticidal proteins: novel proteins for control of corn pests. Pages 109-121 in Advances in insect control: the role of transgenic plants, edited by N. Carozzi and M. Koziel. Taylor and Francis, London, UK. Williams, w.P., P.M. Buckley, and F. Davis. 1987. Tissue culture and its use in investigations of resistance of maize. Agriculture, Ecosystems and Environment 18: 185-190. Yunes,A.N., T.M. deAndrade, P.M. Sales, R.A. Morais, VS. Fernandez, VM. Gomes, and J. Xavier- Filho. 1998. Legume seed vicilins interfere with the development of the cowpea weevil (Cal- losobruchus maculatus). Journal of Agricultural and Food Chemistry 76: 111-116. 222 Digitized by Google 3.5 I nsecticidal activities of the African yam bean seed lectin on the development of the cowpea beetle and the pod-sucking bug O.G. Okeola 1, J. Machuka2, and 1.0. FasidP Abstract The cowpea beetle, Callosobruchus maculatus, and pod-sucking bug, Clavigralla tomentosicollis, are two ofthe major insect pests of cowpea in Africa. A lectin was purified from the seeds ofthe African yam bean (AYB), Sphenostylis stenocarpa, by affinity chromatography on Galactose-Sepharose 4B. The purifiedAYB lectin (A YBL) was tested on the two insect pests of cowpea. When C. maculatus larvae were fed on artificial cowpea seed containing 0.2, 2, and 5% (w/w) of dietary lectin, larval mortality ranged from 30 to 88% and delay in number of days to first emer- gence from 4-13 days. When A YBL was tested on C. tomentosicollis, nymphal mortalities ranged from 76 to 81 % at 1 % and 87 to 94% at 2%. From 4 to 8%, no nymph survived up to six days after infestation. The results of these insect bioas- says provided a scientific basis for isolating a lectin gene fromAYB for the trans- formation of cowpea. Introduction The cowpea beetle, Callosobruchus maculatus (F.) (Coleoptera: Bruchidae) and the pod- sucking bug, Clavigralla tomentosicollis Stal (Hemiptera: Coreidae) are two major insect pests of cowpea in Africa. C. maculatus attacks cowpea during storage (Jackai and Adalla 1997). F arm storage for six months is accompanied by about 30% loss in weight with up to 70% of the seeds infested and unfit for consumption (Singh and lackai 1985). C. tomentosicollis feeds primarily on the developing pods in the field where it causes extensive damage to pods and seeds (J ackai 1984). These insect infestations cause weight and quality losses that lead to a reduction in commercial value and seed viability. In recent times, control of crop insect pests has focused mainly on the use of genetic engineering to develop transgenic plants that express insecticidal proteins. For example, the toxic proteins produced by Bacillus thuringiensis conferred resistance on cotton against pink bollworm (Wilson et al. 1992) while cowpea inhibitor genes conferred resistance on tobacco against corn earworm (Hoffman et al. 1991). Other control agents are peroxidases, chitinases, and plant lectins (Duck and Evola 1997; Machuka et al. 1999). Lectins are a large and heterogenous group of proteins (Van Damme et al. 1998) possessing at least one noncatalytic domain, which binds reversibly to a specific mono- or oligosaccharide (peumans and Van Damme 1995). In a preliminary investigation, Omitogun et al. (1999) conducted bioassays on C. maculatus and C. tomentosicollis using 1. Biotechnology Research Unit, International Institute of Tropical Agriculture, Ibadan, Nigeria. 2. PO Box 347, Kilifi, Kenya. 3. Department of Botany and Microbiology, University ofIbadan, Ibadan, Nigeria. 223 Digitized by Google Biotechnology for cowpea crude lectin extracts from 20 resistant Vigna and non-Vigna legumes. The extracts from the African yam bean (AYE), Sphenostylis stenocarpa (Harms) was toxic to the insect pests at 5% (w/w) concentration. However, the result of their investigation was inconclusive as it could not be ascertained whether the lectin or other protein contaminants present in the extracts were responsible for the toxicity. This study was conducted to investigate the effect of purified AYE lectins on C. macu- latus and C. tomentosicollis. Materials and methods Seed Seeds of AYE were obtained from the Genetic Resources Unit of International Institute of Tropical Agriculture (I1TA), Ibadan. Seeds of three AYE accessions were used in this study; EN953, EN982, and UMUE9832. EN953 was used by Omitogun et al. (1999) while EN982 and UMUE9832 were recent collections. Seeds ofIfe Brown, a cultivated cowpea variety, were used as the susceptible control and a wild Vigna (~~ vexillata) accession (TVnu 72), was the resistant control. Insects The insects were obtained from the insect rearing laboratory at I1TA, Ibadan. This labora- tory is maintained at 26 ± 2 °C and 70-80% RH. Preparation of affinity matrix Galactose was coupled to Sepharose 4B by the divinyl sulphone coupling method accord- ing to Peumans et al. (1995). Lectin purification Five hundred grams of AYE seeds were milled in a Warren blender. The seed meal was then extracted in 2 L of ascorbic acid solution by stirring for at least one hour at room temperature. The homogenate was then centrifuged at 5000 rpm for four to five min- utes. The supernatant was saved, adjusted to pH 7.5 with sodium hydroxide, and then filtered on Whatman filter paper (No.1). The resulting filterate was applied onto a column (5 x 2 cm) of Galactose-Sepharose 4B that had been equilibrated with 1M NH4SO 4' Unbound proteins were washed off with 1M NH4SO 4 until A2so fell below 0.3. The lectin was then eluted with 20 mM 1, 3-diamino-propane (DAP), desalted on Sephadex G 25 column and lyophilized. Artificial seeds Artificial cowpea seeds (ACS), as previously described by Shade et al. (1986), were used as the delivery method in these insect bioassays. ACS were prepared by milling decor- ticated Ife Brown seeds, adding aqueous solutions containing 0.2, 1,2,4, 5,6, and 8% (w/w) lectins to make a paste. The paste was injected into a teflon mold, frozen, and then lyophilized for 24 hours. The resulting pellets were hydrated at 29°C and 60 ± 5% RH and coated with 8% gelatin solution (w/w). Insect bioassays ACS containing 0.2,2, and 5% (w/w) lectin were tested with C. maculatus. Two adult insects (male and female) were placed in a Petri dish containing five artificial or control seeds. The dishes were incubated on a shelf at 27 ± 2 °C and 65 ± 2% RH for 24 hours. 224 Digitized by Google Insecticidal activities of the African yam bean seed lectin This was to allow the insects to mate and lay eggs, after which they were removed. After seven days, the eggs on the seeds were counted for each sample. After two weeks, the vari- ous treatments were examined daily for adult emergence. Emerged adults were counted and removed daily. Observations were terminated two weeks after the first adult emerged. For C. tomentosicollis, we employed a bioassay previously developed in the Insect Rearing Unit of I1TA. One ACS of each lectin treatment (l, 2, 4, 6, and 8%) was placed in a separate box (6.5 cm x 6.5 cm x 2.5cm). An inverted lid of a 1O-dram vial with a slightly moistened cotton wool swab was placed in the box to provide water for the insects. Five first instar nymphs of C. tomentosicollis were placed inside each box and covered. Other treatments with either blank ACS without lectin and intact seeds (Ife Brown and TVnu 72) were included as controls. Seeds and cotton swabs were changed only if mold started to grow on them. The boxes were left undisturbed on laboratory shelves (10: l4h; light dark; 26 ± 2°C; and 70 to 80% RH) until the end of the experiment. Each treatment was replicated five times. The following variables were determined from the bioassays: number of eggs per seeds (for C. maculatus), number of emerged adults, mortality (the total number of hatched eggs/first instar nymphs used for infestation for each treatment, minus the total number of emerged adults for each treatment, divided by the number of hatched eggs/first instar nymphs for each treatment, times a hundred), and total develop- ment time (TDT). Statistical analysis The data were analyzed using General Linear Model (GLM) procedures and the means were separated by Duncan's Multiple Range Test (DMRT) (SAS 1989). Percentage data were transformed using arcsin transformation prior to analysis. Results and discussion AYB lectin is a tetrameric protein of about 122 kDa. It is composed of four subunits with molecular mass of about 27,29, 32, and 34 kDa, respectively (Fig. 1). During purification, higher and lower molecular weight-contaminating proteins were successfully removed (Fig. 1). The tetrameric nature of AYB lectin was similar to that of Glycine max (soy- bean) and Phaseolus vulgaris (red kidney bean) in that they also have four subunits with molecular weights of 115-140 and 120 kDa, respectively (Sharon 1973). However, not all plant lectins are tetrameric proteins. Some are dimeric, containing only two subunits. For example, the lectin from the greater celandine contains two subunits with molecular weights of 9.5 and 11.5 kDa, respectively (peumans et al. 1985). When bruchid beetles were fed onACS containing 0.2,2, and 5% (w/w) dietary lectin, larvae mortality of C. maculatus ranged from 30 to 88% (Fig. 2) whereas low mortality (5%) was observed for larvae fed on Ife Brown (Fig. 2). In an earlier study, Huesing et al. (1991) observed that larval mortality ranged from 8 to 12% when fed on susceptible cowpea lines. When EN953 and UMUE9832lectins were increased from 0.2 to 2%, nymph mortal- ity increased from 30 to 33.3% with EN953 and from 29.67 to 33.33% with UMUE9832. However, when the lectin concentration from the two AYB accessions was increased to 5%, the percentage mortality increased between 2-fold and 2.5-fold (Fig. 2). AYB lectin greatly reduced C. maculatus progeny and delayed the TDT, compared to the susceptible control (Fig. 3). Sixteen adults in all emerged from Ife Brown, whereas only 10 adults emerged from ACS containing EN953 and 11 adults from UMUE9832 at 0.2% (w/w) lectin. The toxic effect of AYB lectin was more pronounced at 5% dietary 225 Digitized by Google Biotechnology for cowpea .., ~ J. II ",DI ... .... ... A B C • {: - ..1 1 - -- Figure 1. SDS-PAGE (12%) of Sphenostylis stenocarpa (EN982) seed proteins. M = Molecular mass reference proteins 1 = Affinity purified lectin 2 = Non-lectin fraction 3 = Total protein Identical pattern was visualized for all accessions of AVB. 100 90 80 70 >- .<.::: 60 -; - .. 50 0 E :.e 0 40 30 20 10 0 \;;)'Y & ':'>~ ~'" P\~ # ~ «; Lectin source (concentration in %) Figure 2. Effect of lectin from Sphenostylis stenocarpa, cowpea, and V. vexillata on the mortality of C. maculatus. 226 Digitized by Google 8 ell 7 =: == "t:I 6 IU "t:I 5 Q,j ell .. Q,j 4 E Q,j .... 3 0 .. Q,j 2 .J::J E == z 0 8 ell =: 7 == "t:I 6 IU "t:I 5 Q,j ell .. Q,j 4 E Q,j 3 .... 0 .. 2 Q,j .J::J E == z 0 8 ell 7 =: == "t:I 6 IU "t:I 5 ~ .. Q,j 4 E Q,j 3 .... 0 .. 2 Q,j .J::J E == z 0 Insecticidal activities of the African yam bean seed lectin -+- Ife Brown _ EN953 (0.2%) ~ UMUE9832 (0.2%) 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Duration (days) -+- Ife Brown _EN953 (2%) ~ UMUE9832 (2%) 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Duration (days) -+- Ife Brown _EN953 (5%) ~ UMUE9832 (5%) 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Duration (days) Figure 3. Emergence patterns of C. maculatus from artificial cowpea seeds containing different concentrations of s. stenocarpa seed lectin. 227 Digitized by Google Biotechnology for cowpea lectin concentration. Only six adults emerged from ACS containing EN953 lectin and three from ACS containing UMUE9832 lectin (Fig. 3). With the exception of EN953 (0.2%), delay in number of days to first emergence also ranged from 4 to 13 days in all the different lectin concentrations (Fig. 3). Similar effects of plant lectins on C. maculatus had been reported by Murdock et al. (1990). Lectins from Arachis hypogaea, Solanum tuberosum, Datura stramonium, Triticum aestivum, and Maclura pomifera were found to cause a significant delay in C. maculatus developmental time at a dietary level of 1 %. The presents study shows that AYE lectin is insecticidal to C. maculatus. The deleterious effect of AYE lectins on C. tomentosicollis is shown in Figure 4. Nymph mortality ranged from 76 to 81% at 1% and from 87 to 94% at 2% dietary lectin concentrations. From 4 to 8% lectin concentrations, no nymph survived more than six days. C. tomentosicollis nymphs survived on Ife Brown and blankACS. No adult emerged from the resistant control seed and ACS containing 4% lectin (Fig. 4). This lectin was obviously toxic to C. tomentosicollis. Furthermore, there was a significant difference in the TDT obtained when C. tomen- tosicollis nymphs were fed on blankACS (17.65 ± 0.47 days) as compared to the intact susceptible seed treatment (13.77 ± 0.29 days) (P < 0.05). This was unexpected. These insects are different in their feeding mode from cowpea bruchids for which the ACS was originally developed. AlthoughACS has been previously used on C. tomentosicollis (Omi- togun et al. 1999; Koona 1999), this delivery system has some defects for bioassays on C. tomentosicollis. Possibly by reducing the concentration of gelatin used in making theACS, insect development similar to that observed when insects are fed on intact susceptible seeds could be obtained. Plans are underway to examine the optimum gelatin concentration that will be required for making ACS for bioassays on C. tomentosicollis. 100 ,--- ,- 90 80 70 >- .<.::: 60 "iU - 50 .. 0 E 40 :.e .. 30 20 10 0 Lectin source (concentration in %) Figure 4. Effect of Sphenostylis stenocarpa lectin on the mortality of C. tomentosicollis. 228 Digitized by Google Insecticidal activities of the African yam bean seed lectin Reports on lectin bioassays on C. tomentosicollis are not common, as most lectin bio- assays have been on C. maculatus (Gatehouse et al. 1984; Murdock et al. 1990; Huesing et al. 1991; Zhu et al. 1996). Considering the high larval/nymph mortality rate and delay in TDT observed in these bioassays, AYE lectin seems to be biologically active and a promising candidate for genetic transformation of cowpea against C. maculatus and C. tomentosicollis. Acknowledgements The authors thank Tayo Bamidele for his excellent technical assistance. We also acknowl- edge the assistance of A.O. Odeseye, 1. Asiwe, S. Adekunle, and A. Olaife (Mrs). References Duck, N. and S. Evola. 1997. Use oftransgenes to increase host plant resistance to insects; oppor- tunities and challenges. Pages 1-18 in Advances in insect control, edited by N. Carizzi and M. Koziel. Taylor and Francis, London, UK. Gatehouse, A.M.R, EM. Dewey, 1 Dove, H.A. Fenton, and A. Putztai. 1984. Effect of seed lectins fromPhaseolus vulgaris on the development oflarvae ofCallosobruchus maculatus: a mechanism oftoxicity. Journal of Science and Food Agriculture 35: 373-380. Hoffman, M.P., F.G. Zalom, J.M. Smilanick, L.T. Wilson, LD. Malyj, 1 Kiser, VA. Hilder, and W.M. Barnes. 1991. Field evaluation oftransgenic tobacco containing genes encoding Bacillus thuringiensis endotoxin or cowpea trypsin inhibitor efficacy against Helicoverpa zea (Lepidop- tera: Noctuidae). Journal of Economic Entomology 85: 2516-2522. Huesing, J.E., L.L. Murdock, and RE. Shade. 1991. Rice and stinging nettle lectins: insecticidal activity similar to wheat germ agglutinin. Phytochemistry 30: 3565-3568. Jackai, L.E.N. 1984. Studies on the feeding behaviors of Clavi grail a tomentosicollis (Stal.) (Hemip- tera: Coreidae) and their potential use in bioassays for host plant resistance. Journal of Applied Entomology 98: 344-350. Jackai, L.E.N. and C.B. Adalia. 1997. Pest management practices in cowpea. Pages 240-259 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (ilTA) and Japan Inter- national Research for Agricultural Sciences (nRCAS). ilTA, Ibadan, Nigeria. Koona, A.A. 1999. Anatomical and biochemical basis of resistance of wild and cultivated Vigna species to the Coreid bug, Clavigralla tomentosicollis Stal. PhD thesis, University of Ibadan, Ibadan, Nigeria. 213 pages. Machuka, 1, E.J.M. Van Damme, W.l Peumans, and L.E.N. Jackai. 1999. Effect of plant lectins on survival development ofthe pod borer Maruca vitrata. Entomologia Experimentalis etAppli- cata 93: 179-187. Murdock, L.L., J.E. Huesing, S.S. Nielsen, RC. Pratt, and RE. Shade. 1990. Biological effects of plant lectins on the cowpea weevil. Phytochemistry 29: 85-89. Omitogun, O.G., L.E.N. Jackai, and G. Thottappilly. 1999. Isolation of insecticidal lectin enriched extracts from African yam bean, Sphenostylis stenocarpa (Harms) and other legume species. Entomologia Experimentalis et Applicata 90: 301-311. Peumans, w.J., M. de Ley, H.M. Stinissen, and W.F. Broekaert. 1985. Isolation and partial charac- terization of a new lectin from seeds of the greater celandine (Chelidonium majus). Plant Physiology 78: 379-383. Peumans, W.J. and E.J.M. Van Damme. 1995. Lectins as plant defence proteins. Plant Physiology 109: 347-352. Peumans, w.J., H.C. Winter, V Berner, F. Van Leuven, I.l Goldstein, P. Truffa-Bachi, and E.J.M. Van Damme. 1995. Isolation of a novel plant lectin with unusual specificity from Calsepa sepium. Glycoconjugate Journal 14: 259-265. 229 Digitized by Google Biotechnology for cowpea SAS User's Guide. 1989. SAS Institute Inc., Cary, NC, USA. Shade, RE., L.L. Murdock, D.E. Foard, and MA. Pomeroy. 1986. Artificial seed system for bioas- say of cowpea weevil (Coleoptera: Bruchidae) growth and development. Environmental Ento- mology 15: 1286-1291. Sharon, N. 1973. Giycoproteins of higher plants. Pages 235-252 in Carbohydrate biochemistry, edited by lB. Pridham. Academic Press, London, UK. Singh, Rand L.E.N. Jackai. 1985. Insect pests of cowpea in Africa: their life cycle, economic importance and potential for control. Pages 217-231 in Cowpea research, production and utiliza- tion, edited by S.R. Singh and K.O. Rachie. John Wiley and Sons, Chichester, UK. Van Damme, E.J.M., W.l Peumans, A. Putzai, and S. Bardocz. 1998. Handbook of plant lectins: properties and biomedical applications. John Wiley and Sons, Chichester, UK. 452 pages. Wilson, F.D., H.M. Flint, W.R Deton, D.A. Fischoff, F.I Perlak, J.A. Armstrong, RL. Fuchs, S.A. Berberich, N.I Parks, and V.B.R Stapp. 1992. Resistance of cotton lines containing a Bacillus thuringiensis toxin to pink bollworm (Lepidoptera: Gelichiidae) and other insects. Journal of Economic Entomology 85: 1516-1521. Zhu, K., IE. Huesing, RE. Shade, RA. Bressan, P.M. Hasegawa, and L.L. Murdock. 1996. An insecticidal N-acetylglucosamine-specific lectin gene from Griffonia simplicifolia (Leguminosae). Plant Physiology 110: 195-202. 230 Digitized by Google Section IV Cowpea contributions to farming systems/agronomic improvement of cowpea production Digitized by Google Digitized by Google 4.1 Cowpea as a key factor for a new approach to integrated crop-livestock systems research in the dry savannas of West Africa S.A. Tarawali 1,3, B.B. Singh2, S.c. GuptaS, R. Tabo6, F. Harris?, S. Nokoe1, S. Fernandez-Rivera4, A. BationoB, V.M. Manyong1, K. Makinde1, and E.C. Odion9 Abstract Agriculture in the dry savannas is intensifYing in response to increasing populations of humans and livestock. As a result, increased productivity demands are placed upon integrated crop-livestock systems and more emphasis is on the roles of legumes such as cowpea. Cowpea has the potential to function as a key integrating factor in intensifYing systems through supplying protein in the human diet, and fodder for livestock, and bringing nitrogen into the farming system through nitro- gen fixation. This paper describes the development and evaluation of integrated "best-bet" options which maximize the benefits of cowpea and addresses aspects of improved crop varieties, crop and livestock management, nutrient cycling, and soil fertility. The approach used includes a multicenter, multidisciplinary approach to working with farmers which combines complementary strengths of previous component research involving crops and livestock by key international and national research institutions in the region. Introduction Cowpea is an important crop for farmers in much of the West African region, particularly in the dry savannas. Estimates of world hectarage of cowpea is in the range of 12.5 mil- lion, with about 8 million in West Africa, the majority of these being in Niger and Nigeria (Singh et al. 1997). Current FAO estimates for 1999 are lower than these figures, although the proportions are similar (FAO 2000). The same database estimates average cowpea grain production in West Africa as 358 kg/ha whereas Singh et al. (1997) estimate 240 kg/ha as an average for northern Nigeria. The apparent popularity of the crop may seem paradoxical if only the relatively low grain yields on farmers' fields are considered. Perhaps this is related to the fact that cowpea is a legume with the potential for multiple contribu- tions not only to household food production, but also as a cash crop (grain and fodder), livestock feed, and soil ameliorant. In this context, it is a crop that may have a wide role 1. International Institute of Tropical Agriculture (lIT A), Ibadan, Nigeria. 2. I1TA-Kano, Nigeria. 3. International Livestock Research Institute (lLRI), Ibadan, Nigeria. 4. ILRI, Addis Ababa, Ethiopia. 5. International Crops Research Institute for the Semi-Arid Tropics (lCRISAT), Kano, Nigeria. 6. ICRISAT, Bamako, Mali. 7. School of Earth Sciences and Geography, University of Kingston, UK. 8. Tropical Soils Biology and Fertility Programme (TSBF), Nairobi, Kenya. 9. Institute of Agricultural Research (IAR), Zaria, Nigeria. 233 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production in contributing to food security, income generation, and the maintenance of the environ- ment for millions of small-scale farmers who grow it in the region. In order to place such contributions in context, this paper will begin by considering the ongoing evolution of farming systems in West Africa, especially the integration of crop and livestock produc- tion, with reference to the particular features of the dry savannas where these scenarios are prominent. The potential role that cowpea can play in addressing the opportunities posed will also be addressed and as part of this, ongoing research which includes the utilization of such multiple benefits of cowpea will be considered. The changing face of agriculture In sub-Saharan Africa, the population may reach 1.2 billion by 2025 and be combined with a demographic shift from about 30% of the population (in 1990) in urban areas to at least 50% (Winrock 1992). These changes will mean an increasing demand for crops and livestock and even if production expands at the rate of 3% annually, which would be necessary to meet this demand (Winrock 1992), it is likely that at least 21 % of the children, about 39 million, will remain undernourished (Badiane and Delgado 1995). Recent studies have indicated that through both natural accretion and the change in requirements related to urbanization (Ehui et al. 1998), livestock demand in particular is likely to increase dramatically, ranging between an increase of 2.5% for mutton, pork, and poultry, to 4.2% for beef between 1993 and 2020 (Delgado et al. 1999). Within sub-Saharan Africa, more than 40% of the region's current population is in West Africa (based on FAO estimates for 1999; FAO 2000) meaning that the opportunities and challenges presented by the intensification scenario will be heightened in this region. One of the responses offarming systems to agricultural intensification is the integration of crop and livestock production (McIntire et al. 1992). As crop farmers seek to increase produc- tion, their cropping activities spread onto marginal land, fallow periods become reduced or absent, and consequently, the demand for nutrient inputs is raised. In the absence of reliable and cheap supplies of inorganic fertilizers, manure from transhumant livestock becomes more important. At the same time, as livestock keepers enlarge their herds, crop residues from crop farmers increasingly become the major feed resource because there is no longer marginal or fallow land for grazing. Estimates have shown that ignoring crop residues as a feed resource would result in serious feed shortages (Naazie and Smith 1997). In these scenarios, crop farmers may begin to own their own livestock for ready access to manure and simultaneously sell off some of the marginal land to livestock keepers, who settle and begin crop farming, using the manure from their animals (and possibly traction) as an input (Okike et al. 2001). In the dry savannas of West and Central Africa, crop-livestock integration is already a common feature of the farming systems. Dry savannas The dry savannas consist of the drier part of the northern Guinea savanna, plus the Sudan savanna representing more than 50% of the total land area of sub-Saharan Africa, with a significant proportion located in West Africa. Over 40% of the total ruminant livestock in West and Central Africa are in this region (Winrock 1992). Annual rainfall is less than 1000 mm with a growing period of 180 days or less meaning that much of the region experiences a long (7-9 months) harsh dry season. The growing period shortens on a 234 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research south-north axis. The sandy soils are generally poor, with low organic carbon, and cation exchange capacity, and are deficient in nutrients, especially nitrogen and phosphorus. Cropping is cereal-based with sorghum and millet dominating, and the former decreasing in prominence towards the north. Intercropping cereals with grain legumes is common in over 90% of fields, with cowpea and groundnut being the most common legume components. As well as grain, the residues from cropping, especially from cowpea (and groundnut), are important components of the farming systems in particular as fodder resources for the ruminant livestock which are also an integral part of the farming systems. Cattle, sheep, goats, and to a lesser extent camels, provide milk, meat, traction, manure, and cash. Major constraints to agricultural productivity in the region include the long dry season, which results in crop stress due to drought at the beginning and/or end of the wet season and a shortage of ruminant fodder during the harsh dry period. The poor soils and inci- dences of pests and diseases also have negative effects on crop production (both grain and fodder). In much of sub-Saharan Africa, inputs such as fertilizers and pesticides to counteract these negative forces are generally scarce or priced well above the means of the smallholder farmer. F arm sizes in the region are generally small, ranging from about 3 to 6 ha; each field is usually 1 ha or less and one farmer rarely owns contiguous fields (Ogungbile et al. 1999). A typical cropping pattern is as follows (Singh and Tarawali 1997). At the onset of the rainy season, cereal (millet or sorghum) is sown in rows with wide interrow spaces; two-three weeks later, a grain type of cowpea (short duration) is sown in alternate interrow spaces, followed by a fodder (or dual-purpose, late maturing) type of cowpea in the remaining interrows about three weeks later (Fig. 1). The cropping layout may be complicated by replacing some of the cowpea rows with groundnut and the timing of planting (but not the order) may vary, with the interval between planting the crops often much shorter than three weeks. Cereals will mature and be harvested first, together with the grain type cowpea, which will give a reasonable grain yield, but virtually no crop residue. The remaining dual-purpose/fodder type cowpea is left to grow over the rest of the field, until the rains cease and the leaves begin to show signs of wilting. At this stage, any grain on the plants is harvested, and the residue is cut and rolled up for storage on house roofs or in tree forks. The stored residue is fed to ruminants during the dry season, or, in some cases, sold in local markets where the high price during this period of feed scarcity means it will make a substantial contribution to a farmer's income. The cereal stalks remaining after harvest are fed to ruminants, but often, the leaves may be stripped off and fed to animals and ---------------------Cereal ............................................................................................................ ·········Grain cowpea ---------------------Cereal ..................................................................................................................... Fodder cowpea ---------------------Cereal Figure 1. Schematic representation of common cropping pattern in the dry savannas. Spacing between the cereal rows can be as much as 3 m. 235 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production the stalks used as building or fencing materials. Ruminants within farm compounds are supplemented with the cowpea residues and, within the compound, the manure is collected with household waste. At the start of the next cropping season, the "compost" of manure and household waste is spread on the crop fields, before land preparation. Thus, in the dry savannas, crop and livestock enterprises are closely integrated, with reciprocal benefits from crop residues as livestock fodder, and the latter providing manure and in some cases, traction, that contribute directly to crop production. While the benefits of such integration are recognized, and mixed crop-livestock farming systems, which currently contribute over 50% of the world's meat and over 90% of its milk (ILRI 2000) are recognized to have the greatest potential for intensification (de Haan et al. 1997), food demands of expanding populations place increased pressure on these systems to raise productivity. Such productivity increases, if they are to be sustainable, need to be achieved without damaging the natural resource base. In some cases, where production of mixed farming systems has intensified, the full implications have not been considered as, for example, soil is mined and severely degraded and livestock waste products become a problem, etc. (Delgado et al. 1999). In this context, the situation in the dry savannas of West Africa, where integrated crop and livestock production systems have existed for many decades, but now face the pressure to produce more, is ripe for interventions that address these opportunities. Cowpea, which can contribute both to crop-livestock produc- tion systems, and directly to soil fertility, has the potential to make major contributions in this respect. Contributions of cowpea towards increased and sustainable productivity in mixed systems As a legume, cowpea can contribute to soil fertility, mainly through its nitrogen fixing abilities. Part of the nitrogen fixed will remain in the soil in the roots, and thereby contribute to the soil fertility for subsequent crops. Some fixed nitrogen will eventually return to the soil as manure after residues are fed to livestock. In terms of the direct effects of cowpea in rotation with cereals, Manu et al. (1994) report a comparison of on-station and on-farm studies in Niger where cowpea-millet intercrop and cowpea-millet rotations were used. Their results are summarized in Table 1. On farmers' fields, rotation with cowpea gave 2.6 times more millet grain and 3.3 times more residue, than the intercropped, nonrotated treatment. Bagayoko et al. (1998) reported that cowpea can supply 35-40 kg N/ha in a cowpea-millet rotation, and Carsky and Berner (1995) presented similar figures for cowpea rotations with maize. See also Carsky et al. this volume. Table 1. Summary of results comparing cowpea intercropping with rotation in farmer- and researcher-managed fields. Yield (kg/ha) Cropping system Farmer-managed Researcher-managed Traditional intercropping Rotation Millet grain Millet residue Millet grain Millet residue Source: Extracted from Manu et al. (1994). 236 62 162 163 538 172 827 308 1531 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research There is some evidence that cowpea may help to reduce the number of viable Striga hermonthica seeds in the soil through stimulating suicidal germination of the seed. S. hermonthica is parasitic on cereal plants, and causes huge crop losses (Berner et al. 1996). Carsky and Berner (1995) report that rotation with selected cowpea varieties has a substantial and rapid effect on reducing S. hermonthica, with the number of attached S. hermonthica plants per maize plant being reduced by at least 50% when maize was grown after cowpea. Farmers' awareness of these roles of cowpea for soil fertility and S. hermonthica reduction is, to some extent, demonstrated by the fact that they usually rotate the legume and cereal rows within fields in alternate years. This means that the cereal and cowpea rows are interchanged each year, and the cereal will benefit at this "microlevel" from the cowpea grown in the previous year. Cowpea residue is an important fodder resource for ruminant livestock (Tarawali et al. 1997). Farmers in the dry savannas deliberately grow varieties and use management practices that will ensure some cowpea fodder is available for harvest at the end of the growing season, even at the expense of grain production. Harvesting at the end of the wet season, before the dry season becomes severe, gives the best quality, and this is preserved throughout the storage period. If the fodder is harvested late, when the dry season is already underway, quality is poor (Tarawali et al. 1997). Recognition of the importance of fodder from cowpea led to the initiation of joint IITA-ILRl research in 1990 when fodder quan- tity and quality parameters were included in the breeding and selection program. These efforts resulted in the identification of promising dual-purpose cowpea varieties suitable for the dry savannas (Singh and Tarawali 1997). Cowpea fodder as a feed supplement increases animalliveweight gain during the dry season. Schlecht et al. (1995) report an experiment where Zebu cattle (bulls of about 250 kg, equivalent to 1 TLU-Tropical Livestock Unit) were supplemented with 1 kg cowpea hay at night and 0.5 kg fresh rice feed meal in the morning per day/animal during the second half of the dry season. The animals were allowed to graze as usual for the rest of the day. From February 1988 to September 1989 the supplemented group gained 95 kg compared to 62 kg for the unsupplemented group. Taking animal numbers into account, this worked out to be equivalent to a difference of 67 g/animal/day. In many regions, cowpea fodder is particularly valued as a supplement in the period leading up to Muslim festivals when sheep are traditionally slaughtered. Some farmers sell cowpea fodder during the dry season when feed shortage is critical, and there have been suggestions that income from fodder sales makes a substantial contribution to the annual income in such cases (lCRlSAT 1991). In addition to the direct benefits of improved livestock production and health that result from feeding cowpea fodder, the quantity and quality of manure from such better fed animals will be improved and therefore, when returned to the land at the beginning of the growing season, contribute more towards the maintenance of soil fertility. In the same experiment referred to above, although not significant in this particular trial, the manure nitrogen, in g N/TLU/day was on average 25% higher in animals receiving supplements. Indications are that from 1 ha of improved cowpea, a farmer could benefit by an extra 50 kg meat per annum from better nourished animals, with over 300 kg more cereal grain as a result of improved soil fertility directly from the cowpea and morelbetter manure from the animals (Tarawali, unpublished). Of course, considerations of the time scale-increased 237 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production crop yields-would be realized only the next year and the distribution of manure should be taken into account. It is, however, noteworthy that these preliminary calculations have not considered all the potential benefits, for example, better fed traction animals would work harder, meaning more timely land preparation and better crop yields; better fed ruminants would give more milk and are likely to be more productive (increased weight gains mean young animals come into oestrus earlier). Providing more nutritious fodder also means that the comparatively indigestible parts of cereals (stalks, etc.) that are used as fodder are likely to be better consumed-intake of more fibrous material usually improves with the addition of better quality material to the diet. The potential impact of reduced S. hermonthica because of rotation with cowpea has also not been quantified. Some of the potential contributions of cowpea described above are summarized in Figure 2. In view of these contributions of cowpea and the availability of improved variet- ies, when seeking to address the opportunities posed by the intensification of crop-livestock systems in the dry savannas, it was apparent that a key component should be improved dual-purpose cowpea varieties. What was equally clear, however, was that cowpea, live- stock, or cereal crops never function in isolation in farm fields or households in the dry savannas; likewise, there is a complex of interactions between the biophysical, economic, social, and policy environments that influence farmers' decisions in these environments. As a result of such considerations, in the late 1 990s, international and national institutions working on various aspects of component research in the dry savannas began to develop Figure 2. Schematic representation of the potential contributions of cowpea in crop-livestock systems in the dry savannas. Not all potential interactions are shown for simplicity. For example, dussa is a regular household product which can contribute to livestock feed. Similarly, other crops and weeds in the system are not shown. Dussa is the testa of the grain which is separated from the endosperm by soaking prior to pounding and winnowing. 238 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research a new approach designed to bring together some of these key elements. This strategy is presented by Tarawali et al. (2000) in the context of natural resource management. In this paper, the emphasis is on the role of cowpea in promoting food and feed production as well as sustainable agriculture. Development of the research paradigm The three international research centers with interest in various aspects of the system began meeting to consider how best to initiate such an integrated approach. Scientists from I1TA, with the world mandate for cowpea research, ILRI for livestock, and ICRISAT for cereals and groundnut as well as the majority of the dry savanna ecoregion began to plan joint research in 1997. International Fertilizer Development Centre, Niger, with an interest in the soils component of the system, and Center for Overseas Research and Development, University of Durham, UK, with scientists from national research and development institu- tions have also joined this group more recently. From the outset, there has been consensus among the institutes that the aim of this joint research should be to "improve the lives of farm families in the dry savanna and Sahel of West Mrica through sustainable management of the natural resource base for food security and income generation." The first step in implementing the joint research was the establishment of an experi- ment at one location in 1998, using existing resources from the institutes involved. At the meeting to plan this research, two major principles were elucidated: first, the idea of "best-bet" options and secondly, a holistic, on-farm approach to evaluate these options. Combining the best of each aspect of the integrated crop-livestock system, varieties, crop geometry, crop residue/manure management, and livestock feeding constituted the best- bet options and it was recognized that these would differ from region to region within the dry savanna, depending on the dominant crop species and management practices. In some regions, sorghum and cowpea would be appropriate, in others, millet and cowpea, etc. Corralling livestock on crop fields may be suitable in some cases but not in others. It was further recognized that, depending on, among other things, market access, it would not be unrealistic to anticipate that some inputs would be available to farmers, and that the options offered, both in terms of the crops used and their arrangement in the field, should seek to maximize the use of available inputs. Implementing this research in a holistic manner meant that not only would crop grain and residue yields be measured, but that the animal performance when fed this fodder and the manure produced to return to the field would be assessed. Furthermore, aspects of nutrient cycling, and the social and economic circumstances and implications of these best-bet options would need to be assessed as a whole. Implementation of research The challenges posed by the best bet approach were recognized and so, the initial strategy was to start small and in 1998 the trial was established at just one location in northern Nigeria in Bichi Local Government (8 °19'E; 12 °12'N). This is about 50 km from Kano, on a good road. It was selected because information on village characterization (Ogungbile et al. 1999) from a survey carried out by ICRISAT and IAR in late 1996 was available. Originally, the intention was to use this survey dataset to define various groups of farmers so that representatives of each group could be selected to participate in the trial. However, after describing the aims of the trial to farmers from the village, only 11 volunteered to 239 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production participate and provided land; it was therefore decided to work with these 11 for the first year. In 1999, an additional 13 farmers participated. A total of three treatments were established by the participating farmers and in all cases one treatment consisted of the traditional field of sorghum and cowpea (L). Two best-bet options were used; both had improved varieties of cowpea (IT90K -277 -2) and sorghum (lCSV 400) and the rows were planted 75 cm apart with four rows of cowpea to two rows of sorghum, in contrast to the farmers' 1 to 1.5 m row spacing and one: one cereal: cowpea geometry. One best-bet option (BB+) included minimum inputs in the form of fertilizer, with nitrogen (N) applied only to the sorghum rows, and insecticide spray (for post-flowering insect pests) applied only to the cowpea; the other best-bet option (BB) had no inputs. It was anticipated that, in addition to maximizing the benefits from cowpea to the soil and minimizing the detrimental effects of sorghum shading on the cowpea, this row arrangement would allow optimal use of scarce inputs. The farmers appreciated the inputs (even though they were required to pay for them) so that in 1999, the BB treatment was modified to include local sorghum but with the same inputs of fertilizer and pesticide. Part of the best-bet options also included the concept of double cropping the cowpea-planting another crop of the same cowpea variety after harvesting the grain and fodder of the first. Previous trials had shown that this could give a good fodder yield with some grain, depending on the rainfall pattern (Singh and Tarawali 1997). All treatment plots received 3 tlha of manure (1.6% Nand 0.7% P) at the start of the 1998 growing season. All operations, land preparation, planting, weed- ing, application of inputs, harvesting, etc. were carried out by the farmers themselves with some technical guidance from technicians and scientists. Prior to planting, bulked soil samples were collected from the top 20 cm of soil and analyzed for C, N, and P. Plots were sampled for grain and stover at maturity, using ran- domly placed quadrants (of about 20 m2), at the same time they were harvested by the farmers. Samples of grain and biomass were taken for analysis of N and P. When all the sorghum and cowpea residues were dry in the field, they were weighed, collected, and stored in treetops or on house roofs prior to use in the feeding trial. Residues from differ- ent treatments were kept separately. On-farm livestock feeding During the first part of the dry season, farmers usually release their small ruminants into the fields once the grain harvest is completed to enable them to graze the remaining crop residues and weeds. Once these resources are used up, usually by the middle of the dry season, the animals are tethered within the homestead and fed with the stored crop resi- dues. The initial intention was to tether animals on the respective treatment plots early in the dry season, but farmers indicated that there would be no way to prevent other animals from grazing the plots also, as livestock roam freely once the crop harvest is complete. It was therefore decided to follow the farmers' usual practice and allow free grazing until the weeds and crop residue remaining in situ were used up. Harris (1998) reported that manure deposition on crop fields from free grazing animals is fairly insignificant at an estimated 17 kg/ha. Accordingly, the period for feeding the crop residues harvested from the present experiment began in early February in 1999 and early March in 2000, when the animals were confined to the compounds. By using estimates of 10 kg dry matter per TLU (TLU = Tropical Livestock Unit = 250 kg animalliveweight) per day for a period 240 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research of 180 days, the recommended liveweight of animals to be fed using the available residue was estimated. The 10 kg daily allowance was made up of a mixture of sorghum and cowpea residues in proportion to the available total weight of biomass of each compo- nent on a plot by plot basis. At their suggestion, the farmers provided areas within their compounds where the animals were tethered. In those cases where a farmer had more than one treatment, the area was divided to separate different treatment groups. Animals were tagged; and tags, bowls for feed, and ropes to tie the fodder were color-coded according to treatment. It was recognized that for the L treatment, the fodder was unlikely to be sufficient and farmers were not prevented from providing their own inputs to animals on these treatments, once the material from the experimental plots had been used up. In these instances, the material provided, amounts, and costs were monitored. Even for the animals on BB+ and BB treatments, some farmers opted to provide additional feed resources in the form of dussa from millet or sorghum grain. In these instances, the quantities fed were estimated, and samples taken for analysis of Nand P. The animals were weighed at the start of the feeding period and thereafter every two weeks. Manure and urine produced during the course of the feeding trial were allowed to accumulate in situ, and kept in the treatment compartment, together with any feed refusals. At the end of the feeding period, in late May, samples of this manure/compost were collected for analysis of N and P. The manure/compost collected during the feeding period was applied to the same treatment plots shortly before planting in 1999. The costs of inputs used were recorded on a plot by plot basis, and included the plant- ing material, fertilizer, pesticides, purchased manure, and labor. Local market prices for grain and fodder were recorded year round. Information on the sociocultural circumstances relating to farmers' crop-livestock management was also collected during the experiment, largely through village-based technicians and extension officers who interacted closely with both participating and nonparticipating farmers. In 1999, in addition to the farmers at Bichi, a similar experiment commenced at Ungu- wan Zangi (8 °OS'E, 11 °lS'N), a village 60 km northeast of Zaria, in northern Nigeria, with 23 farmers participating. Unguwan Zangi is further south than Bichi, has a longer growing season, and slightly poorer market access. Treatments were the same as for Bichi in 1999, but the varieties were cowpea IT86D-7l9 and sorghum KS V 8. Unguwan Zangi had been characterized in the medium to high resource use intensity domain as part of a survey carried out in 1997 within the context of the Ecoregional Program for the Humid and Subhumid Tropics of Africa (EPHTA) (Manyong et al. 1998). Preliminary results Crop yields The estimated quantities of cowpea grain and fodder in the BB treatments were greater than those in the local treatment (Fig. 3). The most dramatic difference was for cowpea grain at Bichi in 1998 where the BB+ treatment yielded more than double the BB and about 16 times the L. Fodder yields for BB+ were one and a half times more than BB and five times more than L. In 1999, these differences were less marked, partly because the yields from L were higher. In many instances, although not quantified, this could be related to an increase in the number of farmers adopting some aspects of the best-bet options-varieties and/or cropping patterns. In terms of quantity, the grain and fodder from improved sorghum did not differ much from the local sorghum, but the farmers 241 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Dry-matter yield (kglha) 5 grain 5 fodder CP grain CPfodder 2500 2000 1500 1000 500 0 5 grain CP grain 2500 2000 1500 1000 500 0 S grain 5 = sorghum, CP = Cowpea Figure 3. Estimates of dry-matter yields of grain and fodder. From top to bottom, Bichi, 1998; Bichi, 1999; Unguwan Zangi, 1999. Diagonal hatching: BB+; Solid shading: BB; Vertical hatching: L. 242 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research indicated a preference for the improved sorghum, both in terms of cooking quality and time for the grain, and the fodder quality. The farmers' observation of the latter was backed up by analysis that showed about 30% of the local sorghum fodder, which had tall and thick stems, to be edible, compared to at least 60% of the improved, with shorter, thinner stems. Comparing actual fodder yields for both cowpea and sorghum in 1998 indicated that there were considerable losses of the dry fodder during transportation and storage. In some instances, the actual fodder yield when converted to kg/ha was as little as 20% of that predicted from the quadrant harvests. These losses were, to some extent, reduced in 1999 with careful handling, and minimized movement of the fodder for weighing. Double cropping was not fully implemented to date. In 1998, farmers were reluctant to harvest the first cowpea crop, as the rains, atypically, continued later than usual. This had two effects; one was that the farmers wanted to continue picking the ripe pods and the other was that they did not want to harvest fodder when the environment was still wet meaning the fodder would not dry, but become rotten and be unpalatable to the animals. This limitation was further emphasized by labor requirements for harvesting tomato and pepper on other parts of the farm at the time the second cowpea crop was to be planted. A few farmers at Bichi in 1999 and 1998 implemented double cropping and were able to harvest both grain and fodder. At a recent field day, samples of fodder from the second cowpea crop were compared visually with those from the first. Farmers agreed that the second crop was clearly of better quality, based on a visual comparison of the leafiness and greenness-criteria they usually use to assess fodder quality. Livestock productivity F or livestock feeding, using the fodder harvested in 1998 to feed small ruminants during the 1998/99 dry season, only eight farmers at Bichi were able to participate so the results should be viewed with some caution, considering also the farm-to-farm variation. These preliminary data indicated that animals on the BB+ treatment gained significantly more weight during the last six weeks of the l6-week feeding period than those on BB or L (Fig. 4). Overall, the average liveweight gains (averaged over all farmers) were 3.54 kg per animalforBB+, 0.91 kg (BB) and2.l9kg (L). While manure quantities produced by animals on the different treatments (manure here is used to refer to the manure plus feed refusals-all that was collected and returned to the field) did not differ significantly, the N content was 1.35% (BB+) 1.09% (BB) and 0.80% (L). P contents were estimated as 0.28% (BB+) 0.27% (BB) and 0.25% (L). These values are within the ranges reported by Tarawali et al. (2001). Figure 4 shows the preliminary results from livestock feeding trials in the 1999/2000 dry season at Bichi (17 farmers participating) and Unguwan Zangi ( 11 farmers). At Bichi, again the BB+ was superior to BB or L, but at Unguwan Zangi it appeared that the two best-bet options were better than the local, but not different from each other. Average weight changes (kg) per animal over the entire feeding period at Bichi were 1.75 (BB+), 0.28 (BB), and 0.03 (L), representing gains of 8, 1.3, and 0.1 %. At Unguwan Zangi, there were slight weight losses for BB+ (0.74 kg) and L (0.78 kg), whereas animals on BB gained an average of 0.9 kg per animal. 243 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production ~ ii E ·c ~ : :c .!!P ~ Q/ ~ ii E ·c 10 25:.--------------------------------------------------. 24 23 22 21 20 19 .. ." ...... .... ----- .. , . ~ ..:o....I . -~.-.p..-=-=-:-tI!!~ _ , .-A--l~-.., ... , .............-.......... & .... 18,1-~~_,~=__,--~_,,_~--,_~--,_~--,_,_~_,,_~~ " n U M 6 ffi V ill 25 24 23 22 21 20 19 18 time (fortnightly weighing) ••••• .. ......................... .. ., .... ~ .....•.. "'I"" .. ~~ I.~~ __ ••• ...-..-•••••• A t1 t2 U t4 t5 t6 time (fortnightly weighing) 25:,--------------------------------------------------, .. 24 --~~~--~. ~ .. ......... :;7 ... , ....... ~- .. - TI •• 22 .... • • • •• ... ........... . . . ... . ! ............... .£. 21 20 19 18,~-------,--------,_------_,------_.--------,_----~ t1 t2 U t4 t5 t6 time (fortnightly weighing) [ • ... BB+ ___ BB - .. - L Figure 4. Average liveweight (kg/animal) for livestock feeding trials. Upper graph Bichi 1998/1999; center graph Bichi 1999/2000; lower graph Unguwan Zangi 1999/2000. 244 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research Nutrient dynamics U sing data from the eight farmers who participated in the feeding trial at Bichi in 1998, it is possible to look at some aspects of nutrient dynamics in these integrated options (Table 2). In simple terms, for nitrogen (N) and phosphorus (P), the inputs have been considered as the soil status for these elements at the time of trial establishment, the manure and fertilizer added, and a small input ofP from the harmattan dust (Harris 1998). Outputs are the nutrients removed in grain and fodder. At present, there has been no attempt to take account of nutrient loss through leaching, volatization, etc. These figures are within the range reported by Harris (1998) for similar farmers' fields in the Kano region and indicate that both Nand P balances were positive at the end of the growing season. It would appear that the cowpea removed more nutrients than the sorghum, or this could be interpreted that the cowpea used the added nutrients more effectively than the sorghum. The strong positive balances are surprising and could be attributed to a number of factors. As indicated above, the N and particularly the P concentrations in the applied manure were quite high, compared to results in other reports (Tarawali et al. 2001). Furthermore, since leaching and volatilization were not considered, it may be inappropriate to include the initial soil Nand P and the contribution from P in the harm at- tan dust. If these factors are excluded, and the manure N contents reduced to 1.5 and P to 0.2%, then the balances are only just positive (Table 2). This information is at present inadequate to enable estimation of the role of cowpea in promoting nutrient cycling, and the nutrient balances need to be monitored for several more seasons, including the returns to the system from the manure and crop residue refusals, removal of subsequent crop harvests, etc. At this point, the emphasis is that nutrient dynamics is being monitored in these studies and should provide quantitative information on whether nutrients are being mined by this more intensive production system, if the applied nutrients are being opti- mally used, and how the improved options compare with farmers' traditional systems. Economics The objective of the economic evaluation is to compare the costs, returns, and profits among the three treatments as a basis for further assessing the desirability of introducing the best-bet options. Although a whole system analysis is planned, as an example, only a partial result on the treatments is presented here, based on the results of the crop yields in 1999 at Bichi. This approach will subsequently be expanded to include an estimation of the value of the livestock products (increased liveweight and manure nutrients), rather than, as treated in this example, considering the monetary value of the crop residues as if they were all sold. In order not to bias the comparison between the improved and local varieties, average market prices for the study area were used for inputs and outputs. Labor data were collected separately for hired and family labor and include the cost of ridging, planting, spraying, fertilizer application, weeding, remolding, and harvesting. Material costs include fertilizers, insecticide, seeds, and manure. Results of the partial economic analyses are summarized in Table 3. Because farmers use a lot of family labor (about 70% of the total for most operations), the cost of which is often not estimated, figures are presented for both total costs which includes an estimate of family labor, and the actual costs where this value is excluded. One of the most striking features is the difference in costs for labor and materials between BB+ and BB. In 1999, the only difference between these two options was that BB+ had improved sorghum and 245 Digitized by Google Table 2. Estimated nitrogen and phosphorus inputs and outputs (kglha) during the first year of the trial at Bichi (1998). BB+ BB Nitrogen Phosphorus Nitrogen Phosphorus Nitrogen Inputs Soil 7.9 0.1 7.8 0.1 8.3 Man u re (1.6% N; 0.7% P) 48.0 21.0 48.0 21.0 48.0 Inorganic fertilizer 35.0 15.0 0.0 0.0 0.0 Harmattan dust 0.8 0.8 Total inputs 90.9 37.0 55.8 22.0 56.3 N Outputs ~ 0'\ Sorghum grain 6.6 0.9 5.0 0.7 6.1 Sorghum fodder 3.8 0.8 3.4 1.0 5.0 Cowpea grain 27.8 2.2 19.7 1.6 1.7 Cowpea fodder 16.5 2.0 12.1 1.2 10.4 Total outputs 54.6 5.9 40.1 4.6 23.2 0 Balance 36.4 31.1 15.7 17.4 33.2 0" "" N" CD Balance with 1.5% Nand 0.2% P, 25.4 15.1 4.9 1.4 21.8 D- o- excluding soil and harmattan '< C"') BB+ = Best bet option with inputs. 0 BB = Best bet options without inputs. ~ L = Traditional sorghum and cowpea. ......... (i) l Phosphorus 0.1 21.0 0.0 0.8 21.9 0.9 0.8 0.1 1.1 2.9 19.1 3.1 bl ~ ~ § ~ 5' c:: g. ::J '" o ~ 3 s· OQ '" ~ ~ ;;;- l o :3 ?i. i" (§ :3 ~ o ...., 8 ~ ~ "tl a @- n g. Cowpea as a key factor for a new approach to integrated crop-livestock systems research Table 3. Summary of partial economic analyses for the three treatments. Total costs include the value of family labor, which is not accounted for in the values for actual costs. BB+ BB L Total cost Total revenue 32069 35181 19872 Materials 8746 10796 4767 Labor 12581 16675 10644 Total costs 21327 27471 15411 Gross margin 10742 7710 4461 Benefit: cost ratio 1.50 1.28 1.29 Actual cost Total revenue 32069 35181 19872 Materials 8746 10796 4767 Labor 3355 3617 3004 Total cost 12101 14413 7771 Gross margin 19968 20768 12101 Benefit: cost ratio 2.65 2.44 2.56 Values are all in Naira/hectare (at the time of writing, No100 = U5$1.00). BB local sorghum. Closer analysis of the information reveals that BB has 23% more material costs, with the highest component of this being a 30% increase in the cost of seed. Labor costs were even more different, with BB having 32% more labor costs than BB+. Within these costs, BB had higher costs than BB+ for remolding (86%), harvesting (34%), and weeding (39%). It can be speculated that these differences are related to the higher yield of the local sorghum and its tall stature (this could have necessitated more remolding to make sure the tall stalks did not get blown over late in the season). Because the local sorghum plants are generally bigger than the improved variety, they may have been planted less densely and therefore more space between plants could have meant more weeding. Alternatively, moving through these taller plants to weed could have been more difficult and therefore more time consuming. While BB+ required 38% more inputs than L, the revenue was 77% more, indicating that increased yields amply compensated for the investment in fertilizers and insecticides. Total revenue from the crop enterprise (grain and fodder) was highest for BB, followed by BB+, representing increases of 77 and 61 % respectively, over L. Income differences related almost entirely to differences in yield. All treatments, in both scenarios including and excluding family labor gave positive gross margins and benefit cost ratios greater than one, indicating that the system as a whole is quite profitable. BB+ had the highest benefit-cost ratio. For both the best-bet treatments, about 70% of the revenue was from cowpea grain and fodder, with the balance being contributed by the sorghum component. By contrast, 59% of the revenue in the L treatment was obtained from cowpea. About one-fifth of the cowpea revenue in BB+ and BB was contributed by cowpea fodder, but as much as 25% of the cowpea revenue in the L treatment was from fodder. Such considerations suggest that it may be more profitable for a farmer to grow only cowpea, if maximum profit is the aim. Indeed, hypothetical calculations comparing potential partial budgets from 100% cowpea or 100% sorghum fields, based on these figures, give higher benefit-cost ratios 247 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production for cowpea only 1.82 (BB+); 1.42 (BB), and 1.46 (L). If only sorghum were to be grown, benefit-cost ratios fall to 1.20 (BB+), 1.28 (BB), and 1.3 (L). Nevertheless, it is important to keep these hypothetical examples in the context of the family needs; no farmer could afford not to grow some sorghum because it is the staple family diet. This stresses the importance of considering not only the economic values, but the social context of the introduced technologies. It could also be argued that maintaining the intercropping system used by farmers ensures some degree of risk diversification. A win-win situation? In Nigeria, with an estimated 4 million ha planted annually to cowpea (FAO 2000), ifwe were to estimate that the best-bet options would be appropriate for one-third of this, and take the lower figure of a doubling in grain yield and apply it to the 538 kg/ha average national yield (FAO 2000), the implication would be an increase of 0.7 million tonnes of cowpea grain. Applying similar speculations to livestock figures, Winrock (1992) estimates 56% of the goats and 64% of the sheep in sub-Saharan Africa are in the dry savannas. If these estimates are applied to current FAO figures for the numbers of sheep and goats in Nigeria (FAO 2000), then an estimate is obtained of 13.6 million goats and 13.1 mil- lion sheep in the dry savannas of Nigeria. From the livestock feeding trials carried out in Bichi in 1998/99, those animals on BB+ gained 1.6 times more weight than the local treatment animals. If the intervention were to reach one-third of the small ruminants in the Nigerian dry savanna, this would mean 8.9 million animals gaining an extra 1.35 kg each per annum, a total of 11.6 million kg liveweight-in the region of 5 million kg of extra meat, or 0.6 million animals. If these 0.6 million animals produced manure at the rate of 1 kg/day ITL U and a nitrogen content of 7%, this could represent about 12000 tonnes of nitrogen (although this figure does not take account of volatilization or leaching). Clearly, these figures are really speculation, and it is not possible to put a time scale on the adop- tion of these interventions at this point. Furthermore, these are based on calculations of productivity alone, and it is important to recollect that the aim of the best-bet options is not solely to increase productivity, but to do so in a way that is sustainable and does not destroy the natural resource base, as well as being economically and socially attractive to farmers. In this context, it is important to take into consideration the nutrient dynamics, and to ask whether we are really intensifying production without mining the soil. This ques- tion requires several years of data to answer, and there are opportunities to continue to optimize the nutrient use. In order to identify what some of these options might be, complementary trials have been carried out in Niger, where, in farmer-managed trials involving 10 farmers in the Sahelian zone at Sadore, hill placement of small quantities of fertilizers and broadcasting of phosphate rock of Tahoua were compared with farm- ers' practices in continuous, intercropping, and rotation systems. The farmers' practices without any input yielded 497 kg/ha of millet grain whereas about an additional 300 kg/ha was obtained with broadcasting oflocally available phosphate rock of Tahoua plus 4 kg P /ha of compound P fertilizers. With the addition of nitrogen fertilizers, whereas in continuous cropping, 881 kg/ha of millet grain was harvested, 1135 kg/ha was obtained when millet was rotated with cowpea. In the intercropping system, in addition to 858 kg/ha of millet grain, 234 kg/ha of cowpea grain was harvested. It is important to note 248 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research that the benefit of selling the cowpea grain will be enough to purchase the needed external inputs in this case. The calculations of partial budget data, based on the crop yields only, suggest that the best -bet options are profitable for farmers. Including the livestock values in the calculations is likely to enhance this even further. In trials established in 2000, the introduction of improved cowpea grain storage methodology, using a simple triple bagging method (Murdock et al. 1997) is anticipated to increase income from cowpea grain even more. By storing the cowpea grain without fear of insect attack, farmers can keep the grain for at least three months when the price could increase by as much as threefold. Semistructured interviews with participating farmers are planned during 2000 and 2001 in order to assess the social context into which these interventions fit, and to better elucidate farmers' perceptions and priorities. Acknowledgements Funding from the Systemwide Livestock Program (SLP) of ILRI for the joint institute research on crop-livestock systems in the dry savannas during the development phase described in this paper is gratefully acknowledged, together with the support of the SLP coordinator, Jimmy Smith. Outstanding technical assistance was provided by A. Adediran, H. Ajeigbe, T. Ayedogbon, Z.B. Jamagani, S. Mohammed, A. Musa, S. Odeh, and Ben I. Yusuf. References Badiane, O. and C.L. Delgado. 1995. A 2020 vision for food, agriculture, and the environment in sub-SaharanAfrica. International Food Policy Research Institute (IFPRl), Washington DC, USA. 56 pp. Bagayoko, M., S.C. Mason, and S. Traore. 1998. The role of cowpea on pearl millet yield, N uptake and soil nutrient status in millet--cowpea rotation in Mali. Pages 109-114 in Soil fertility manage- ment in West African land-use systems, edited by G. Renard,A. Neef, K. Becker, and M. von Oppen. Margraf Verlag Weikersheim, Germany. Berner, D.K., R.J. Carsky, K.E. Dashiell, J. Kling, and V.M. Manyong. 1996. A land management- based approach to integrated Striga hermanthica control in sub-Saharan Africa. Outlook on Agriculture 25: 157-164. Carsky, R.J. and D.K. Berner. 1995. Benefits of crop rotation with soybean and cowpea in savanna cereal-based systems. Pages 391-402 in Technology options for sustainable agricultural produc- tion in sub-SaharanAfrica, edited by T. Bezuneh,A.M. Emechebe, J. Sedgo, and M. Ouedraogo. Semi-Arid Food Grain Research and Development (SAFGRAD), Ouagadougou, Burkina Faso. Delgado, C., M. Rosegrant, H. Steinfeld, S.K. Ehui, and C. Courbois. 1999. Livestock to 2020. The next food revolution. Food,Agriculture and the Environment Discussion Paper 28. International Food Policy Research Institute (IFPRl), Washington DC, USA; Food and Agriculture Organiza- tion of the United Nations (FAO), Rome, Italy; and International Livestock Research Institute (lLRl), Nairobi, Kenya. de Haan, C., H. Steinfeld, and H. Blackburn. 1997. Livestock and the environment. Finding a bal- ance. Report of a study coordinated by FAO, USAID, and the World Bank. FAO, Rome, Italy. Ehui, S., H. Li Pun, V. Mares, and B. Shapiro. 1998. The role of livestock in food security and environmental protection. Outlook on Agriculture 27: 81-87. FAO. 2000. FAOSTAT Database. http://apps.fao.org Food and Agriculture Organization of the United Nations, Rome, Italy. Accessed June 2000. 249 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Harris, F. 1998. Farm-level assessment of the nutrient balance in northern Nigeria. Agriculture, Ecosystems and Environment 71: 201-214. ICRISAT. 1991. ICRISAT West African program annual report. 1990. International Research Institute for the Semi-Arid Tropics (lCRISAT) Sahelian Center, Niamey, Niger. ILRI. 2000. ILRI strategy to 2010. Making the livestock revol ution work for the poor. International Livestock Research Institute (ILRI), Nairobi, Kenya. McIntire, J., D. Bourzat, and P. Pingali. 1992. Crop-livestock interaction in sub-Saharan Africa. World Bank, Washington DC, USA. Manu, A., T.L. Thurow,A.S.R. Juo, I. Zanguina, M. Gandah, and I. Mahamane. 1994. Sustainable land management in the Sahel: a case study of an agricultural watershed at Hamdallaye, Niger. TropSoils Program, Soils and Crop Sciences Department, Texas A & M University, USA. Manyong, VM., K.O. Makinde, and 10. Olukosi. 1998. Delineation of resource-use domains and selection of research sites in the northern Guinea savanna ecoregional benchmark area, Nigeria. Paper presented during the launching of the northern Guinea savanna ecoregional benchmark area, 2 December 1998. Institute of Agricultural Research (lAR), Zaria, Nigeria. I1TA,Ibadan, Nigeria. Murdock, L.L., RE. Shade, L.w. Kitch, G. Ntoukam, 1 Lowenberg-DeBoer, J.E. Huesing, W. Moar, O.L. Chambliss, C. Endondo, and J.L. Wolfson. 1997. Postharvest storage of cowpea in sub- Saharan Africa. Pages 302-312 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropi- cal Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Naazie,A. and lW. Smith. 1997. Modelling feed resources budgets in the moist savannas of West Africa. Pages 197-198 in Proceedings ofthe XVIlI International Grassland Congress, June 1997, Winnipeg and Saskatoon, Canada. Ogungbile,A.O., R Tabo, and N. van Duivenbooden. 1999. Multiscale characterization ofproduc- tion systems to prioritize research and development in the Sudan savanna zone of Nigeria. (sum- mary in English and French). Information Bulletin no. 56. International Crops Research Institute for the Semi-Arid Tropics, Patancheru 502324, Andhra Pradesh, India. 112 pp. Okike, I., MA. Jabbar, V M. Manyong, J.w. Smith, J.A. Akinwunmi, and S.K. Ehui. 2001. Agri- cultural intensification and efficiency in the West African savannas: Evidence from northern Nigeria. Socioeconomics and Policy Research Working Paper 33. International Livestock Research Institute (lLRI), Nairobi, Kenya. 54 pp. Ousman Badiane, O. and C.L. Delgado (editors). 1995. A 2020 vision for food, agriculture and the environment. International Food Policy Research Institute (lFPRI), Washington DC, USA. Schlecht, E., F. Mahler, M. Sangare, A. Susenbeth, and K. Becker. 1995. Quantitative and qualita- tive estimation of nutrient intake and faecal excretion of Zebu cattle grazing natural pasture in semiarid Mali. Pages 85-97 in Livestock and sustainable nutrient cycling in mixed farming systems of sub-Saharan Africa, edited by 1M. Powell, S. Fern{mdez-Rivera, T.O. Williams, and C. Renard. International Livestock Centre for Africa (lLCA), Addis Ababa, Ethiopia. Singh, B.B., O.L. Chambliss, and B. Sharma. 1997. Recent advances in cowpea breeding. Pages 30-49 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). IlIA, Ibadan, Nige- na. Singh, B.B. and SA. Tarawali. 1997. Cowpea and its improvement: key to sustainable mixed crop-livestock farming systems in West Africa. Pages 79-100 in Crop residues in sustainable mixed crop-livestock farming systems, edited by C. Renard. International Crops Research Insti- tute for the Semiarid Tropics (lCRISAT), International Livestock Research Institute (ILRI), and CAB International, Wallingford, UK. 250 Digitized by Google Cowpea as a key factor for a new approach to integrated crop-livestock systems research Tarawali, SA., B.B. Singh, M. Peters, and S.F. Blade. 1997. Cowpea haulms as fodder. Pages 313-325 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (ilTA) and Japan International Research Center for Agricultural Sciences (J1RCAS). ilTA, Ibadan, Nige- na. Tarawali, SA., J.w. Smith, P. Hiernaux, B.B. Singh, S.C. Gupta, R. Tabo, F. Harris, S. Nokoe, S. Fernandez-Rivera, and A. Bationo. 2000. Integrated natural resource management-putting livestock in the picture. Paper presented at the Integrated Natural Resource Management meeting, 20-25 August 2000, Penang, Malaysia. Tarawali, S.A.,A. Larbi, S. Fernandez-Rivera, andA. Bationo. 2001. The role of livestock in the maintenance and improvement of soil fertility. Pages 281-304 in Sustaining soil fertility in West Africa, SSSA Special Publication No. 58. Soil Science Society of America and American Soci- ety of Agronomy, Madison, USA. Winrock. 1992. Assessment of animal agriculture. Winrock International, Morrilton, Arkansas, USA. 251 Digitized by Google 4.2 Cowpea rotation as a resource management technology for cereal-based systems in the savannas of West Africa R.J. Carskyl, B. Vanlauwe2, O. Lyasse3 Abstract A synthesis of results from the savanna zone of West Africa suggests that cowpea rotation can be considered to be an effective resource management technology in cereal-based systems. Part ofthe N requirement of cereal crops can be satisfied by cowpea crop rotation. Furthermore, benefits of cowpea rotation are sometimes higher than expected based on the N content of the cowpea crop alone. Reasons for this include substantial root biomass and N, substantial N-sparing by the legume, and other benefits such as reduction in Striga hermonthica, or pests and often diseases, and possibly access to sparingly soluble P. The characteristics to be encouraged to maximize the N benefit are the amount of nitrogen derived from the atmosphere and the amount ofN returned in the residues. In addition the data sug- gest that (1) the maturity class ofthe cowpea variety should be as late as possible, (2) the cereal should be planted as soon as possible after cowpea has been harvested, and (3) minimum soil requirements for optimum cowpea growth should be respected. These can be considered as recommendations to be followed ifappropri- ate for local agroecological and socioeconomic circumstances. Introduction Herbaceous legumes as cover crops occupy land meant for food production, therefore, grain legumes are usually more acceptable to farmers than cover crops (Schulz et al. 2001). However, the potential benefit to the soil and subsequent crops from grain legumes is less. We reviewed the literature to learn more about the benefits of cowpea to cereal-based cropping systems in the savannas of West Africa to help design better systems. Several examples of short-term rotation trial results (Table 1) show a clear benefit of cowpea rota- tion. The benefit may be due to N supply by the legume, non-N effects, or a combination of the two. First we explore the nitrogen contribution and then non-N benefits of cowpea rotation. Based on this we give recommendations to optimize the benefits of cowpea rota- tion. The recommendations relate to the choice of the cowpea variety to use and how to manage the cowpea crop, with special emphasis on P fertilizer management. Evidence and estimates of N benefit of cowpea rotation The N benefit of cowpea includes the contribution to the soil-plant system through bio- logical N fixation (Fig. 1). Because legumes fix N from the atmosphere, we expect an N contribution to subsequent cereal crops. The direct contribution to the soil-crop system is 1. International Institute of Tropical Agriculture, I1TA-Benin, B.P. 08-0932, Cotonou, Benin. 2. Tropical Soil Biology and Fertility Institute ofCIAT, PO Box 30677, Nairobi, Kenya. 3. VVOB, Londenstraat, 30, Paramaribo, Suriname. 252 Digitized by Google Cowpea rotation as a resource management technology Table 1. Benefit of cowpea rotation to cereal with low « 30 kg/hal or no N application. Control crop Test crop Sorghum maize Sorghum maize Millet sorghum Grass maize Millet millet Millet millet Millet millet Maize maize Maize maize Maize maize Maize maize Harvest BNF Test crop yield (kg/ha) Control After cowpea Source 2394 2690 1945 2795 482 835 650 943 410 1590 760 2190 1148 1685 952 1857 1830 2740 1167 1879 1220 1940 Harvest/burning/grazing • Decomposition Jones (1974) Nnadi et al. (1981) Stoop and van Staveren (1982) Carsky et al. (1999) Rodriguez (1986) Reddy et al. (1994) Bagayoko et al. (1997) Dakora et al. (1987) Horst and Hardter (1994) Osei-Bonsu and Asibuo (1997) Jeranyama et al. (2000) Volatilization Atmospheric deposition Soil Fertilizer Leaching, runoff, erosion Figure 1. Schematic of major and minor N fluxes to be found in a cowpea sole crop. Major N inputs to the soil-plant system are biological N fixation (BNF) and major N outputs are harvest of grain and export of vines. the amount of N in the legume crop derived from the atmosphere (Ndfa). The rest of the N content of the cowpea plant is absorbed from the soil. Part of the plant N is exported in the grain, which has a high N content (3-4%). The N in the cowpea residue is then available to the soil and subsequent crop. Thus, it is important to know the partitioning of N within the plant. The proportion of N in the grain to the total aboveground N is the 253 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production nitrogen harvest index (NHI). As a rule of thumb, a legume increases the soil N pool if the proportion ofN fixed from the atmosphere exceeds the NHI (Giller et al. 1994). The N balance from a cowpea rotation is an estimate of the N accrual to the soil-plant system. Some examples of N balance from the literature are presented in Table 2. This synthesis shows that N balance is generally positive because approximately 60 to 70% of N is derived from the atmosphere (not the soil) and the N exported in the grain is gener- ally less than 50%. The low Ndfa observed by Carsky et al. (2001) was for a very early local variety on relatively poor soils, although soil P was apparently adequate. On low P soil, Sanginga et al. (2000) found that N balance of cowpea increased slightly with P application (Table 3). The available estimates of N balance take only the aboveground cowpea crop into consideration. The roots will contain some N derived from the soil and some N derived Table 2. Examples of N balance calculations for cowpea in the West African savanna. Uptake Aboveground Grain N Aboveground N from soil Ndfa removed N in residue balance (kg/ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha) Source 29 17 19 27 -2 1 25 201 76 150 125 2 29 70 50 49 20 3 64 93 63 94 30 4 (a) 36 61 48 49 13 4 (b) 62 92 88 66 4 5 Sources: 1. Carsky et al. (2000): mean ofthree early maturing cowpea varieties (60-70 days); estimate of soil uptake from nearby fallow plots. 2. Dakora et al. (1987): one medium-late duration (approximately90-110 DAP) cowpea crop grown in the Guinea savanna of Ghana. 3. Eaglesham et al. (1982): mean offour cowpea varieties grown in pots. 4. Horst and Hardter (1994): two consecutive early duration (60 DAP) crops grown in the Guinea savanna of Ghana during 1984(a) and 1985(b). 5. Awonaike et al. (1990): three cowpea varieties at 57 DAP in the derived savanna of Nigeria. Table 3. Effect of P application on the N-balance (kg N/ha)t of cowpea lines grown in a low P soil (source: Sanginga et al. 2000). P application (kg P/ha) Cowpea lines 0 20 40 60 Non-P-responders IT81 D-715 -2.9 0.1 5.5 11.1 Danila 3.2 -5.8 -2.3 6.3 IT90K-59 -2.6 -10.6 -9.3 -3.6 IT89KD-349 -2.4 -4.0 -8.4 8.1 P-responders IT89KD-374 -1.6 7.6 -0.1 0.9 IT82D-716 -9.3 1.9 -9.0 -2.1 IT82KD-391 -4.8 0.9 2.8 -5.0 IT82D-849 -10.4 -6.4 -1.8 7.7 LSDo.os (P level) 6.0 LSDo.os(cowpea line) 8.5 tN balance is calculated from the difference between total N fixed and total N exported in seeds. 254 Digitized by Google Cowpea rotation as a resource management technology from the atmosphere and thus, accounting for root N may make the N balance more positive. The belowground cowpea biomass may be a source of N for a subsequent cereal crop. Estimates of cowpea root dry matter are extremely variable ranging from 0.3 Mg/ha (Carsky 2000) to 2.9 Mg/ha (Groot et al. 1995). Poulain (1980) assumed 0.5 Mg/haof cowpea roots as a probable range. Root N concentration was 1.5 and 2.5% for two varieties grown and sampled by Nnadi and Balsubramanian (1978). Root N, if measured, may help to explain the beneficial effect of cowpea rotation when above- ground N balance does not appear sufficient. Franzluebbers et al. (1994a) estimated the contribution of the cowpea roots to the following sorghum crop to be in the range of one-fifth of the whole cowpea plant used as green manure. In contrast to this, in a field study conducted by John et al. (1992), the aboveground cowpea material was removed and cowpea roots only accounted for an increase in soil mineral N content, but did not affect the yield of the subsequent rice crop. When the aboveground cowpea biomass was included, however, the rice yield increased significantly. All or part of the cowpea residue may be exported as animal feed or it may be grazed off or burned off during the dry season. In these cases, the recycled cowpea residue consists only of leaves fallen before harvest (i.e., the litter) and the roots. Estimates of cowpea litter in the literature are rare and those shown in Table 4 indicate extremely variable results in different trials, ranging from less than 0.1 to more than 1 Mg/ha and from less than 5% to more than 60% of total aboveground residue. The nitrogen concentration of cowpea litter in the Nigeria study was 1.7% compared with N content in leaves of 2% (Carsky et al. 2001). The nitrogen fertilizerreplacement value (NFRV) is an estimate of the benefit oflegume rotation for the farmer. It compares cereal yield after a legume to cereal yield after a cereal or fallow control treatment. N fertilizer applied to the control allows estimates of the N benefit of the legume (Fig. 2). The N benefit consists of N derived from the atmosphere (the aboveground and below ground cowpea crop), the N-sparing effect of the cowpea crop and other non-N benefits, and therefore, overestimates the N contribution of the rotation (Wani et al. 1995). The N-sparing effect may result in more N in the soil for a subsequent crop if the N is not lost from the soil profile before the subsequent cereal crop (e.g. by leaching). Although it is an apparent benefit to the subsequent cereal crop, it is not a contribution to the soil-plant system. While N supply is the major benefit of cowpea rotation with cereals, non-N benefits are possible. In order to ascertain whether there are non-N benefits, there should be a full Table 4. Haulm and surface litter of cowpea measured in West Africa. P applied Haulm Litter Site; year (kg/ha) (kg/ha) (kg/ha) Nigeria, 1996+ n.a. 1249 46 Nigeria, 1997+ n.a. 1601 63 Benin, 1998* 0 534 1273 Benin, 1998* 3()§ 1038 1971 + On-farm trials at lOo24'N; 7°42'E, mean ofthree early varieties in two replicated trials measured at harvest approximately 10 weeks after planting. Source: Carsky et al. (unpublished data). * Research station field at 6°36'N, 2°14'E, variety NI-86-650-3, measured at 12 weeks after planting. Source: Vanlauwe et al. (unpublished data). § Application of 30 kg P/ha as triple superphosphate. 255 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production 2500 ----.. Ii 2000 ,k'- -- .s::. -~ ,k' 'C , Qj , , , 's.. 1500 , , , , .-II~ C. , , --- --0 , --- --- ... , --- u .... ..:::.--- -1/1 1000 ~ --Cereal 500 --.- Legume 1 __ • __ Legume2 0 0 20 40 60 N applied to test crop (kg/ha) Figure 2. Hypothetical response of cereal to previous cereal or legumes, legume 1 having only N effects and legume 2 having Nand non-N effects. range ofN levels after both preceding crops (cowpea and control) as shown in Figure 2. If the curves converge as N fertilizer is applied (as for legume 1 in Fig. 2), then one can characterize the benefit as being due to soil N supply and NFRVs can be estimated from those studies. We will examine results from short (:'S 5 months) and long (> 5 months) ramy season zones. In the short rainy season zone, it is only possible to grow one sole crop per year or one relay intercrop. In Zimbabwe, Jeranyama et al. (2000) grew cowpea and Crotalariajuncea as relay intercrops with maize for two years and in the third year calculated an NFRV of 36 kg/ha compared to continuous maize (Fig. 3). The cowpea yield after maize and after legumes converged at higher N levels, suggesting that only N benefits were realized. At Cinzana, Mali (latitude 13 oN) Bagayoko et al. (1997) grew millet after cowpea for four years with a continuous millet control and found that 40 kg N/ha applied to the continuous millet gave a yield similar to cowpea rotation (Fig. 4). Thus, they estimated the NFRV to be approximately 40 kg/ha. It should be noted, however, that soil N was not higher in the cowpea system than the continuous millet system after four years. The two previous estimates of NFRV are substantial, approaching 40 kg/ha. In contrast, the mean NFRV from two sites in northern Nigeria was only 9 kg/ha for one season of cowpea in the first year followed by one season of maize in the second year in the Guinea savanna of northern Nigeria at latitude lIoN (Carsky et al. 1999; Fig. 5). In this case, the cowpea effect was compared to native fallow rather than a continuous cereal control. A continuous cereal control is likely to give higher estimates of NFRV than fallow because of higher N export by the cereal. When the rainy season is six months long or longer it is possible to grow a cowpea and a cereal crop in succession in the same year. Dakora et al. (1987) grew cowpea in the first growing season followed immediately by maize in the second season after 256 Digitized by Google 4.0 Ii .J:. 3.0 ~ CI ~ C .~ CI CII N 'iii :::E 2.0 1.0 Cowpea rotation as a resource management technology -----+-- After legume --<>--- After maize 0.0 +-----------,---------,---------------, o 60 120 N applied (kg/ha) Figure 3. Effect of previous legume (cowpea and crota/aria data combined) relayed into maize for two years on response of subsequent maize to N fertilizer in Zimbabwe compared to continuous maize. Points derived from equations published by Jeranyama et al. (2000). 20 -----+--After cowpea ~After millet 40 N applied (kg/ha) Figure 4. Effect of cowpea rotation on response of millet to N fertilizer in Mali from 8agayoko et al. (1997). incorporating cowpea (and maize control) residues into the soil. Their estimate of NFRV was approximately 60 kg/ha (Fig. 6). In a similar set of conditions in Nigeria, Carsky et al. (2001) observed that the yield of maize after cowpea was not significantly different from the yield of maize with 0.30 kg N/ha applied after a previous fallow. The apparent NFRV of 30 kg/ha is slightly greater than the N content of the aboveground cowpea residues (Carsky et al. 2001). Root N was not measured in that study and may have accounted for the NFRV observed. Although an early cowpea variety was used ("Achishuru" described 257 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production 2.5 2.0 Ii .s::. -CI 1.5 ~ c 'iii ... 1.0 CI -----+-- After cowpea -~ --<>-After grass i 0.5 0.0 0 20 40 N applied (kg/ha) Figure 5. Response of maize to N after cowpea and after natural grass fallow at two northern Guinea savanna sites in northern Nigeria (Carsky et al. 1999). 2.5 2 Ii .s::. -CI ~ 1.5 c 'iii ... CI CII N 1 'iii :::E -----+--After cowpea 0.5 --<>-After maize 0 0 30 60 30 120 N applied (kg/ha) Figure 6. Effect of cowpea in the first growing season on maize response to N in the second growing season in northern Ghana (Dakora et al. 1987). by Blahut and Singh 1999), it is easy to imagine that immediate incorporation into the soil followed soon by the cereal is responsible for maximizing the benefit. In a study conducted in Niger, Franzluebbers et al. (l994b) observed that at 26 days after incorporation, 13 to 26% of the N released from cowpea was already found in the shoots and roots of the subsequent sorghum crop. How important are non-N benefits of cowpea rotation? Cereal yields are almost always higher after a cowpea crop than after a cereal crop, but this may not be due entirely to N supply. The benefit of cowpea rotation, often thought to 258 Digitized by Google Cowpea rotation as a resource management technology be due solely to biological N fixation, may be related to its influence on pest and disease problems of cereals (including reduction in Striga hermonthica) and other soil benefits such as access to sparingly soluble P. Experimental data supporting these aspects are discussed below. An analysis of the results presented in Table 1 shows that the effect of cowpea rota- tion is greater when the control system was continuous mono-specific cereal (i.e., millet after millet or maize after maize). Yield increase after cowpea compared with continu- ous cereal of the same species was 80% while it was only 31 % for continuous cereal of differing species (i.e., maize followed by sorghum or sorghum followed by millet). This suggests that a mono-specific continuous cereal control may have more pest and disease problems than a different-species continuous cereal control. If this is true, then the benefit of cowpea (providing a break in pest and disease cycles) could also be provided by many other non-leguminous crops. Cowpea would not be the only solution. A study reported by Reddy et al. (1994) clearly shows a non-N benefit of cowpea rotation (Fig. 7). The curves for previous cereal and previous cowpea do not converge. In this case the effect of cowpea rotation appeared to be related to incidence of Striga hermonthica on the cereal test crop as there was more Striga hermonthica on millet after millet than on millet after cowpea. It is not clear whether cowpea actually reduced Striga hermonthica incidence or whether it simply did not result in build-up as the millet did. Ariga et al. (1994) showed how a preceding crop of cowpea variety TVx 3236 reduced Striga hermonthica density on a subsequent maize crop and increased maize yield. The effect increased with the duration of growth of the cowpea crop. However, in condi- tions of very low soil fertility, any source of N may increase emergence and growth of Striga hermonthica (pieterse and Verkleij 1991). For example, cowpea rotation (or N application) was shown to increase Striga hermonthica density on subsequent maize in northern Nigeria (Carsky et al. 1999). 2.5 'ii' 2 ~ .c - • • Cl • C c 1.5 f! Cl I!:I ii ::E 0.5 --+-After cowpea ~After millet 0 0 N applied (kg/ha) Figure 7. Response of millet to N fertilizer after cowpea compared with continuous millet (Reddy et al. 1994). 259 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production It has been reported that some legumes improve bioavailability of sparingly soluble soil P and the same effect might be expected for sparingly soluble P fertilizers (Vanlauwe et al. 2000). In a test of cowpea in southern Benin, application of P as rock phosphate did not increase cowpea biomass (Table 5), but subsequent maize yield after cowpea with rock phosphate is similar to the yield of maize after triple superphosphate. This may have been due to the release of P from rock phosphate over time independent of the cowpea crop. A maize control treatment, with and without rock phosphate, is needed to estimate the cowpea effect on rock phosphate. Arhou and Adomou (2000) conducted such a study for two years (two cycles of cowpea-maize rotation) in southern Benin. The response of maize to rock phosphate in a cowpea-maize rotation was not signifi- cantly different from the response in a maize-maize control system. Research should be conducted using many cowpea varieties to see if there is any potential for cowpea to make P more available (see Bationo et al. this volume). The ability to make sparingly soluble P available may be a heritable trait. Lyasse et al. (2001) tested four cowpea varieties to assess the genotype specific potential to utilize rock phosphate as P source in a P-deficient soil in the derived savanna zone of Nigeria. Significant genotypic variation in terms of both P uptake and grain yield were observed in this study, and one variety was identified to react positively to the appli- cation of RP (Fig. 8). A similar trend was observed for the N-fixation as well as the biomass production at peak physiological growth stage (data not shown). Krasilnikoff et al. (2002) calculated that the same variety (IT90K-59) was also able to deplete the stable P fraction (non-Olsen P) in the rhizosphere. Organic matter replenishment is often mentioned as a possible benefit of cowpea rotation. But the amount of organic matter generated by a cowpea crop is usually not as great as a cereal crop. Furthermore, cowpea residues, because of higher N concentra- tion, may decompose more rapidly than low N cereal residues. This is good for sup- plying N and other nutrients to subsequent cereals but is not conducive to the build up of soil organic matter. In N'Dounga (Niger), Franzluebbers et al. (l994b) found that cumulative C loss from decomposing cowpea residues from the time of incorporation until the end of the rainy season was 78% of the initial cowpea C and no additional soil organic carbon build up was observed when compared to the control treatment without organic amendment. It is possible to calculate the amount of residues needed to maintain or increase SOC given an estimate of soil carbon mineralization rate and the rate of carbon loss from applied residues (De Ridder and van Keulen 1990). Assuming that 6% of soil carbon is mineralized each year (De Ridder and van Keulen 1990) and that 0.35 kg of humus carbon is generated from every kg of residue carbon (Himes 1997), it can be estimated that 3.2 Mg/ha of cowpea dry matter would be necessary to maintain soil organic carbon at 0.3% and 6.4 Mg/ha would be needed to maintain soil organic carbon at 0.6% (Table 6). Typical observations of one to four Mg/ha of aboveground cowpea dry matter (Table 7) and 0.5 to 2 Mg/ha belowground dry matter (as mentioned above) indicate that cowpea, if recycled, could maintain soil organic C at low levels but not at moderate levels. It can be seen from these calculations that it is not possible to increase soil organic C using cowpea. When the cereal yield following cowpea is greater than that following a non-cowpea control even at high N, then non-N benefits should be suspected. When this occurs, follow-up research should be planned carefully after narrowing down the plausible 260 Digitized by Google Cowpea rotation as a resource management technology Table 5. Maize grain yields as affected by previous application of 30 kg P/ha as triple superphosphate (TSP) or 90 kg P/ha as rock phosphate (RP) to cowpea at Sekou (6°36'N; 2°14'E), southern Benin, 1998. P applied to cowpea OP 30TSP 90RP SED Cowpea haulms and litter (kg DM/ha) 1808 3009 1853 329 Source: B. Vanlauwe et al. (unpublished data). 1.2 1 SED I Ii .s::. -CI 0.8 ~ c 'iii "- 0.6 CI CII N 'iii :::E 0.4 0.2 0 IT90K-59 I T89KD-391 IT89KD-288 IT90K-277-2 Maize grain (kg DM/ha) 611 1024 1192 181 Figure 8. Grain yield of four cowpea cultivars as affected by RP application on a low-P soil in the derived savanna of Nigeria at Fashola, (Lyasse et aI., 2001). [SED = standard error of the difference]. Table 6. Quantities of annual C loss by mineralization and organic residues (Mglhalyr) needed to maintain soil organic carbon at initial levels. InitialOC Closs Residue C Residue DM ("!o) (Mglha) (Mglha) (Mglha) 0.3 0.50 1.44 3.20 0.6 1.01 2.88 6.40 1.2 2.02 5.76 12.80 Assumes 1. 6% loss of soil organic carbon per year (De Ridder and Van Keulen 1990). 2. One hectare at 0.20 m depth weighs 2 800000 kg. 3. 0.35 kg humus C for every kg of residue C (Himes 1997). 4. 0.45 kg residue C for every kg of residue dry matter. 261 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 7. Aboveground dry matter (Mglha) of cowpea residue (after harvest) as a function of maturity class and insecticide treatment for several sites and years (number of observations in parentheses). Maturity class Early Medium late Source: Schulz et al. (2001). Spray 2.2 (29) 2.5 (73) 3.3 (99) No spray 2.3 (34) 3.9 (41) 3.9 (69) non-Nbenefits. Trials should be designed to isolate and understand individual effects. This can lead to manipulation of these phenomena to improve the ability of cowpea rotation to maintain soil productivity. Recommendations/strategies to optimize cowpea rotation benefits Choice of variety The first condition for a cowpea rotation benefit is good growth of cowpea. For a posi- tive N effect, the cowpea must nodulate well and fix N from the atmosphere. Varietal characteristics that determine N fixation in grain legumes are discussed in Chapter 8 of Giller (2001). High biological N fixation will not lead to net N benefits if the N harvest is also high. To improve the benefit to the soil, a variety that puts more N in vegetation is preferred although the farm household may require more grain. Data derived from Schulz et al. (2001) show that aboveground residue dry matter increases with the maturity class of the cowpea (Table 7). Therefore, the benefit of cowpea rotation can be expected to increase in varieties that mature later, even if the harvest index remains constant. Indeed, Stoop and van Staveren (1982) demonstrated that the impact on subsequent millet increased as maturity cycle of preceding cowpea increased. We therefore recommend the variety with the longest agronomically appropriate maturity cycle. Phosphorus deficiency is commonly observed in legumes in the savanna zone. As Sanginga et al. (2000) observed large varietal differences in P requirements for cowpea growth and N2 fixation, this suggests the need to take the P requirements of these cowpea lines into account in plant introduction and plant selection for the moist savanna zone soils. The possibility to use less soluble and much cheaper P-sources (e.g., low reactive rock phosphate) in combination with selected P-efficient cowpea breeding lines could alleviate P depletion. Another important consideration for the choice of cowpea variety is that it should be resistant to biotic and abiotic stresses that reduce aboveground biomass, including Striga gesnerioides and drought. In addition, an obvious advantage in a cereal-based system is the ability to promote Striga hermonthica seedbank reduction. Management of soil and crop In order to optimize the benefit of cowpea rotation with cereals, it is of utmost importance to have sufficient soil P. This is clearly shown by the improvement in cowpea biomass with TSP application in southern Benin (Table 5). This benefit is believed to be related to improved nodulation as shown by many studies where P application increases nodule numbers and nodule fresh weight. The importance of P supply was also shown in a study of cowpea-maize rotation in Nigeria (Fig. 9) in which the effect of cowpea rotation was not important until plant available P was increased above 5 mg/kg. The critical level 262 Digitized by Google Cowpea rotation as a resource management technology of plant available P (the level above which P fertilizer application is not economically justified) has been estimated at 10.6 mg/kg by Aune and Lal (1995) from a small data set. This level should be confirmed as it would eventually be used to guide P fertilizer application by farmers. It is sometimes observed that cowpea, if not adequately protected from insect damage, produces less grain and more leaf and vine dry matter as suggested by the data of Schulz et al. (2001), which is summarized in Table 7. The subsequent benefit to a cereal in rotation may be increased as was observed by Carsky et al. (1999). Although this should not be a goal of cropping systems development, it may provide an internal recovery mechanism for farmers who suffer from insect losses. It may be questioned why cowpea intercropping is not recommended as a resource management technology. Whereas intercropping does benefit the need of the household for balanced food production and risk avoidance, rotation of sole cowpea generally has a much greater effect on a subsequent maize crop than rotation with a cereal-cowpea intercrop. This was shown by Bagayoko et al. (1997) in Mali and by Rodriguez (1986) in Burkina Faso (Table 8). The benefits of cowpea cereal rotation may be realized in a "within-field" rotation where cowpea and cereal rows are swapped over time especially in a spatial arrangement of two rows of cereal and four rows of cowpea. The time between cowpea harvest and cereal planting is obviously important. One only needs to look at the large effect of cowpea rotation in the trials of Dakora et al. (1987) and Carsky et al. (2001) when cowpea was grown in the first growing season followed immediately by maize. Thus, first season cowpea should be considered if the length of growing season permits two crops. Possible problems with the system, which can not be ignored, include loss of cowpea grain quality when harvested mid-season and the need for the household to produce cereals first. Researchers should be aware of these as good reasons to grow cowpea in the second part of the season rather than the first. 1.2 0.2 o o • Cowpea grain <> Maize grain increase from cowpea • <> 2 4 6 Bray-1 P (mg/kg) • 8 <> • 10 Figure 9. Effect of soil P on cowpea grain yield (Mg/ha) and increase in maize grain yield (Mglha) from preceding cowpea compared with preceding native fallow in north- ern Nigeria from Carsky et al. (2001). Cowpea and natural (grass dominated) fallow from May to July were followed by maize from August to October. Plant available P was 6.8 mg/kg at a moderate-P site and 1.4 mglkg at a low-P site where is was increased to 9.1 mg/kg by applying 200 kg of SSP/ha. 263 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 8. Comparison of sole cowpea (C) and millet-cowpea intercrop (M-C) on millet yield (Mglha) under low input conditions. M-M C-M M-C-M Source 0.41 1.59 0.98 Rod ri gu ez (1986)+ 1.17 1.88 1.91 Osei-Bonsu and Asibuo (1997)* 1.38 1.69 1.46 Bagayoko et al. (1997)' Notes: tYield in fourth year only with N application rate of 28 kglha; averaged for two cowpea cultivars. * Yield of maize in second year without N applied. 'Yield averaged for four years and across N application rates of 0, 20, and 40 kglha. Conclusion Adoptability of cowpea is high. Dembele (2000), for example, recorded grain legume sys- tems adoption in Mali to be many times higher than adoption of forage legumes. Oyewole et al. (2000) fOWld that farmers preferred cowpea-maize to Mucuna-maize double cropping to keep grain producing cowpea in the system although the benefit of cowpea was less than that of Mucuna. It appears from our synthesis that cowpea rotation should be considered as an important resource management technology. However, for this to fWlction, systems should be designed that optimize the benefit of cowpea to the soil-plant system. These will include: (l) cowpea varieties with the longest agronomically acceptable maturity cycles, (2) maintenance of adequate P supply, and (3) the shortest possible time between the cowpea and cereal crops in rotation. It will be possible to pursue these strategies in some, but not all socioeconomic and agroecological circumstances. References AIhou, K. and M. Adomou. 2000. Contribution du phosphore a I'amelioration de I 'assimilation de I'azote par Ie maIs en rotation avec Ie mucuna et Ie niebe. Pages 138-147 in Cover crops for integrated natural resource management in West Africa. Proceedings of a regional workshop October 1999, Cotonou, Republic of Benin, edited by R.J. Carsky, JD.H. Keatinge, VM. Man- yong, and A.C. Eteka. I1TA, Ibadan, Nigeria. Ariga, E.S., D.K. Berner, and J. Chweya. 1994. Effects of previous season cotton and cowpea on Striga hermonthica parasitism on maize. Phytopathology 84: 1151. Aune, J.B. and R. La!. 1995. The tropical soil productivity calculator-A model for assessing effects of soil management on productivity. Pages 499-520 in Soil management: experimen- tal basis for sustainability and environmental quality, edited by R. La!. and B.A. Stewart. CRC Lewis Publishers, Boca Raton, Florida, USA. Awonaike, K.O., K.S. Kumarasinghe, and S.K.A. Danso. 1990. Nitrogen fixation and yield of cowpea (Vigna unguiculata) as influenced by cultivar and Bradyrhizobium strain. Field Crops Research 24: 163-171. Bagayoko, M., S.C. Mason, and S. Traore. 1997. The role of cowpea on pearl millet yield, N uptake, and soil nutrient status in millet-cowpea rotation in Mali. Pages 109-114 in Soil fertility management in West African land-use systems, edited by G. Renard, A. Neef, K. Becker, and M. von Oppen. Margraf-Verlag, Weikerseim, Germany. Bationo, A., B.R. Ntare, S.A. Tarawali, and R. Tabo. 2002. Soil fertility management and cowpea production in the semiarid tropics. Pages 299-316 in Challenges and opportunities for enhancing sustainable cowpea production, edited by C.A. Fatokun, S.A. Tarawali, B.B. Singh, P.M. Kormawa, and M. Tamc). I1TA, Ibadan, Nigeria. Blahut, G.R. and B.B. Singh. 1999. Achishuru cowpeas in central Nigeria. I. Origin, diversity and production practices. Samaru Journal of Agriculture 15: 21-28. Carsky, R.J., B. Oyewole, and G. Tian. 1999. Integrated soil management for the savanna zone of W. Africa: legume rotation and fertilizer N. Nutrient Cycling in Agroecosystems 55: 95-105. 264 Digitized by Google Cowpea rotation as a resource management technology Carsky, R.J. 2000. Potential of herbaceous legume cover crop fallow systems in the savanna zone. Pages 594-602 in La Jachere en Afrique tropicale. Proceedings of the International Seminar, April, 1999, Dakar, edited by C. Floret and R. Pontanier. John Libbey Eurotext, Paris. Carsky, R.J., B.B. Singh, and B. Oyewole. 2001. Contribution of early-season cowpea to late- season maize in the savanna zone of West Africa. Biological Agriculture and Horticulture. 18: 303-315. Dakora, F.D., R.A. Aboyinga, Y. Mahama, and J. Apaseku. 1987. Assessment ofN fixation in ground nut (Arachis hypogaea L.) and cowpea (Vigna unguiculata L. Walp.) and their rela- tive N contribution to a succeeding maize crop in northern Ghana. MIRCEN Journal. 3: 389-399. Dembele, E. 2000. Activites liees a I'utilisation des legumineuses herbacees au Mali-sud. Pages 254-258 in Cover crops for integrated natural resource management in West Africa, edited by R.J. Carsky, J.D.H. Keatinge, V.M. Manyong, and A.C. Eteka. I1TA, Ibadan, Nigeria. De Ridder, N. and H. van Keulen. 1990. Some aspects ofthe role of organic matter in sustainable intensified arable farming systems in the WestAfrican semiarid tropics (SAT). Fertilizer Research 26: 299-310. Eaglesham, A.R.J., A. Ayanaba, V. Ranga Rao, and D.L. Eskew. 1982. Mineral N effects on cowpea and soybean crops in a Nigerian soil. II. Amounts ofN fixed and accrual to the soil. Plant and Soil 68: 183-192. Franzluebbers, K., R.W. Weaver, andA.S.R Juo. 1994a. Mineralization oflabeled N from cowpea (Vigna unguiculata [L.] Walp.) plant parts at two growth stages in sandy soil. Plant and Soil 160: 259-266. Franzluebbers, K.,A.S.R. Juo, andA. Manu. 1994b. Decomposition of cowpea and millet amend- ments to a sandy alfisol in Niger. Plant and Soil 166: 255-265. Giller, K.E., J.F. McDonagh, and G. Cadisch. 1994. Can nitrogen fixation sustain agriculture in the tropics? Pages 173-191 in Soil science and sustainable land management in the tropics, edited by J.K. Syers and D.L. Rimmer. CAB. International, Wallingford, UK. Giller, K.E. 2001. Nitrogen fixation in tropical cropping systems. 2nd edition. CAB International, Wallingford, UK. 423 pp. Groot, J.J.R., D. Kone, M. Traore, and N. Kamissoko. 1995. Description du systeme racinaire de I 'Andropogon gayanus, du Vigna unguiculata et du Stylosanthes hamata en zone soudano- sahelienne. Rapports Production Soudano-Sahelienne (PSS) No.8. AB-DLO, Wageningen, The Netherlands. Himes, F.L. 1997. Nitrogen, sulfur, and phosphorus and the sequestering of carbon. Pages 315-319 in Soil processes and the carbon cycle, edited by R. Lal, J.M. Kimble, R.F. Follett, and B.A. Stewart. CRC Press, Boca Raton, Florida, USA. Horst, W.J. and R. Hardter. 1994. Rotation of maize with cowpea improves yield and nutrient use of maize compared to maize monocropping in an Alfisol in the northern Guinea savanna of Ghana. Plant and Soil 160:171-183. Jeranyama, P., O.B. Hestermann, S.R. Waddington, and RR. Harwood. 2000. Relay intercrop- ping of sunnhemp and cowpea into a smallholder maize system in Zimbabwe. Agronomy Journal 92: 239-244. John, P.S., RK. Pandey, R.J. Buresh, and R Prasad. 1992. Nitrogen contribution of cowpea green manure and residue to upland rice. Plant and Soil 142: 53-61. Jones, M.J. 1974. Effect of previous crop on yield and nitrogen response of maize at Samaru, Nigeria. Experimental Agriculture 10: 273-279. Krasilnikoff, G., T.S. Gahoonia, and N.E. Nielsen. 2002. Phosphorus uptake from sparingly available soil-P by cowpea (Vigna unguiculata) genotypes. Pages 239-250 in Integrated 265 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production nutrient management in sub-Saharan Africa: from concept to practice, edited by B. Vanlauwe, J. Diels, N. Sanginga, and R Merckx. CAB International, Wallingford, UK. In press. Lyasse, 0., B.K. Tossah, B. Vanlauwe, J. Diels, N. Sanginga, and R Merckx. 2002. Options for increasing P availability from low reactive rock phosphate. Pages 225-237 in Integrated nutrient management in sub-Saharan Africa: from concept to practice, edited by B. Vanlauwe, J. Diels, N. Sanginga, and R Merckx. CAB International, Wallingford, UK. In press. Nnadi, L.A. and V. Balasubramanian. 1978. Root nitrogen content and transformation in selected grain legumes. Tropical Agriculture (Trinidad) 55: 23-32. Nnadi, L.A., L. Singh, and V. Balasubramanian. 1981. Effect of grain legumes and sorghum on soil nitrogen status and the yield of subsequent maize crop. Samaru Journal of Agriculture 1: 183-190. Osei-Bonsu, P., and J.Y. Asibuo. 1997. Studies on Mucuna (Mucuna pruriens var. utilis) in Ghana. Pages 435-441 in Technology options for sustainable agriculture in sub-Saharan Africa, edited by T. Bezuneh,A.M. Emechebe, J. Sedogo, and M. Ouedraogo. OAU/STRC- Semiarid Food Grain Research and Development, Ouagadougou, Burkina Faso. Oyewole, B., R.J. Carsky, and S. Schulz. 2000. On-farm testing of Mucuna and cowpea double cropping with maize in the Guinea savanna of Nigeria. Pages 137-147 in Cover crops for integrated natural resource management in West Africa, edited by RJ. Carsky, J.D.H. Keatinge, V.M. Manyong, and A.C. Eteka. I1TA, Ibadan, Nigeria. Pieterse, A.H. and J.A.C. Verkleij. 1991. Effect of soil conditions on Striga development: a review. Pages 329-339 in Proceedings ofthe 5th international symposium of parasitic weeds, edited by J.K. Ransom, L.J. Musselman,A.D. Worsham, and C. Parker. CIMMYT, Nairobi, Kenya. Poulain, J.-F. 1980. Crop residues in traditional cropping systems of West Africa-effects on the mineral balance and level of organic matter in soils-proposals for their better manage- ment. Pages 38-71 in Organic recycling in Africa. Papers presented at the FAO/SIDA workshop on the use of organic materials as fertilizers in Africa. FAO Soils Bulletin 43, FAO, Rome, Italy. Reddy, K.C., P.L. Visser, M.C. Klaij, and C. Renard. 1994. The effects of sole and traditional intercropping of millet and cowpea on soil and crop producti vity. Experimental Agriculture 30: 83-88. Rodriguez, M. 1986. Agronomie du mais. Pages BI-B45 in Rapport Annuel 1986 du projet "Recherche et developpement des cultures vivrieres dans les zones semiarides d' Afrique". SAFGRADIlITA, Ouagadougou, Burkina Faso. Sanginga, N., O. Lyasse, and B.B. Singh. 2000. Phosphorus-use efficiency and nitrogen balance of cowpea breeding lines in a low P soil of the derived savanna zone in West Africa. Plant and Soil 220: 119-128. Schulz, S., RJ. Carsky, and S. Tarawali. 2001. Herbaceous legumes: the panacea for West African soil fertility problems? Pages 179-195 in Soil fertility maintenance in West Africa, edited by G. Tian et al. ASA, Madison WI. Stoop, W.A. and J.P. van Staveren. 1982. Effect of cowpeas in cereal rotations on subsequent crop yields under semiarid conditions in Upper Volta. Pages 653-657 in Biological nitrogen fixation technology for tropical agriculture, edited by P.H. Graham and S.C. Harris. CIAT, Cali, Colombia. Vanlauwe, B., J. Diels, N. Sanginga, RJ. Carsky, J. Deckers, and R Merckx. 2000. Utilization of rock phosphate by crops on a representative toposequence in the northern Guinea savanna zone of Nigeria: response by maize to previous herbaceous legume cropping and rock phosphate treat- ments. Soil Biology and Biochemistry 32: 2079-2090. Wani, S.P., O.P. Rupela, and K.K. Lee. 1995. Sustainable agriculture in the semiarid tropics through biological nitrogen fixation in grain legumes. Plant and Soil 174: 29-49. 266 Digitized by Google 4.3 Advances in cowpea cropping systems research 0.0. 0lufajo1 and B.B. Singh2 Abstract Cowpea (f/igna Wlguiculata) [L.] Walp.) is a major component of the traditional cropping systems inAfrica,Asia, and Central and SouthAmerica where it is widely grown in mixtures with other crops in various combinations. The productivity of cowpea in these mixtures is low, mainly due to low plant population, competition under intercropping, and lack of crop protection measures. Studies have shown that the productivity of cowpea in these systems could be enhanced through the use of improved varieties, appropriate date of planting with respect to the cereal, higher plant populations, improved soil fertility, and suitable spatial arrangements. This paper highlights recent research leading to improvements in cowpea cropping systems. These include improved productivity as a result of early cowpea planting, strip cropping, dense planting, and appropriate soil fertility management. For example, in West Africa, the use of high yielding improved varieties in a strip cropping system with two cereal rows: four cowpea rows offers an opportunity for selective input application and appears to be economically superior to the traditional one cereal row: one cowepea row. Introduction Cowpea (Vigna unguiculata [L.] Walp.) is a major component of the cropping systems of the drier parts of the tropics, particularly sub-Saharan Africa. West and Central Africa account for over 64% of the estimated 12.5 million ha cultivated to cowpea worldwide (Singh et al. 1997). This is followed by Central and South America (19%), Asia (10%), and East and Southern Africa (6%). Cowpea is mainly grown in mixtures with other crops and a great diversity of crop mixtures has been reported (Mortimore et al. 1997). In a recent survey, Henriet et al. (1997) reported the existence of up to 43 crop mixtures in the Sudan savanna of Nigeria with a millet-cowpea mixture being predominant, representing 22% of the fields sampled (Table 1). Other dominant crop mixtures included millet-sorghum-cowpea (18.6%), sorghum- cowpea (10.4%) and millet-cowpea-groundnut (7.6%). The importance of cowpea in the cropping systems of the dry savanna is well illustrated by the fact that this crop occurred in 71.4% of the fields sampled. However, the cowpea grain yields in these systems ranged from 0 to 132 kg/ha (Table 2) compared with a sole yield potential of 1500 to 3000 kg/ha under optimwn management (Muleba and Ezumah 1985). 1. Department of Agronomy, Institute for Agricultural Research (lAR), Ahmadu Bello University, PMB 1044, Zaria, Nigeria. 2. International Institute of Tropical Agriculture (lIT A), Kano Station, Sabo Bakin Zuwo Road, PMB 3112, Kano, Nigeria. 267 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 1. Major cropping systems identified at several locations in the Sudan savanna eco- logical zone of Nigeria in 1992 and 1993. Crop mixtures Millet-cowpea Millet-cowpea-groundnut Millet-sorghum-cowpea Sorghum-cowpea-groundnut Sorghum-cowpea Millet Millet-sorghum Millet-groundnut Sorghum Sorghum-groundnut Millet-sorghum-cowpea-sesame Millet-sorghum-cowpea-groundnut Otherst % of different cropping systems 1992 1993 22.0 22.5 15.4 7.6 12.4 18.6 9.7 2.8 8.0 10.4 5.4 6.0 2.7 4.7 5.4 4.5 2.7 2.5 7.0 2.1 8.1 18.3 'Others: those involving sesame, cassava, okra, maize, and bambara nut. Source: Henriet et al. 1997. Table 2. Mean grain yield of cowpea in different crop mixtures in farmers' fields in parts of the Nigerian Sudan savanna (1992 and 1993). Crop mixtures Millet-cowpea Millet-cowpea-groundnut Millet-sorghum-cowpea Sorghum-cowpea-groundnut Sorghum-cowpea 'Figures in parentheses represent the range in values. Source: van Ek et al. 1997. Grain yield (kg/ha)t 1992 22 (5-29) 18 (0-36) 18 (0-40) 13 (0-25) 29 (22-39) 1993 42 (6-129) 40 (8--103) 54 (16-132) 63 (16-104 52 (30-84) The maj or yield-limiting factors of cowpea cropping systems are low plant population, low yield potential of local cultivars, insect pests and diseases, shading by the cereals, drought stress, and low soil fertility. In this respect, there are opportunities for improved management practices that overcome these production constraints and enhance cowpea productivity. These include sowing date, row geometry, pest incidences, and variety improyement. The objective of this paper is to highlight developments in these management areas for cropping systems research mainly in West Africa conducted with cowpea as one of the component crops thus complementing earlier reviews by Muleba and Ezumah (1985) -, I I I I I and Blade et al. (1997). Relative sowing date in intercropping Date of sowing is dictated by many factors including weather, soil moisture, time, and labor constraints faced by the farmer, variety, and crop production system. Cowpea is generally grown as the understorey crop in a cereal- or tuber-based system. In the West African savannas, cowpea is usually relay planted into the cereal crops. It has been noted 268 Digitized by Google Advances in cowpea cropping systems research that in the Sahel, millet yield is reduced if millet and cowpea are planted simultaneously (Ntare 1990; Ntare and Williams 1992). However, Reddy and Visser (1997) concluded that intercropped cowpea should be sown simultaneously or soon after millet for maximum yield of cowpea. They found that, compared to simultaneous sowing, delaying cowpea sowing to seven weeks after millet led to significantly lower crop growth rates (19 to 10 kg/ha), lower grain (1110 to 100 kg/ha), and dry matter (2110 to 560 kg/ha) yields of cowpea. In contrast, grain yield of intercropped millet did not vary significantly with cowpea interplanting time (Table 3). Terao et al. (1997) also advocated simultaneous plant- ing of cowpea and millet if there is no severe competition for water. Over two years, millet yield was not reduced when millet and cowpea were sown simultaneously at Minjibir (800 mm rainfall) while millet yield reduction at Mallam Madori (426 mm) was only 16%. In the Sudan savanna, Blade et al. (1997) found that delaying cowpea planting by two or three weeks resulted in a reduction of cowpea grain yield of over 50% in comparison to simultaneous millet and cowpea planting. In choosing the appropriate time to introduce cowpea into millet, an important consideration is the objective of the farmer-which is to have a full millet grain yield with some additional cowpea grain and fodder. Thus, farmers would be reluctant to adopt any practice that may reduce millet grain yield. Most of the reported work on maize-cowpea mixtures indicated a reduction in cowpea yields while maize yields were unaffected (Haizel1974; Isenmilla et al. 1981; Olufajo 1988; Cardoso et al. 1993). However, the competitive effects from the maize component could be reduced by sowing cowpea early. Myaka (1995) showed that when sown four weeks after maize, cowpea yields were 67% less than cowpea planted two weeks after maize. In both cases, maize yields were not affected by the cowpea component. Being strongly competitive, cowpea reduces cotton yields when grown as an inter- crop and the extent of yield reduction depends on the cowpea sowing date. Results from Endondo and Samatana (1999) suggest that cowpea should be sown five to six weeks after cotton in a cotton-cowpea intercrop. With simultaneous sowing, the intercropped cotton yield was 50% of the sole crop yield whereas the cotton yield was reduced by Table 3. Grain yield (kg/ha) of cowpea and millet as affected by the planting date of cowpea in a millet/cowpea intercropping system in Kolo, Niger Republic, in 1986 and 1988. Cowpea grain yield Millet grain yield Time of planting cowpeat 1986 1988 1986 1988 Simultaneous 1920 690 450 520 8--14 days after millet 1410 550 530 710 15-20 II II II 1340 420 560 720 26--28 II 680 240 610 680 31-36 II 770 80 520 700 42-47 II II 250 0 570 730 56 300 0 590 580 Cowpea sole 2850 1360 Millet sole 480 580 Mean 1190 420 540 650 CV(%) 20 24 21 18 SE 139 50 65 59 tCowpea interplanting dates varied between years depending on the occurrence of a rain of12 mm or more to ensure cowpea germination. Source: Reddy and Visser (1997). 269 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production 16% and cowpea yield by 54% when cowpea was sown five to six weeks after cotton. However, year-to-year differences in the response of cotton-cowpea intercrop have been reported (Natarajan and Naik 1992). In wetter years, Myaka and Kabissa (1996) fOWld that cowpea yield was reduced when cowpea sowing was delayed from two to four weeks after cotton, whereas in drier years, cowpea yield was not affected by sowing date. An important consideration with respect to cotton-cowpea intercropping is the time of insec- ticide application to the cotton component. Since farmers routinely apply insecticide to cotton whereas cowpea rarely receives insecticide protection, the main advantage of this mixture is the "incidental" benefit derived by the cowpea crop from the insecticide applied directly to cotton. It is noteworthy that the increase in cowpea grain yield as a result of the insecticide applied to cotton could be as high as 400% (Table 4). Further improvement in cowpea grain yield in this mixture could probably be achieved by using early maturing cowpea varieties whose reproductive phase would coincide with the period of insecticide application to cotton. In order to avoid contamination of cowpea, it is important to apply nonpersistent chemicals to cotton. The only recent data on a cassava-cowpea intercrop are those of Okoli et al. (1996) who reported significant reductions in cassava yield whereas intercropped cassava had no effect on cowpea yield. However, cowpea yields were higher when established simultane- ously with cassava than when introduced later into cassava. Row arrangement and density The traditional production system involves varied arrangements of the component crops in time and space with implications for crop and livestock productivity, and sustainability (Shetty et al. 1995). Spatial arrangements and densities of the component crops have been manipulated in order to enhance complementarity and to reduce competition between the Table 4. Plant density, total aboveground biomass, pod weight, 1000 grain weight, and gross return of cowpea as a sole crop and when interplanted between cotton rows in Zimbabwe. Plant Time of density planting established cowpea ('OOO/ha) Sole cowpeat Simulta- neous 69.0 Cotton- cowpea 1 : 1 37.6 Cotton- cowpea 1 : 1 Staggered* 37.9 Cotton- Simulta- cowpea 1 : 2 neous Cotton- cowpea 1 : 2 Staggered SE tNot sprayed. *Cowpea sown two weeks after cotton. Source: Natarajan and Naik (1992). 68.3 69.8 1.8 270 Total above- ground biomass (tlha) 1.16 3.24 1.93 2.97 1.93 0.32 100 Pod Grain seed Gross weight weight weight return (tlha) (tlha) (g) (Z$/ha) 0.31 0.19 127 376 1.32 0.96 135 1866 0.71 0.49 129 948 1.26 0.87 127 1695 0.77 0.52 128 1012 0.14 0.11 6 213 Digitized by Google Advances in cowpea cropping systems research component crops so that the physiological advantage from combining crop components is maximized (Willey and Osiru 1972; Willey 1979; Ofori and Stern 1 987a, b). In a millet-cowpea intercrop, Odo and Bibinu (1998) reported optimum spatial arrange- ments of one: three and three: three (millet: cowpea rows). Myaka (1995) showed that in a maize: cowpea intercrop, cowpea yields were 57% higher in two : two (maize: cowpea rows) compared with one: one (maize: cowpea rows). Asafu-Agyei et al. (1997) found that two: two (maize: cowpea rows) gave higher yields of maize and cowpea, land equiva- lent ratio (LER) and net benefit than one: one (maize: cowpea rows). Obuo et al. (1998) investigated the effect of intrarow spacing on cowpea-sorghum intercrop and found that yields of both components were highest at 60 x 20 cm inter- and intrarow spacing. Myaka and Kabissa (1996) found that alternating single rows of cotton with single rows of cowpea was superior to two: two or one: two (cotton: cowpea) in terms of crop yield and control of cowpea pests by insecticide applied directly to the cotton component in cotton-cowpea intercrop. Bezerra-Neto and Robichaux (1996) studied the effect of spatial arrangement and density on cotton-cowpea-maize intercrop and reported that the land equivalent ratio for yield was higher in the spatial arrangement of single rows of cowpea and maize between single rows of cotton. Land equivalent ratios for total bio- mass and grain yields were not affected as cotton density increased from 25000 to 75000 plants/ha. However, Bezerra-Neto and Robichaux (1997) noted that component yields and biomass production could be significantly affected by alteration of spatial arrangement and density. They therefore concluded that the most appropriate sowing arrangements in cotton-cowpea-maize intercrop should be determined by individual requirements for total biomass and grain yields. Attempts have been made to plant the component crops in strips. This is advantageous in terms of ease of crop management, fertilizer and insecticide application, weeding, and reduction of the shading effect of cereal on cowpea. There is evidence that strip cropping with two rows cereal: four rows cowpea offers an opportunity for selective input applica- tion and better economic advantage than the traditional one row cereal: one row cowpea spatial arrangement (Singh and Emechebe 1998; Singh and Ajeigbe, this volume). Mensah (1997) noted that alternating three rows of cowpea with two or three rows of sorghum and one to two insecticide applications gave a yield advantage of 58 to 69% and proposed this as the most productive method to be adopted by subsistence farmers. Soil fertility Soil fertility management has a major influence on the overall productivity of the inter- cropping system. In the traditional cereal-cowpea systems of the dry savanna of West Africa, millet is planted in less fertile fields with little or no fertilizer while sorghum is planted in relatively more fertile soils and with application of farmyard manure and fertil- izers when available (van Ek et al. 1997). In the Sahel, phosphorus is the most limiting soil nutrient; nitrogen increases crop yield only in the presence of adequate phosphorus (Fussell et al. 1987; 1992). The application of a small quantity of phosphorus (13 kg/ha) has been suggested for increased productivity of the millet-cowpea system (Shetty et al. 1995; Subbarao et al. 1999). Bielders (1998) obtained 27% increase in cowpea and 52% increase in millet grain yield, as a result of the application of Tahoua rock phosphate. Millet benefited from the residual effects of rock phosphate applied to cowpea when millet and cowpea rows were rotated. Buerket et al. (1998) found thatthe application of Tahoua rock 271 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production phosphate led to between 25 and 78% increase in total dry matter of millet and sorghum, and between 12 and 46% increase for cowpea. In cereal-cowpea intercrops, N application generally favors cereals, resulting in decreased cowpea yield due to shading by the cereal crop (Blade et al. 1997). However, in a farmer-managed trial, the yield of intercropped maize and cowpea increased by 35% and cowpea by 24%, with the application of 120-26-50 kglNPK/ha compared with 60- 13-25 kglNPK/ha (Olufajo et al. 1997). In rotation and as intercrops with cereals, cowpea provides N and it contributes to overall fertility improvement. However, subsequent ben- efits from cowpea in sole cropping were greater than from intercropped cowpea (Carsky and Vanlauwe 2002). A further elaboration of soil fertility issues in relation to cowpea production is given by Bationo et al. (2002). Implications for pest incidence Intercropping has long been known to be a major component of integrated pest control. Singh and Emechebe (1998) screened ten cowpea breeding lines under intercropping with millet as well as sole cropping with and without insecticide application. They found that intercropped cowpea grain yields were generally higher than yields from the sole crop when no insecticide was applied, indicating less insect damage under intercropping (Table 5). Mensah (1997) reported a low population density of post-flowering pests (Maruca vitrata and a complex of pod-sucking insects) but a high population density of flower pests (Megalurothrips sjostedti) in a crop mixture consisting of one row of sorghum alternated with two rows of cowpea. Although he observed a reduction in pests and damage to cowpea in mixtures compared with monoculture, he recommended one to two insecticide applications to maximize cowpea yields. Agboh-Noameshie et al. (1997) studied pest populations on cowpea intercropped with cassava and found that the micro- environment created by the intercrop reduced the populations of flower thrips (M sjostedti) and pod-sucking bugs (Heteroptera) but increased those of the pod borer (M vitrata). Intercropped maize, pepper, and cassava have also been reported to reduce the popula- tion of cowpea flower thrips while maize, cassava, and pigeonpea intercrops reduced the incidence of blister beetles (Mylabris sp.) on cowpea (Emeasor and Ezueh 1997). It is, Table 5. Grain and fodder yields of promising medium maturing cowpea varieties in different cropping systems at Minjibir, Nigeria, 1995. Sole crop 2 sprays Variety Grain Fodder IT93K-23 2739 IT90KD-277-2 2571 IT92KD-371-1 2316 IT90K-391 2278 IT90K-365 2026 IT93 K-621-7 1944 Dan lIa 1835 IT90K-372-1-2 1499 IT89KD-349 1981 IT93K-734 1866 LSD 5% 458 Source: Singh and Emechebe 1998. 3277 1492 3590 3423 2588 2818 1429 960 2713 1022 521 Sole crop no spray Grain Fodder 21 163 36 8 272 28 347 157 261 33 160 180 4416 4250 4139 4000 3861 1833 1222 1000 2861 1431 2314 Intercrop no spray Grain Fodder 144 293 42 423 237 117 151 133 393 207 129 406 437 171 703 437 406 265 265 656 203 535 Digitized by Google Advances in cowpea cropping systems research however, noteworthy that none of these companion crops significantly reduced cowpea damage by the mung moth, M. vitrata, African pea moth, Cydia ptychora (Leguminivora ptychora), and the pod-sucking bug complex, all of which constitute the major pests of the cowpea crop. J ackai and AdalIa (1997) reviewed the effect of intercropping on insect pests of cowpea and emphasized that intercropping does not necessarily reduce the pest load in any given situation; it depends on the crop(s) and pest(s) in question. Although intercropping can contribute to the control of a pest in an integrated control context, in most cases, pest damage to intercropped cowpea is generally no less than that to the monocrop at the time of harvest. Improved varieties adapted to intercropping systems Variety development is discussed in detail in other parts of this conference by Singh et al. (2002). However, considering the fact that variety selection is a key to modifications that can be made in the cropping system, it is appropriate to consider the efforts cur- rently being made to develop cowpea varieties that are suitable for intercropping. This is especially relevant, as different plant traits are required for cultivars intended for use under intercropping compared to those intended for use under sole cropping (Nelson and Robichaux 1997). The local cowpea varieties are highly adapted to intercropping but they have a low harvest index. Terao et al. (1997) concluded that the type of cowpea adapted to intercrop- ping is the spreading type of cowpea, improved to retain a substantial root system and high translocation efficiency. The number of branches and nodes and increased internode length are plant traits that are important under intercropping (Nelson and Robichaux 1997). Thus, the cultivar with a bush-type habit has been reported to be higher yielding under sole cropping, whereas the cultivar with a spreading habit was higher yielding under intercropping (Nelson and Robichaux 1997). Subsistence farmers require crop varieties, which produce acceptable grain and fodder yields under a wide range of environmental conditions. New cowpea breeding lines are currently being evaluated under three systems at I1TA, Kano: (i) pure crop with two sprays of insecticide, (ii) pure crop with no insecticide, and (iii) intercrop with no insecticide. Singh and Emechebe (1998) noted good performance of a number of improved varieties, particularly IT90K-277-2 under both sole- and intercropping. Thus, there is a good poten- tial for increasing intercropped cowpea grain and fodder yields through the introduction of appropriate improved varieties. Implications and research needs Future research should focus on the following: To enable better understanding of the dynamics of the complex farming system and ensure the introduction of innovations that are compatible with the farmers' socioeco- nomic environment, work must continue on the characterization of the farming system. It is important to involve farmers in technology testing and validation to enhance the process of technology dissemination and adoption. Since cowpea is mainly grown in mixture with cereals, there is need for proper inves- tigation of cereal--cowpea genotypic interactions in order to identify plant types of both components that will contribute to increased efficiency and production of the cropping 273 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production systems. Improved genotypes of both cowpea and cereals adapted to intercropping and amenable to management that suit the farmers' objectives should be developed. There is need for better understanding of soil fertility management in intercropping systems. Areas such as the possible contribution of some cowpea cultivars to P uptake under low soil P conditions including the effect of mycorrhizal association on P uptake, as well as the identification of cowpea cultivars with enhanced N2 -fixing efficiency need special attention. (see also Bationo et al. 2002). Conclusions Cowpea is predominantly a crop of drier areas. For the foreseeable future, its produc- tion will continue to be based mainly on the diverse and complex intercropping systems. Over the years, food requirements have increased while land availability has become less. Thus, the only way to increase agricultural production is to increase the yield of individual crops. Being the understorey crop in most intercropping systems, growth and yield of cowpea are usually suppressed by the dominant crop. Complementarity in an intercropping situation can occur when the growth patterns of the component crops differ in time or when they make better use of resources in space. It is evident that the depressive effect of cereals on cowpea could be reduced by planting cowpea simultaneously or soon after cereals, particularly if there is no severe competition for water. Denser planting also improves productivity. Consistent increases in cowpea grain yields under strip cropping with cereals have also been reported, although there is scope for a better understanding of the physiological mechanisms that could limit the production and efficiency of such cropping systems. Concurrent with the development of cowpea genotypes that are adapted to intercropping, there is need to screen and develop cereal and tuber crops that are suit- able for intercropping with cowpea. This will contribute to the overall improvement of the cropping system. References Agboh-Noameshie,A., L.E.N. Jackai,A.A.Agboola, and H.C. Ezumah. 1997. Manipulating canopy structure in cassava intercropped with cowpea and its effects on cowpea insect population densi- ties. Tropical Agriculture 74: 210-215. Asafu-Agyei, J.N., K. Ahenkora, B. Banful, and S. Ennin-K wabiah. 1997. Sustaining food produc- tion in Ghana: the role of cereal-legume-based cropping systems. Pages 409-416 in Technology options for sustainable agriculture in sub-Saharan Africa, edited by T. Bezuneh,A.M. Emechebe, J. Sedogo, and M. Ouedraogo. Semi-Arid Food Grain Research and DevelopmentAgency (SAF- GRAD) ofthe Scientific, Technical and Research Commission ofOAU, Ouagadougou, Burkina Faso. Bationo, A., B.R. Ntare, S.A. Tarawali, and R Tabo. 2002. Soil fertility management and cowpea production in the semiarid tropics. Pages 299-316 in Challenges and opportunities for enhancing sustainable cowpea production, edited by C.A. Fatokun, SA. Tarawali, B.B. Singh, P.M. Kor- mawa, and M. Tamo. I1TA, Ibadan, Nigeria. Bezerra-Neto, F. and RH. Robichaux. 1996. Spatial arrangement and density effects on an annual cotton-cowpea-maize intercrop. I. Agronomic efficiency. PesquisaAgropecm'tria Brasileira 31: 729-741. Bezerra-Neto, F. and RH. Robichaux. 1997. Spatial arrangement and density effects on an annual cotton-cowpea-maize intercrop. II. Yield and biomass. Pesquisa Agropecm'tria Brasileira 32: 1029-1037. 274 Digitized by Google Advances in cowpea cropping systems research Bielders, C.L. 1998. Improving the productivity of millet--cowpea intercrops through the application of inputs to cowpeas and rotation between rows. Pages 101-107 in Soil fertility management in West African land-use systems, edited by G. Renard, A. Neef, and K. Becker. University of Hohenheim,ICRISAT Sahelian Centre and INRAN, Niamey, Niger. Blade, S.F., S.V.R Shetty, T. Terao, and B.B. Singh. 1997. Recent developments in cowpea cropping systems research. Pages 114-128 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropi- cal Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Buerkert, A., M. Bagayoko, A. Bationo, and H. Marschner. 1998. Site-specific differences in the response of cereals and legumes to rock phosphate, crop residue mulch and nitrogen in the Sudano-Sahelian zone of West Africa. Pages 53-59 in Soil fertility management in West African land-use systems, edited by G. Renard, A. Neef, and K. Becker. University of Hohenheim, ICRISAT Sahelian Centre and INRAN, Niamey, Niger. Cardoso, M.J., F.R. Freire Filho, V.Q. Ribeiro, A.B. Frota, and F. de B. Melo. 1993. Plant density of maize--cowpea intercrops under irrigation. PesquisaAgropecm'tria Brasileira 28: 93-99. Carsky, RJ., B. Vanlauwe, and O. Lyasse. 2002. Cowpea rotation as a resource management tech- nology for cereal-based systems in the savannas of West Africa. Pages 250-264 in Challenges and opportunities for enhancing sustainable cowpea production, edited by CA. Fatokun, SA. Tarawali, B.B. Singh, P.M. Kormawa, and M. Tamo. I1TA, Ibadan, Nigeria. Emeasor, K.C., and M.1. Ezueh. 1997. The influence of companion crops in the control of insect pests of cowpea in intercropping systems. Tropical Agriculture 74: 285-289. Endondo, C. and M. Samatana. 1999. Effect of cowpea sowing date on cotton yield in a cotton- cowpea intercrop. CahiersAgricultures 8: 215-217. Fussell, L.K., P.G. Serafini, A. Bationo and M.C. Klaij. 1987. Management practices to increase yields and yield stability of millet in Africa. Pages 255-268 in Proceedings ofthe International Pearl Millet Workshop, 7-11 April 1986. ICRISAT, Patancheru, India. Fussell, L.K., M.C. Klaij, C. Renard, and B.R. Ntare. 1992. Millet-based cropping systems for improving food production in the southern Sahelian zone. Pages 109-127 in Proceedings ofthe workshop on appropriate technologies for developing sustainable food production systems in the semiarid regions of sub-Saharan Africa. Purdue University Press, West Lafayette, Indiana, USA. Haizel, KA. 1974. The agronomic significance of mixed cropping. I. Maize interplanted with cowpea. Ghana Journal of Agricultural Science 7: 169-178. Henriet, J., GA. van Ek, S.F. Blade, and B.B. Singh. 1997. Quantitative assessment oftraditional cropping systems in the Sudan savanna of northern Nigeria. I. Rapid survey of prevalent cropping systems. Samaru Journal of Agricultural Research 14: 37-45. Isenmilla,A.E., O. Babalola, and G.O. Obigbesan. 1981. Varietal influence of inter cropped cowpea on the growth, yield, and water relations of maize. Plant and Soil 62: 153-156. Jackai, L.E.N., and C.B. Adalia. 1997. Pest management practices in cowpea: a review. Pages 240-258 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). IlIA, Ibadan, Nige- na. Mensah, G.W.K. 1997. Integrated pest management in cowpea through intercropping and minimal insecticide application. Annals of Plant Protection Sciences 5: 1-14. Mortimore, M.J., B.B. Singh, F. Harris, and S.F. Blade. 1997. Cowpea in traditional cropping sys- tems. Pages 99-113 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute ofTropicalAgriculture (I1TA) and Japan International Research for Agricultural Sciences (JiRCAS). I1TA, Ibadan, Nigeria. 275 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Muleba, N., and H.C. Ezumah. 1985. Optimizing cultural practices for cowpea in Africa. Pages 289-295 in Cowpea research production and utilization, edited by S.R. Singh and K.O. Rachie. John Wiley and Sons Ltd., Chichester, UK. Myaka, F.A. 1995. Effect oftime of planting and planting pattern of different cowpea cultivars on yield of intercropped cowpea and maize in tropical sub-humid environment. Tropical Science 35: 274-279. Myaka, F.A. and J.C.B. Kabissa. 1996. Fitting short duration cowpea into a cotton-based crop- ping system in Tanzania: effect of planting pattern, time of planting cowpea, and insecticide application to the cotton. Experimental Agriculture 32: 225-230. Natarajan, M. and D.M. Naik. 1992. Competitive effects ofa short duration bush type cowpea when intercropped with cotton in Zimbabwe. Experimental Agriculture 28: 409-416. Nelson S.C. and R.H. Robichaux. 1997. Identifying plant architectural traits associated with yield under intercropping: implications of genotype-cropping system interactions. Plant Breeding 16: 163-170. Ntare, B.R. 1990. Intercropping morphologically different cowpea with pearl millet in short season environment in the Sahel. Experimental Agriculture 26: 41-47. Ntare, B.R. and J.H. Williams. 1992. Response of cowpea cultivars to planting pattern and date of sowing in intercrops with pearl millet in Niger. Experimental Agriculture 28: 41-48. Obuo, J.E., E.Adipala, and D.S.O. Osiru. 1998. Effect of plant spacing on yield of cowpea-sor- ghum intercrop. Tropical Science 38: 67-73. Odo, P.E. andA. T.S. Bibinu. 1998. Effects of sowing date and planting pattern in millet/legume mixtures. Pages 114-119 in Pearl millet in Nigerian agriculture: production, utilization, and research priorities, edited by A.M. Emechebe, M.C. Ikwelle, O. Ajayi, M. Aminu-Kano, and A.B. Anaso. Lake Chad Research Institute, Maiduguri, Nigeria. Ofori, F. and W.R. Stern. 1987a. Cereal-legume intercropping systems. Advances in Agronomy 41: 41-90. Ofori, F. and W.R. Stern. 1987b. Relative sowing time and density of component crops in maize-cowpea intercrop system. Experimental Agriculture 23: 41-52. Okoli, 0.0., M.A. Hossain, A.F.K. Kissiedu, and A. Asare-Bediako. 1996. Effect of planting dates and growth habits of cassava and cowpea on their yield and compatibility. Tropical Agriculture 73: 169-174. Olufajo, 0.0. 1988. Effects of component populations on the productivity of maize-cowpea intercrop. Nigerian Agricultural Journal 23: 25-34. Olufajo, 0.0., A.O. Ogungbile, and B. Ahmed. 1997. On-farm testing of variety and NPK fertilization for maize-cowpea mixture in the Nigerian savanna. Pages 235-246 in Technol- ogy options for sustainable agriculture in sub-Saharan Africa, edited by T. Bezuneh, A.M. Emechebe, J. Sedgo, and M. Ouedraogo. Semi-Arid Food Grain Research and Development Agency (SAFGRAD) ofthe Scientific, Technical, and Research Commission ofOAU, Oua- gadougou, Burkina Faso. Reddy, K.C. and P.L. Visser. 1997. Cowpea intercrop growth and yield as affected by time of planting relative to millet. African Crop Science Journal 5: 351-357. Shetty, S.v.R., B.R. Ntare,A. Bationo, and C. Renard. 1995. Millet and cowpea in mixed farm- ing systems ofthe Sahel: A review of strategies for increased producti vity and sustainability. Pages 293-303 in Livestock and sustainable nutrient cycling in mixed farming systems of sub-Saharan Africa, edited by J.M. Powell, S. Ferm'tndez-Rivera, and T.O. Williams. Inter- national Livestock Centre for Africa (ILCA), Addis Ababa, Ethiopia. Singh, B.B., O.L. Chambliss, and B. Sharma. 1997. Recent advances in cowpea breeding. Pages 30-49 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashi- ell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) 276 Digitized by Google Advances in cowpea cropping systems reseach and Japan International Research Center for Agricultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Singh, B.B. and A.M. Emechebe. 1998. Increasing productivity of millet-cowpea intercropping systems. Pages 68-75 in Pearl millet in Nigerian agriculture: production, utilization and research priorities, edited by A.M. Emechebe, M.C. Ikwelle, O. Ajayi, M. Aminu-Kano, and A.B. Anaso. Lake Chad Research Institute, Maiduguri, Nigeria. Singh, B.B. and HA.Ajeigbe. 2002. Improving cowpea-cereals-based cropping systems in the dry savannas of West Africa. Pages 276-284 in Challenges and opportunities for enhancing sustain- able cowpea production, edited by CA. Fatokun, SA. Tarawali, B.B. Singh, P.M. Kormawa, and M. Tamo. IlIA, Ibadan, Nigeria. Singh, B.B., J.D. Ehlers, B. Sharma, and F.R Freire Filho. Recent progress in cowpea breeding. 2002. Pages 22-40 in Challenges and opportunities for enhancing sustainable cowpea production, edited by CA. Fatokun, S.A. Tarawali, B.B. Singh, P.M. Kormawa, and M. Tamo. I1TA, Ibadan, Nigeria. Subbarao, G.V, C. Renard,A. Bationo, N. van-Duivenbooden, and C. Bielders. 1999. Alternative technologies for Sahelian crop production systems in West Africa. Pages 121-132 in Recent advances in management of arid ecosystem, edited by A.S. Faroda, N.L. Joshi, and S. Kattju. Arid Zone Research Association ofIndia, Jodhpur, India. Terao, T., I. Watanabe, R Matsunaga, S. Hakoyama, and B.B. Singh. 1997. Agrophysiological constraints in intercropped cowpea: an analysis. Pages 129-140 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (llRCAS), IlIA, Ibadan, Nigeria. van Ek, GA., J. Henriet, S.F. Blade, and B.B. Singh. 1997. Quantitative assessment of traditional cropping systems in the Sudan savanna of northern Nigeria. II. Management and productivity of major cropping systems. Samaru Journal of Agricultural Research 14: 47-60. Willey, RW. 1979. Intercropping-its importance and research needs. Part 1. Competition and yield advantages. Field Crop Abstracts 32: 1-10. Willey, RW. and D.S.O. Osiru. 1972. Studies on mixtures of maize and beans (Phaseolus vulgaris) with particular reference to plant population. Journal of Agricultural Science (Cambridge) 79: 519-529. 277 Digitized by Google 4.4 Improving cowpea-cereals based cropping systems in the dry savannas of West Africa B.B. Singh 1 and H.A. Ajeigbe 1 Abstract Most of the farmers in the dry savannas of West Africa plant local varieties of cowpea, millet, sorghum, and groundnut in various intercropping systems with little or no purchased inputs. In this system, the cowpea and groundnut yields are low due to shading by cereals and lack of plant protection measures. The cereal yields are low mainly due to lack of fertilizer. Efforts are being made, therefore, to develop a combination of improved varieties and improved cropping systems for higher productivity and profitability with a minimum use of insecticides and fertilizers. We evaluated four cereal--cowpea intercropping row arrangements involving one cereal : one cowpea, one cereal : four cowpea, two cereal : four cowpea intercrops, and sole crops of improved and local varieties of millet, cowpea, and sorghum with selective application of two sprays of insecticides on cowpea only and 100 kglha fertilizer (N.P.K.lS: 15: 15) basal and 20 kgNlha top-dressed to cereals only. The results indicated sole crop improved cowpea to be most profitable followed by the two cereal: four cowpea intercrop system. Farmer participatory evaluation ofthe improved intercrop system involving two rows of sorghum: four rows of improved cowpea with inputs as indicated above, gave 100 to 300% gross economic superiority over the traditional intercropping systems. Smalholder farm- ers prefer the improved intercropping system over sole crops because it provides them with sufficient sorghum and cowpea for home use and additional cowpea for cash income. Introduction Crop production in West and Central Africa is still based on traditional intercropping systems which may be quite diverse and complex (Norman 1974; Mortimore et al. 1997). In the Sudan savanna and the Sahelian zones, these systems involve intercropping of sor- ghum, millet, cowpea, and groundnut in various spatial and temporal arrangements and have evolved over centuries of experience to ensure maximum use of rainfall and avail- able resources for sustainable production of food and fodder. The cereals are the staple diet, complemented with cowpea as a source of protein. Most of the groundnut and some cowpea are sold for cash. Cereal stovers are used as building material, for fencing, fodder, and fuel but the haulms of cowpea and groundnut are always used as fodder, being the most valuable source of livestock feed during the dry season. The cropping systems depend on a number of factors, including local traditions, level of technology, resource availability, and physical environment. The general objective of farmers is a sustained production (Baker and Norman 1975) of reasonable levels, at minimal risk, to satisfy subsistence and commercial needs (Beets 1990). These needs have increased due to the rise in population and consequent reduction in arable land on 1. International Institute of Tropical Agriculture (lIT A), Kano Station, PMB 3112, Kano, Nigeria. 278 Digitized by Google Improving cowpea-cereals based cropping systems in the dry savannas of West Africa a per capita basis. Therefore, an important approach to increase agricultural production would be through improving the yield of individual crops per unit area. Farmers with few resources at their disposal have a limited capacity to tolerate production failure. They attach a risk factor to their assessment of agricultural innovations, preferring incremental changes to radical departures from existing practices (Edwards 1993). Therefore, research must build on these farming practices and aim at risk-free increases in productivity even if these have to be gradual. Cowpea is an integral component of the traditional cropping systems due to its beneficial effect on sustainability and as a source of nutritious food and fodder (Benriet et al. 1997). The International Institute of Tropical Agriculture (I1TA) with a global mandate for cowpea has been working on the improvement of cowpea varieties as well as the improvement of cropping systems to increase total productivity with limited use of purchased inputs. The strategy has been to study the role of cowpea in major cropping systems, identify the production constraints in traditional systems, and then develop improved cowpea varieties and improved systems (Singh 1993; 1994). A quantitative assessment of the traditional cropping system revealed 22 types of crop mixtures in northern Nigeria of which millet-cowpea, sorghum-cowpea, millet-sor- ghum-cowpea, and millet-sorghum-groundnut-cowpea were predominant (Benriet et al. 1997). The mean grain yields in these systems ranged from 0 to 132 kg/ha for cowpea, 0 to 197 kg/ha for groundnut, 131 to 2600 kg/ha for millet, and 0 to 4903 kg/ha for sorghum, depending on the fertility level of the fields (van Ek et al. 1997). The major production constraints in the intercropping system were low fertility, low population, lack offertilizer and pesticides, shading of cowpea and groundnut by the millet and sorghum, as well as late maturity and poor yield potential of local varieties. Efforts are being made, therefore, to develop a combination of improved varieties and improved cropping systems for higher productivity and profitability with limited use of insecticides and fertilizers. This paper describes improved varieties and optimum dates of planting for intercropping and the superiority of an improved strip cropping system that maximizes the benefits of limited fertilizers and pesticides, and minimizes competition between cereals and legumes. Improving productivity of traditional intercropping system Traditional intercropping involves planting cowpea with millet and/or sorghum in a one row cereal : one cowpea row arrangement. Also, the cereals are planted at the onset of rains and cowpea is planted three to four weeks later between cereal rows when the rains have stabilized. Thus, cowpea is shaded by the cereals throughout the growing period. This causes severe reduction of shoot as well as root growth of cowpea resulting in very low grain and fodder yields. Recent studies have shown that even though cowpea occupies 50% of land area under intercropping, its grain and fodder yields are between 10 to 20% of those of sole crop cowpea (Singh et al. 1997; Terao et al. 1997). Even though sole crop cowpea is most profitable, most subsistence farmers plant cowpea as an intercrop with millet and sorghum. This is primarily because the land is limited and they want to pro- duce a sufficient quantity of cereals for home consumption; and partly because sole crop cowpea requires one or two sprays of insecticide to control recalcitrant insect pests such as Maruca and pod bugs; furthermore, chemicals are often not available even if farmers have financial resources. Efforts are, therefore, being made to develop shade-tolerant cowpea varieties with resistance to diseases, insect pests, and parasitic weeds, giving grain and 279 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production fodder yields Wlder intercropping with millet and sorghwn. The general approach is to screen all the new breeding lines under intercropping with millet and sorghum, improve selected local varieties by defect elimination using the backcross method, and develop new improved cowpea varieties specifically adapted to intercropping without insecticide application. Improved cowpea varieties for intercropping Selected improved cowpea breeding lines were screened under traditional one: one inter- crop with millet without insecticide. The millet rows were 2 m apart and planted about three weeks before cowpea to reflect the actual farmers' practice. The improved varieties performed significantly better than the local variety, Dan Ila, under sole crop as well as intercrop (Table 1). However, grain yields under intercrop were less than 10% of the sole crop and fodder yields ranged from 10 to 20% even though expected yields of cowpea are 50% of the sole crop. This is primarily because millet grows faster, shades cowpea, and competes for nutrients and water, thereby reducing cowpea grain and fodder yields. Thus, the traditional intercropping of one row cereal: one row cowpea is less productive for cowpea even though the yields of improved varieties are three to four times higher than of the local variety. The promising lines selected for good performance under intercropping in 1999 were separately evaluated with maize and sorghum at Samaru in 2000. IT95K-193-l2 and IT95K-222-3 gave the highest grain yield followed by others with an average ranging from 300 kg to 500 kg/ha, compared to zero yield of the local varieties (Table 2). Table 1. Performance of promising cowpea breeding lines under sole crop and intercrop with millet. Variety Extra-early maturing cowpea varieties IT95K-627-34 IT98K-463-7 IT98K-205-8 Dan Iia SED Early-matu ri ng cowpea varieties IT97K-499-39 IT97K-508-2 IT97K-608-14 Dan Iia SED Medium-maturing cowpea varieties IT95K-193-12 IT98K-131-1 IT98K-494-3 Dan Iia SED -Grain yield (kg/ha)-- Sole crop Intercrop 2335 1968 2041 1694 349 2715 2265 2357 1823 258 2381 2409 2079 846 237 280 237 134 120 82 31 177 182 134 41 29 277 233 175 58 49 - Fodder yield (kg/ha)- Sole crop Intercrop 1662 612 1422 2188 267 1406 2074 1102 2104 321 1308 2689 1982 1091 343 458 208 146 677 73 125 396 448 363 89 417 573 365 406 106 Digitized by Google Improving cowpea-cereals based cropping systems in the dry savannas of West Africa In another trial at Minjibir, selected promising varieties were evaluated under inter- cropping with local and improved varieties of millet and sorghum to ascertain whether improved varieties of cereal would cause less competition with cowpea. The improved varieties yielded higher than the locallandrace with almost 200 to 300% superiority in grain yield (Table 3). The mean grain yield of cowpea varieties was similar under improved and local sorghum but it was consistently less under local millet compared to the improved millet although the differences were not significant. These results indicated that the improved cowpea varieties are more productive than the local variety under millet, sorghum, and maize intercropping. Effect of date of planting cowpea as intercrop in millet As indicated earlier, under traditional intercropping, farmers normally plant millet first at the onset of rains in the beginning of June; about three weeks later, they plant cowpea as an intercrop. This causes shading of cowpea by the fast growing millet. An experiment was, therefore conducted to assess the effect of different dates of planting cowpea as an intercrop in millet. The treatments included planting of cowpea simultaneously with millet, and at three, six, and nine weeks after millet. There was a significant genotype x date of planting interaction (Table 4). The early- and medium-maturing cowpea varieties such as IT93K-452-1, IT90K-277-2, and Dan Ila had highest grain and fodder yields when Table 2. Performance of improved cowpea varieties under intercrop at Samaru, 2000. Grain yield (kg/ha) Variety Cowpea Maize Cowpea Sorghum IT95K-193-12 571 5574 365 1950 I T95K-222-3 414 3193 420 3042 IT97K-1129-51 365 3842 425 2175 IT98K-279-2 403 2651 362 3240 IT97K-207-21 325 2137 375 3519 IT97K-461-4 277 2315 396 3300 I T95k-627-34 303 3900 305 1693 IT97K-499-39 232 3411 385 2165 IT97K-499-38 282 2937 298 2386 IT97K-819-118 362 2758 179 1981 Dan lIa 0 1647 0 2064 Aloka 0 1656 0 2150 SED 99 1073 26 373 Table 3. Performance of promising cowpea varieties for intercropping with millet and sorghum (lCSV-lll) at Minjibir, 2000. -Cowpea grain yield (kg/ha)- _Cowpea fodder yield (kg/ha)_ Sorghum Millet Sorghum Millet Variety Local ICSV Local Sosat Local ICSV Local Sosat IT95K-231-1 335 353 214 261 417 567 317 283 IT97K-356-1 320 434 194 200 350 367 317 217 IT97K-608-14 392 245 171 178 517 583 250 267 IT97K-499-39 349 306 148 167 200 367 183 200 IT97K-207-21 253 375 139 143 350 357 317 217 Dan lIa 105 126 120 155 417 500 350 283 281 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production simultaneously planted with millet, whereas the late-maturing varieties IT89KD-288 and IT89KD-349 had highest yields when planted three weeks after millet. However, all the varieties showed drastic yield reduction when planted six and nine weeks after millet. This may be due to severe shading and competition for nutrients in later plantings, as the millet already had well developed stem and root systems. The results confirm earlier observa- tions and suggest that for maximum yield under intercrop, cowpea should be planted as soon as millet has been sown in wider rows to reduce shading. Improved strip cropping system for higher productivity Since the overall productivity under traditional intercropping is very low, due to shading and severe competition for nutrients, efforts were made to develop alternative systerns which will minimize shading and maximize gains from limited application of fertilizer and agrochemicals. Among several systems evaluated, a strip cropping system involving two rows of densely planted cereal : four rows of densely planted cowpea appeared to be significantly more productive, particularly when limited amounts of fertilizer was applied to the cereal and one or two sprays were given to cowpea. Figure 1 illustrates this system. Productivity of different cropping systems was tested on-station involving improved cowpea varieties in sole crop and intercrop systems using one row of millet: one row of cowpea; and two rows of millet: four rows of cowpea with a minimum basal application of 15 kgN; 15 kg P P 5' and 15 kg Kp, top dressing of the cereals only at the rate of 30 kgIN/ha. The cowpea was sprayed with insecticide twice-at flowering and at full podding. The performance of the improved cowpea variety was superior to the local cowpea in sole crop as well as in two: four intercrop system and similar to local one: one intercrop system (Table 5). However, the gross economic value of the two: four system was signifi- cantly higher than the one: one system and very close to that of the sole crop cowpea. The improved cowpea varieties IT89KD-391 and IT90K-277-2 appeared to be most promising for the two : four system. During farmer field days, farmers showed great interest in the two: four system because it provides them with sufficient millet for home consumption and a large amount of additional cowpea, part of which can be used as nutritious food at home and part can be sold for cash. Table 4. Grain and fodder yields of different cowpea varieties planted as intercrop at different dates in millet field. Yield (kg/ha) under intercrop with millet Cowpea Simultaneous -3weeks __ 6weeks __ 9 weeks_ variety Grain Fodder Grain Fodder Grain Fodder Grain Fodder IT90K-277-2 214 491 94 175 35 75 4 33 I T93K-452-1 200 342 43 125 31 58 7 20 Dan lIa 199 625 132 175 20 133 28 45 IT89KD-349 109 650 153 242 43 117 7 38 IT89KD-288 75 600 107 267 17 142 45 35 SED 35 51 35 51 35 51 35 35 282 Digitized by Google Improving cowpea-cereals based cropping systems in the dry savannas of West Africa A. Traditional intercropping system 1m ~ Constraints • Low plant population • Shading of legumes by cereals • Difficulty in selective input application • Lack of integration with livestock component • Low productivity B. Improved strip cropping systems cereals legumes cereals legumes cereals 000000000000000000000000000000000000000 2 rows of 000000000000000000000000000000000000000 cereals X*~E*X~*,e< cowpea "tiXXXXXXXXX)oo(XX~Je • ..-.-= ~ ... : .. '- J#' .. - ... -• .. ". '~ ... ' 7-:\"~ ~ .".: _ ~Io eN Figure 2. Field screening for drought tolerance. The plants on the left are TVu 7778, (drought susceptible) those on the right, TVu 12349 (drought tolerant). Figure 3. Pot screening for drought tolerance. Left to right: drought susceptible, drought tolerant, drought susceptible, and drought tolerant. 292 Digitized by Google Cowpea varieties for drought tolerance the possibility of using a controlled environment, and the ability to screen large numbers of lines or plants. Also, field screening for drought tolerance may be complicated due to differences in root length and architecture of the test materials. The shallow box method described here eliminates the effects of roots and thereby permits the identification of plants with shoot drought tolerance. The box method is simple, nondestructive, and offers flexibility in terms of size of operation as boxes can be larger or smaller depending upon the need. The test materials can be homozygous lines or segregating populations and the drought-tolerant plants can be saved and transplanted for further progeny testing and selection. Mechanisms of drought tolerance and its inheritance A close observation of cowpea plants and its inheritance in the boxes showed two types of drought-tolerance mechanisms (Mai-Kodomi et al. 1 999a). Under drought stress, Type 1 drought-tolerant lines such as TVu 11986 and TVu 11979 stopped growth and conserved moisture in all the plant tissues, stayed alive for over two weeks, and gradually the entire plant parts dried. The Type 2 drought-tolerant lines such as Dan Ila and Kanannado continued slow growth of the trifoliates. However, with continued moisture stress, the unifoliates of these varieties showed early senescence and dropped off but the growing tips remained turgid and alive even longer (Fig. 4), suggesting that the moisture was being mobilized from the unifoliates to the growing tips. Using the box screening method, the inheritance of drought tolerance in cowpea was studied (Mai -Kodomi et al. 1 999b ). Three cowpea lines: TVu 11986 with Type 1 drought tolerance, Dan Ila with Type 2 drought tolerance, and TVu 7778 as susceptible to drought were crossed in all possible combinations. The genetic segregation revealed that drought Figure 4. Two types of drought tolerance. Left to right, Type 1 (1 plant), Type 2 (2 plants), Type 1 (2 plants), and a susceptible line. 293 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production tolerance is a dominant trait and both Type 1 and Type 2 reactions are controlled by a single dominant gene but the genes are independent in the two types. These are being designated as Rds1 (resistance to drought stress) and Rds2. Test of allelism indicated that Type 1 is dominant over Type 2 and the F 2 population between the two types segregated to 3 Type 1 : 1 Type 2 indicating that the genes Rds1 and Rds2 are either closely linked or are allelic at the same locus. The simplicity of the wooden-box method and the inheritance of drought tolerance in this study may be due to its focus on only the shoot drought tolerance without involving the contribution of roots and other factors. Most of the earlier studies on drought tolerance have been conducted in the field where different mechanisms contribute to the overall drought tolerance of the plants and make it appear to be a complex trait. Screening for dehydration tolerance of the shoots only in the seedling stage using the wooden-box method should be related primarily to the stomatal behavior or osmotic adjustments as other mechanisms would not be operative. Once the plants sense water stress, it is likely that the genes controlling stomatal behavior or osmotic adjustments would be activated. The opening and closing of stomata, permitting solutes to accumulate in the cells may be simple phenomena, and therefore, they may be under major gene control as suggested by the results of this study. Relative shoot drought tolerance of major crops The relative drought tolerance of some of the major crops grown in the semiarid trop- ics was reported by Singh et al. (1999b). They studied ten crop species for their relative shoot drought tolerance at the seedling stage using the box screening method. Based on percentage of dead plants at various time intervals and days taken to 100% dead plants for each of the crops, soybean appeared to be the most drought susceptible and cowpea the most drought tolerant (Table 3). The overall ranking of the crops in the increasing order of drought tolerance was soybean followed by blackgram, greengram, groundnut, maize, sorghum, millet, bambara nut, lablab, and cowpea. The water stress in the wooden-box Table 3. Proportion of dead plants of different crops at various intervals after terminating wateringt. Days after terminating watering Crop 7 9 11 15 19 23 Cowpea: IT90K-59-2 0 0 0 0 29 100 Cowpea: TVu 11979 0 0 0 13 53 100 Cowpea: TVu 7778 0 0 0 27 94 100 lablab bean 0 0 0 17 66 100 Bambara nut 0 0 6 33 44 100 Groundnut 14 59 100 100 100 100 Pearl millet 14 28 68 100 100 100 Sorghum 0 0 93 100 100 100 Greengram 8 17 86 100 100 100 Blackgram 14 75 100 100 100 100 Maize 17 50 100 100 100 100 Soybean 63 100 100 100 100 100 lSD5% 46 56 23 31 50 NS t Source: Singh et al. 1999b. 294 Digitized by Google Cowpea varieties for drought tolerance method using higher sand content was too drastic for crops other than cowpea and lab lab. With increased clay content and gradual water stress, it may be possible to use this method to detect variable differences in crops such as maize, soybean, millet, and sorghum, which are less drought tolerant than cowpea. Screening for root characteristics Screening for root characteristics is difficult because of the underground distribution of roots and associated soil variations. Several methods have been used to estimate root length, density, volume, and distribution in the field (Krishnamurthy et al. 1996). The "auger method" provides for a three-dimensional volumetric measure of soil-root rela- tionship; however, this has large sampling variations. For the "monolith method", soil samples of an area of 20 x 30 cm to a depth of 10 or 15 cm are successively recovered and the roots are washed in a 1 mm sieve. This method is less variable because it involves a larger sample size. However, these methods are suitable only for limited comparisons. The rhizotron or minirhizotron methods are more efficient and permit nondestructive continuous studies of root systems but these involve expensive setups and equipment and are not practical for screening large numbers of segregating populations. Also, the root density estimates using the minirhizotron method do not compare well with the auger or monolith methods (Krishnamurthy et al. 1996). Recently, the use of the "root-box pin board" method has permitted a two-dimensional study of root systems of large numbers of plants or progenies with limited resources and great simplicity. Results of these screen- ing methods are highly correlated with actual field observations using in situ root pits and monolith methods. Root-box pin board method for root study Recently, a simple box method for studying root architecture was developed at I1TA, Kano Station (unpublished), which permits fast screening of cowpea lines for root length, root density, and root spread. The two-dimensional distribution of roots can be studied using a thin wooden box made by nailing two plywood sheets of 80 cm length, 60 width, and 5 mm thickness on a frame made of 5 cm thick square wooden stakes. One sheet is fixed with soft nails for easy removal and the other sheet is fixed with hard nails. Before fixing nails, the inner sides ofthe plywood sheets are lined with polyethylene and one side ofthe frame is open leaving a 5 cm gap. The 5 cm gap is then filled with a mixture of sand and topsoil (50 : 50) and watered. Three to five handpicked good seeds of the test crop are planted in a single hole in the middle, and after germination thinned to one to three plants, depending on the objective of the study. The box is watered daily for three to five weeks after which the roots can be studied. This is done by removing one side of the box (the soft nail side) and fixing the nail board (Fig. 5) in its place. The box is then turned over so that all the soil and the plant with its roots lie on the nail board when the sheet on the hard nail side is lifted out. The soil is washed off gently using sprinkling water leaving the roots on the polythene sheet on the nail board. The polythene sheet (with the roots) is lifted out and studied. Varietal or species differences are studied using one box for each with three to four replications. Major varietal differences have been observed in cowpea root architecture (Fig. 6). Some varieties have a well-spread deep root system while others have concentrated roots only on the upper soil strata. These differences affect 295 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Figure 5. Box screening for root architecture. Left shows box with growing plant; right shows the two-dimensional structure of the roots on the nail board. Figure 6. Varietal differences in root architecture. Left shows variety with well spread, deep root system; right shows variety with roots only in upper strata. 296 Digitized by Google Cowpea varieties for drought tolerance plant ability to absorb water from the receding water after the rains cease. Well distributed deep roots permit plants to survive longer than those with shallow roots. Breeding for drought tolerance Using the box screening method for shoot drought tolerance and the root-box pin board screening method for root architecture, it has been possible to identify cowpea varieties with enhanced levels of shoot drought tolerance (Figs. I and 3), and varieties with well- distributed deep root systems (Fig. 6). These have been crossed to combine the two charac- teristics in order to develop new improved varieties with a high level of drought tolerance and the segregating progenies are being screened. In the meantime, the available improved breeding lines have been screened using box screening and a number of drought-tolerant lines have been identified (Table 4) and are being tested in drought-prone areas. Table 4. Reaction of improved cowpea breeding lines to drought. Drought-tolerant Type 1 IT93K-452-1, IT95K-627-34, IT96D-711, IT96D-724, IT97K-1068-27, IT97K-1068-7, IT97K- 1075-7, IT97K-1106-6, IT97K-1129-51-1, IT97K-1133-7, IT97K-207-21, IT97K-491-4, 114, IT97K-499-38, IT97K-499-39, IT97K-634, IT98K-1025-6, IT98K-1106-3, IT98K-128-3, IT98K- 131-1, IT98K-251-5, IT98K-258-19, IT98K-308-12-1, IT98K-310-7-1, IT98K-368-24-1, IT98K- 368-43-1, IT98K-394-20, IT98K-402-2-2, IT98K-406-12, IT98K-406-3, IT98K-463-7, IT98K-471, IT98K-568-11, IT98K-580, IT98K-96-4, IT99K-1238, IT99K-1362, IT99K-254-2, IT99K-316-2, IT99K-332-3, IT99K-368-1, IT99K-429-2, IT99K-429-4, IT99K-467-7, IT99K-494-6, IT99K-499- 5, IT99K-529-2, IT99K-536-5, IT99K-536-6, IT99K-539-1, IT99K-544-1, IT99K-564-2, IT99K- 790-2. Drought-tolerant Type 2 Aloka local, Dan Ila, Gorom local, IAR1696, IT89KD-288-40, IT89KD-288-42, IT89KD- 349, IT89KD-374-57, IT95K-105-2, IT95K-1072-57, IT95K-181-9, IT95K-222-3, IT95K-223-19, IT95K-357-2, IT95K-426-2, IT95K-825-3, IT96D-602, IT96D-604, IT97K-1021-9, IT97K-1069-2, IT97K-1069-6, IT97K-209-4-1, IT97K-338-7, IT97K-377-4, IT97K-569-9, IT97K-573-1, IT97K- 608-14, IT97K-8119-154, IT97K-819-118, IT97K-819-170, IT97K-819-172, IT97K-819-178, IT97K-819-220, IT97K-819-84, IT97K-820-13, IT97K-820-8, IT97K1025-18, IT98D-1219, IT98D-1232, IT98D-1300, IT98D-1355, IT98K-1091-2-1, IT98K-1091-3, IT98K-1093-5-2, IT98K-1108-4, IT98K-1399,IT98K-143-14, IT98K-210-1, IT98K-234-5, IT98K-317-8, IT98K-415- 6, IT98K-418-2-2, IT98K-557-1, IT98K-690, IT99K-1000,IT99K-1016, IT99K-1152-14, IT99K- 1235, IT99K-1260, IT99K-1288, IT99K-1296, IT99K-210-2, IT99K-210-3, IT99K-270, IT99K- 298-2, IT99K-363-3, IT99K-364-2, IT99K-381-6, IT99K-411-2, IT99K-411-4, IT99K-412-1, IT99K-412-6, IT99K-415-2, IT99K-421-4, IT99K-421-5, IT99K-445-3,IT99K-451-4, IT99K-466-3, IT99K-476-2, IT99K-541-1, IT99K-556-2, IT99K-562-1, IT99K-636-7, IT99K-687, IT99K-695, IT99K-720-3, IT99K-723-13, IT99K-818-27, IT99K-826-3, IT99K-835, IT99K-957, IT99K826-2, Suvita-2, IT98K-412-13, IT98K-415-1. Drought susceptible IT82E-16, IT95K-1133-6, IT95K-231-1, IT95K-238-3, IT95K-356-1, IT95K-398-14, IT97K-1042- 3, IT97K-356-2, IT97K-399-32, IT97K-419-3, IT97K-467-7, IT97K-556-4, IT98K-1079-10, IT98K- 1088-5, IT98K-1107-2, IT98K-1107-8, IT98K-1110-2, IT98K-1111-1, IT98K-1128-18, IT98K- 279-6, IT98K-311-8-1, IT98K-311-8-2, IT98K-399-1, IT98K-399-32, IT98K-439-3, IT98K-491-4, IT98K-555-1, IT98K-589-2, IT98K-598-1, IT98K-625-1, IT98K-642-2, IT99K-1122, IT99K-1152- 23, IT99K-1152-8, IT99K-1256, IT99K-1366, IT99K-195-8, IT99K-390, IT99K-407-3, IT99K-573- 2, IT99K-820-7, TVu-7778. 297 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Evaluation of selected drought-tolerant varieties A number of selected drought-tolerant and susceptible varieties based on box screening were evaluated in the field at Minjibir towards the end of the rainy season and at Zinder, (Niger Republic) where rainfall is normally low. From the rainfall pattern, there was good level of moisture stress at both locations and cowpea varieties differed in their response. Generally, the drought-tolerant varieties had significantly higher grain yields than susceptible varieties at both locations but both had similar fodder yields (Table 5). The most promising drought-tolerant varieties were IT98K-452-1, IT97K-819-154, and IT98K -205-8. These results indicate thatthe box method can be used to screen for drought tolerance of new breeding lines to reduce their numbers before field testing. Table 5. Performance of drought-tolerant cowpea varieties, 2000. Grain yield (kg) Fodder yield (kg/ha) Variety Zinder Minjibir Zinder Minjibir Reaction to drought IT98K-452-1 1209 650 1253 919 tolerant IT97K-819-154 1075 599 1614 1091 tolerant IT98K-20-8 1017 313 1252 946 tolerant IT89KD-349 730 669 1364 768 tolerant Aloka 903 597 1141 668 tolerant TVu 7778 583 214 751 356 susceptible I T95K-238-3 500 152 1280 752 susceptible IT92KD-357-2 330 271 1587 752 susceptible SED 167 96 187 258 Conclusion The traditional approach of studying drought tolerance on a whole plant basis makes it appear as a complex trait and therefore, difficult to manipulate by plant breeders. The studies described here indicate that it is possible to simplify this by separating shoot drought tolerance from the influence of roots and vice versa. Using the box screening method for cowpea, major varietal differences have been observed for shoot drought tolerance and the trait seems to be simply inherited. Similarly, using the root-box pin board method, major varietal differences have been observed for root architecture in cowpea. These methods have provided a simplified approach to the study of drought tolerance in cowpea and may lead to faster progress in breeding for drought tolerance in other crops. References Acevedo, E. and S. Ceccarelli. 1989. Role of physiologist-breeder in a breeding program for drought resistance conditions. Pages 117-139 in Drought resistance in cereals, edited by W.G. Baker. CAB International, Wallingford, UK. 298 Digitized by Google Cowpea varieties for drought tolerance Arraudeau, MA. 1989. Breeding strategies for drought resistance. Pages 107-116 in Drought resistance in cereals, edited by W.G. Baker. CAB International, Wallingford, London, UK. Ashley, J. 1993. Drought and crop adaptation. Pages 46--67 in Dryland farming in Africa, edited by lRJ. Rowland. Macmillan Press Ltd, UK. Bates, L.S., RP. Waldren, and 1.0. Teare. 1973. Rapid determination offree proline in water stress studies. Plant and Soil 38: 205. Blum,A. 1985. Breeding crop varieties for stress environments. Critical Reviews in Plant Sciences 2: 199-238. Boyer, 1 S. 1996. Advances in drought tolerance in plants. Advances in Agronomy 56: 189-218. Hall,A.E, RG. Mutters, K.T. Hubick, and GD. Farquhar. 1990. Genotype differences in carbon isotope discrimination by cowpea under wet and dry field conditions. Crop Science 30: 300- 305. Hall,A.E., S. Thiaw,A. Ismail, and ID. Ehlers. 1997. Water-use efficiency and drought adaptation of cowpea. Pages 141-146 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agri- culture (ilTA) and Japan International Centre for Agricultural Sciences (J1RCAS). ilTA, Ibadan, Nigeria. Krishnamurthy, L.C, C. Johansen, and O. Ito. 1996. Genotypic variation in root system development and its implication for drought resistance in Chickpea. Pages 235-250 in Roots and nitrogen in cropping systems of the semiarid tropics, edited by O. Ito, C. Johansen, 11 Adu-Gyamfi, K. Katayama, J.VK. Kumar Rao, and T. J. Rego. J1RCAS and ICRISAT, Hyderabad, India. Lawan, RJ. 1983. Responses of four grain legumes to water stress in higher plants. Annual Review of Plant Physiology 35: 299-319. Mai-Kodomi, Y, B.B. Singh, O. Myers Jr., J.H. Yopp, P.J. Gibson, and T. Terao. 1999a. Two types of drought tolerance in cowpea. Indian Journal of Genetics 59 (3): 309-313. Mai-Kodomi, Y, B.B. Singh, T. Terao, O. Myers Jr., J.H. Yopp, and P.J. Gibson. 1999b. Inheritance of drought tolerance in cowpea. Indian Journal of Genetics 59 (3): 317-232. Morgan, 1M. 1984. Osmoregulation and water stress in higher plants. Annual Review of Plant Physiology 35: 299-319. Singh, B.B. 1987. Breeding cowpea varieties for drought escape. Pages 299-306 in Food grain production in semiarid Africa, edited by 1M. Menyonga, T. Bezuneh, andA. Youdeowei. OAUI STRC-SAFGRAD, Ouagadougou, Burkina Faso. Singh, B.B. 1993. Cowpea breeding archival report (1988-1992) of Grain Legume Improvement Program, ilTA, Ibadan, Nigeria. Singh, B.B. 1994. Breeding suitable cowpea varieties for West and Central African savanna. Pages 77-85 in Progress in food grains research and production in semiarid Africa, edited by J.M. Menyonga, J. B. Bezuneh, 1 Y Yayock, and I. Soumana. OA U/STRC-SAFGRAD, Ouagadougou, Burkina Faso. Singh, B.B., O.L. Chambliss, and B. Sharma. 1997. Recent advances in cowpea breeding. Pages 30-49 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (ilTA) and Japan International Research Centre for Agricultural Science (J1RCAS). ilTA, Ibadan, Nigeria. Singh, B.B., Y Mai-Kodomi, and T. Terao. 1999a. A simple screening method for drought tolerance in cowpea. Indian Journal of Genetics 59 (2): 211-220. Singh, B.B., Y Mai-Kodomi, and T. Terao. 1999b. Relative droughttolerance ofmajorrainfed crops of the semi-arid tropics. Indian Journal of Genetics 59: 1-8. Subbarao, G.V, C. Johansen,A.E. Slinkard, RC. Nageswara Rao, N.P. Saxena, and YS. Chauhan. 1995. Strategies for improving drought resistance in grain legumes. Critical Reviews in Plant Sciences 14: 269-523. Turk, K.l and A.E. Hall. 1980. Drought adaptation of cowpea. Influence of drought on plant water status and relations with seed yield. Agronomy Journal 72: 421-427. 299 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Walker, D.W. and J.C. Miller Jr. 1986. Intraspecific variability for drought resistance in cowpea. Scientia Horticulturae 29: 87-100. Watanabe, I., S. Hakoyama, I. Ierao, and B.B. Singh. 1997. Evaluation methods for drought toler- ance in cowpea. Pages 141-146 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute ofIropical Agriculture (IlIA) and Japan International Research Centre for Agricultural Sciences (J1RCAS). IlIA, Ibadan, Nigeria. Yadava, RB.R and B.D. Patil. 1984. Screening of cowpea (Vigna unguiculata L.) varieties for drought tolerance. Zeitschrift Fur Acker Und Pflanzenbau 93: 259-262. 300 Digitized by Google 4.6 Soil fertility management and cowpea production in the semiarid tropics Bationo, A. \ B.R. Ntare2, S.A. Tarawali3, and R. Tabo2 Abstract Cowpea (Vigna unguiculata [L.] Walp.) is an important grain legume in the semi- arid zone of West Africa as it is a major source of dietary protein for the people. It is usually grown as an intercrop with the major cereals, namely millet and sorghum. Despite its importance, its yields are very low due to several constraints including poor soil, insect pests, and drought. The soils in semiarid West Africa are inherently low in nitrogen and phosphorus. Soil, water, and nutrient management practices are inadequate to sustain food production and to meet the food requirements ofthe fast growing population. Research results show that proper management of organic amendments such as crop residues and manure, which are essential complements to mineral phosphorus fertilizers, can increase yields of cowpea and associated cereals more than three fold. Direct application of indigenous phosphate rocks can be an economical alternative to the use of imported, more expensive soluble phos- phorus fertilizers for cowpea production in the region. The agronomic effectiveness of indigenous phosphate rock is about 50% compared to the imported single super- phosphate. Furthermore, when the unreactive phosphate rocks are partially acidu- lated at 50%, their agronomic effectiveness can increase to more than 70%. Stud- ies on cereal--cowpea rotation revealed that yields of cereals succeeding cowpea could, in some cases, double compared to continuous cereal cultivation. With efficient soil fertility management, cowpea can fix up to 88 kg Nlha and this results in an increase of nitrogen use efficiency on the succeeding cereal crop from 20% in the continuous cereal monoculture to 28% when cereals are in rotation with cowpea. Furthermore, the use of soil nitrogen increased from 39 kg Nlha in the continuous cereal monoculture to 62 kg N/ha in the rotation systems. Future research needs to focus on understanding the factors affecting phosphorus uptake from different sources of natural rock phosphate. There is also a need to quantity the below-ground nitrogen fixed by different cowpea cultivars. The increase of cowpea productivity in the cropping systems in this region will improve the nutri- tion of people, increase the feed quantity and quality for livestock, and contribute to soil fertility maintenance. This should contribute to reduction in poverty and environmental degradation. Introduction Cowpea (Vigna unguiculata [L.] Walp.) is an important grain legume in the West African Semiarid Tropics (WASAT), where it occupies 6 million hectares. Cowpea is an important component of the predominantly cereal/legume production systems in the region. The 1. Tropical Soil Biology and Fertility (TSBF) c/o UNESCO, PO Box 30592, Nairobi, Kenya. 2. International Crops Research Institute for the Semi-Arid Tropics (lCRISAT), BP 320, Bamako, Mali. 3. International Livestock Research Institute/International Institute of Tropical Agriculture (lLRIIIITA), PMB 5320, Ibadan, Nigeria. 301 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production most important cereals are sorghwn and pearl millet and cowpea is often intercropped with these cereals (Steiner 1984). Cowpea grain contains about 22% protein and constitutes a major source of protein for resource-poor rural and urban people. It is estimated that cowpea supplies about 40% of the daily protein requirements to most of the people in Nigeria (Muleba et al. 1997). The crop residues from cowpea constitute an important source of livestock feed especially in the dry savannas of WAS AT. The principal reasons for farmers to intercrop are flexibility, profit maximization, risk minimization, soil conservation and maintenance, weed control, and nutritional advantages (Norman 1984; Swinton et al. 1984; Shetty et al. 1995; Fussell and Serafini 1985). In mixed cropping systems, cowpea yields are very low due to low soil fertility, low plant- ing densities, and pests and diseases (Ntare 1989, Reddy et al. 1992). Cowpea grain yield varies between 50 kg/ha and 300 kg/hain farmers' fields in marked contrast to over 2000 kg/ha obtainable on research stations and by large-scale commercial enterprises in pure cropping. In the mixed farming systems of the WASAT, increasing legume component in the farming systems is important in order to increase the availability offodder as livestock feed while increasing soil fertility. Rotation of cereals with legumes has been extensively studied in recent years. Use of rotational systems involving legumes is gaining importance throughout the region because of economic and sustainability considerations. The beneficial effect of legumes on suc- ceeding crops is normally exclusively attributed to the increased soil N fertility as a result of N2 fixation. The amount of N2 fixed by legwninous crops can be quite high, although it has been demonstrated that legumes can also deplete soil nitrogen (Rupela and Saxena 1987, Blwnenthal et al. 1982). Most of the data reported on the quantity of N fixed by legume crops in the WASAT concerned the aboveground part of the legume and very little is known about the nitrogen fixed by the roots. Where much of the legume biomass is returned to the soil as green manure, a positive N balance is to be expected. However, this may not be true for cowpea, where the bulk of above biomass is removed from the system. Nevertheless, there are many other positive effects of grain legwnes such as the improvement of soil biological and physical properties and the ability of some legumes to solubilize occluded phosphorus and highly insoluble Calcium-bounded phosphorus by roots exudates (Arihara and Ohwaki 1989). Other advantages of crop rotation include soil conservation (Stoop and Staveren 1981), organic matter restoration (Spurgeon and Grimson 1965), and pest and disease control (Curl 1963). While considerable information is available on fertilizer requirements for sole cropping of various crops, it is limited for intercropping and rotations. This paper will review the cowpea production environment, the effect of soil fertility improvement, and will conclude with new research opportunities. Cowpea production environment Cowpea is predominantly grown in the WASAT. This zone is characterized by a grow- ing period of 60-150 days. The rainfall is low, variable, and undependable. One striking feature of the soils is their inherent low fertility expressed in low levels of organic carbon (generally less than 0.3%), total and available phosphorus and nitrogen, and effective cation exchange capacity (ECEC) (Table 1). About 98% of the soil nitrogen is stabilized in organic matter. Thus, the total nitrogen in the soil and the amount of nitrogen released 302 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics Table 1. Means and ranges of selected physical and chemical properties of West African semiarid soils from 30 representative sites. Parameter Range Mean pH-H20 (2: 1 water: soil) 3.95-7.6 6.17 pH-KCI (2: 1 water: soil) 3.41-7.0 5.05 Clay ("!o) 0.7-13 3.9 Sand ("!o) 71-99 88 Organic matter ("!o) 0.14-5.07 1.4 Total nitrogen (mglkg) 31-226 446 Exchangeable bases (cmol/kg) Ca 0.15-16.45 2.16 Mg 0.02-2.16 0.59 K 0.03-1.13 0.20 Na 0.01-0.09 0.04 Exchangeable AI (cmol/kg) 0.02-5.6 0.24 Effective cation exchange capacity (cmol/kg) 0.54-19.2 3.43 Base saturation ("!o) 36--99 88 AI saturation ("!o) 0-46 3 Total phosphorus 25-941 136 Available phosphorus 1-83 8 Maximum P sorbed 27-406 109 Source: Bationo et al. (2000). for plant nutrients uptake will depend on the organic matter level of the soil. Total and available P levels are very low and P deficiency is the most limiting soil fertility factor for cowpea production. Apart from low P stocks, the low-activity nature of these soils results in a relatively low capacity to fix added phosphorus (Bationo et al. 1995). Phosphorus sorp- tion maxima of the WASAT soil ranged from 27 to 405 mg P/kg with a mean of 109 mg P/kg. Low quantities of P need to be added to the soil to maintain 0.2 ppm P in the soil solution. At present most cultivated land in the region lose more N, P, and K than gained and continuous cultivation has led to nutrient mining and loss of topsoil by wind and/or water erosion (Table 2). Under these conditions, productivity levels of both cereals and legumes are too low to sustain food production and to meet food requirements of the fast growing human populations. Although organic amendments such as crop residue, manure, or compost are essential in the sustainability of the cropping systems, they cannot prevent nutrient mining. The addition of organic amendments corresponds in most cases to a recycling process, which cannot compensate for nutrient exported through crop products. As a result, the use of external inputs such as inorganic plant nutrients or local sources of P such as phosphate rock are essential requirements for soil productivity. Effect of soil fertility improvement on cowpea production Research results in the region have shown the importance of the improvement of soil fertility for crop production (Mokwunye and Vlek 1986; Pieri 1989; Van Reuler and Jansen 1989; Van derHeide 1989; Bationo and Mokwunye 1991; Sedogo 1993). In the Sahelian zone, soil fertility appears to be more limiting to crop and fodder production than rainfall and the use of fertilizer will increase water-use efficiency (Penning de 303 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 2. Annual nutrient losses for some West African countries. Losses for the region (1()3 tonnes) Country Area ('000 hal N P,05 K,O Benin 2972 41388 10366 32499 Burkina Faso 6691 95391 27754 78764 Ghana 4505 137140 32313 90474 Mali 8015 61707 17888 66725 Niger 985 176120 55331 146617 Nigeria 2813 1107605 316687 946157 Source: Adapted from Stoorvogel and Smaling (1990). Vries and Djiteye 1991, Breman and de Wit 1983). The use of mineral fertilizers can significantly increase water-use efficiency. Significant cowpea responses to nitrogen applied as urea have been obtained in different agroecological zones of the WASAT (Table 3). These significant responses indicate that the predominantly sandy soils of the WASAT may be deficient in molyb- denum required for efficient symbiotic fixation (Hafner et al. 1992). For example, on the sandy acid soil at Bengou in the Sudanian zone, significant molybdenum response was obtained at different levels of soil fertility management for cowpea (Fig. 1). Legumes such as cowpea have a high P requirement. P is reported to stimulate root and plant growth, initiate nodule formation, as well as influence the efficiency of the rhizobium-legume symbiosis. It is also involved in reactions with energy transfer, more specifically ATP in nitrogenase activity (Israel 1987). Research conducted at Ikenne in the humid zone and Kamboinse in the Sudanian zone of West Africa indicated a strong differential response to P by cowpea cultivars (Fig. 2). The local Kamboinse variety is a fodder type and the application of P resulted in higher fodder yield but lower grain production. As reported by several scientists such as Dwivedi et al. (1975); Khan and Zende (1977); Stukenholtz et al. (1966); Takkar et al. (1976); and Youngdhal et al. (1977), the application of P resulted in significant decrease of zinc concentration in the cowpea grain which can affect the nutritional quality (Buerkert et al. 1998). Despite the importance of P in these soils, the use of commercial P fertilizers in the WASAT is limited due to the high cost of imported fertilizers. Several countries in the region, however, are known to have natural phosphate deposits. Direct appli- cation of indigenous phosphate rocks (PR) can be an alternative to the use of more expensive water-soluble phosphorus fertilizers. This practice would also promote savings in scarce foreign exchange. The effectiveness of PR depends on its chemical and mineralogical composition, soil factors, and the crops to be grown (Khasawneh and Doll 1978; Lehr and McClellan 1972; Chien and Hammond 1978). The relative agronomic effectiveness of Tahoua PR and Kodjari PR in different agroecological zones of the WASAT has been evaluated (Table 4). The data indicate that Tahoua PR outperformed Kodjari PR in agronomic effectiveness at two of the three sites. These results are in agreement with the chemical composition of the two rocks where the molar PO/C04 ratio is 25 for Kodjari PR and 4.9 for Tahoua PR. The agronomic cowpea is not better than that of the cereal pearl millet crop. This is in contradiction to other reports where legumes have highest strategy to solubilize PR than cereals by rhizosphere acidulation (Aguilar and Van Diest 1981; Kirk and Nye 1986; Hedley et al. 1982) and exudation of organic acids (Ohwaki and Hirata 1992). 304 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics Table 3. Effect of nitrogen on cowpea yield at three sites in 1988. N rates (kg N/ha) o 15 30 45 S.E. (D.F.27) CV(%) -------Cowpea fodder------- Sadore 4069 4474 4288 4264 218.3 15 Bengou 2213 2510 2548 3008 153.7 17 Tara 2974 2963 3025 3500 161.3 15 Source: Bationo and Ntare (2000). 3000 .-. 2000 fU ..c ~ ~ .. CII "C "C .S! 1000 ~ Q. ~ o U o 5000 fa' 4000 ..c ~ ~ .. QJ 3000 "C "C .S! -== C 2000 "C C == e \,j 1000 _ With molybdenum c=JWithout molybdenum Control SSP TSP Control SSP TSP SE = 112 PRT CR+SSP lime+SSP ssP = Supper single phosphate TSP = Triple superphosphate PRT = Phosphate rock of Thaoua CR = Crop residue SE = 85 PRT CR+SSP lime+SSP Figure 1. Effects of different phosphorus sources, crop residue, lime, and molybdenum on cowpea and groundnut fodder yield, Tara, Niger, 1993. Source: Bationo, (unpublished data). 305 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production 1500- a) Ikenne SE = 118 ~ 1000 .>. --+-- Ife brown ~ TYX 1193-7D ------A- Shaki local ~Vita4 c -i! \,j 500~--------------,-------------,-------------, 0.00 0.01 0.10 1.00 Phosphorus concentration in soil solution at sowing (ppm). 2500 b) Kamboinse 'ii ..c ~ SE=25 ~ 2000 "t:I -+- Kamboinse "ii .>. ~ TYX 1193-7D c -i! -----fr- Vita4 \,j 1500 1000-4--------------,--------------,--------------, o 0.01 0.10 1.00 Phosphorus concentration in soil solution at sowing (ppm). Figure 2. Relationship between grain yield and phosphorus concentration in soil solution at sowing in sandy loam Paleustatif Oxic Paleustalf at (a) Ikenne and (b) Kamboinse. 306 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics Table 4. Relative agronomic effectiveness for pearl millet and cowpea as compared to single superphosphate (SSP) (%) of Tahoua phosphate rock (TPR) and Kodjari phosphate rock (KPR) in three agroecological zones of Niger. Cowpea fodder (kg/ha) Cowpea total dry matter (kg/ha) Source: Mahamane et al. (1997). Sad ore TPR KPR 43 56 28 40 Goberi TPR KPR 73 72 51 51 Gaya TPR KPR 42 52 42 55 The response of cowpea grain and stover yield to different sources of P fertilizers is presented in Figure 3. The application of P fertilizers can triple cowpea stover produc- tion. The relative agronomic effectiveness of Phosphate rock annual application (PRA) indigenous to Niger varied from 42 to 54% as compared to the water soluble single superphosphate (SSP) (Table 5). The acidulation of PR at 50% (pAPR 50) with sulfuric acid can increase the relative agronomic effectiveness to 96% for cowpea stover produc- tion. For fodder production, triple superphosphate (TSP) relative agronomic effectiveness varied from 77 to 91 % indicating that sulfur is needed for cowpea growth. Research at ICRISAT-Niger has focussed on the placement of small quantities of P fertilizers at planting in order to develop optimum farmer-affordable P application rec- ommendation for increased crop yield. For cowpea stover production, phosphorus-use efficiency increased from 44 with the addition of Kodjari PR to 93 kg/kg phosphorus plus 4 kg P/ha as 15-15-15, respectively, (Table 6). Long-term experiments are a practical means of addressing the difficult issues asso- ciated with quantitative assessment of sustainability in agriculture. In summarizing the results of long-term soil fertility management in Africa, Pieri (1986) concluded that soil fertility in intensive arable farming in the WASAT can only be maintained through efficient cycling of organic materials in combination with mineral fertilizers and with rotation with leguminous N 2-fixing species. Results from a long-term experiment at Sadore in Niger indicated that the application of small quantities of fertilizers and crop residues resulted in an increase of cowpea fodder yield from 1700 to 5300 kg/ha (Fig. 4). In on-farm trials, pocket applications of small quantities of manure (3 tlha) plus 4 kg/ha ofP at seedling time increased cowpea yield from 180 kg/ha in the control plot to 400 kg/ha (Fig. 5). Effect of cowpea production on soil fertility improvement Despite the recognized need to apply chemical fertilizers for high yields, the use of min- eral fertilizers in West Africa is limited by lack of capital, inefficient distribution systems, poor enabling policies, and other socioeconomic factors. Cheaper means of improving soil fertility and productivity is therefore necessary. Cereal-legume rotation effects on cereal yields have been reported for the WASAT (Bagayoko et al. 1996; 2000; Bationo et al. 1998; Klaij and Ntare 1995; Nicou 1977; Stoop and Staveren 1981; Bationo and Ntare 2000). In all these studies, the yield of cereal after cowpea was significantly higher than in continuous cereal cultivation. Cowpea yield also significantly responded to crop rotation, indicating that factors other than N alone contributed to the yield increases in the cereal-legume rotations. 307 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Grain yield {kg/hal 400 40 20 P applied (kg p,o,tha) P applied (kg p,o,tha) 4 Fodder yield {kg/hal 4000 P applied (kgP/ha) Padded {kg/hal 40 P uptake (kg P/ha) P soun:es + PRA = Phosphate rock annual application <> PAPR25 = Phosphate rock acidulated at 25% • PAPR 60 = Phosphate rock aacidulated at 50% o TSP = Triple superphosphate • SSPN = Single superphosphate with nitrogen b. SSP-N = Single superphosphate without nitrogen ... PRB = Phosphate rock basal application 8 P uptake {kg/hal Figure 3. Relationship between cowpea grain and fodder yield with P applied, and between phosphorus applied and phosphorus uptake, Sadore, Niger, 1983. Source: Bationo (unpublished data). 308 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics Table 5. Relative agronomic effectiveness of different sources of phosphorus on cowpea. --1993-- --1994-- P sources Grain % Fodder Grain % Fodder Phosphate rock annual 70 54 49 42 application (PRA) Partially acidulated phosphate rock 45 58 61 75 at 25% (PAPR 25) Partially acidulated phosphate rock 72 92 88 96 at 50% (PAPR 50) Triple superphosphate (TSP) 68 91 65 77 Single superphosphate (SSP) 74 87 86 91 Phosphate rock based application Table 6. Effect of different sources of phosphorus and their placement** on cowpea yield and Phosphorus-use efficiency (PUE), Karabedji, (1998 rainy season). Grain Fodder Yield PUE Yield PUE P sources and method (kg/ha) (kg/ha) (kg/ha) (kg/ha) of application P applied P applied Control 505 1213 SSP broadcast 1073 44 2120 70 SSP broadcast+SSP HP 1544 61 3139 113 SSP HP 1050 136 2021 452 15-15-15 broadcast 1165 51 2381 90 15-15-15 broadcast+15-15-15 HP 2383 110 3637 142 15-15-15 HP 1197 173 2562 337 PRT broadcast 986 37 2220 77 PRT broadcast+SSP HP 1165 68 3127 113 PRT broadcast+15-15-15 HP 1724 72 3163 115 PRK broadcast 920 32 1791 44 PRK broadcast+SSP HP 1268 45 2588 81 PRK broadcast+15-15-15 HP 1440 55 2792 93 S.E. 164 313 *SSP Single superphosphate; 15-15-15 compound fertilizer containing 15% N, 15% P20 5, 15% K20; Tahoua phosphate rock (TPR), Kodjari phosphate rock (KPR). HP signifies hill placement of fertilizer. **For broadcast, 13 kg P/ha was applied. ** For HP, 4 kg P/ha as hill placement. Source: Bationo (unpublished data). Bationo and Ntare (2000) studied nitrogen dynamics in different cropping systems. In order to determine N availability, the soil was incubated and mineral nitrogen deter- mined at 7,21, and 35 days (Keeney 1982). Crop rotation significantly affected mineral nitrogen release (Fig. 6). The fallow millet rotation supplied more nitrogen than the cowpea-millet rotation, but the latter was more productive for millet production. Isotopic dilution method with 15N was used to determine the nitrogen fixed by cowpea using pearl millet as a non-fixing crop. Nitrogen derived from the atmosphere by cowpea varied from 65 to 89% and the total nitrogen fixed by cowpea depended on the level of soil fertility improvement (Table 7). The quantity of nitrogen fixed by 309 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production c=J Pearl millet (SE = 450) _ Cowpea (SE = 275) 0-'------'-- Control Crop residue Fertilizer Crop residue + fertilizer Figure 4. Long-term crop residue management at Sadore, Niger, 1996. Source: Bationo et al. (2000). (2 tlha of crop residue was applied as mulch in crop residue treatment and 4 tlha of crop residue was applied as mulch in the crop residue plus fertilizer treatment; fertilizer was applied at 30 kg N/ha and 13 kg P/ha). cowpea varied from 26 kg/ha in the control plot to 87 kg/ha in the treatment where the soils were amended with mineral and agronomic plant nutrients. In order to determine 15N recovery from different cropping systems, labeled nitrogen fertilizers were applied to microplots where pearl millet was grown continuously (M-M) in rotation with cowpea (C-M), in rotation with groundnut (G-M), intercropped with cowpea (CIM-CIM), and intercropped with groundnut (GIM-GIM). Nitrogen-use effi- ciency increased from 20% in continuous pearl millet cultivation to 28% when pearl millet was rotated with cowpea (Bationo, unpublished data). Nitrogen derived from the soil was better used in rotation systems than with continuous millet cultivation. In another trial on interaction between phosphorus fertilizers and different cropping systems, the application of P had a significant effect on yield of cowpea and pearl millet and rotation performed better than continuous cultivation of both crops (Fig. 7). A higher level of organic carbon was also found in the rotation systems compared to the continuous cropping systems, probably due in part to fallen cowpea leaves (Fig. 8). 310 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics 1000 ~ 800 ~ c ·E 600 DO ~ ... ---+--Control _ F1M1 = 3 t/ha of manure broadcast ------A------ F1M2 = 3 t/ha of manure hill placed ---e--- F2M1 = 6 t/ha of manure broadcast -----.-- F2M2 = 6 t/ha of manure hill placed SE = 31 8 400~----~~-=~~---------=~ 200 O~ _______________________ -' ________________________ 'I 13.0 0.0 6.5 Figure 5. Effects of fertilizer and manure placement on cowpea grain yield, Karabedji, 1999. Source: Bationo (unpublished data). ---+-- Fallow/Fallow 12 ------'Efr----- Fallow/Millet --e- Millet/Millet ----e--- Cowpea/Millet ---A-- GroundnutlMiliet 10 14 21 28 35 Days of incubation Figure 6. Relationship between cumulative mineral nitrogen and time of incubation of soils from different crop rotations pooled over three sites. Source: Bationo and Ntare (2000). 311 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 7. Nitrogen derived from the air (Ndfa) and total N fixed by cowpea stover using 15N dilution technique, Sadore, Niger, (1991 rainy season). Yield Nyield Treatment (Uha) N (%) (kg/ha) NdFF (%) Ndfa (%) Control 1.75 2.18 38 2.43 65 Molybdenum 3.08 2.28 71 1.37 80 Carbofuran 2.58 2.19 57 2.04 71 Manure 2.42 2.44 60 0.79 89 Phosphorus 3.58 2.01 65 1.56 78 Complete 3.75 2.66 100 0.80 89 SE ±0.47 ±0.09 ±10.39 ±0.18 ±2.56 CV(%) 28 6 27 20 6 Source: Bationo (unpublished data). 500 ~ !' 4000 ~ Ql .>. t 3000 ~ ~ ~ ..s 0; ~ 2000 Cropping systems ~ C-M=millet following cowpea -+- M-M = millet following millet --7<- C/M-M = millet following millet intercropped with cowpea -+- C-c/M = millet intercropped with cowpea following cowpea ~ M-c/M = millet intercropped with cowpea following millet ----- c/M-c/M = millet intercropped with cowpea following millet intercropped with cowpea S.E = 512 1000+---------------------------------- o 15 30 Phosphorus applied (kglP,O,lha) N fixed (kg/ha) 26 58 41 53 51 89 ±9.06 29 Figure 7. Effects of phosphorus and nitrogen on different cropping systems over four years, Sadore, Niger. Source: Bationo (unpublished data). The application of phosphorus, nitrogen, crop residue, ridging, and rotation of pearl millet with cowpea was evaluated to determine phosphorus-use efficiency. The results showed that soil productivity of the sandy Sahelian soils can be significantly increased with the adoption of improved crop and soil management technologies. Whereas the absolute control recorded 33 kg/ha of grain yield, 1829 kg was obtained when phos- phorus, nitrogen, and crop residue were applied to plots that were ridged and in rotation 312 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics 0.30 0.28 ~ c 0.26 0 -+- C-M = millet following cowpea .." Ii ----I:>- M-M = millet following millet u .1.< Iii 0.24 2." -A-M/C-M = millet following millet intercropped with cowpea ~ M/C-M/C = continuous millet-cowpea intercropped 0 ·0 0.22 II> 0.20 0.0 6.5 13.0 P applied (kg P/ha) Figure 8. Effects of phosphorus and cropping system on soil organic carbon, after four years of cultivation, Sadore, Niger, 1995. Source: Bationo (unpublished data). with cowpea. The plots without rotation yielded 1146 kg/ha. Results indicated that for grain yield, phosphorus-use efficiency increased from 46 kg/kg P with only phosphorus application, to 133 kg/kg phosphorus when phosphorus was combined with nitrogen and crop residue application and the crop was planted on ridges (Table 8). Conclusion and research opportunities In the traditional cropping systems, cowpea is grown between cereals at very low density, as the farmers' primary goal is to produce cereal for family subsistence, cowpea being an additional benefit. This means that farmers need to be assured of sufficient cereal harvest to feed their families before integrating more cowpea in the cropping systems. Cowpea grain yield in the mixed systems is very low, varying between 50 and 300 kg/ha in marked contrast to over 2000 kg/ha on-station and by large-scale commercial enterprises in sole cropping. In addition to the low planting densities, pests and disease control, the inher- ent low fertility of the soil in the WASAT (particularly phosphorus) is one of the major constraints to cowpea production. Thus, soil fertility replenishment should be an integral part of any program aimed at reversing the downward trend in cowpea production and the conservation of the environment. Phosphorus is the most limiting plant nutrient for cowpea production in the WASAT and there is ample evidence that indicates marked differences between cowpea genotypes for phosphorus uptake. Understanding the factors affecting phosphorus uptake such as the ability of plants to (i) solubilize soil P through acidification of the rhizosphere and the release of chelating agents and phosphate enzymes, (ii) explore a large soil volume, and (iii) absorb phosphorus from low phosphorus solution would help increase cowpea production and yield in the semiarid tropics. The available and total phosphorus values are very low in the region. With these extreme low values of total phosphorus, selecting cultivars adapted to low phosphorus condition 313 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) w .... .l>- Table 8. Effect of mineral fertilizers, crop residue (CR), ridging, and crop rotation on pearl millet and phosphorus-use efficiency (PUE) wastes, Sadore, Niger (1998 rainy season). Without CR, without N Without CR, without N Without CR, without N TOM Grain TOM Grain TOM Grain Yield PUE Yield PUE Yield PUE Yield PUE Yield PUE Yield PUE Treatment kglha kglha P kg/ha kg/ha P kglha kglha P kg/ha kglha P kg/ha kg/ha P kglha kg/ha P Control 889 33 2037 58 995 61 13 kg P/ha 2704 140 633 46 4339 177 1030 75 4404 185 726 51 n kg P/ha + ridge 2675 137 448 32 4057 155 946 68 3685 210 785 56 13 kg P/ha + rotation 5306 340 1255 94 6294 327 1441 106 5392 338 1475 109 13 kg! P/ha + ridge + rotation 5223 333 1391 104 5818 291 1581 117 6249 404 1702 126 SE 407 407 407 407 407 407 CR: Crop residue; N: Nitrogen; TDM: total dry matter; PUE: phosphorus-use efficiency (kg grain/kg P); yield: g!ha). Source: 8ationo (unpublished data). Without CR, without N TOM Grain Yield PUE Yield PUE kg/ha kg/ha Pkglha kglha P 1471 98 240 4594 1212 86 4530 235 1146 81 6124 358 1675 121 7551 468 1829 133 407 407 bl ~ ffi 8 :J q- 5' c: g. :J '" 6" ill' 3 s· OQ '" ~ ~ I\;- OQ a :J o :3 ?i. i· (§ :3 ~ o ...., 8 ~ ffi -0 a ~ n g. Soil fertility management and cowpea production in the semiarid tropics would not be feasible as one cannot mine what is not there. Direct application of indigenous PR can be an economic alternative to the use of more expensive imported water-soluble P fertilizers. The effectiveness of mycorrhizal in utilizing soil P has been well documented (Silberbush and Barber 1983; Lee and Wani 1991; Daft 1991). An important future research opportunity is the selection of cowpea genotypes that can efficiently associate with vesicular- arbuscular mycorrhizal (VAM) for better utilization of P from applied PR. Cereal-cowpea rotations have led to increased cereal yields at many locations in the WASAT. Factors such as mineral nitrogen (VAM) for P nutrition improvement and plant parasitic nematodes have been identified as mechanisms accelerating the enhanced yield of cereals in rotation with cowpea. Most of the research quantified the aboveground N fixed by different cowpea cultivars, but very little is known about the below-ground N fixed by cowpea. In the WASAT, most of the aboveground cowpea biomass is used for animal feed and not as green manure. Furtherresearch should focus more on on-farm quantification of the below-ground N fixed by cowpea in order to identify the best cultivar for soil N buildup. The identification and alleviation of technical and socioeconomic constraints in order to increase cowpea in the present cropping systems needs attention in future. As a cash crop, farmers will increase their purchasing power to acquire external inputs such as fertilizers. The enhancement of cowpea in the present cropping systems will not only improve the soil conditions for the succeeding cereal crop, but will provide good quality livestock feed, and the manure produced will be of better quality for soil fertility improvement. References Aguilar,A.S. and A. Van Diest. 1981. Rock-phosphate mobilization included by the alkaline uptake pattern oflegumes utilizing symbiotically fixed nitrogen. Plant and Soil 61: 27-42. Arihara,1 and Y Ohwaki. 1989. Estimation of available phosphorus in vertisol and alfisol in view of root effects on rhizosphere soil in XI Colloquium on plant nutrition-physiology and applications, 30 July to 4 August 1989. Wageningen, The Netherlands. Bagayoko, M., S.C. Mason, S. Traore, and K.M. Eskridge. 1996. Pearl rnillet'cowpea cropping systems yield and soil nutrient levels. African Crop Science Journal 4: 453-462. Bagayoko, M., A. Buerkert, G. Lung, A. Bationo, and V Romheld. 2000. Cereal!legume rotation effects on cereal growth in Sudano-Sahelian West Africa: soil mineral nitrogen, mycorrhizae, and nematodes. Plant and Soil 218: 103-116. Bationo,A. andA.U Mokwunye. 1991. Alleviating soil fertility constraints to increased crop produc- tion in West Africa. The experience in the Sahel. Fertilizer Research 29: 95-115. Bationo,A.,A. Buerkert,M.P. Sedogo,B.C. Christianson, andA.U Mokwunye. 1995. A critical review of crop residue use as soil amendment in the West African semi-arid tropics. Pages 305-322 in Livestock and sustainable nutrient cycling in mixed farming systems of sub-Saharan Africa, edited by 1M. Powell, S. Fernandez Rivera, T.O. Williams, and C. Renard. Proceedings International Conference, ILCA, Addis Ababa, Ethiopia. Bationo,A., F. Lompo, and S. Koala. 1998. Research on nutrient flows and balances in West Africa: State-of-the-art. Pages 19-36 in Nutrient balances as indicators of production and sustainability in sub-SaharanAfrican agriculture, edited by E.MA. Smaling.Agriculture, Ecosystems and Environ- ment 71: 1,2, and 3. Bationo,A. and B.R. Ntare. 2000. Rotation and nitrogen fertilizer effects on pearl millet, cowpea, and groundnut yield and soil chemical properties in a sandy soil in the semi arid tropics, West Africa. Journal of Agricultural Science, (Cambridge) 134: 277-284. Bationo,A., S.P. Wani, C.L. Bielders, P.L.G. Vlek, and U Mokwunye. 2000. Crop residue and fertil- izer management to improve soil organic carbon content, soil quality, and productivity in the desert margins of West Africa. Pages 117-145 in Global climate change and tropical ecosystems, edited 315 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production by R Lal, 1M. Kimble, and B.A. Stewart. Advances in Soil Science. CRC Press, Washington DC, USA. Blumenthal, M.J, v.P. Quach, and P'G.E. Searle. 1982. Effect of soybean population density on soybean yield, nitrogen accumulation, and residual nitrogen. Australian Journal of Experimental Agriculture 28: 99-106 Breman, H. and C.T. de Wit. 1983. Rangeland productivity and exploitation in the Sahel. Science 221: 1341-1347. Buerkert,A., C. Haate, M. Ruckwied, and H. Marschner. 1998. Phosphorus application affects the nutritional quality of millet grain in the Sahel. Field Crops Research 57: 223-235. Chien, S.H. and L.L. Hammond. 1978. A simple chemical method for evaluating the agronomic potential of granulated phosphate rock. Soil Science Society of America Journal 42: 615--617. Curl, E.A. 1963. Control of plant diseases by rotation. Botanical Review 29: 413-479. Daft, M.l 1991. Infl uences of genotypes, rock phosphate, and plant densities on mycorrhizal devel- opment and the growth responses offive different crops. Agriculture, Ecosystems and Environ- ment 35: 151-169. Dwivedi, RS., N.S. Randwawa, and RL. Bansal. 1975. Phosphorus-zinc interaction. L Sites of immobilization of zinc in maize at high levels of phosphorus. Plant and Soil 43: 639--648. Fussell, L.K. and P.G. Serafini. 1985. Associations de cultures dans les zones tropicales semi-arides d'Afrique de l'Ouest: strategies de recherche anterieures et futures. (In Fr.) Pages 254-278 in Technologies appropriees pour les paysans des zones semi-arides de I' Afrique de I 'Ouest, edited by H.W. Ohm and IG. Nagy. Purdue University, West Lafayette, Indiana, USA. Hafner, H., B.J. Ndunguru,A. Bationo, and H. Marshner. 1992. Effect of nitrogen, phosphorus, and molybdenum applications on growth and symbiotic N2 fixation of groundnut in an acid sandy soil in Niger. Fertilizer Research 156: 164-176. Hedley, M.J., P.H. Nye, and RE. White. 1982. Plant-induced changes in the rhizosphere of rape (Brassica napus var. Emerad) seedlings. IT. Origin ofthe pH change. NewPhytologist 91: 31- 44. Israel, D.N. 1987. Investigation of the role of phosphorus in symbiotic nitrogen fixation. Plant Physiology 84: 835-840. Khan, A.A. and G.K. Zende. 1977. The site for Zn-P interactions in plants. Plant and Soil 46: 259-262. Keeney, D.R. 1982. Nitrogen availability indices. Pages 711-730 in Methods of soils analysis, edited by A.L. Paye et al. American Society of Agronomy, Madison, Wisconsin, USA. Khasawneh, F.E. and E.C. Doll. 1978. The use of phosphate rock for direct application to soils. Advanced Agronomy 30: 155-206. Kirk, G.JD. and P.H. Nye. 1986. A simple model for predicting the rate of dissolution of sparingly soluble calcium phosphate in soil. L The basic model. Journal of Soil Science 37: 529-540. Klaij, M.C. and B.R. Ntare. 1995. Rotation and tillage effects on yield of pearl millet (Pennisetum glaucum) and cowpea (Vigna unguiculata), and aspects of crop water balance and soil fertility in semi-arid tropical environment. Journal of Agriculture Science (Cambridge) 124: 39-44. Lee, K.K. and S.P. Wani. 1991. Possibilities for manipulating mycorrhizal associations in crops. Pages 107-166 in Phosphorus nutrition of grain legumes in the semi-arid tropics, edited by C. Johansen, K.K. Lee, and K.L. Sahrawat. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT). Patancheru, AP 502324, India. Lehr, J.R. and G.H. McClellan. 1972.A revised laboratory reactivity scale for evaluating phosphate rocks for direct application. Bulletin 43, Tennessee Valley Authority, Muscle Shoals, Alabama, USA. Mahamane, L,A. Bationo, F. Seyni, and Z. Hamidou. 1997. Acquis recents des recherches sur les phosphates naturels du Niger. Pages 73-78 in Soil fertility management in West African land use systems, edited by G. Renard,A. Neef, K. Becker, and M. von Oppen. Margraf Verlag, Weiker- sheim, Germany. 316 Digitized by Google Soil fertility management and cowpea production in the semiarid tropics Mokwunye, A.U., and P.L.G. Vlek (editors). 1986. Management of nitrogen and phosphorus fertilizers in sub-Saharan Africa. Martinus Nijhoff, Dordrecht, The Netherlands. Muleba, N., C. Dabire, J.B. Suh, I. Drabo, and J.T. Ouedraogo. 1997. Technologies for cowpea production based on genetic and environmental manipulations in the semi-arid tropics. Pages 195-206 in Technology options for sustainable agriculture in sub-Saharan Africa, edited by T. Bezuneh,A.M. Emechebe, J. Sedgo, and M. Ouedraogo. Publication ofthe Semi-Arid Food Grain Research and Development Agency (SAFGRAD) of the Scientific, Technical and Research Commission ofOAU, Ouagadougou, Burkina Faso. Nicou, R.N. 1977. Le travail du sol dans les terres exondees du Senegal. Motivations constraintes. ISRA-CNRA, Bombay, India. 51 pp. Norman, D.W. 1974. Rationalizing mixed cropping under indigenous conditions: the example of northern Nigeria. Journal of Development Studies 11: 3-21. Ntare, B.R. 1989. Intercropping morphologically different cowpea with pearl millet in a short season environment in the Sahel. Experimental Agriculture 26: 41-47. Ohwaki, Y. and H. Hirata. 1992. Differences in carboxylic acid exudation among P-starved leguminous crops in relation to carboxylic acid contents in plant tissues and phospholipid levels in roots. Soil Science and Plant Nutrition 38: 235-243. Penning de Vries, F.W.T. and M.A. Djiteye. 1991. La productivite des paturages saheliens: une etude des sols, de la vegetation et de l'exploitation de cette resource naturelle. Center for Agricultural Publishing and Documentation (Pudoc-DLO), Wageningen, The Netherlands. Pieri, C. 1986. Fertilization des cultures vivrieres et fertilite des sols en agriculture paysanne subsaharienne. Agronomie Tropicale 41: 1-20. Pieri, C. 1989. Fertilite des terres de savanes : Bilan de trente ans de recherches et de developpe- ment, Centre de cooperation internationale en recherche agronomique pour Ie developpement (CIRAD) et Ie Ministere de la Cooperation, Paris, France. 443 pp. Reddy, K.C., P. Visser, and P. Buekner. 1992. Pearl millet and cowpea yields in sole and intercrop system, and their after-effects on soil and crop productivity. Field Crops Research 28: 315- 326. Rupela, O.P. and M.C. Saxena. 1987. Nodulation and nitrogen fixation in chickpea. Pages 191-206 in The Chickpea Farnham Royal UK, edited byM.C. Saxena and K.B. Singh. Com- monwealth Agricultural Bureaux International and International Center for Agricultural Research in the Dry Areas. Sedogo, M.P. 1993. Evolution des sols ferrugineux lessives so us culture: influence des modes de gestion sur la fertilite: These de Doctorat Es-Sciences, Abidjan, Universite Nationale de COte d'Ivoire. Shetty, S.v.R., B.R. Ntare, A. Bationo, and C. Renard. 1995. Millet and cowpea in mixed farm- ing of the Sahel. A review of strategies for increased productivity and sustainability. Pages 293-304 in Livestock and sustainable nutrient cycling in mixed farming systems of sub- Saharan Africa, edited by J.M. Powell, S. Fernandez Rivera, T.O. Williams, and C. Renard. Proceedings International Conference, ILCA, Addis Ababa, Ethiopia. Silberbush, M. and S.A. Barber. 1983. Sensitivity of simulated phosphorus uptake to parameters used by a mechanistic-mathematical model. Plant and Soil 74: 93-100. Spurgeon, W.I., and P.H. Grimson. 1965. Influence of cropping systems on soil properties and crop production. Mississippi Agricultural and Forestry Experiment Station Bulletin No. 710. Steiner, K.G. 1984. Intercropping in tropical smallholder agriculture with special reference to West Africa. Stein, West Germany. 304 pp. 317 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Stoop W.A. and J.P.v. Staveren. 1981. Effects of cowpea in cereal rotations on subsequent crop yields under semi-arid conditions in Upper Volta. Pages 653--657 in Biological nitrogen fixation technology for tropical agriculture, edited by P.C. Graham and S.C. Harris. Centro Inter- national de Agricultura Tropical (CIAT), Cali, Colombia. Stoorvogel, 1.1. and E.M.A. Smaling. 1990. Assessment of soil nutrient depletion in sub-SaharanAfrica 1983-2000. Report 28, The Winand Staring Centre for Integrated Land, Soil and Water Research (SC-DLO), Wageningen, The Netherlands. Stukenholtz, D.D., R1. Olsen, G. Gogan, and RA. Olsen. 1966. On the mechanism of phosphorus- zinc interaction in corn nutrition. Soil Science Society of America Proceedings 30: 759-763. Swinton, S.M., G. Numa, and L.A. Samba. 1984. Les cultures associees en milieu paysan dans deux regions du Niger: Filingue et Madarounfa. Pages 183-194 in Proceedings of the Regional Workshop on Intercropping in the Sahelian and Sahelo-Sudanian zones of West Africa, 7-10 November 1984, Institute du Sahel, Niamey, Niger. Takkar, P.N., M.S. Mann, RL. Bansal, N.S. Randwawa, and H. Singh. 1976. Yield and uptake response of corn to zinc as influenced by phosphorus fertilization. Agronomy Journal 68: 942- 946. Van der Heide, 1. (editor). 1989. Nutrient management for food crop production in tropical farming systems. Institute for Soil Fertility Research (IB-DLO), Haren, The Netherlands. Van Reuler, H. and B.H. Jansen. 1989. Nutritional constraints in secondary vegetation and upland rice in southwest Ivory Coast. Pages 371-382 in Mineral nutrients in tropical forest and savanna ecosystems, edited by 1.H. Proctor. Special Publication 9 of The British Ecological Society, Blackwell Scientific Publications, Oxford, England. Youngdhal, L.J., L.v. Svec, W.C. Liebhardt, and M.R. Tel. 1977. Changes in the zinc-65 distribution in corn root tissue with phosphorus variable. Crop Science 17: 66--69. 318 Digitized by Google 4.7 Differential response of cowpea lines to application of P fertilizer G.O. Kolawole" G. Tian 2, and B.B. Singh 1 Abstract Phosphorus is important for cowpea production in many tropical African soils with inherent low P fertility. Most farmers in Africa, however, do not have access to P fertilizer. Selection of cowpea lines that produce good yield under low soil P or those with high P-use efficiency can be a low input approach to solving this problem. Pot and field trials were conducted at the International Institute of Tropical Agriculture (ilTA), Ibadan, southwestern Nigeria, to assess the differ- ential P responses of cowpea lines obtained from the germ plasm collection at I1TA. Thirty-five lines were assessed for P response in a pot trial using surface (0-15 cm) soil of a P-deficient Alfisol (Oxic Paleustalf). Seventeen lines (com- prising of 12 lines selected from the pot trial and five not included in the pot trial) were further assessed in the field. In the pot trial, P fertilizer significantly enhanced shoot, root, and grain dry weights. More than 60% of the cowpea lines also had greater nodule weight with P. Response of some of the cowpea lines was more pronounced for shoots than roots. In the field trial, more than 50% ofthe cowpea lines showed significant response to P. Compared with the pot trial, there were considerable variations in the pattern of responses of the cowpea lines to P. The cowpea lines were classified on the basis of their dry grain weights in the pot trial into four groups. Based on our results, we recommend that lines IT 90K- 284-2, IT 96D-724, and IT 93K-637-1 can be selected for further testing without P fertilizer. Lines IT 87D-941-1, IT 86D-719, and Dan Ila may perform very well without P fertilizer and give a high return when P is applied. When P fertilizer is available, line IT 87D-941-1 is recommended. These varieties should be tested at multiple sites to truly extend the results to breeding cowpea lines that could be targeted towards various soil P conditions. Introduction Phosphorus (P) is among the most needed elements for crop production in many tropical soils. However, many tropical soils are P-deficient (Adetunji 1995). The deficiency can be so acute in some soils of the savanna zone of western Africa that plant growth ceases as soon as the P stored in the seed is exhausted (Mokwunye et al. 1986). Soil P-deficien- cies primarily result from either inherent low levels of soil P or depletion of P through cultivation. Phosphorus, although not required in large quantities, is critical to cowpea yield because of its multiple effects on nutrition (Muleba and Ezumah 1985). It not only increases seed yields but also nodulation (Luse et al. 1975; Kang and Nangju 1983) and thus N fixation. Phosphorus application influences the contents of other nutrients in cowpea leaves (Kang 1. International Institute of Tropical Agriculture, Oyo Road, PMB 5320, Ibadan, Nigeria. 2. Institute of Ecology, University of Georgia, 106 Ecology Annex, Athens, GA 30602-2360, USA. 319 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production and Nangju 1983) and seed (Omueti and Oyenuga 1970). Application of P is therefore recommended for cowpea production on soils low in P (Sellschop 1962; Rachie and Rob- erts 1974). However, inorganic P fertilizers are often expensive and not readily available to resource-poor farmers. Furthermore, fertilizer P can be fixed into forms unavailable to plants by Fe and Al oxides found in tropical soils (Sample et al. 1980). Application of inorganic P fertilizers can therefore not be relied upon to adequately alleviate P-deficiency for improved cowpea production. Genotypic differences in the effect ofP on nodulation (Ankomah et al. 1995) and yield (Jain et al. 1986; Tenebe et al. 1995; Sanginga et al. 2000) of cowpea have been previously reported. However, mechanisms by which these cowpea varieties exhibit differential abilities to grow at low or high P supply are not completely understood. A better understanding of cowpea varietal differences in P nutri- tion may help in breeding new lines for areas where fertilizers are scarce and expensive. One of the options for overcoming the reliance on P fertilizers for improved cowpea production in P -deficient soils would be the selection of low soil P-tolerant cowpea lines that could access a greater proportion of the total soil P pool. There is, however, a paucity of information on variability in P responses among cowpea varieties. This paper reports the results of the responses of cowpea lines obtained from the germplasm collection at the International Institute of Tropical Agriculture (I1TA) to P fertilizer. Materials and methods Pot trial The trials were carried out at I1TA, Ibadan, southwestern Nigeria. For the pot trial, surface (0-15 cm) soil of a P-deficientAlfisol (Oxic Paleustalf) that was collected from Fashola village, Oyo State, southwestern Nigeria was used. The soil has the following properties: pH-H20 6.0 organic C; 6.5 g/kg total N; 0.5 g/kg extractable P; 7.5 mg/kg exchangeable (cmol (+)/kg soil) K 0.26; Ca 3.68, and Mg 0.96, respectively. The experiment was a factorial combination of 35 cowpea lines and two P application rates in a randomized complete block design with three replications. Table 1 lists 35 cowpea lines from the germplasm collection at I1TA grown in soil (3.5 kg/pot) with two levels of phosphorus (SSP): 0 (control) and 30 kg pps/ha. All the pots received basal dressing of 50 K (KCl); 50 Mg (MgS047HP); 5 Zn (ZnSOJ; 10 Mn (MnC124HP); 5 Cu (CuSOJ; 5 Mo [(NHJ~oP244HP]; and 5 P (NaHl047HP) in mg/kg soil. Four seeds were sown in each pot on 5 October 1998. Two weeks after planting, the seedlings were thinned to two plants per pot. A mixture of Karate® 2.5 E.C. (a.i. 25 g lambda-cyhalothrin per liter; 4 ml in 1 liter of water) and Vertimec® (a.i. 1.8% w/v abamectin (18 g/liter; 1.5 ml in 1 liter of water) insecticides was sprayed twice to con- trol insect pests during the experiment. The plants were grown to maturity. At maturity, pods were harvested from all pots. Dry pods were threshed by hand and grain weight determined. The plant shoots were cut at ground level. Roots were washed free of soil with water, using a screen with l-mm openings. Nodules were collected from the roots and counted. Plant shoots, roots, and nodules were oven dried at 65°C for 48 hours for dry weight determination. Litter was collected as part of the shoot biomass. Plant shoots were ground in a Wiley mill to pass through a 60-mesh size sieve and later analyzed for Nand P concentrations using the procedure described by Okalebo et al. (1993). Data collected were subjected to analysis of variance using the SAS package (SAS 1985). 320 Digitized by Google Table 1. Effect of P application on shoot, root, and grain dry weights (glpot) of 35 cowpea lines. Shoot Root Grain Lines -p +p Mean -p +p Mean -p +p Mean IT 90K-277-2 3.6 4.2 3.9 efg 0.57 0.78 0.68 defghi 0.1 1.2 0.7 hijklmn IT 90K-284-2 2.1 6.5 4.3 edefg 0.49 1.01 0.75 edefg 0.2 2.6 1.4 abed tJ IT90K-76 2.6 5.1 3.8 g 0.34 0.76 0.55 j 0.4 1.7 1.1 defghijklm ~ IT90K-59-2 3.7 4.1 3.9 fg 0.68 0.78 0.73 edefgh 0.2 1.1 0.7 ghijklmn ~ IT 93K-513-2 3.0 4.8 3.9 fg 0.45 0.73 0.59 ij 0.5 2.4 1.5 abed ~ IT 93K-693-2 2.8 5.5 4.2 defg 0.59 0.79 0.69 defghi 1.0 2.6 1.8 a i1l IT 93K-734 4.1 4.9 4.5 bedefg 0.44 0.89 0.67 efghij 0.3 1.2 0.8 fghijklmn '" 1:J IT 93K-452-1 2.5 5.9 4.2 edefg 0.42 0.82 0.62 hij 0.4 2.2 1.3 abedefg 0 w :J N IT 93K-637-1 3.5 6.3 4.9 abedefg 0.67 1.10 0.89 ab 0.6 2.6 1.6 abc '" ..... rt> IT 94K-437-1 4.0 7.2 5.6 abc 0.57 0.70 0.64 ghij 0.6 1.8 1.2 bedefghi 0 ...., IT 94K-440-3 4.3 6.8 5.5 abed 0.44 0.86 0.65 fghij 0.6 2.5 1.5 abed 8 IT 95K-1406 2.8 6.1 4.5 bedefg 0.55 0.75 0.65 fghij 0.1 1.3 0.7 ghijklmn ~ IT 95K-1156-3 3.7 6.3 5.0 abedefg 0.53 0.87 0.70 defghi 0.2 0.8 0.5 mn rt> I\) 0 IT 95 K-1 090-12 2.8 5.8 4.3 edefg 0.63 0.94 0.78 bede 0.2 1.1 0.7 ghijklmn S 0" IT 95 K-1 096-7 4.0 6.5 5.2 abedefg 0.48 1.02 0.75 edefg 0.1 0.8 0.5 mn ffi "" N" IT 95K-1464 3.1 5.2 4.2 defg 0.63 0.75 0.69 defghi 0.2 1.4 0.8 fghijklmn 0 CD a. IT 95 K-1 090-1 3.3 7.3 5.3 abedef 0.78 1.08 0.93 a 0.1 0.8 0.4 n I\) 0- :g '< IT 95 K-1 05-2 3.5 7.0 5.3 abedef 0.73 1.10 0.92 a 0.3 1.4 0.8 fghijklmn C"') IT 95 K-1 091-3 3.2 6.3 4.8 abedefg 0.78 0.70 0.74 edefg 0.1 1.4 0.8 fghijklmn ?f I\) 0 IT 95 K-1 095-2 3.3 8.4 5.9 ab 0.63 0.95 0.79 bed 0.02 1.1 0.5 mn g. ~ :J ..... continued 0 ...., ......... "'1:l (i) is' ... ...,. ~ rt> ... bl ~ ~ n 0 :J ~ Table 1. (continued) 5' c:: g. Shoot Root Grain :J '" Lines -p +p Mean -p +p Mean -p +p Mean 0 ~ IT 95 K-l 088-4 3.1 7.1 5.1 abedefg 0.64 0.87 0.75 edefg 0.03 1.3 0.7 ghijklmn 3 s" IT 95K-1384 3.2 5.6 4.4 edefg 0.46 0.83 0.65 fghij 0.1 1.3 0.7 ghijklmn OQ '" IT 95K-1543 3.8 7.1 5.5 abed 0.73 1.12 0.93 a 0.2 2.0 1.1 bedefghijkl ~ IT 95K-1491 4.3 5.0 4.7 abedefg 0.70 0.85 0.78 bede 0.4 2.4 1.4 abede fi) :3 IT 96D-666 2.8 6.3 4.5 bedefg 0.54 0.73 0.64 ghij 0.3 1.7 1.0 defghijklmn '" IT 96D-740 3.2 6.1 4.7 abedefg 0.72 0.66 0.69 defghi 0.2 1.5 0.8 efghijklmn ;;;- w a N IT 96D-748 2.5 6.1 4.3 edefg 0.70 0.78 0.74 edefg 0.1 2.3 1.2 bedefghi N :J IT 96D-757 3.2 6.2 4.7 abedefg 0.51 0.97 0.74 edefg 0.3 1.9 1.1 bedefghijkl 0 IT 96D-759 4.5 7.6 6.0 a 0.54 1.13 0.84 abc 0.1 1.1 0.6 klmn :3 ?i" IT 89KD-374-57 3.5 4.5 4.0 efg 0.49 0.75 0.62 hij 0.3 1.1 0.7 ghijklmn 3" IT 86D-719 3.7 5.5 4.6 abedefg 0.66 0.87 0.77 edef 0.4 2.9 1.7 ab -0 IT 86D-715 3.3 6.9 5.1 abedefg 0.56 0.85 0.71 defghi 0.3 2.1 1.2 bedefgh a 0 IT 89KD-288 3.0 7.1 5.1 abedefg 0.64 0.66 0.65 fghij 0.5 0.7 0.6 jklmn (§ 0" :3 "" IT 87D-941-1 3.4 5.8 4.6 bedefg 0.64 0.81 0.73 edefgh 0.4 1.9 1.1 bedefghijkl rl) N" ~ CD Dan Iia 2.5 5.2 3.8 g 0.62 0.73 0.68 defghi 0.02 2.6 1.3 abedef D- o- Mean 3.3 6.1 0.59 0.86 0.3 1.7 0 '< ...., C"') n 0 0 Means followed by the same letter(s) within a column are not significantly different (DMRT) at 5% probability level. ~ ~ ~ -0 a ......... @-(i) n g. :J Differential response of cowpea lines to application of P fertilizer Field trial Seventeen cowpea lines were selected from the pot trial as follows: six lines with high grain yield without P and high productivity with P, IT 86D-7l9, IT 94K-437-l, IT 87D- 941-1, IT 94K-440-3, IT 93K-693-2, and IT 93K-637-l (Category 1); five with low grain yield without P and high productivity with P, Dan Ila, IT 96D-748, IT 90K-284-2, IT 96D-757, and IT 86D-7l5 (Category 2); one line with high grain yield without P and low productivity with P, IT 89KD-288 (Category 3); and five lines not included in the pot trial, IT 96D-739, IT 96D-772, IT 89KD-349, IT 97K-820-l8, and IT 96D-724 (Category 4) were sown in weed-free plots measuring 3 m x 3 m on 25 August 1999. Planting distance was 0.75 m between and 0.20 m within the row with two seeds per hole. There were two factors; 17 cowpea lines and two phosphorus levels, 0 and 30 kg Plha laid out as a factorial in a randomized complete block design (RCBD) with three replicates. The fertilizer (SSP) was band applied along the planting row at planting. The seedlings were later thinned to one per stand at two weeks after planting (WAP). The plots were weeded twice at three and six WAF. Karate insecticide was sprayed twice to control insect attack. The plants were grown to maturity. At maturity, when the pods were dry, they were threshed by hand. The grain was then oven-dried for 24 hours at 65°C and weighed. Data collected were analyzed as in the pot trial. Results Pot trial Shoot, grain, and root dry weights The effect ofP fertilizer on shoot, grain, and root dry weights of cowpea lines is presented in Table 1. Generally, application of phosphorus fertilizer had positive effects on shoot, grain, and root dry weights. Variability among cowpea lines in shoot, grain, and root dry weight response to P was pronounced and more so for shoots than roots. Lines IT 96D- 759 and IT 95K-1095-2 produced the highest shoot weight while lines IT 95K-1090-l, IT 95K-l052-2, and IT 95K-1543 had the highest root dry weights. Interactions between cowpea lines and P levels on root and grain weights were significant (Table 2). Total aboveground dry matter (TOM) TDM consists of the total of shoot and pod (including grains) dry weights. Phosphorus fertilizer had positive effects on TDM and the cowpea lines did not show significant varia- tions in their responses of TDM to P (Table 2). Nodulation Generally, P fertilizer significantly enhanced nodule dry weights of the cowpea lines, but nodule number was depressed by P (Table 3). There were variations among the cowpea lines in the responses of nodulation to P. Interactions between cowpea lines and P levels on nodulation were significant (Table 2). Line IT 94K -437 -1 produced the highest nodule weight. Nutrient accumulation Phosphorus fertilizer had positive effects on Nand P accumulation by the cowpea lines (Table 4) and there was variation in the P accumulation but not N between varieties (Table 2). 323 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 2. Probabilities (P 0.05) of the F test for the analysis of variance for biomass, nodu- lation, and Nand P content variables of cowpea lines. Shoot Root dry dry Nodule Nodule Grain Grain N P Source TDM weight weight number weight (pot) (field) yield yield Cowpea lines (C) 0.3714 0.1137 0.0001 0.0026 0.0001 0.0001 0.0001 0.0239 0.6319 Phosphorus rate (P) 0.0001 0.0001 0.0001 0.0018 0.0001 0.0001 0.2611 0.0001 0.0001 CxP 0.3860 0.2539 0.0001 0.0291 0.0001 0.0092 0.1792 0.1274 0.8098 Table 3. Effect of P application on nodulation of 35 cowpea lines. Nodule number (no.lpot) Nodule weight (mglpot) lines -p +p Mean -p +p Mean IT 90K-277-2 15 20 18 b edefg 6.5 15.7 11.1 jklm IT 90K-284-2 18 12 15 defg 13.1 9.5 11.3 ijklm IT90K-76 18 15 16 edefg 8.7 15.9 12.3 hijklm IT90K-59-2 18 18 18 abedef 7.9 16.8 12.4 hijklm IT 93K-513-2 14 13 14 fg 11.2 25.3 18.3 edef IT 93K-693-2 16 14 15 defg 10.5 15.9 13.3 ghijkl IT 93K-734 12 21 17 bedefg 5.6 12.7 9.2 1m IT 93K-452-1 18 14 16 edefg 16.7 12.6 14.7 efghij IT 93K-637-1 12 14 13 fg 7.2 12.9 10.1 kim IT 94K-437-1 16 12 14 efg 16.4 38.8 27.6 a IT 94K-440-3 17 18 17 bedefg 17.2 18.9 18.1 edef IT 95K-1406 17 22 20 abed 22.5 21.4 21.9 be IT 95K-1156-3 22 16 19 abedef 12.4 18.3 15.4 efghij IT 95 K-1 090-12 19 14 17 bedefg 17.3 13.9 15.6 efghij IT 95 K-1 096-7 25 17 21 abc 16.9 25.3 21.1 bed IT 95K-1464 15 15 15 defg 12.2 24.9 18.6 ede IT 95 K-1 090-1 19 27 23 a 22.6 12.1 17.3 defg IT 95 K-1 05-2 17 13 15 defg 17.1 25.7 21.4 bed IT 95K-1091-3 27 16 22 ab 12.7 15.4 14.0 fghijk IT 95 K-1 095-2 17 18 18 bedefg 12.9 18.9 15.9 efgh IT 95 K-1 088-4 16 11 14 fg 7.2 12.1 9.6 kim IT 95K-1384 20 17 18 abedef 12.3 12.3 12.3 hijklm IT 95K-1543 18 17 18 abedef 11.1 20.4 15.8 efghi IT 95K-1491 16 14 15 defg 12.0 20.3 16.2 efgh IT 96D-666 15 16 16 defg 14.6 11.4 13.0 ghijkl IT 96D-740 18 21 20 abed 22.1 25.7 23.9 ab IT 96D-748 16 11 13 fg 22.2 15.5 18.8 ede IT 96D-757 13 14 14 fg 5.7 14.4 10.1 kim IT 96D-759 19 16 18 abedefg 18.6 13.0 15.8 efghi IT 89KD-374-57 23 15 19 abede 22.6 12.2 17.4 edefg IT 86D-719 19 10 15 defg 17.8 14.7 16.3 efgh IT 86D-715 18 14 16 defg 12.1 18.6 15.4 efghij IT 89KD-288 12 13 13 fg 9.1 7.5 8.3 m IT 87D-941-1 13 15 14 fg 10.3 7.6 9.0 1m Dan lIa 26 16 21 abc 11.8 14.7 13.2 ghijkl Mean 18 16 10.4 10.7 Means followed by the same letter(s) within a column are not significantly different (DMRT) at 5% probability level. 324 Digitized by Google Differential response of cowpea lines to application of P fertilizer Table 4. Effect of P fertilizer on Nand P accumulation (mglpot) of 35 cowpea lines. N p Lines -p +p Mean -p +p Mean IT 90K-277-2 71 72 72 fgh 3.9 4.9 4.4 abed IT 90K-284-2 35 124 80 bedefgh 2.2 5.9 4.0 abed IT90K-76 45 87 66h 2.3 6.5 4.4 abed IT90K-59-2 76 64 70 fgh 2.9 3.4 3.2 d IT 93K-513-2 69 71 70 fgh 3.2 4.2 3.7 cd IT 93K-693-2 53 98 76 edefgh 2.7 5.7 4.2 abed IT 93K-734 78 105 91 bedefgh 3.5 4.7 4.1 abed IT 93K-452-1 47 101 74 efgh 2.2 5.5 3.9 bed IT 93K-637-1 67 115 91 bedefgh 2.8 5.1 4.0 abed IT 94K-437-1 73 120 96 bedefg 3.0 6.3 4.7 abed IT 94K-440-3 85 125 105 abc 3.3 6.6 4.9 abc IT 95K-1406 53 94 73 fgh 2.8 5.5 4.1 abed IT 95K-1156-3 76 132 104 abede 2.9 6.7 4.8 abed IT 95 K-1 090-12 59 128 93 bedefgh 2.1 5.2 3.7 cd IT 95 K-1 096-7 78 121 99 bedefg 3.7 5.5 4.6 abed IT 95K-1464 64 100 82 bedefgh 2.6 4.6 3.6 cd IT 95 K-1 090-1 63 154 109 ab 2.5 6.4 4.4 abed IT 95 K-1 05-2 65 127 96 bedefgh 3.5 5.3 4.4 abed IT 95K-1091-3 67 123 95 bedefgh 2.8 4.8 3.8 cd IT 95 K-1 095-2 66 143 105 abed 3.6 7.6 5.6 ab IT 95 K-1 088-4 65 118 91 bedefgh 2.5 5.8 4.1 abed IT 95K-1384 62 103 82 bedefgh 2.7 5.7 4.2 abed IT 95K-1543 72 104 88 bedefgh 3.2 6.8 5.0 abc IT 95K-1491 65 89 77 edefgh 4.0 5.9 4.9 abc IT 96D-666 62 121 91 bedefgh 2.4 6.3 4.4 abed IT 96D-740 55 118 86 bedefgh 2.7 7.9 5.3 abc IT 96D-748 43 113 78 edefgh 1.8 5.9 3.8 cd IT 96D-757 70 109 90 bedefgh 2.7 4.9 3.8 cd IT 96D-759 101 164 133 a 4.5 7.0 5.7 a IT 89KD-374-57 67 93 80 bedefgh 2.8 4.4 3.6 cd IT 86D-719 77 108 92 bedefgh 2.9 5.1 4.0 bed IT 86D-715 55 139 97 bedefg 2.4 6.1 4.2 abed IT 89KD-288 50 150 100 bedef 2.9 6.6 4.8 abed IT 87D-941-1 75 107 91 bedefgh 2.4 5.3 3.9 cd Dan lIa 48 102 75 defgh 2.5 4.9 3.7 cd Mean 65 113 2.9 5.7 Means followed by the same letter(s) within a column are not significantly different (DMRT) at 5% probability level. Classification of cowpea lines The cowpea lines were classified on the basis of their dry grain weights into four groups. Ten lines were classified as having high grain yield without P and high productivity with P application (Table 5). Field trial In the field trial, more than 50% of the cowpea lines showed significant response to P. Compared with the pot trial, there were considerable variations in the pattern of response of grain yields of the cowpea lines to P (Table 6). Lines IT 87D-941-1 and IT 90K-284-2 325 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 5. Classification of cowpea lines on the basis of response of grain weights to P fertilizer. 1 IT90K-76 IT 93K-513-2 IT 93K-693-2 IT 93K-452-1 IT 93K-637-1 IT 94K-437-1 IT 94K-440-3 IT 95K-1491 IT 86D-719 IT 87D-941-1 2 IT 89KD-288 3 IT 90K-284-2 IT 95K-1543 IT 96D-666 IT 96D-748 IT 96D-757 IT 86D-715 Dan Iia 4 IT 90K-277-2 IT90K-59-2 IT 93K-734 IT 95K-1406 IT 95K-1156-3 IT 95K-1090-12 IT 95K-1096-7 IT 95K-1464 IT 95K-1090-1 IT 95K-105-2 IT 95K-1091-3 IT 95K-1095-2 IT 95K-1088-4 IT 95K-1384 IT 96D-740 IT 96D-759 IT 89KD-374-57 1. High yield without P (yield> 0.3 glpot) and high productivity with P application (yield> 1.7 glpot). 2. Low yield without P (yield < 0.3 glpot) and high productivity with P application. 3. High yield without P and low productivity with P application (yield < 1.7 glpot). 4. Low yield without P and low productivity with P application. Table 6. Effect of P fertilizer on grain yield (kglha) of cowpea lines in the field. lines IT 86D-719 IT 94K-437-1 IT 87D-941-1 IT 94K-440-3 IT 93K-693-2 IT 93K-637-1 Dan lIa IT 96D-748 IT 90K-284-2 IT 96D-757 IT 86D-715 IT 89KD-288 IT 96D-739 IT 96D-772 IT 89KD-349 IT 97K-820-18 IT 96D-724 Mean -p 263 123 450 60 164 344 303 62 523 95 140 284 219 277 179 185 354 237 +p 346 225 704 65 140 226 358 168 405 104 289 182 177 221 309 148 369 261 Mean 305 cde 174 fghij 577 a 63 j 152 ghij 285 cdef 331 cd 115 hij 464 ab 99 ij 215 defghi 233 defgh 198 efghi 249 cdefg 244 cdefg 167 fghij 362 be Means followed by the same letter(s) within a column are not significantly different (DMRT) at 5% probability level. produced significantly highest grain yield. Only lines IT 86D-719, IT 87D-941-1, IT 86D-715, and IT 89KD-288 maintained their classification in conformity with the results obtained in the greenhouse. Some of the lines that performed well in the pot trial exhibited a dismal performance in the field, especially lines IT 94K-440-3 and IT 96D-757. 326 Digitized by Google Differential response of cowpea lines to application of P fertilizer Discussion The clear response to P application observed in terms of shoot, root, grain weights, and nodule dry matter and N and P production of the cowpea lines confirms that P is an important nutrient element affecting the yields of cowpea (Anonymous 1977). There are, however, differential responses among the cowpea lines studied. Okeleye and Okelana (1997) also observed significantly increased nodulation, grain yield, and total dry matter for cowpea varieties in response to P application. The decreased nodule number with P addition observed in this study contradicts the findings ofLuse et al. (1975), that reported increased nodule number in cowpea due to P application. The observed increased cowpea grain yield with P application agrees with the results of Luse et al. (1975) but contradicts the results obtained by Agboola and Obigbesan (1977), who observed that P application did not significantly increase cowpea yield but rather enhanced nodulation and P content of leaf and stem. Osiname (1978) also did not observe a significant effect on cowpea yield with P application at Ibadan. The observed differential performances of the cowpea lines under no P application could provide a basis for selecting lines with greater agronomic efficiency in P-deficient soils and so reduce fertilizer costs. The observed variations in the performance of some of the cowpea lines in the pot and field trials is a pointer to the fact that pot trial screening methodology (which does not represent the real-life situation) may not be a very good methodology for evaluating varieties for farmer release. However, it could be used for an initial assessment of large numbers of breeder lines. Sanginga et al. (2000), reported that about 42% of the cowpea breeding lines (18 out of 43 lines tested) screened for P-use efficiency and N balance had the same grouping for the field and pot experiments. Watanabe et al. (1997), observed a high correlation coefficient (0.666**) of scores between field evaluation and pot evaluation of drought tolerance of cowpea in Nigeria. However, they stated that the highly significant correlation observed between scores evaluated by the two methods was beyond expectation and so suggested further testing of the methodologies. Variability noted in response to P could be important for selecting lines suitable for a range of soil P conditions or farmer production systems. We recommend that lines IT 90K-284-2, IT 96D-724, and IT 93K-637-l can be selected for further testing without Pfertilizer. Lines IT 87D-94l-l, IT 86D-7l9, and Dan Ilamay perform very well without P fertilizer and give a higher return when P is applied. When P fertilizer is available, line IT 87D-94l-l is recommended. These varieties should be tested at multiple sites to truly extend the results to breeding cowpea lines that could be targeted towards various soil P conditions. References Adetunji, M.T. 1995. Equilibrium phosphate concentration as an estimate of phosphate needs of maize in some tropical Alfisols. Tropical Agriculture 72: 285-289. Agboola, A.A. and G.O. Obigbesan. 1977. Effect of different sources and levels of P on the performance and P uptake of Ife-Brown variety of cowpea. Ghana Journal of Agricultural Science 10 (1): 71-75. Ankomah, A.B., F. Zapata, G. Hardarson, and S.K.O. Danso. 1995. Yield, nodulation, and N2 fixation by cowpea cultivars at different phosphorus levels. Biology and Fertility of Soils 22: 10-15. Anonymous. 1977. Notes on the cowpea and grain legume research program, cropping scheme meeting. Institute for Agricultural Research, Ahmadu Bello University, Zaria, Nigeria. 327 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Jain, VK., YS. Chauhan, and P.C. Jain. 1986. Effect of different doses of phosphorus on growth, yield, and quality of cowpea (Vigna unguiculata [L.] Walp.). MadrasAgricultural Journal 73 (4): 199-202. Kang, B.T. and D. Nangju. 1983. Phosphorus response of cowpea (Vigna unguiculata [L.] Walp.). Tropical Grain Legume Bulletin 27: 11-16. Luse, RL., B.T. Kang, RL. Fox, and D. Nangju. 1975. Protein quality in grain legumes grown in the lowland humid tropics, with special reference to West Africa. Pages 193-201 in Fertilizer use and protein production. XIth Colloquium, International Potash Institute, 1975. Ronne-Born- holm, Denmark. Mokwunye,A.U., S.H. Chien, and E. Rhodes. 1986. Phosphorus reaction with tropical African soils. Pages 253-281 in Management of nitrogen and phosphorus fertilizers in sub-Saharan Africa, edited by A.U. Mokwunye and P.L.G. Vlek. Martinus NijhoffPublishers, Dordrecht, The Neth- erlands. Muleba, N. and H.C. Ezumah. 1985. Optimizing cultural practices for cowpea in Africa. Pages 289-295 in Cowpea research, production, and utilization, edited by S.R Singh and K.O. Rachie. John Wiley and Sons Ltd, Chichester, UK. Okalebo, J.R., K.w. Gathua, and P.L. Woomer. 1993. Laboratory method of soil and plant analysis: a working manual. Tropical Soil Biology and Fertility Programme (TSBF), Nairobi, Kenya. 88p. Okeleye, KA. and M.A.O. Okelana. 1997. Effect of phosphorus fertilizer on nodulation, growth, and yield of cowpea (Vigna unguiculata) varieties. Indian Journal of Agricultural Science 67(1): 10-12. Omueti, 10. and VA. Oyenuga. 1970. Effect of phosphorus fertilizer on the protein and essential components ofthe ash of groundnut and cowpeas. West African Biology and Applied Chemistry Journal 13 (1): 299-305. Osiname,O.A. 1978. The fertilizer (NPK) requirement ofIfe-Brown cowpea (Vigna unguiculata [L.] Walp.). Tropical Grain Legume Bulletin No. 11/12: 13-15. Rachie, K.O. and L.M. Roberts. 1974. Grain legumes ofthe lowland tropics. Advances inAgronomy 26: 44-61. Sample, E.C., RJ. Soper, and G.l Racz. 1980. Reactions of phosphate fertilizers in soils. Pages 263-310 in The role of phosphorus in agriculture, edited by F.E. Khasawneh, E.C. Sample, and E.J. Kamprath. American Society of Agronomy, Madison, Wisconsin, USA. Sanginga, N., O. Lyasse, and B.B. Singh. 2000. Phosphorus use efficiency and nitrogen balance of cowpea breeding lines in a low P soil ofthe derived savanna zone in West Africa. Plant and Soil 220: 119-128. SAS. 1985. SAS user's guide. Statistical Analysis System Institute, Cary, NC, USA. Sellschop, IP.F. 1962. Cowpeas, Vigna unguiculata (L.) Walp. Field Crops Abstracts 15: 259- 266. Tenebe, VA., Y Yusuf, B.K. Kaigama, and 1.0.E. Aseime. 1995. The effects of sources and levels of phosphorus on the growth and yield of cowpea (Vigna unguiculata [L.] Walp.) varieties. Tropical Science 35: 223-228. Watanabe, I., S. Hakoyama, T. Terao, and B.B. Singh. 1997. Evaluation methods for drought toler- ance of cowpea. Pages 141-146 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropi- cal Agriculture (IlTA) and Japan Research Center for Agricultural Sciences (JIRCAS). IlTA, Ibadan, Nigeria. 328 Digitized by Google 4.8 Farmer participatory evaluation of newly developed components of cowpea and cotton intercropping technology F.A. Myakal, J.C.B. Kabissa2, D.F. Myakal, and J.K. Mligol Abstract A technology verification experiment was carried out in farmers' fields in eastern Tanzania in 1997 and 1998. An erect short-duration cowpea variety Vuli-l was intercropped with the cotton variety IL 74. Cotton was planted in single rows alternating with either single or double rows of cowpea. In an alternating single row intercrop, cowpea was planted either two weeks or four weeks after cotton. In the intercrop where a single cotton row alternated with double rows of cowpea, cowpea was planted two weeks after cotton. After harvesting, farmers were asked to assess and rank the technology components using an open-ended questionnaire and pair-wise ranking. Statistical analysis showed cotton and cowpea yield differ- ences between technology components. Farmers' assessment revealed variation in terms of technology component preferences and showed that farmers rejected the one: two cotton: cowpea row configuration. Famers accepted the one: one row configuration, and cowpea planted two or four weeks after cotton. However, it was evident that the adoption of these acceptable technology components will depend on whether certain cotton production constraints are solved. Introduction Cowpea (Vigna unguiculata [L.] Walp.) is an important grain legume in Tanzania where its tolerance to moisture stress makes it suitable for cultivation in semiarid areas. Its leaves and seeds are consumed as an important supplement to a staple cereal diet. In Tanzania, cowpea is grown in almost all the areas below 1500 m above sea level (price et al. 1982). It is usually found intercropped with cereals or other crops, although it is sometimes grown as a monocrop. However, its productivity is limited by high infestation with insect pests so that spraying against such pests is important for good yield. Cotton (Gossypium hirsutum L.) is an important cash crop for smallholder farmers in eastern and western Tanzania. It is currently rated third after cashew and coffee in terms offoreign exchange earnings. Like cowpea, insect pests limit its productivity. Thus, insec- ticide application is recommended for optimal yield. However, profit margins for cotton have recently been reduced as a result of the rising cost of insecticides. Consequently, some farmers opt not to apply insecticide, thereby reducing cotton yield and quality. Therefore, for both cotton and cowpea, technologies are needed to increase returns in order to make production more attractive. l. llongaAgricultural Research Institute, Private Bag, K.ilosa, Tanzania. 2. Tanzania Cotton Lint and Seed Board, PO Box 9161, Dar es Salaam, Tanzania. 329 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production One way of optimizing profit margins would be to intercrop cotton and cowpea so that the cowpea benefits from insecticide sprays applied on cotton thus reducing production cost (Natarajan and Naick 1992; Myaka and Kabissa 1993; 1996). Willey (1979) outlined other advantages of intercropping including greater yield stability over different seasons and better use of growth resources. Previous work on cotton and cowpea intercropping by Myaka and Kabissa (1996) showed that optimal yield depends on moisture regime, time of planting cowpea, and planting pattern. On this basis, it could be anticipated that the most appropriate time for planting cowpea should therefore be different in wet and dry areas. Furthermore, their results showed the possibility that cowpea yield could be raised by increasing plant den- sity, especially in dry areas. However, these results were not tested on farmers' fields to verify such technology components. A technology verification experiment was therefore initiated in 1997 on farmers' fields in three districts in eastern Tanzania with the follow- ing objectives to: verify the on-station results on cotton and cowpea intercropping by Myaka and Kabissa (1996) on farmers' fields with the input of farmers. create farmers' awareness on the possibility of intercropping cotton and cowpea and on the use of the electrodyne sprayer as safe for the user and the environment. have farmers assess the technology to confirm its compatibility with the farming system. recommend acceptable cotton and cowpea technology components for wider adop- tion. Materials and methods The experiment was conducted during 1997 and 1998 cropping seasons on farmers' fields in Mangae and Fulwe, Morogoro rural district; Magamba and Mzundu, Handeni district; and Kisiwani of same district. These locations were selected on the basis of prevailing moisture regimes and history of cotton cultivation; Mangae and Kisiwani were classified as dry while Fulwe, Mangae, and Mzundu were classified as wet locations. Prior to field experimentation, baseline data were collected in the target areas in an informal survey involving participatory methods (Rhoades 1995). Data collected included the production system involving the two crops, whether farmers were practising cotton and cowpea intercropping, farmers' knowledge of insecticide applicators, and cotton/cowpea production constraints. Fields where the trials were conducted lie between 500 and 750 m above sea level. Fulwe and Mangae have a monomodal to weak bimodal rainfall pattern. Fulwe experiences higher rainfall than Mangae. At these locations, rains usually fall between OctoberlNovember and May with a dry spell from January to February. The trials at Fulwe and Mangae were conducted during the main rains. Mzundu and Magamba experience a bimodal rainfall pattern. The short rainy season is between October and December and the long rainy season from mid-February to June. Trials at Mzundu and Magamba were conducted during the long rainy season. Kisiwani experiences a monomodal rainfall pattern with rains falling between February and May. The trial at this location was conducted during this period. In collaboration with staff from the extension service, farmers were selected as follows: Mangae, five farmers, Fulwe, two, Mzundu and Magamba, six, and Kisiwani eight. Criteria for selecting farmers were their willingness to participate and their accessibility to land. 330 Digitized by Google Farmer participatory evaluation of cowpea and cotton intercropping technology In the 1997 season, the trial was laid out in a block of six plots of 10 mx 10m replicated twice in each farmer's field. Treatments were as follows: 1. Cotton and cowpea intercropped in alternate single rows and cowpea planted two weeks after cotton. 2. Cotton and cowpea intercropped in alternate single rows and cowpea planted four weeks after cotton. 3. Cotton and cowpea intercropped in one: two cotton to cowpea row ratio and cowpea planted two weeks after cotton. 4. Sole cropped cotton. 5. Sole cropped cowpea planted two weeks after cotton. 6. Sole cropped cowpea planted four weeks after cotton. In the 1998 season, the trial was laid out in a single block (no replications) in response to farmers' observation that the replicated experiment in 1997 (two replications per farmer) was too complicated for them. Cotton was sown in hills spaced 0.3 m apart within the row and thinned to one plant per hill three weeks after sowing, while the cowpea was sown in hills spaced 0.2 m apart with two plants per hill. Spacing between rows for each component crop was 0.9 m. This gave a target population of 37 000 plants/ha for cotton and 110000 plants/ha for cowpea. In one: two row configuration, the target density for cowpea was 200 000 plants/ha. Sole cotton was planted at the same density as in the intercrop. Sole cowpea was planted two or four weeks after cotton with a space of 0.5 m between rows and 0.2 m between plants within the row, and the plants were subsequently thinned to two plants per hill with the aim of achieving a population of 200 000 plants/ha. Cotton variety IL 74 and cowpea variety Vuli -1 were used. IL 74 is an indeterminate, late-maturing ( 180 days) cotton cultivar. Vuli -1 is a determinate, erect, and early -maturing cowpea cultivar. The trial was farmer-managed. Table 1 shows the allocation of res pons i- bilities for the main operations and management of nonexperimental variables. After harvesting, yield data were recorded and subjected to analysis of variance using MSTATC statistical software package. In 1998, farmers were treated as replications. During both seasons, the analysis was done on a village basis. Farmers' assessment was done through individual farmer interviews and in groups using an open-ended questionnaire, and farmers used pair-wise ranking to rank the technology components. Table 1. Allocation of responsibilities for the main operations and management of nonexperimental variables of the field experiment. Field operations land preparation layout of experiment Planting Thinning Weeding Insecticide application Harvesting "Did the main job; VEO = village extension officer. 331 Implementers Farmers ResearcherNEO * Jfarm er YEO and farmer* YEO and farmer* YEO and farmer* ResearcherNEOlfarmer* YEO and farmer* Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Results and discussion In 1997, rainfall was assessed as normal and within expectations in amount and distri- bution pattern, while in 1998, the rains started early in all locations (Table 2). In 1997, there was a clear difference between dry and wet locations in terms of rainfall amount and distribution. This conformed to the location classification, contrary to 1998 when all locations received similar rainfall in terms of amount and distribution (Table 2) except Magamba where that year's rainfall was not recorded. In 1998, rainfall was abnormally higher compared to long term averages. This abnormally high rainfall was probably due to the El nino phenomenon. Therefore, the hypothesis that a suitable cotton and cowpea intercropping pattern would depend on the moisture regime could not be tested during this season. Several seasons of evaluation will be needed to further investigate this. The objective of the baseline data collection was to have information on farmers' knowledge on the technology and also to have some basis for future impact assessment. The baseline data collected revealed that, with the exception of Mangae, farmers were not aware of cotton and cowpea intercropping. Furthermore, it was apparent that prior to the present study, farmers had no knowledge of the electrodyne sprayer. Cotton production constraints mentioned and prioritized by farmers are listed in Table 3. Both men and women do most of the operations involving these two crops. However, only men sell the cotton. The same applies to cowpea but only when the harvest is large. When the cowpea harvest is small, the work is left for the women. For both crops, only men do the spraying. This shows that there is some gender balance in the execution of most of the field operations. However, only men control the income realized from these crops. Sole-cropped cowpea yielded higher than intercropped cowpea during both years and at all locations except at Fulwe in 1998 (Tables 4 and 5). This indicates that intercropping affected cowpea yield. These results are in agreement with the on-station results reported Table 2. Rainfall totals (mm) and number of rainy days (in parentheses) over successive monthly periods at the experimental sites in eastern Tanzania in 1997. Year Month Magamba Mzundu Fulwe Mangae Kisiwani 1997 January 0.0 (0) 0.0(0) 0.0(0) 0.0(0) February 0.0 (0) 8.0 (1) 30.0 (1) 40.5 (2) March 72.0 (6) 49.5 (4) 76.5 (5) 247.5 (6) April 144.1 (15) 183.2 (14) 237.8 (12) 52.9 (6) May 118.6 (10) 100.1 (8) 44.4 (4) 10 (4) June 21.4 (5) 99.2 (13) 0.0(0) 2 (1) July 0.0 (0) 0.0(0) 0.0(0) 0.0(0) August 0.0 (0) 0.0(0) 0.0(0) 0.0(0) 1998 January 196.5 (8) 208.4 (5) 264.8 (8) 419.1 (21) February 38.0 (3) 177.5 (5) 107.1 (3) 135.3 (10) March 0.0(0) 195.9 (6) 170.1 (6) 90.0 (8) April 171.0 (9) 225.0 (7) 146.0 (11) 169.7 (12) May 93.5 (6) 30.1 (1) 40.0 (5) 49.3 (5) June 0.0(0) 0.0(0) 0.0(0) 0.0(0) July 0.0(0) 0.0(0) 0.0(0) 0.0(0) August 5.5 (2) 0.0(0) 0.0(0) 0.0(0) 332 Digitized by Google Farmer participatory evaluation of cowpea and cotton intercropping technology Table 3. Cotton and cowpea production constraints in order of priority as prioritized by farmers using pair-wise ranking in eastern Tanzania in 1998. location Rank Production constraints Handeni (A) 1 2 3 4 5 6 7 Handeni (B) 1 2 3 4 5 6 Morogoro (wet) 1 2 3 4 5 6 Morogoro (dry) 1 2 3 4 5 6 Same 1 2 3 4 lack of market for cotton Cotton seed not available on time; when available has poor germination Insecticides not available on time; when available are expensive and ineffective Cotton buyers come late lack of harvesting and storage bags· Hunger stress during weeding time (people spend most of their time looking for food) Brown-seeded cowpea have no market lack of market for cotton Cotton seed not available on time Insecticides not available on time; when available are expensive and ineffective low cotton price lack of market for brown cowpea lack of harvesting and storage bags· lack of market for cotton low cotton price lack of credit facility Insecticides are brought too late and they are expensive Tractors not enough, resulting in late land preparation Vermin Cotton seed not available on time Insecticides not available on time; when available are expensive Batteries for sprayers are unaffordable Cotton buyers come late Sprayers are not enough Tractors for hire are not available Insecticides not available on time; when available are expensive and ineffective lack of market for cotton Aphids on both cotton and cowpea come early before the anticipated date of fi rst spray lack of harvesting and storage bags· "The cooperatives used to supply bags but they have stopped. by Myaka and Kabissa (1996). Cowpea intercropped with cotton in one: two row ratio yielded higher than that which was intercropped in one : one ratio with cotton. In 1997, higher cotton yields were observed from wet locations. In 1998, cotton was not affected by intercropping at all locations except Kisiwani. Technology ranking was variable between locations. When ranks were pooled across moisture regime classifications, intercropping was scored high in wet locations while in dry locations, farmers preferred cowpea monocropping (Table 6). It is interesting to note that cotton monocropping was ranked low at all locations. Various comments on technology components from farmers are indicated in Table 7. When asked if they would continue intercropping these crops, farmers at all locations agreed to continue except at Mangae 333 Digitized by Google Table 4. Yield (kglha) of cotton and cowpea planted at four locations in eastern Tanzania in 1997. bl ~ Mangae Kisiwani Mzundu Fulwe t't) II> Treatment Cotton Cowpea Cotton Cowpea Cotton Cowpea* Cotton Cowpea* 8 ::J 1 : 1 intercrop 2 weeks 1153 338 663 375 1517 2000 g: 1 : 1 intercrop 4 weeks 1017 283 850 267 1433 1700 c:: g. 1 : 2 intercrop 2 weeks 1297 315 713 475 1455 1900 ::J <.n Sale cotton 1097 1250 1300 1200 6" Sale cowpea 2 weeks 630 467 Qj> Sale cowpea 4 weeks 480 900 ~ CV 33 31 37 67 17 11 S· O'Q SE 129 34 113 81 62 <.n Ci p 0.63 0.007 0.007 0.03 0.49 0.08 fir :3 ·Cowpea yield during this season wasnot recorded. <.n ;;;- W O'Q w a .j::o. ::J 0 Table 5. Yield (kglha) of cotton and cowpea planted at four locations in eastern Tanzania in 1998. :3 ;::;. Mangae Kisiwani Mzundu Fulwe ~. a 0 Treatment Cotton Cowpea Cotton Cowpea Cotton Cowpea Cotton Cowpea (§ 0" :3 "" 1 : 1 intercrop 2 weeks 407 80 750 208 733 450 550 N" ~ CD 1 : 1 intercrop 4 weeks 347 90 863 291 567 500 613 D- o- 1 : 2 intercrop 2 weeks 360 225 500 236 633 450 700 0 '< ...... C"') n Sale cotton 420 1200 633 325 0 0 Sale cowpea 2 weeks 410 597 825 ~ t't) ~ Sale cowpea 4 weeks 310 500 450 II> "tl ......... CV 10 33 30 35 15 31 23 a (i) SE 13 46 92 58 160 47 14 2- P 0.19 0.03 0.01 0.02 0.29 0.64 0.28 n g. ::J Farmer participatory evaluation of cowpea and cotton intercropping technology Table 6. Pair-wise ranking results of intercropping components and sole crops as ranked by farmers in eastern Tanzania in 1998. Wet Dry Treatment Mzundu Magamba Fulwe Mangae Kisiwani 1 : 1 intercrop 2 weeks 1* 2* 2* 4* 4* 1 : 1 intercrop 4 weeks 3 5 4 1 3 1 : 2 intercrop 2 weeks 4 1 5 5 2 Sole cotton 5 4 6 3 5 Sole cowpea 2 weeks 2 3 1 2 1 Sole cowpea 4 weeks 3 "Technology component rank within a location. Scale: 1 = very important, 6 = less important. Table 7. Summary of general comments given by farmers during the ranking exercise of technology components in eastern Tanzania in 1998. Location Magamba Mzundu Fulwe Mangae Same Comments on technology components In one: one row intercrop two weeks, there is no competition When cowpea is delayed to four weeks, cotton affects the cowpea; one: two row intercrop not preferred because it is difficult to weed and spray as rows are very close Cowpea planted two weeks after cotton becomes vegetative When cowpea is delayed to four weeks, it is overshadowed by cotton In one: two intercrop, cowpea yielded higher than in the other intercrops but is not preferred because of difficulty in weeding and spraying All farmers preferred intercropping to monocropping because when they intercrop, they get both crops When cowpea is planted two weeks after cotton, there is good synchro- nization of weeding for both crops; one: two row intercrop not preferred because it is difficult to weed Preferred intercrop to cotton sole-crop Preferred sole cowpea to intercrop and sole cotton; there is no market for cotton Preferred to plant cowpea four weeks after cotton because both crops reach spraying time at the same time one: two row intercrop not preferred because it is difficult to weed and there are too many insects Preferred sole cowpea because it is marketable Preferred intercrop to sole crop where they preferred to grow both crops as sole crops because they had observed that intercropping reduced the yield of both crops (Table 8). Farmers cautioned that they would intercrop provided that there would be a cotton market and that insecticides would be provided on credit. Uncertain cotton markets and low cotton prices were priority constraints at all locations (Table 3). Farmers preferred alternate single rows and rejected the one : two cotton and cowpea row configuration (Table 6). This was in conformity with results reported by Myaka and Kabissa (1996). The latter reported 47% cotton yield reduction when intercropped with cowpea in one: two cotton and cowpea row configuration when compared to sole cropped cotton. In the present study, farmers rejected this pattern because it was difficult to walk through the rows during 335 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 8. Farmers' perception on whether to continue with cotton and cowpea intercropping or not in eastern Tanzania in 1998. Situation Magamba ("!o) To continue 33 Not to continue 67* Number of farmers (n) 3 Reason why no Yield of both Mzundu ("!o) 100 0* 9 crops reduced Reason why yes You get two crops Less production cost - Fulwe Mangae ("!o) ("!o) 100 0 0** 100 2 12 *Farmers are ready to divide the field into half; one half for cotton, the other for cowpea. Kisiwani ("!o) 95** 5 20 **Farmers will continue on condition that market for cotton is assured and insecticide is obtained on credit. spraying. They also complained that this configuration hampered weeding. The cowpea variety Vuli -1, which was used in the present study is brown-seeded. There is an indication that this type of seed does not fetch a good market price (Table 3). Research is needed to develop a short-duration cowpea suitable for intercropping with cotton like Vuli-l but with acceptable cream color seed. We conclude that the one : one row configuration and cowpea planted two or four weeks after cotton are acceptable to farmers and compatible with the existing farming system. Although farmers' preferences for these technology components were variable, their adoption will depend on the removal of production constraints for cotton which is the main crop in this intercropping system. The use of cotton insecticide for cotton: cowpea intercrop does, however, need to be developed with caution. Inappropriate chemicals or timing of application could result in excessive contamination of cowpea food products. Acknowledgements The authors thank the Tanzania National Agricultural Research Fund for funding this research. The Ireland Aid through EZCORE project made it possible to present this paper during the 3rd World Cowpea Research Conference. We also thank the Zonal Director for Agricultural Research and Development, Eastern Zone, staff of the grain legume research subprogram and the Agricultural Extension Departments in Morogoro rural, Handeni, and Same districts for their assistance in undertaking this research. The interest and enthusi- asm of participating farmers inspired the team. Thanks to the Director of Research and Development who gave us permission to publish this work. References Myaka, FA. and J.C.B. Kabissa. 1993. Cotton and cowpea relay intercropping: preliminary results on its economics and effects on some agronomic characters of both crops. Pages 60-67 in Trends in cowpea research. Proceedings of Cowpea Research Seminar, 25-26 September 1991, Harare, Zimbabwe. Myaka, F.A. and lC.B. Kabissa. 1996. Fitting short duration cowpea into a cotton-based cropping system in Tanzania: Effect of planting pattern, time of planting cowpea and insecticide applica- tion to cotton. Experimental Agriculture 32: 225-230. 336 Digitized by Google Farmer participatory evaluation of cowpea and cotton intercropping technology Natarajan, M. and D.M. Naick. 1992. Competitive effect of short duration bush type cowpea when intercropped with cotton in Zimbabwe. Experimental Agriculture 28: 409-416. Price, M., F. Machange, and JA. Assenga. 1982. Improved cultivation of cowpea (Vigna Wlguicu- lata) in Tanzania. Tanzania Ministry of Agriculture, Dar es Salam, Tanzania. 42p. Rhoades, R.E. 1995. The art ofthe informal agricultural survey. I1TA Research Guide 36. Interna- tional Institute of Tropical Agriculture, Ibadan, Nigeria. 57p. Willey, R.w. 1979. Intercropping: its importance and research needs. Part 1: Competition and yield advantages. Field Crop Abstracts 32: 1-10. 337 Digitized by Google 4.9 Cowpea dissemination in West Africa using a collaborative technology transfer model J.O. 0lufowote 1 and P.W. Barnes-McConneW Abstract Improved cowpea culti vars from the Bean-Cowpea Collaborative Research Support Program (CRSP) and from other sources were introduced into the cereal cropping system in Chad, Ghana, Mali, Niger, and Senegal to ameliorate the declining soil fertility in the Sudan-Sahelian zone of West Africa and contribute to food security. The project was implemented through the formation of country technology trans- fer teams involving CRSP, World Vision International (WVI), the national agricul- tural research systems (NARS), the national agricultural extension services, other nongovernmental organizations (NGOs), and leader farmers. Interventions involved the dissemination of appropriate improved cowpea cultivars grown in association or in rotation with cereal cultivars suitable for each environment. To minimize postharvest losses usually associated with cowpea in the region, appropriate stor- age technologies were introduced through the training oftechnicians and farmers. Indications are that those interventions have the potential to contribute to improve- ments in soil fertility and increased food security in the subregion. Introduction Soils of most of West Africa are characterized by relatively low inherent fertility. This is due to the type of their parent material, high degree of weathering, lack of volcanic rejuvenation, and intensive leaching. In Africa, 65% of the agricultural land, 31 % of the permanent pasture land, and 19% of the forest and woodland are affected by human-induced soil degradation. It is estimated that about 332 million hectares of African drylands are subject to soil degradation, with nutrient depletion being a major factor influencing this degradation (Bationo and Lompo 1996). Farmers traditionally practice shifting cultivation on these low fertility soils. In this agricultural practice, cropped lands are left fallow to restore fertility after they have been cultivated to a point where crop yields had declined to uneconomic levels. Increasing population pressure in the subregion has continued to reduce the fallow period, thereby limiting the effectiveness of shifting cultivation in restoring soil fertility. Compounding the problem is the fact that inorganic fertilizer is now beyond the reach of many small- holder farmers, either due to nonavailability or high prices. The Collaborative Research Support Projects (CRSPs) working in West Africa have as their main thrust, technology development. For several years, these bilateral research 1. Food Security Program, Africa Region, World Vision International, No.3, Kotei Robertson Street, North Industrial Area, PO Box 1490, Kaneshie, Accra, Ghana. 2. Bean-Cowpea Collaborative Research Support Program, 200 International Center, Michigan State University, East Lansing, MI 48824-1035, USA. 338 Digitized by Google Cowpea dissemination in West Africa teams have developed natural resource management (NRM) technologies for specific environments in individual countries. Many of these technologies have the potential of being adapted for use throughout West Africa. However, many are not yet widely adopted. While CRSPs' capacity for transfer of these technologies is limited, national and regional efforts are constrained by inadequate collaboration and linkages. Project objectives and outputs The NRM InterCRSP initiative in West Africa has a specific charge to transfer NRM tech- nologies in West Africa (Anonymous 1997). The major objectives of the project are: To develop a model for CRSPINGO collaboration that will mobilize the existing knowledge, technologies, and capacity of CRSPs for major regional impact. To use this intervention model to improve natural resource management, reduce natural resource degradation, and improve farmers' food security and incomes in West Africa through regional adaptation and transfer of sustainable NRM technologies. The expected outputs are: A model regional mechanism for collaborative adaptive research and transfer of CRSP NRM technologies in West Africa. Strengthened and mutually reinforced West African institutions and professional resources for NRM technology adaptation and transfer. The successful functioning of the model mechanism, leading to more productive exchanges among CRSPs, the national agricultural research system (NARS), the international agricultural research centers (lARCs), the national agricultural exten- sion system (NAES), nongovernmental organizations (NGOs), and farmers in West Africa. Improved regional technology adaptation and transfer, leading to more sustainable yields and greater profitability for farmers. Collaborating institutions The initial CRSPs are the Bean-Cowpea CRSP, which is the lead CRSP and the Sor- ghum-Millet CRSP (lNTSORMIL). Scientists from both the USA and host countries collaborate in the project. World Vision International (WVl) maintains programs in eight countries in West Africa, with a comparative competency in West African regional technology transfer. WVl maintains healthy collaborative relationships with the NARS and NAES in the countries in which it works. Collaborative adaptive research and technology transfer teams, comprising CRSP, NARS, NAES, WVI, other NGOs, and farmer collaborators, were formed for each par- ticipating country. Each team, with a coordinator, prepared and implemented a work plan, setting targets for the adaptation and transfer of selected technologies. The CRSP and NARS team members implement adaptive research activities, while WVl, NAES, other NGOs, and farmer team members undertake transfer activities. The Bean-Cowpea CRSP and WVl facilitate the exchange ofNRM technologies among team members and between country teams. They complement internal and external link- ages with additional regional collaborative relationships with the following networks in the region: the West and Central Africa Sorghum Research Network (WCASRN), the West and 339 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Central Africa Millet Research Network (ROCAS), the West and Central Africa Cowpea Research Network (RENACO), and the cowpea protection network at I1TA, Protection ecologiquement durable du niebe (PEDUNE). Current activities The project document stated that initially the NRM technologies adapted and extended would be confined to genetic resources (i.e., cowpea, sorghum, and millet varieties and storage technologies) developed by the Bean-Cowpea and INTSORMIL CRSPs. These activities are summarized in this paper. However, it should be noted that the mechanisms described, especially the interactions between multiple partners, have provided a means for subsequent transfer of additional technologies. Technologies currently being promoted are mainly in the areas of dissemination of improved genetic materials of cowpea and sorghum, and millet and cowpea storage tech- nologies. Cowpea is an important grain legume in West Africa, providing an inexpensive source of protein for both the urban and rural poor. Incorporation of cowpea into the cropping system is crucial for sustainable crop pro- duction in sub-Saharan Africa. Two major cereals grown in the target areas of the project are sorghum and millet. It is hoped that incorporating cowpea in the cropping system, either as a sole crop or intercrop with sorghum and millet will go a long way to improve the fertility of those degraded soils and hence, contribute to NRM. Cowpea improves the soil through the fixation of atmospheric nitrogen. Where soil degradation is a major constraint to crop production, inclusion of cowpea into the crop- ping system is crucial as it helps to replenish soil nitrogen. Cowpea rotation is an effec- tive resource management technology in cereal-based systems, since part of the nitrogen requirement of cereal crops can be met by cowpea intercropping and/or rotation. Studies on cereal-cowpea rotation (Bationo et al. 2000) show that grain yields of cereals suc- ceeding cowpea can, in some cases, double compared to continuous monoculture. The authors claimed that in an efficient soil fertility management system, cowpea can fix up to 88 kg N/ha and this results in an increase of nitrogen-use efficiency on the succeeding cereal crop from 20% in the continuous cereal monoculture to 28% when cereals are in rotation with cowpea. The authors also found that the use of soil nitrogen increased from 39 kg N/ha in the continuous cereal monoculture to 62 kg N/ha in the rotation systems. Similarly, cowpea, when intercropped with cereals, helps reduce the menace of Striga hermonthica, a major problem confronting smallholder farmers in the region. The inclusion of cowpea in the cropping system will improve the nutrition of the people, increase the feed quantity and quality for livestock, and contribute to soil fertil- ity maintenance. This will lead to increased food security and reduced environmental degradation. The project also emphasizes the dissemination of cowpea storage technologies. A maj or deterrent to cowpea production is the problem of insect pests, which occur during post- flowering, preharvest, and in storage. Storage of cowpea seed is particularly problematic, due to high damage by storage insect pests. Indeed, this is often the major reason adduced by many smallholder farmers for not growing cowpea. Hence, along with the introduction of improved varieties, the project emphasizes disseminating cowpea storage technologies. These technologies were developed by the CRSP projects in Cameroon and Senegal, in 340 Digitized by Google Cowpea dissemination in West Africa collaboration with the CRSP cowpea storage project at Purdue University, USA. (Kitch and Ntoukam 1991; Kitch et al. 1992; Ntoukam and Kitch 1991; Kitch et al. 1997). Activities are currently in progress in Chad, Ghana, Mali, Niger, and Senegal. These participating countries cut across the Sudan-Sahelian zone where annual rainfall is between 200 and 1200 mm. Strategies and progress 1. Collaboration. The project has created a five-country network of over 50 collabo- rators from more than 15 different organizations, encouraging both national and international collaboration. Within each country, scientists and the entire country team members from different disciplines are working together to identify/develop, test, and disseminate technologies best suited to local conditions. There have been improved working relationships between WVI, NARS, NAES, and farmers. Apart from country team meetings to select appropriate technologies for testing, the team as a whole monitors progress on the field, participates in field days, and jointly ana- lyzes and interprets the data collected. This type of interaction between scientists, extension agents, NGOs, farmers, and processors is a novelty in the subregion. Internationally, technologies are being shared throughout West Africa by NARS, CRSPs, IARCs, and the commodity networks. 2. Mutual learning. Exchange of expertise has been encouraged and facilitated by the project. Because farmers are directly involved in the project, scientists have learned a lot from the farmers. Among indigenous knowledge gained from farmers are the uses of shea butter (from Vitellaria paradoxa more commonly refered to as Butyro- spermum parkii) in Ghana and the powder from the leaves of wild custard apple (Anona senegalensis) in Niger for local cowpea seed preservation. Even though the use of oils and botanicals has been documented by several authors (e.g., Murdock et al. 1997), the Ghana experience showed that a uniform layer of shea butter at the surface of the earthenware pot could be effective. The effective use of the leaf powder from the custard apple is worth further studies and refinement by scientists. 3. Dissemination of technologies. Several technologies have been disseminated by the project (Anonymous 1998, 1999,2000). Some of these technologies are described in this paper. Storage technologies Technicians and farmers from the five participating countries were trained on cowpea storage technologies (solar heater, triple bagging, drum storage, and improved ash stor- age) developed by CRSP scientists (Kitch and Ntoukam 1991; Kitch et al. 1992; Ntoukam and Kitch 1991; Kitch et al. 1997). These technologies consist of (1) using a solar heater to kill bruchids, (2) triple bagging in three layers of hermetically sealed plastic, and (3) mixing cowpea with wood ash for storage. Generally, it is recommended to use a solar heater to kill bruchids and then store by triple bagging or in wood ash. Even when used separately, each of the three techniques allows the storage of cowpea for six months or longer without bruchid infestation or damage. The farmer, therefore, has an option to keep the cowpea and seek higher prices long after harvest. In Senegal, a variation of these techniques combined the use of neem oil with either triple bagging or storage in sealed drums (Murdock et al. 1997). 341 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production The nwnber of people trained in storage technologies between 1997 and 2000 per country are: Chad (2038), Ghana (578), Mali (65), Niger (86), Senegal (290), making a total of 3057. These training activities targeted both men and women, with participants from several communities. Similarly, farmers were supplied with storage materials with which to train more farmers in their communities. Table 1 shows the number of participants in the stor- age training activity in Chad in 1998. WVI also sent two staff members to Institut de Recherche Agricole pour Ie Developpe- ment (lRAD)lBean-Cowpea CRSP project in Maroua, Cameroon, for in-depth training in cowpea storage technologies. These participants coordinate training activities in the subregion. Farmer field schools Specific production training in the form of field schools was arranged for farmers in all the countries during on-farm trials, and for research technicians in Chad. Similarly, nonpar- ticipating farmers were invited to visit the trial plots during field days. These interactions were of particular benefit to the participating and nonparticipating farmers, especially in making varietal preference decisions, based on phenotypic considerations. Varietal and cropping system recommendations Improved cultivars were introduced to farmers through both adaptive (researcher-man- aged) and on-farm trials. The mix of cowpea, sorghum, and millet varieties tested was a combination of advanced breeding lines, improved varieties that have proven successful in different parts of the region, and local check varieties. Varieties with a range of maturation rates, seed colors, and yield potential were selected to match farmer preferences, and the range of weather patterns with which they must contend. Country teams determined the composition of the technology package best suited for each country. Farmers' preferred Table 1. Participants in the cowpea storage technology training workshops, Chad, 1998 cropping season. Farmers who got storage Total technology materials for persons Location Women Men training other farmers trained Laokassy 19 118 15 137 Souley 13 120 15 133 Mango 20 30 5 50 Maibombaye 25 25 5 50 Nangkesse 17 33 5 50 Nassian 23 27 5 50 Koro 30 40 11 70 Gama 70 80 11 150 Mouroum-Touloum 100 150 22 250 Silambi 12 12 12 24 Rakena 15 15 15 30 Danamadji 26 26 26 Total 370 650 147 1020 342 Digitized by Google Cowpea dissemination in West Africa varieties resulting from on-farm trials, field days, and palatability tests are enumerated and typified by data from participating countries (for on-farm trials, Table 2) and Ghana (for adaptive trials, Table 3). Adaptive and on-farm trials in the participating countries were made up of replicated trials on research stations or researcher-managed in outstations (for adaptive trials) and in farmers' fields (for on-farm trials). The design and entries were decided by the country technology transfer teams. The entries were CRSP-developed materials for similar climatic environments, materials developed by the NARS and other NARS collaborators (e.g., the IARCs), and farmers' currently grown varieties. Though the number of participating farm- ers varied from year to year and from country to country, the average had been between 50 and 200 per year between 1997 and 2000. Varietal preferences are made by the farmers participating in the on-farm program, during field days in which other farmers participate, and during palatability tests conducted at the end of the season. During the end-of-year country team meetings (where field data are discussed), decisions are taken, with the help of the farmers in the team, on what farmers' preferred entries and technologies are. These decisions guide the NARS on materials and technologies to officially release. Table 2 shows some of the varieties released in some participating countries. Table 2. Cowpea, sorghum, and millet varieties extended by the InterCRSP project in participating countries (1997-2000). Crop/variety Developed by Developed in Extended to Cowpea Mouride Bean-Cowpea Senegal Senegal, CRSP (HC & USA) Niger, Chad, Melakh Bean-Cowpea CRSP Ghana, Mali Senegal, Niger, (HC& USA) Senegal Chad, Ghana, Mali C93W-24-130 (Lori Niebe) Bean-Cowpea Cameroon Senegal, Chad, CRSP (HC & USA) Ghana C92S-12-58 (CRSP Niebe) Bean-Cowpea Cameroon Cameroon, CRSP (HC & USA) Ghana C93W-2-38 Bean-Cowpea Cameroon Cameroon, Ghana, CRSP (HC & USA) Mali IT89KD-245 IITA Nigeria Mali IT89KD-374 IITA Nigeria Mali Sorghum NAD-1 (hybrid) INTSORMIL Niger Niger (HC& USA) Seguetana Cinzana IER Mali Mali N'tenimissa INTSORMIL Mali Mali (HC& USA) Millet HKP (Hainei-Khiere ICRISAT, INRAN Niger Niger Precoce) He = Host country. 343 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production Table 3. Grain yield (kg/ha) of elite cowpea lines tested across four locations in northern Ghana, 1999. Sites Entries Nyankpala Manga Damongo Wa Average Rank IT87D-829-2 992 678 845 504 755 9 Melakh 1141 573 1237 621 893 1 1 T93K-452-1 881 643 971 592 m 7 IT95-1497 1191 539 857 692 820 5 Bengpla (check) 940 313 966 513 683 11 1 TP-148-1 1118 469 761 604 738 8 SUL518-2 1408 591 845 613 864 3 IT87D-885 888 695 120 597 775 6 IT87D-1951 1013 452 669 537 668 10 IT86D-719 802 695 696 461 664 12 24-130 1135 608 1176 581 875 4 SUL-87KD 1202 382 1121 670 844 2 Mean 1059 553 922 582 779 LSD (0.05) 484 339 330 434 397 CV(%) 31.8 42.5 24.9 20.7 30.0 Some of the highlights of preferred technologies in the participating countries are: Chad Cowpea IT8ID-994, C7-29, Melakh, and C93W-24-l30 are top yielders. Farmers' preference: IT8ID-994 and IT89KD-288. Even though C93W-24-l30 pro- duced more haulms that could be of advantage as fodder, it was not preferred by the farmers. Sorghum Identification of the sorghum variety GRW as promising. GRW was developed by the NARS in Chad. Local selections, such as GRW mentioned above, outperformed newly developed varieties on farmers' fields. Participatory approach in the selection of sorghum varieties best adapted for intercrop- ping with cowpea, with the participation of about 160 farmers in all the 11 agricultural research centers. Sorghum-cowpea intercropping Trials involving 86 participating farmers indicated that sorghum and cowpea grown in alternate hills on the same row or in alternate rows were most effective in reduc- ing the menace of Striga hermonthica resulting in less Striga infestation on the field and higher sorghum yields. These two spatial arrangements showed superiority over the other two treatments: plots with sorghum only and plots where the local cultural practice was to plant sorghum and cowpea seeds were in the same hole. 344 Digitized by Google Ghana Cowpea Cowpea dissemination in West Africa Two CRSP cultivars (Melakh and C93W-24-l30) are now being tested on-farm after being found promising in adaptive trials conducted by the project in collaboration with the Savanna Agricultural Research Institute at four locations (Nyankpala, Manga, Darnongo, and Wa) for three years in Ghana. Table 3 shows the data of the third year adaptive trial in 1999. The three Cameroon-CRSP lines (C93W-2-38, C92S-l2-58, and C93W-24-l30) significantly outyielded all entries (12) in fodder production in the adaptive trials conducted at the above sites for the three years. Sorghum P 9407 (one of the four Striga-resistant cultivars obtained from Purdue University) was used as a source for genetic resistance to Striga hermonthica in the national pro- gram). Integrated Striga management, workshops, and demonstrations were held at four sites in northern Ghana (two sites each at lirapa-Lambusie and Sissala districts) for three years. Agronomic practices and varietal resistance formed the major basis for an integrated management of the pest. Major components were: - Use of varietal resistance - Early weeding before the flowering of Striga plants - Intercropping with trap crops (cowpea and soybean) - Crop rotation - Uprooting Striga plants that emerged in between normal weeding times - Manuring/fertilization Sorghum-legume intercropping Cowpea intercropped with sorghum, irrespective of the pattern reduced Striga her- monthica infestation at the on-farm trials. Soybean intercropped with sorghum was more effective than cowpea in reducing Striga in the on-farm trials. Mali Cowpea Variety Korobalen (lT89KD-374) was early, high grain yielding, and preferred by most farmers. Farmers' specific preferences (with regards to traits) are: IT89KD-245 (high pulse yield, drought-tolerant, Striga-resistant, high haulm yield, and grain whiteness) IT89KD-374 (early maturity and palatability). Better performance (in terms of grain yield) over local varieties of the following CRSP cultivars: C93W-2-38, Mouride, Melakh, and Marne fama. Sorghum Seguetana Cinzana was preferred by most farmers during field days. Some farmers preferred N'tenimissa for the whiteness of its grain and the taste and consistency of its porridge, but complained of its weak stems and consequent lodging. 345 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production With the exception of one of the sites, the local varieties outyielded the tested varieties (N'tenimissa, 96CZF498, and 96CZF499). Medium-duration millet There was no significant difference in the average yield of the varieties (Guefore CMDT 16, Tontoro 21, and Indiana 05), although particular varieties topped at specific sites (Indiana 05 at Dakoumani and Tonto ADPs; Guefore CMDT 16 at N'Torosso- Sokourani; and the local variety at Parana-Boho). Cereal-cowpea intercropping Intercropping of sorghum or millet with the improved cowpea cultivar IT89KD-245 in row intercropping or in alternate rows gave the best results in combating the menace of Striga hermonthica, compared to the farmers' traditional method of mixing cereal and cowpea and dibbling seeds from both crops. The farmers' traditional method gives very low cowpea stands, with resultant low grain yield. Farmers, however, complained about the laborious nature of interrow and alternate row intercropping. Seed multiplication During the 1999 cropping season, the identified promising cultivars (millet: Guefoue, Tontoro 21; sorghum: N'tenimissa, Seguetana; cowpea: IT89KD-245, IT89KD-374) were multiplied by 97 farmers in three WVI ADPs: Diaramana, Bani Valley, and Yangasso. The following quantities of seed were made available to the farmers: millet and sor- ghum 169 kg and cowpea: 84 kg. The total area planted was l3.75ha,dividedbetween sorghum: 8 ha; millet: 4 ha; and cowpea: 1.75 ha. Niger Cowpea Top performance ofMouride (lSRAlCRSP), IT89KD-349, and IT89KD-374 for grain yield and the local variety (TN5-78) for fodder production (Table 4). Table 4. Grain and straw weight of cowpea varieties in on-farm trials at Zinder, Tera, and Maradi (Kornaka), Niger, 1999. Sites Zinder Tera Kornaka straw grain straw grain straw grain Varieties Origin (kg/ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha) IT90KD-372-1-2 IITA 1524 428 152 428 1523 426 I T89KD-349 IITA 2100 552 2100 555 2102 553 Mouride ISRA/CRSP 2200 703 2200 704 2202 315+ Melakh ISRA/CRSP 2050 480 2050 479 2051 306+ Local (TN5-78) INRAN 2700 436 2700 436 2500 434 Mean 2114 520 2114 521 2134 408 LSD 0191 72.1 0191 37.42 0186 31.04 CV(%) 514 7.89 514 4.09 498 4.33 'Reduced yield of Mouride and Melakh due to overnight pilfering before the field day by farmers anxious to have seeds of both varieties. 346 Digitized by Google Sorghum Top performance of the hybrid NAD-l. Millet Top performance of the variety HKP. Senegal Millet Cowpea dissemination in West Africa Identification of GBS 8735 as an overall performer. Cowpea Mouride and Melakh were top yielders preferred by farmers. Conclusions Improved cultivars developed initially for specific countries stand a chance of adaptability and acceptability in other countries with similar environments. This is exemplified by the varieties Mouride and Melakh bred in Senegal, but now grown in Chad, Ghana, Mali, and Niger. Similarly, C93W-24-l30 (Lori Niebe) bred in Cameroon is now widely grown in Chad and Ghana. Mouride and Melakh (bred in Senegal) and C92S-l2-58 (CRSP Niebe), bred in Cameroon are now widely grown in Mali. These beyond-territory movements of CRSP cultivars would not have been possible without this intervention. NRM can be improved in West Africa under the framework of current InterCRSP-WVI initiative that has involved the mobilization of existing capacities within the Bean/Cowpea CRSP and INTSORMIL. Specifically, the initiative has increased the cultivation and pro- ductivity of cowpea in participating countries, thus increasing effective resource manage- ment in cereal-based systems. Increased productivity of the introduced cultivars is likely to have increased the food security of our target communities. An excellent demonstration of collaborative technology development and transfer is typified by the current West Africa NRM InterCRSP. The technology transfer aspect, which is the focus of this initiative, and which has come to be referred to as the "CRSP-NGO Model" appears to emerge as a model for the future. The model has reinforced the current state extension services with appropriate adaptive research projects, training, logistic, and evaluation tools. Outlook There is the need to expand the number ofNRM technologies included in the technology packages, particularly in the areas of integrated pest management and soil and water con- servation. Two additional CRSPs being targeted as sources of such interventions are: SANREM (Sustainable Agriculture and Natural Resource Management) CRSP IPM (Integrated Pest Management) CRSP However, relevant technologies from the IARCs and other advanced institutions will be included in a future expanded project. Acknowledgements This work was funded by USAID-Africa Bureau. We are grateful to all collaborators in the country technology transfer teams who have contributed to the achievements of the project so far. 347 Digitized by Google Cowpea contributions to farming systems/agronomic improvement of cowpea production References Anonymous. 1997. Adaptive Research with InterCRSPNatural Resource Management Technologies for Regional Transfer in West Africa. Project proposal submitted by Bean-Cowpea CRSP, INTSORMlL, and World Vision International, 14 February 1997. Anonymous. 1998. Adaptive Research Trial Results-First Year (March 1997-28 February 1998). Bean-Cowpea CRSP, Michigan State University, Michigan, USA. Anonymous. 1999. Adaptive Research Trial Results-Second Year (March 1998-28 February 1999). Bean-Cowpea CRSP, Michigan State University, Michigan USA. Anonymous. 2000. Adaptive Research Trial Results-Third Year (March 1999-28 February 2000). Bean-Cowpea CRSP, Michigan State University, Michigan, USA. Bationo, A., and F. Lompo. 1996. Technologies disponibles pour combattre la perte en elements nutritifs des sols en Afrique de I 'ouest. Proceedings organized by Centre International pour la Gestion de la Fertilite des Sols-Afrique, Lome, Togo. Bationo, A., B.R. Ntare, S. Tarawali, and R Tabo. 2000. Soil fertility management and cowpea production in the semiarid tropics of West Africa. Paper presented at the World Cowpea Research Conference III held at the International Institute of Tropical Agriculture, Ibadan, Nigeria, 4-7 September 2000. Kitch, L.w. and G. Ntoukam. 1991. Storage of cowpea in ash. Technical Bulletin 1. Bean-Cowpea CRSP, East Lansing, Michigan, USA. Kitch, L.w., G. Ntoukam, RE. Shade, J.L. Wolfson, and L.L. Murdock. 1992. A solar heater for disinfecting stored cowpea on subsistence farms. Journal of Stored Products 28(4): 261-267. Kitch, L.w., H. Bottenberg, and J.L. Wolfson. 1997. Indigenous knowledge and cowpea pest man- agement in sub-Saharan Africa. Pages 292-301 in Advances in cowpea research, edited by B.B Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (ilTA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Murdock, L.L, RE. Shade, L.w. Kitch, G. Ntoukam, J. Lowenberg-DeBoer, J.E. Huesing, W. Moar, O.L. Chambliss, C. Endondo, and J.L.Wolfson. 1997. Postharvest storage of cowpea in sub- Saharan Africa. Pages 302-312 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropi- cal Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Ntoukam, G. and L.W. Kitch. 1991. Solar heaters for improved cowpea storage. Technical Bulletin 2. Bean-Cowpea CRSP, East Lansing, Michigan, USA. 348 Digitized by Google Section V Cowpea postharvest and socioeconomic studies Digitized by Google Digitized by Google 5.1 The economics of cowpea in West Africa O. Coulibalyl and J. Lowenberg-DeBoer2 Abstract The contribution of cowpea to food security and poverty reduction can be substan- tial in West Africa if both biological and socioeconomic constraints are addressed. While some attention has been given to genetics, agronomy, and pest control, such economic issues as access to input, marketing, and consumer preferences are key research areas which contribute to the adoption and wide diffusion of improved cowpea technologies among small farmers. An area neglected in cowpea research but which is becoming important is consumer appreciation of improved cowpea grain. Results from the hedonic pricing analysis showed, for example, that consum- ers prefer larger grain size and seeds with low level of bruchid damage. Another area which needs to be investigated is the financial and economic profitability of chemical-intensive cowpea technologies. Cowpea is very sensitive to pests and chemical protection of the crop is financially profitable. However, this financial profitability may substantially decrease if hidden costs, such as the opportunity costs of capital, health hazards, and environmental costs are taken into consider- ation. This calls for the adoption of more environmentally sound and health con- scious crop protection techniques such as the use of botanicals and an integrated pest management approach for cowpea research. The study also reviews the eco- nomic impact of cowpea research and concludes that the integration of biological and social science in cowpea research will lead to sustainable technology develop- ment for food security and poverty reduction. Introduction Cowpea (Vigna unguiculata [L.] Walp.) has the potential to contribute to food security and to poverty reduction in West Mrica. The demand for cowpea in this region is increasing because of high population growth, mainly from the urban areas, and also because of pov- erty and the demand for low-cost food. Moreover, cowpea yields can increase if technical and socioeconomic constraints are addressed. The high protein content of cowpea and its use as a staple in the diets of Sahelian and coastal populations make it also a crop with high potential for food security in these regions. Cowpea forage contributes significantly to animal feed mainly during the dry season when the demand for feed reaches its peak. The largest producer and consumer of cowpea in West Mrica (and in the world) is Nigeria where a dense population creates an enormous demand for the crop. Niger is the largest cowpea exporter in West Mrica (and in the world) with an estimated 215 000 MT exported annually, mainly to Nigeria. Substantial amounts of cowpea also come to Nigeria from other neighboring countries, especially Cameroon and Chad. A large proportion of cowpea 1. International Institute of Tropical Agriculture, Plant Health Management Division, 08 BP 0932 Tripostal, Cotonou, Benin. 2. Department of Agricultural Economics, 1145 Krannert Building Purdue University, West Lafayette IN 47907-1145 USA. 351 Digitized by Google Cowpea postharvest and socioeconomic studies from Burkina Faso and Mali is imported into Cote d'Ivoire, and also Nigeria. Despite this importance of cowpea in food security, trade, and therefore poverty reduction, increased cowpea production, storage, and marketing face many constraints that need attention from research and development. The objective of this paper is to review past and ongoing studies to assess some of these key constraints and make recommendations for further research. The paper is divided into four sections. The first section deals with cowpea production, the second with marketing, the third analyzes cowpea pest management, and the fourth examines the impact of new cowpea technologies. Cowpea production Aggregate production World cowpea production was estimated at 3 319375 MT and 75% of that production (Fig. 1) is from Africa (FAOSTAT 2000). West Africa is the key cowpea producing zone, mainly in the dry savanna and semiarid agroecological zones. The principal cowpea pro- ducing countries are Nigeria, Niger, Senegal, Ghana, Mali, and Burkina F aso (FA OSTAT 2000). Among these countries, Nigeria and Niger are ahead with a production of 2 099 000 and 641 000 MT, respectively, in 1999 (FAOSTAT 2000). Nigeria, the largest cowpea producer in West Africa, also has the highest level of consumption (Table 1) with a popu- lation of 113.8 million and a per capita consumption of 23 kg per year (Table 1). The domestic supply of cowpea does not meet the demand, leading to a deficit of 518 400 MT per year. This is partly met by imports from neighboring countries, mainly Niger, where the estimated consumption of 78 200 MT in 1999 was far below the production figure of 641 020 MT (Table 1). South America (21%) Europe (1%) Asia- (2%) North America_._----;;:::::::::....J (1%) Figure 1. Share of Africa in world cowpea production. Source: Adapted from FAOSTAT (2000). 352 Digitized by Google The economics of cowpea in West Africa Table 1. Cowpea production and consumption in West Africa. Produc- Consumption/ Popula- tion capita/year tion Countries (t) (kg) (MI) Burkina 10,000 5.2 Mali 110,060 7.4 Niger 641,024 7.82 Nigeria 2,099,000 23.00 Source: Computed from FAO data, FAOSTAT (2000). 11.6 11 10 113.8 Consump- tion (t) 60,320 81,400 78,200 2,617,400 Cropping systems and adoption of related improved technologies Cowpea cropping system Surplus/ deficit (t) -50,320 28,660 562,824 -518,400 In West Africa, cowpea is grown mostly in subsistence farming systems and on a small scale in the lowland dry savanna and Sahelian regions. Traditionally, cowpea is grown in association or in relay cropping with cereals such as sorghum, millet, and maize mainly in the Sahelian regions (pedune Mali 1999). However, cowpea cropping systems are moving towards monocropping as the crop's economic importance increases. For example, cowpea monocrop has taken off in central Mali, thanks to an integrated rural development proj- ect which supplied improved seed, fertilizer, and pesticides on credit (Coulibaly 1987). Cowpea monocrop is frequent in cotton producing zones and in inland valleys and the Lake Chad basin in Cameroon (pedune Cameroon 1999). The increase in cowpea produc- tion is linked to the use of improved technologies including high yielding varieties and improved crop protection and production practices. A key issue behind the wide use of the improved cowpea technologies is their profitability. A few studies have been carried out on the profitability of cowpea technologies (Coulibaly 1995; Lowenberg-Deboer et al. 1994; Aitchedji 2001) and a case study in Benin is reviewed in this paper. Cost of production of cowpea with new technologies The profitability of the cowpea cropping system depends mainly on the varieties used (local or improved), the cropping practices and management (use of chemicals including fertilizers and pesticides), and the access to input and output markets. In this section, we review the financial profitability of cowpea production systems with improved produc- tion and protection technologies in Benin (Aitchedji 2001). The study was carried out in southern Benin on 35 farms with different combinations of improved technologies including improved variety, neem extract used as insecticide, chemical insecticide, and plastic bagging after solar drying (Table 2). The study used a Policy Analysis Matrix (PAM) developed by Monke and Pearson (1989) to assess the financial profitability of the improved cowpea technologies. Since access to capital has been reported by farmers as a key constraint (Aitchedji 2001), a sensitivity analysis was carried out with three scenarios linked to opportunity cost of the capital at 0, 25, and 50%. These rates are within those computed by Lowenberg-DeBoer et al. (1994) in the rural areas in West Africa which vary from 0 to 100%. The results (Table 2) show that only improved cowpea technologies are profitable even under tight credit constraints, compared to local technologies mainly local cultivars and 353 Digitized by Google o 0" "" N" CD a. -:?" C"') o ~ ......... (i) w I.n -I:>. Table 2. Financial profitability of cowpea production systems in South Benin: application of policy analysis matrix (FCFA Iha): 3 scenarios with opportunity cost of capital (respectively 0%, 25%, and 50%). Financial budget Financial budget Financial budget Production systems (opportunity cost of capital 0%) (opportunity cost of capital 25%) (opportunity cost of capital 50%) Costs of Costs of Costs of Costs of Costs of Costs of non Gross tradable nontradable Gross tradable nontradable Gross tradable tradable Technologies revenue input input profits revenue input input profits revenue input input profits Improved variety IT 86 D-364 + botanical insecticide + local storage 212600 29975 128115 54510 212600 37483 135838 39279 212600 44972 143590 24038 Local variety + botanical insecticide + chemical insecticide + local storage 168000 76775 128965 -37740 168000 95975 136918 -M893 168000 115165 144870 -92035 Local variety + chemical insecticide + improved storage 168000 162360 122965 -117325 168000 202959 130918 -165877 168000 243540 138870 -214410 Local variety + chemical insecticide + local storage 140000 52095 129215 -41310 140000 65124 138743 -63867 140000 78146 148245 -86391 Local variety + botanical insecticide + local storage 140000 23550 129115 -12665 140000 29438 137107 -26545 140000 35326 145095 -40421 Improved variety Vita 5 + chemical insecticide + local storage 200000 68130 121915 9955 200000 85168 129586 -14754 200000 102196 137290 -39486 Source: A'itchedji et al. (2001). Note: Physical quantity of inputs per hectare, tradable inputs: fertilizer (100 kg), chemical insecticide (5 I), storage bags (19, 5, 0, 7, 2, and 2 units), storage barrel or tank (0, 5, 8, 0, 0, and 3 units), Sofagrain (0, 10,0,0, 7, and ° sachets), Gastoxin (19, 0, 0, 0, 0, and ° tablets), Tarpaulin (0, 0, 15, 0, 0, and ° units); Nontradable inputs: Seed (34, 44, 44, 44, 44, and 34 kg), Labor (243, 243, 228, 228, 243, and 228 man-day), Soap (4, 4, 0, 0, 4, and ° units), Pepper (0, 0, 0, 25, 0, and ° kg), Small equipment (990 FCFA), Land (18675 FCFA); Output: average yields (1063, 1200, 1200, 1000, 1000, and 1000 k/¥ha). Profit = Gross revenue: (cost of tradable inputs + cost of nontradable inputs). For more information about PAM model, see Manke and Pearson (1989), Adesina and Coulibaly (1998) . (;\ ~ re "0 ~ S- II> ... ~ II> :J 0... '" o n O· tb § o :3 ri' '" e- ~ '" The economics of cowpea in West Africa local storage techniques. The most profitable combination is improved varieties of cowpea sprayed with neem or papaya extract followed by improved varieties of cowpea sprayed with chemical insecticide. The difference in gross margins of cowpea production systems showed in Table 2 is explained by the difference in the type of varieties (more resistance to pests and diseases, drought and heat tolerance, and higher yield). Also, the type of pes- ticide used for treatment and the storage technology used can make a significant difference (Aitchedji 2001). This profit will be even higher if we include the costs of environment and public health. The misuse of pesticides in general has been causing deaths every year in the cotton and cowpea producing areas. In Benin, a study conducted by the Ministry of Health in the northern Borgou province in 1999 revealed that 37 people died due to endosulfan poisoning, while another 36 people experienced serious ill health (Pesticides News 2000). In view of the importance of the Borgou province to national cotton cultiva- tion, it is felt that at least 70 people might in fact have died (peter et al. 2000). Health hazards should be evaluated and discounted from the benefits of using pesticides. The loss in work days and the health costs can be substantial and are not considered by farmers in assessing the benefits of using pesticides. Cowpea marketing During the past 20 years, the Bean-Cowpea Collaborative Research Support Program (CRSP) and international and national research institutions have made substantial contri- butions to cowpea production and protection technology. Beside new varieties, improved methods for controlling pests in the field and in storage have been developed. These technologies could dramatically increase cowpea production and grain quality in West Africa. The questions now are, Who will buy those cowpea? At what price? And what kind of cowpea would consumers prefer? Cowpea trade in West Africa Cowpea markets in West Africa are part of an ancient trade that links the humid coastal agroecological zones with the semiarid interior. This ancient trade is based on the com- parative advantage in food production of each zone. In the humid coastal areas, it is relatively easy to produce carbohydrates (e.g., cassava, maize, rice), but because of pests and diseases, it is difficult to produce animal or vegetable protein. Lack of rainfall limits grain production in the interior, but creates good conditions for livestock, cowpea, and groundnut. In the traditional cowpea growing countries of the Sudano-Sahelian zone, there is a well developed network of village buyers who assemble small quantities from farmers into 100 kg bags and merchants who transport and store the bags. These trade linkages can be illustrated with Ghana which though a maj or producer of cowpea imports about 10000 MT annually (Langyintuo 1999). About 30% of the Ghanaian imports are from Burkina F aso (Table 3) and the rest from Niger. According to Langyintuo (1999), in Accra, the large, rough coated Nigerien cowpea (cowpea from Niger) sells for a premium, but they need to be marketed quickly because they do not store well in the humid coastal climate. 355 Digitized by Google Cowpea postharvest and socioeconomic studies Table 3. Official imports of cowpea into Ghana, 1992-1998. Total imports Imports from Burkina-Faso Imports from Niger Year (MT) (MT) (% of total) (Mn (% of total) 1992 2055.34 592.00 28.80 1463.34 71.20 1993 2640.80 637.92 24.16 2002.88 75.84 1994 11798.98 2898.95 24.57 8900.03 75.43 1995 13086.29 3295.95 25.19 9790.34 74.81 1996 6816.80 3077.79 45.15 3739.01 54.85 1997 NA NA NA NA NA 1998 10167.18 3050.15 30.00 7117.03 70.00 Source: Langyintuo 1999. Cowpea trade in Benin The traders interviewed indicated that there is an active trade between Benin, Niger, Togo, Nigeria, and Gabon. They noted that cowpea imported from Niger, Togo, and Nigeria are later exported to Gabon and Togo. Specifically, between 40 and 60 traders send cowpea totaling about 50 000 MT to Gabon and about half that quantity to Togo each year. Yet there were no cowpea trade statistics at the national level because the government of Benin considers cowpea trade a minor commercial activity. The traders consider this a favorable condition for their activities because they do not pay any tax on cowpea, unlike maize exporters who pay 100 FCFAon each bag exported as fiscal exit tax (taxe de sortie) (Langyintuo 2000). Cowpea trade in Togo Togo is very active in cowpea trade. Togolese traders frequently export cowpea to Gabon and sometimes to Congo. Traders from these countries also purchase cowpea directly from Togo. Exports to Ghana and Benin are mainly by Ghanaian and Beninois traders. It was estimated that on a given market day, between 20 and 40 Ghanaian traders pur- chase between 10 and 20 bags of cowpea each. Most of the traders from Ghana come from Akakyi, Agbozome, and Aflao in the Volta region, and Accra in the Greater Accra region. Gabonese and Beninois traders in the Akodesewa market number up to 20 from each country. All Togolese traders exporting grain are expected to indicate the quantity being exported on their travel document (laissez-passer). Non-Togolese traders, on the other hand, are not obliged to do so. Consequently, the government does not keep track of grain shipped out of or into the country. In Ghana, on the other hand, the Ghana Plant Protection and Regulatory Service (PPRS) of the Ministry of Food and Agriculture (MoFA) subjects grain imported or exported to phytosanitary inspection. As a routine, the quantities of grain per trader are recorded, thus providing an opportunity for the tracking of grain movement in and out of the country. Available export data show that between 1990 and 1998, Gabon was the only country importing cowpea from Togo annually (Table 4). Imports averaged 12.25 MT. Exports to Gabon increased dramatically from 20.18 MT in 1997 to 46.53 MT in 1998. Togo also imports cowpea from Senegal, Nigeria, Niger, Burkina F aso, Ghana, and Benin. In 1998, 356 Digitized by Google The economics of cowpea in West Africa Table 4. Imports and exports of cowpea in and out ofTogo (Mn. Imports Exports Burkina Year Benin Ghana Senegal Faso Niger Nigeria Gabon Congo Ghana 1990 10.70 0.14 78.08 6.20 5.48 1991 0.14 0.02 19.68 210 1992 6.20 0.28 73.80 468.62 0.55 1993 1.20 5.70 1.05 0.55 1994 30.30 0.72 0.55 7.00 2.00 149 1995 1.00 8.00 8.00 2.00 400 1996 0.36 1.80 61 1997 - 14.07 20.18 720 1998 55.60 9.72 35.10 1.41 46.53 334 Source: Langyintuo 2000. Note: - = Zero or data unavailable. for instance, Togo imported 101.830 MT of cowpea, 55% from Benin, 34% from Burkina Faso, 10% from Ghana, and 1 % from Niger (Langyintuo 2000). The hypothesis at the beginning of the cowpea marketing research was that most cowpea from northern Cameroon were marketed into Nigeria. Surveys in Cameroon showed that in fact most of the northern Cameroon production went to southern Cameroon, and that some was exported from there to Gabon and Congo. In northern Senegal as the climate grew drier in the 1980s and the groundnut parastatal declined, cowpea increasingly replaced groundnut as the legume of choice. Some cowpea is exported to Mauritania and Gambia, but the transportation cost and lack of market links limit access of Senegalese cowpea to the large markets in Ghana, Nigeria, and elsewhere along the African coast. Senegal is the only country in the region with a substantial cowpea processing industry. Faye et al. (2000) identified five companies producing cowpea-based weaning food, cowpea flour, and cowpea-based crackers. All the products are made from recipes developed by the National Institute of Agriculture Research (lSRA) in Senegal's Food Technology Institute (ITA). In addition, there is a cracker manufacturer in Nouak- chott, Mauritania, who uses primarily cowpea from Senegal. Consumer preferences Knowledge of consumer preferences is essential to developing cowpea markets. Breeders need to know what characteristics consumers want. Integrated pest management specialists need an estimate of the level of grain damage acceptable to consumers. The Bean-Cowpea CRSP cowpea price and quality study was launched in Maroua, Cameroon, in September 1996, and later extended to four markets in northern Cameroon, three in Nigeria, two in Niger, three in northern Ghana, three in Mali, and six in Senegal using a common data collection protocol. Every month, CRSP researchers and technicians buy five samples per market from randomly selected sellers. They note the gender and other seller char- acteristics. In the laboratory, they record the 100 grain weight, average length and width of grains, number of bruchid holes per 100 grains, color, and texture of the testa, and eye color. The data are analyzed using a hedonic pricing regression model. 357 Digitized by Google Cowpea postharvest and socioeconomic studies Initial results from a hedonic pricing analysis carried out by Langyintuo et al. 2000 and Faye et al. 2000, indicated that consumers in almost all areas prefer larger grain size. Consumers are more sensitive to bruchid damage than hypothesized. It was thought that West African consumers would tolerate a certain level of damage, but the data indicate that cowpea prices are discounted from the first appearance of damage. Results from the same study indicated that women in Cameroon appear to sell at a higher price than men, probably because women sell in small quantities for immediate consumption. In Senegal, consumers appear to pay a premium of 20 FCFAlkg for the traditional black speckled varieties. The Hedonic Model Framework: A review of literature and application The conceptual basis for estimating consumer demand for a good's quality is Lancaster's (1966) model of consumption theory which regards properties of the good and not the good itself as the direct object of utility. Using this concept, Ladd and Suvannunt (1976) developed the consumer goods characteristics model which describes the price of a good as a linear summation of the implicit value of its attributes. For cowpea, the consumer goods characteristics model can be expressed mathemati- cally as: m Pc = L XC} P qj j=l Where Pc = price of cowpea; XCj = quantity of cowpea grain characteristic j, such as size of grain, testa texture, eye color, and damage by weevils. P cj = implicit price of characteristic j. Hedonic pricing models have received wide applicability in the scientific world. In his estimation of quality adjusted price index for computer processors in 1989, Dulberger concluded that hedonic prices could be useful in estimating quality adjusted indexes for the output of complex products manufactured in an industry characterized by technological change. The effects of milling and premilling operations on rice quality were examined by Bonifacio and Duff (1989) using a hedonic pricing model. The results indicated insig- nificant differences in paddy quality by mill type and confirmed that mill type affects milled rice quality and that millers attach economic significance to certain grain quality characteristics. Abansi et al. (1990) used the hedonic pricing model to evaluate consumer preferences for rice quality in the Philippines. They found that rice consumers attach economic significance to quality considerations. Walburger and Foster (1994) used data on boar performance traits from Purdue University Boar Test Station and auction sales data to estimate the implicit prices for back fat, loin eye area, average daily gain, and feed efficiency of boars in the US, using a hedonic pricing model. All of these variables have a 358 Digitized by Google The economics of cowpea in West Africa significant impact on the auction prices of boars. In 1995, Naik used the hedonic pricing theory to show that only 76% of the variations in silk price in India were explained by the quality characteristics, suggesting a poor linkage between quality and price. Hedonic relationship and implicit prices: A review of work by Langyintuo et al. 2000; Faye et aI. 2000 Langyintuo et al. (2000) and Faye et al. (2000) have undertaken a hedonic pricing study for each of three markets in Ghana (Table 5) and four markets in Cameroon (Table 6). The following hedonic equation was specified and estimated in a seemingly unrelated regression model: Where P = price of cowpea Y = yearly dummy M = monthly dummy X cowpea characteristics variables e disturbance term a constant term ~,'I' and y are parameters to be estimated i = 1,2, ... ,n r=2,3 1= 1,2,3 j = 1,2, ... ,K t = 2,3, ... ,T Cowpea price as the dependent variable was measured in FCFAlkg in Cameroon and Cedislkg in Ghana. These were entered in the model as absolute values. Similarly, grain size measured as weight of 100 grains and number of insect holes as indepen- dent variables were also entered as absolute values. Other independent variables in the X-matrix including eye color, seed coat color, and gender of sellers (in the Cam- eroon models) were entered as dummy variables. Cowpea grain color and eye color are important when the intended use requires decortication. Where decortication is required, for example, in making kosa, poor pounding and winnowing may still leave some flecks for which consumers have a low tolerance level. Black flecks tend to be more conspicuous than other colors. A value of 1 was assigned to white grain color and zero otherwise. Similarly, black-eyed grain assumed a value of 1 and zero otherwise. The gender variable was entered as 1 for female and zero otherwise. To account for the effect of time in price variability, yearly and monthly dummies were used. For the yearly dummies, 1997 was used as the base year and each yearly dummy assumed a value of 1 for the year in question and zero otherwise. For the monthly dummies, November was used as the base year since prices in that month are the lowest in both countries. The monthly dummy assumed a value of 1 for the respective month and zero otherwise. 359 Digitized by Google Cowpea postharvest and socioeconomic studies The estimated regression results indicate that seasonal supply, demand, cowpea size, color, and insect damage level explain between 53 and 72% of price variability in the seven markets studied (Tables 5 and 6). Across all markets, cowpea grain size was significant in explaining price variability. Table 5 indicates that besides grain size, color of eye, and number of insect holes are important in explaining cowpea price variability in Cameroon. Unlike grain size that influences price positively, there is an inverse relationship between price and grain eye color or insect holes. Consumers demand discounts (FCFAlkg) for black eye cowpea: 5.5 in Maroua, 4.86 in Mokolo, and 18.88 in Salak. An increase of one hole per every 100 grains leads to a discount of 0.29 FCFAlkg in Maroua and 0.29 FCFAlkg in Salak. Estimates for Mokolo and Banki show a discount for damage, but coefficients are not significant at conventional levels. In Ghana, consumers are equally as sensitive to cowpea damaged by insects as Cameroonians. A discount of up to ¢ l20lkg is demanded for a unit increase in number of damaged grainsll 00 grains in Tamale and Bolgatanga markets but less than a ¢ llkg in Wa market (Table 5). A contrasting feature of consumer demand for grain characteristic between Ghana and Cameroon is observed in the preference for grain eye color. In Ghana, con- sumers are willing to pay a premium of between ¢ 1 09 and ¢226lkg for black-eyed cowpea, with traders in Wa receiving the highest premium. In contrast, consumers in Cameroon discount up to 14 FCFA for black-eyed cowpea. This result reflects the cultural roles of the grain. In Cameroon, one of the main dishes using cowpea Table 5. Estimated model coefficients for selected markets in Ghana. Tamale market Bolgatanga market Wa market Variables Coeff. T-ratio Coeff. T-ratio Coeff. T-ratio Grain weight 11.16 2.06** 11.78 2.15** 13.04 1.99** No. of holes -50.98 -1.34* -120.98 -2.34** -0.989 -1.341* Color of eye 109.60 1.65** 226.68 1.77** 145.69 4.00*** Grain color 21.22 2.29** 4.65 0.04 150.81 2.06** Constant 146.30 1.61 242.17 1.69 745.78 9.24 R-Square 0.66 0.66 0.53 Source: Langyintuo et al. (2000). Note: 'Significant at 10%; "Significant at 5%; "'Significant at 1 %. Table 6. Estimated model coefficients for selected markets in Cameroon (1997-2000). Banki market Variables Coeff. T-ratio Grain size 3.73 3.42*** No. of holes -0.02 -0.15 Color of eye -0.6 -0.08 Grain color 8.48 1.06 Gender of trader 18.53 1.90** Constant 62.45 2.88*** R-Square 0.62 Source: Langyintuo et al. (2000). Maroua market Coeff. T-ratio 2.87 3.83*** -0.40 -2.15** 39.20 -1.76** 1.14 0.18 13.52 1.99** 123.22 4.68*** 0.63 Mokolo market Salak market Coeff. T-ratio Coeff. T-ratio 2.07 1.35* 4.39 4.04*** -0.02 -0.15 -0.29 -1.29* 13.79 1.50* -35.67 -1.64** 1.13 0.13 -5.97 -0.76 -3.87 -0.38 16.76 2.59*** 144.97 4.44*** 107.28 3.19*** 0.68 0.72 Note: 'Significant at 10%; "Significant at 5%; "'Significant at 1%. 360 Digitized by Google The economics of cowpea in West Africa is kosa which is preferred without black flecks. In Ghana, on the other hand, use of cowpea for tubani and a mixture of rice or gari (produced from fermented cassava dough) and beans is more important than kosa. Therefore, black flecks have little impact on demand. In Tamale and Bolgatanga in Ghana, white grain color does not appear to influence price. In Wa, on the other hand, white grain attracts a premium of up to ¢ 1 25/kg because of the role of kosa in the culture of this region. This seems to support the hypothesis that, as the cultural role of cowpea requires more decortication, color plays a significant role in price determination. In Cameroon, grain color is insignificant in explaining price variability. The role of gender appears to be important in grain retail trade in Africa. In Cameroon, female vendors have a competitive edge over their male counterparts. This is reflected in the slightly higher premium of about 13-18 FCFA they receive in Maroua, Salak, and Banki. The hypothesis is that this is a premium for their service for selling in small quantities for immediate preparation. Female traders in Mokolo do not have a similar competitive edge over their male counterparts. Traders receive a premium for storage. Sales made beyond the fourth quarter of the year attract a premium, thus justifying the investment in storage materials. Farmers' pest management It has been shown in most of the cowpea producing countries in West Mrica that field pest problems are substantial, and insects such as flower thrips, mainly Megalurothrips sjostedti Tryb. (Thysanoptera: Pyralidae), and Maruca pod borer, Maruca vitrata F ab- ricus (syn. M. testulalis) (Lepidoptera: Pyralidae) are highly implicated in production losses (Jackai and AdalIa 1997; Tamo et al. 1997; Bottenberg et al. 1997). Without chemical treatment at flowering, for instance, there can be total crop failure. Results from insecticide treatment on improved varieties have shown a substantial yield increase from 30 to 1 00% compared to nontreated cowpea (pedune Nigeria 1999; Pedune Senegal 1999; Pedune Cameroon 1999; Pedune Ghana 1999; Pedune Mali). Most of the pest management research on cowpea in West Africa has focused on developing and testing field and storage pest control technologies. Among these technologies are improved genetic materials (pest and disease resistant and tolerant varieties), insecticide treat- ment, and plant extracts with insecticidal and fungicidal properties. Lately, the focus of research on plant extracts by such research networks as Pedune is primarily related to their low cost and very marginal disturbance of the environment. Also, botanical insecticides may represent a safe substitute for highly toxic pesticides such as cotton or cocoa insecticides, which are very often diverted onto cowpea. In Benin, for example, more than 294 000 farmers use banned insecticides such as organochlorides or organo- phosphates on cowpea (pedune Benin 1999). Death and poisoning were reported from 16 villages in seven out of 12 districts in Benin. If poisoning occurred at the same rate throughout all cotton growing areas, at least 70 people might have died as a result of endosulfan (organochloride) use injust one cotton producing district in Benin (pedune Benin 1999). Cotton insecticides are virtually the only pesticides available in the rural area of northern Benin and the only ones delivered on a credit basis. This may account for some of the hazardous uses of the insecticides, such as on food crops or in storage. In addition, farmers are not adequately informed about the hazards associated with these products. Such inappropriate uses of cotton pesticides in West Africa are well 361 Digitized by Google Cowpea postharvest and socioeconomic studies known to cotton research institutes and should have been considered when selecting insecticides for large-scale application. In Cameroon, a Pedune survey in the Western mid altitude region showed that various chemicals are used for pest control in the field and in storage by farmers and traders (Nkamleu et aI., unpublished). Synthetic chemical products are reported to be used by 46% of traders and 12% of farmers to protect cowpea in storage whereas 17% of the traders and 40% of the farmers reported using traditional methods of treat- ment (no chemicals). Among the traders using chemical control in storage (Fig. 2), 57% reported using Actellic® or Actellic® Super® (pirimiphos methyl, or pirimiphos methyl plus permethrin), and methyl-parafene (22%). Other unidentified chemicals are used by 24% of chemical users for stored cowpea. Malathion® and prohibited DDT® are also fairly often used and are easily obtainable from local dealers. Among farmers using chemicals in storage, 65% reported the use of methy 1 parafene (Fig. 3) and 23% the use of Thimui. Ease of use (tablet or dust formulations) and effectiveness in controlling weevils were cited as the major reasons for the wide use of chemicals in storing cowpea. Farmers also use nonchemical storage technologies that have been developed by both national and international research institutions (national agricultural research 57 60 - fIl 50 .. Q,j -g 40 .. - - 30 0 22 24 ~ 20 RI - 14 - - 7 7 C Q,j 10 u .. a:. 0 n 2 II II ,------, Actellic Malathion Methyl- Percal. M DDT dust Thimul Others parafene Figure 2. Storage chemical pest control by cowpea traders in West Cameroon. Source: Pedune, socioeconomic surveys in Cameroon, 1999. 70 fIl 60 .. Q,j E 50 .. ~ - 40 0 Q,j ~ 30 RI -c 20 Q,j u .. a:. 10 0 23 ,------ Thimul 65 - Methyl- parafene 2 4 2 4 ,------, II ,------, II Tuoden Dysban Actelhc Cypercal. M Figure 3. Storage chemical pest control by cowpea farmers in West Cameroon. Source: Pedune socioeconomic surveys in Cameroon, 1999. 362 Digitized by Google The economics of cowpea in West Africa systems, Purdue University, International Institute of Tropical Agriculture, cowpea research networks). The technologies include solar drying, triple bagging, ash storage, and the use of botanical extracts to store cowpea effectively and at low cost (Mur- dock et al. 1997). For example, the extracts of Boscia senegalensis, a common plant in the Sahel, is shown to cause 75-100% mortality among cowpea bruchids at very low concentration (0.67 gil) in Senegal (pedune Senegal 1999). While the botanicals need further testing for efficacy and adaptability to local agroecological and pest density conditions, solar drying and triple bagging are being largely disseminated. The storage pest management technologies are in high demand by both farmers and traders and would decrease losses and enhance the adoption of cowpea production technologies. Economic impact assessment of cowpea technologies Impact assessment studies in Senegal, Cameroon, and Mali show that research on cowpea production and protection has reached a large number of people and is generating substantial economic benefits. In Senegal, over 80% of stored cowpea are stored with the CRSP drum storage technology (Faye and Lowenberg-DeBoer 1999). In northern Cameroon about 23% of the cowpea area are planted to Vya, BR 1, and BR2, varieties that the CRSP helped develop and popularize (Diaz-Hermelo and Lowenberg-DeBoer 1999). About 10% of cowpea in northern Cameroon are stored with storage technologies developed by the IRADlPurdue CRSP team. The CRSP stor- age technologies developed in Cameroon are now being extended to Nigeria, Niger, Burkina Faso, Mali, Senegal, Chad, Zimbabwe, and Mozambique. Rates of return on cowpea research have varied widely. Schwartz et al. (1993) showed that returns to CRSP and USAID investments in Senegal in the early 1980s had a rate of return over 200% annually because of the large benefit from Operation Cowpea early in the life of the project. Stems and Bernstein (1993) showed an annual rate of return of about 15% on cowpea varietal research in Cameroon in the 1980s and early 1 990s. The annual rate of return on CRSP investment in the ISRAIUniversity of California Riverside team in Senegal for the varietal development and storage after 1985 is about 16%. The rate of return to CRSP breeding and storage research in Cameroon alone is about 5%. In Cameroon and Senegal, the benefits are much higher when the use of the technology outside the country of origin is taken into account. The benefits to society resulting from the multicountry cowpea research and development range from US$l. 3 million to US$12. 3 million per year (Sanders et al. 1995). Conclusion The contribution of social sciences to the development of the cowpea subsector for food security, income, and therefore poverty reduction is important but research is still far behind in this area compared to that in the biological sciences. The review showed that marketing studies are useful in indicating what varieties fit consumers' preferences and are widely adopted and sell for premium prices. Sub-Saharan consumers are more sensitive to bruchid damage than hypothesized, and grain color and size attract premiums according to countries and among consumer groups within countries. Seed production and dissemi- nation will increase the diffusion rate of improved cowpea varieties. This information is useful in guiding entomologists, breeders, biotechnologists, and postharvest specialists in 363 Digitized by Google Cowpea postharvest and socioeconomic studies developing new cowpea technologies to meet the demand. Marketing innovations should be promoted to reduce transactions and other costs that will enhance higher profits for producers and/or lower prices for consumers. The adoption studies carried out showed that farmers would adopt new cowpea technologies with substantial economic benefits. The key is to estimate the economic benefit after deducting all the costs, including transac- tion costs, opportunity cost of capital, and environmental and health costs associated with insecticide use. Biological scientists are challenged to produce low cost/environmentally sound cowpea to meet the increasing demand. The review showed also that major eco- nomic impact has been achieved from cowpea research in Senegal, Cameroon, Burkina F aso, and Mali with improved production and protection technologies. The next challenge is to measure the impact of these technologies on poverty reduction at the country and regional levels. References Adesina, A.A. and O.N. Coulibaly. 1998. Policy and competitiveness of agroforestry-based technologies for maize production in Cameroon. An application of policy analysis matrix. Agricultural Economics 19: 1-13. Abansi, C.L., F.A. Lantica, B. Duff, and I.G. Catedral. 1990. Hedonic model estimation: appli- cation to consumer demand for rice grain quality. The Academy. AItchedji, C.C. 2001. Etude de la rentabilite financiere et economique des nouvelles technologies de la culture du niebe au Benin: Cas du departement du Couffo. Memoire de Maitrise es- Sciences Economiques, FASJEP, Universite Nationale du Benin (UNB). Bonifacio, E.P. and B. Duff. 1989. The impact of postharvest operations on paddy and milled rice quality. Proceedings of the twelfth ASEAN technical seminar on grain postharvest technology held at Surabaya, Indonesia, 29-31 August 1989. 25 p. Bottenberg, H., M. Tamo, D. Arodokoun, L.E.N. Jackai, B.B. Singh, and O.Youm. 1997. Population dynamics and migration of cowpea pests in northern Nigeria: implications for integrated pest management. Pages 271-284 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agri- cultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Coulibaly, O. 1987. F actors affecting the adoption of agricultural technologies in sub-Saharan Africa: the case of new varieties of cowpea around the agricultural research station of Cinzana, Mali. MSc thesis, Michigan State University, East Lansing, Michigan, USA. Coulibaly, O. 1995. Devaluation, new technologies, and agricultural policies in the Sudanian and Sudano-Guinean zones of Mali. PhD thesis, Purdue University, West Lafayette, IN, USA. Dulberger, E.R. 1989. The application of a hedonic model to quality-adjusted price index for computer processors. In Technology and capital formation, edited by D.W. Jorgenson and R. Landau. The MIT Press. Diaz-Hermelo, F. and J. Lowenberg-DeBoer. 1999. Estimating research benefits with both production and postharvest technology: the case of cowpea in Cameroon. Bean-Cowpea CRSP West Africa Regional Social Science Report # 2. FAO. Food and Agriculture Organization. 1998. FAO Production Yearbook, Rome, FAO. FAO. Food and Agriculture Organization. 1999. FAO Production Yearbook, Rome, FAO. FAOSTAT. 2000. Site internet: http//www. Fao. org/statistics Faye, M. and J. Lowenberg-DeBoer. 1999. Adoption of cowpea improved varieties and storage technology in the north central peanut basin of Senegal and economic impact implications. Bean-Cowpea CRSP West Africa Regional Report # 3. 364 Digitized by Google The economics of cowpea in West Africa Faye, M., M. N'diaye, and J. Lowenberg-DeBoer. 2000. Identifying cowpea characteristics which command price premiums in Senegalese markets. Paper presented at the World Cowpea Conference, Ibadan, Nigeria. Jackai, L.E.N. and C.B. Adalia. 1997. Pest management practices in cowpea: a review. Pages 271-284 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Ladd, G. and V. Suvannunt.1976. A model of consumer goods characteristics. American Jour- nal of Agricultural Economics 58: 504-510. Lancaster, K. 1966. Change and innovation in the technology of consumption. American Eco- nomics Review 56: 132-157. Langyintuo, A. 1999. Summer research program in Ghana and Burkina Faso. Trip Report. Department of Agricultural Economics, Purdue University, West Lafayette, IN, USA. Langyintuo, A. 2000. Summer Research Program in Ghana, Togo, Benin, and Burkina Faso, Trip Report. Department of Agricultural Economics, Purdue University, West Lafayette, IN, USA. Langyintuo,A., G. Ntoukam, L. Murdock, and J. Lowenberg-DeBoer. 2000. The market value of cowpea characteristics in Cameroon and Ghana. Paper presented at the World Cowpea Conference, Ibadan, Nigeria. Lowenberg-DeBoer J., T. Abdoulaye, and D. Kabore. 1994. The opportunity cost of capital for agriculture in the Sahel: case study evidence from Niger and Burkina Faso. Staff Paper #94-2, Department of Agricultural Economics, Purdue University, West Lafayette, IN, USA. Monke, E.A. and S.R. Pearson. 1989. The policy analysis matrix for agricultural development. Cornell University Press, Ithaca, USA. Murdock, L.L., R.E. Shade, L.W. Kitch, G. Ntoukam, J. Lowenberg-DeBoer, J.E. Huesing, W. Moar, O.L. Chambliss, C. Endondo, and J.L. Wolfson. 2000. A postharvest storage of cowpea in sub-Saharan Africa. Pages 302-312 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication ofInternational Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Naik, G. 1995. Price-quality relationships in raw silk markets. Indian Economics Review 30(1): 101-119. Pedune-Benin. Production ecologiquement durable du niebe. 1999. Annual country report. Benin. International Institute of Tropical Agriculture, Cotonou, Benin. Pedune-Ghana.Production ecologiquement durable du niebe. 1999. Annual country report. Ghana. International Institute of Tropical Agriculture, Cotonou, Benin. Pedune-Cameroon. Production ecologiquement durable du niebe. 1999. Annual country report. Cameroon. International Institute of Tropical Agriculture, Cotonou, Benin. Pedune-Mali. Production ecologiquement durable du niebe. 1999. Annual country report. Mali. International Institute of Tropical Agriculture, Cotonou, Benin. Pedune-Senegal. Production ecologiquement durable du niebe. 1999. Annual country report. Senegal. International Institute of Tropical Agriculture, Cotonou, Benin. Pedune. Production ecologiquement durable du niebe. 1999. Socioeconomic surveys in Cam- eroon. International Institute of Tropical Agriculture. Cotonou, Benin. Pedune-Nigeria. Production ecologiquement durable du niebe. 1999. Annual country report. Nigeria. International Institute of Tropical Agriculture, Cotonou, Benin. Pesticides News. 2000. Annual report. No. 47. March 2000. 24p. Peter, T., S. Tovignon, and S.D. Vodouhe. 2000. Endosulfan deaths and poisonings in Benin. Pesticides News 47. March 2000. 365 Digitized by Google Cowpea postharvest and socioeconomic studies Sanders, J., T. Bezuneh, and A. Schroeder. 1994. Impact assessment of the SAFGRAD com- modity networks. USAID/AFRIARTSIFARA, Washington, DC, USA. Sanders, J., B. Shapiro, and S. Ramaswamy. 1995. The economics of agricultural technology in semi-arid sub-Saharan Africa. Johns Hopkins University Press, Baltimore, Maryland, USA. Schwartz, L., J.A. Sterns, and J.F. Oehmke. 1993. Economic returns to cowpea research, exten- sion, and input distribution in Senegal. Agricultural Economics 8: 161-171. Sterns, J. and R. Bernstein. 1993. Assessing the impact of cowpea and sorghum research and extension in northern Cameroon. Michigan State University, International Development Working Paper No. 43. Tamo, M., H. Bottenberg, D.Y. Arodokoun, and R. Adeoti. 1997. The feasibility of classical biological control of two major cowpea insect pests. Pages 259-270 in Advances in cowpea research, edited by B.B. Singh, D.R. Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Co pub- lication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (nRCAS). I1TA, Ibadan, Nigeria. Walburger, A. and K. Foster. 1994. Using censored data to estimate implicit values of swine breeding stock attributes. Review of Agricultural Economics 16(2): 259-268. 366 Digitized by Google 5.2 I ndustrial potential of cowpea C. Lamboe Abstract In West Africa, cowpea is a popular legume occupying an important position in local food habits in such countries as Burkina Faso, Niger, and Nigeria where it is a staple food. This observation is, however, not valid in some other West African countries such as Cote d'Ivoire where people are traditionally high consumers of tubers and cereals. Cowpea is also important in some other regions of Africa (Kenya) and the world (India and Brazil), reinforcing its business potential for industrial food companies. The nutritional value of cowpea is mainly from its protein and carbohydrate content. Cowpea's high protein level represents its major advantage for use in nutritional products and compensates for the high proportion of carbohydrates often ingested in African diets. Cowpea is rich in lysine; conse- quently it can be used to balance cereals. Its deficiency in sulphurous amino acids is addressed when it is combined with milk protein and/or cereals known for their high methionine and cystine content. The antinutritional problems linked with the presence of tannins or trypsin inhibitors can be easily avoided with appropriate dehulling and heat treatment. The flatulent sugars are not a limiting factor; cowpea has a lower raffinose content than soybean. Taking into consideration the good nutritional value and the positive image of cowpea for African consumers, it can be concluded that cowpea could be a good source of protein for industrial product manufacturing. It can also be stated that its protein content can still be improved by breeding, using the existing natural diversity in cowpea. The major constraints to the industrial use of cowpea by food companies in Africa include the lack of reliable statistics on production, the strong price fluctuations during the year, the low quality ofthe raw material in terms of physical defects, and the lack of primary processors. Again, the comparative price with soya, which is influenced by world market price, is sometimes not in favor of cowpea. The low quality of cowpea available in the open market is due mainly to the high level of physical defects, but the problem of pesticide residues is also critical for this raw material, which is highly susceptible to insect damage. Processed ingredients based on cowpea are not readily available, forcing food companies to invest in primary processing. There are clear opportunities to develop industrial products using cowpea as a source of protein, but strong support from governments is necessary to promote and to orga- nize the supply chain and the primary processing. The support from research institutions for programs aiming to increase the protein content of cowpea is also required. Introduction Food habits in West Africa are mainly based on tuber crops (cassava, yam) and cereals (maize, rice). Although they have a high nutritional value, grain legumes are a minor component offood diets. Tentative efforts have been made to introduce soybean in African food habits and farmer activities, but with little success. Even in Nigeria, where the annual 1. Project Manager-Agricultural Raw Materials, Nestle Research Center, Abidjan, 01 BP 50, Abidjan, Cote d'Ivoire. 367 Digitized by Google Cowpea postharvest and socioeconomic studies national production of soybean is around 326 000 t, it is still considered an industrial crop, difficult to cook, and with an undesirable taste. Unlike soybean, cowpea is appreciated and is gradually assuming an important position in the food habits of West Africans. Different traditional African meals, foods, and seasonings are prepared from cowpea, among them homemade weaning foods. Cowpea is bought in the market, and processing (soaking, dehulling, milling, etc.,) is done at home. In Africa, cowpea is the most popular legume and the largest part of world production originates from this continent. Cowpea is adapted to stressful environments where other crops either fail or do not perform well. It is a food security crop in the semiarid zone of West and Central Africa (WCA) which ensures farm household subsistence food supply even in dry years. Recently, FAO (1996) estimated the world production area as 5.6 mil- lion ha, of which at least 90% is in West and Central Africa, and the annual world grain production is estimated at 2.7 million tonnes. There are some indications that recent F AO statistics underestimate the production. Cowpea scientists indicate a much larger production area of 12.5 million ha, with 8 million ha (64%) in WCA and an annual world grain production of3 million tonnes (Singh et al. 1997). If the production area is nearly 12 million ha, this is about 45% of the area recorded for dry beans. Production areas are also spread all around the world, making cowpea a global crop. The important cowpea growing countries in WCA are Benin, Burkina F aso, Cameroon, Chad, Ghana, Mali, Niger Republic, Nigeria, Senegal, and Togo. Some countries in East Africa (e.g., Kenya which produces around 250000 tlyear) are also important for cowpea production. National production increased rapidly in Nigeria from 1576000 t in 1993 to 2181000 t in 1999. The same trend was noticed in Burkina Faso (79797 tin 1994 to 327000t in 1998) (Figs. 1 and 2). This reflects consumer appreciation and national support for cowpea. Nutritive value of cowpea The chemical composition of cowpea is similar to that of most edible legumes. It contains about 24% protein, 62% soluble carbohydrates, and small amounts of other nutrients. Thus, most of its nutritional value is provided by proteins and carbohydrates. Many references are available on cowpea nutrient content (Table 1). The high protein content represents a major advantage in the use of cowpea in nutritional products, for infant and children's food, and to compensate for the large proportion of carbohydrates often ingested in African diets. 2500 ; => 2000 => => "- c ~ 1500 u 1000 :::I "'" e 500 .... 0 1993 1994 1995 1996 1997 1998 1999 Figure 1. National production of cowpea in Nigeria. Estimated annual production for 1998 and 1999. Source: Central Bank of Nigeria 1997 and 1998. 368 Digitized by Google 400 ; = 300 = e c 200 c '= U ::::I 100 ", E .... 0 1994 1995 1996 Figure 2. National production of cowpea in Burkina Faso. Source: SAFGRAD unpublished data 1999. Table 1. Cowpea grain composition (gil 00 g). Industrial potential of cowpea 1997 1998 (Bliss 1975) (Omueti and Singh 1987) (Nielsen et al. 1993) Number of cultivars analyzed 8 37 100 Protein 24.1-25.4 20.30-29.05 22.9-32.5 Crude fiber 5.0--6.9 2.7-5.8 Carbohydrate 60.8--66.4 59.7-71.6 Soluble sugar 5.9-8.3 Starch 39.1-54.9 Ash 3.4-3.9 2.9-3.9 Fat (g/100 g) 1.1-3.0 1.66-2.82 1.4-2.7 Since cowpea is a major source of protein in the diet of many people in sub-Saharan Africa, any effort made to increase the level of protein in the seeds would improve the quality of the diet of the population. There are indications that progress can be made through appropriate selection of parental lines for crossing. Considering the existing vari- ability mentioned in literature, the protein content in cowpea can probably be increased by up to 33%. Cowpea is especially rich in lysine, but it is deficient in sulfurous amino acids. Com- pared to other legumes, methionine and tryptophan levels are high. Except for total sulfu- rous amino acids, and to a lesser extent isoleucine, levels of essential amino acids are at least as high as those in soybean (Table 2). The protein quality of cowpea products can be increased significantly by combining it with milk protein and/or cereals known for their high methionine and cystine content (Boulter et al. 1975). Cowpea contains a higher level of flatulent sugars than that found in soybean but its raffinose content (most flatulent sugar) is lower than that in soybean (Table 3). Grain soaking before dehulling and milling decreases levels of the flatulent sugars. Therefore, their presence in cowpea should not limit its use. Levels of trypsin inhibitors are about half the values observed on soybean and are inacti- vated by a heating process. Also, the phytate content of cowpea is half that of soybean. Cowpea procurement Raw materials procurement is crucial in Africa since supply chains for agricultural raw materials are not adapted to industrial needs and specifications. The problems faced are the 369 Digitized by Google Table 2. al amino cowpe , egumes, eals. Protein Lysine Met. Met.+cys. Tryptophan Flour x 6.25) N) (gl1 (gl16 g 16 gN) Cowpe 21.6 1 2.3 .2 Lentil seeds 19.5 8.0 0.94 2.0 1.1 Mung bean 24.9 8.1 1.6 2.7 1.6 Chickpea 22.4 6.9 1.3 2.7 0.8 Millet 10.0 2 4.0 .8 Soybea 40 1 3.0 .4 Wheat 10.5 1 4.4 .4 Milk 34.3 7.8 2.5 3.3 1.4 FAD infant 6.6 4.2 1.7 FAD 2-5y 5.8 2.5 1.1 --- Source: d deWe 1999. Note: Met = Methionine Cys = Cystine Try = Tryptophan Phenyl lalanine Leucine Isoleucine Phenyl. + Tyr Histidine N) gl1 (gl16 gl16 gN 4 9. 3.2 9.0 5.1 9.5 3.0 9.25 5.3 10.2 3.5 7.4 5.8 8.2 2.7 5 7. 1.9 4 9. 2.7 4 8. 2.3 9.8 6.4 9.9 2.6 9.3 4.6 7.2 2.6 6.6 2.8 6.3 --- Threonine Valine 6gN) 9 4.8 5.9 4.2 6.0 3.5 4.95 2 3 3 4.6 6.9 4.3 5.5 3.4 3.5 ~ ~ ~ -0 o '" S- ~ ..... ~ ~ ::J a... '" o (") o· ('[) 8 ::J o :3 n' '" 2 ~ Industrial potential of cowpea Table 3. Antinutritional factors and flatulent sugars in cowpea and soybean. Flatu lent sugarst(%) Raffinose (%) Trypsin inhibitors (TIU/mg N) Phytates (%) --Cowpea-- Dehulled seeds (in-house analyses results) 5.5--6.6 0.36--0.48 228--646 0.37--0.54 • raffinose + stachyose + verbascose. Source: Gaudard de Week, personal communication. Literature value 3.0--7.8 0.4-1.2-2.5 122-440 0.44-1.7 Soybean 2.9-5.5 0.7-1.0 390--1030 1-1.5 high fluctuations in price and quality, the difficulty in identifying reliable intermediaries, and the absence or low development oflocal raw material primary processors. The informal sector is a major force in the food crops market in Africa and, usually, the proportion of the food crop production, which is commercialized, is estimated at 15%. Cowpea procurement strategies The absence of primary processors of grain in the supply chain forces food industries interested in cowpea to look for grain procurement. Different approaches could be con- sidered ranging from purchasing on the market to direct procurement from farmers to contract growing. Studies carried out in Cote d'!voire and Burkina Faso in a collaborative program between Nestle R&D Center, Abidjan, and SAFGRAD (Semi-Arid Food Grains Research and Development) demonstrated the feasibility of the contract growing option. The profitability of cowpea for farmers is guaranteed when they grow improved varieties using fertilizer and insecticide. The break-even point was estimated at 300 kg/ha. Furthermore, the daily income generated by growing cowpea is comparable to the average local daily wage. Hence, it was concluded that contract growing is interesting if a specific variety is required for product manufacture, giving also the advantage of a better quality compared to that of the open market option. Nevertheless, cowpea is available in large quantities on the open market of the different West African countries, a good argument in favor of this procurement strategy. But price and quality fluctuations during the year are important constraints requiring investigations to adapt the strategy accordingly. The main varieties available on the open markets in West Africa are white with black eye. One of them in Nigeria is named Kanannado. Cowpea price fluctuations Close market monitoring for cowpea price registration was carried out in Cote d'!voire and Nigeria. During the 1995/1996 season in Cote d'!voire, it appeared that cowpea price would double within a year with a maximum in September and a minimum just after the harvesting period during November and December. These seasonal fluctuations reduced recently, probably as a result of the rapid increase in national cowpea production in Burkina Faso which is a natural supplier for Cote d'!voire. Cowpea price in November at wholesale level in Abidjan decreased from 250 FCFAlkg in 1995 to 180 FCFAlkg in 1999 (Fig. 3). 371 Digitized by Google Cowpea postharvest and socioeconomic studies Price (FCFAlkg) 300,-------------------------------------------------, 250 200 150 100 50 O~~,_--,_--,_--,_--,_--_,--_,--_,--_,--_,--_,--~ ,,;~ .. OJOJ ... ;:,~ <-.e~ 0- o~ l"' <:::l<:::l '" ' "? 0 ~ <:) '\~<::- Figure 3. Cowpea price fluctuation during the 1999/2000 season at the Abidjan wholesale market. For Nigeria, in 1999/2000, cowpea price fluctuated in six different markets. The average of delivery price to Lagos is indicated in Figure 4. Prices were lower during the harvest period ( October-December) and higher from April to June in 1999. Prices steadily decreased from September 1999 until December 1999, and then increased in January and February 2000. Quality evaluation Kanannado samples were collected from different markets in Nigeria and at different periods during the 1999/2000 season (Table 4, Fig. 5). Laboratory results of samples collected monthly from each location indicated that the average protein content of white cowpea is 23.6%. Humidity levels ranged between 6.6% and 13.5% depending on the period of the year. However, high defect counts, 12.9% on the average, were noticed in the samples collected. The poor quality of cowpea available on the markets is a major constraint for industrial use considering that 3.7% will be lost (stones and waste) and that 9.2% are defective grains. It was estimated that only 40% of cowpea available on the open market is acceptable for industrial use in relation to specifications for physical defects. Considering that cowpea grain is highly susceptible to weevils (Callosobruchus sp.), it was also important to evaluate the level of pesticide residues in the raw material. Cowpea purchased in April or May 1999 showed higher levels of the pesticide residues than normal, especially for chlorpyriphosethyl and pirimiphosmethyl (Table 5). The concentrations of these two organophosphates are such that will result in a calculated exposure exceeding safety standards. The case of chlorpyriphosethyl is unclear, as this pesticide is usually not used for grain storage but for soil treatment and building maintenance. Pirimiphos- methyl is a common pesticide used for grain storage with well known commercial forms (Actellic ). The level of contamination with pesticides was higher for samples purchased in April-May than for samples purchased during the period following harvesting (Nov-Jan). This indicates that the pesticide residues could be a consequence of long storage in non- adapted conditions. It can be suspected that postharvest treatments applied to cowpea are not performed correctly and this leads to variation in the quality of the raw material. 372 Digitized by Google Industrial potential of cowpea Price (N/kg) 70 60 50 40 • 30 20 10 OL, __ -=--~--~----.---~--.---~------~--~--,,--~~ OJOJ «.~ _ ~~ ~~.. _ ~~ ",;:,<::- ,,~ .... ,;:,~ <-.e~ 00 ~o~ <:::l<:::l «.~ ~ ~ ~ , ~ Figure 4. Cowpea price fluctuation-Average of delivery price to factory in Lagos. Table 4. Cowpea quality evaluation (%) overall. Protei n (DM) Moisture Total defects Broken Holes Stones Colored Foreign varieties Waste Average 23.64 8.74 12.89 2.08 5.75 0.16 0.78 0.60 3.52 Waste (27%) Foreign variety--------\ (5%) Colored (6%) Stones (1%) Minimum 22.3 6.6 5.0 0.0 0.0 0.0 0.0 0.0 0.3 Broken (16%) Holes (45%) Maximum 26.1 13.5 20.9 6.2 17.1 1.2 6.7 2.9 8.6 Figure 5. Defects category in Kanannado variety (overall mean). 373 Digitized by Google Cowpea postharvest and socioeconomic studies Table 5. Aluminum and pesticide residues in cowpea-different periods and markets in Nigeria. Period of Norms May April November January the year codex+ 1999 1999 1999 2000 Five Five other other Nigeria Nigeria Markets Katsina markets Katsina markets Aluminum Ppm 10 17 9.2 8.2 12.0 11.3 alpha- Ppb 19 chlordane Gamma Ppb 1000 (db) 210 350 HCH 93 13.7 Chlorpy- Ppb 400 (fb) 1260 940 riphosethyl Diazinon Ppb 100 (g) 98 99 Pirimi- Ppb 1000 (fb) 1560 910 485 168 11 phosmethyl Triazophos Ppb 290 Profenofos Ppb 770 Cyperme- Ppb 100 (fb) 910 58 thrin Deltamethrin, Ppb 1000 (db) 19 31 Decamethrin Permethrin Ppb 100 (db) 200 61 +db = dry bean; fb = fresh bean; g = groundnut. Conclusions and recommendations Cowpea is a source of good quality protein appreciated by Africans. The rapid increase in volume of production is a proof of the acceptance of this legume. However, increased production is affecting the prices of cowpea. The industrial use of cowpea is facing some major constraints: primary processors do not exist, forcing food industries to process the grain; the quality of the grain available on the open market is poor, with a high percentage of physical defects and a risk of pesticide residue contamination; strong price fluctuations along the year are forcing procurement during a short period; the protein content of available cowpea is low compared to that of soybean. Considering these constraints, it can be suggested that national and international research programs on cowpea should be encouraged to promote and support the devel- opment of primary processing of cowpea for food industries. Training and technical support to farmers and wholesalers on the proper application of pesticides is also a priority in order to ensure the safety of the product. The development of varieties resistant to weevil infestation would ensure high quality grain with low levels of pesticide contamination. Increased protein content would make cowpea more attractive for the African food industry as it would then compete with soybean. Furthermore it would have a positive impact on consumers' health through improvement of their diet. 374 Digitized by Google Impact of cowpea breeding and storage research in Cameroon Lowenberg-DeBoer, J. 1999. Trip report: Zimbabwe, Mozambique, Benin, 15 February to IO March 1999. Bean--Cowpea CRSP, East Lansing, MI, USA. Masters, w., B. Coulibaly, D. Sanogo, M. Sidibe, and A. Williams. 1996. The economic impact of agricultural research: a practical guide. Purdue University, West Lafayette, IN, USA. Murdock, L. and R Shade. 1988. I 988 Annual Report, Bean--Cowpea CRSP, Purdue UniversitylIRA- Cameroon Project. Murdock, L., R Shade, and Z. Boli. 1989. 1989 Annual Report, Bean--Cowpea CRSP, Purdue UniversitylIRA-Cameroon Project. Murdock, L.L., RE. Shade, L.w. Kitch, G. Ntoukam, J. Lowenberg-DeBoer, J.E. Huesing, W. Moar, O.L. Chambliss, C. Endondo, and J.L. Wolfson. 1997. Postharvest storage of cowpea in sub- Saharan Africa. Pages 302-312 in Advances in cowpea research, edited by B.B. Singh, D.R Mohan Raj, K.E. Dashiell, and L.E.N. Jackai. Copublication of International Institute of Tropical Agriculture (I1TA) and Japan International Research Center for Agricultural Sciences (JIRCAS). I1TA, Ibadan, Nigeria. Muth, RF. 1964. The derived demand curve for productive factor and the industry supply curve. Oxford Economic Papers 16: 221-234. Ntoukam, G. and L. Kitch. Solar heaters for improved cowpea storage. Technical Bulletin. Bean- Cowpea CRSP, East Lansing, MI, USA. National Cereals Research and Extension (NCRE) Project, Plan of Work, Republic of Cameroon, Ministry of Higher Education and Scientific Research (MESlRES), Institute of Agricultural Research (IRA), 1988, 1991, 1992, 1993, 1994. Oumarou, J.-P. 1999. Etude sur Ie circuit commen.:ial du niebe dans la Province de l'Extreme Nord du Cameroun. Undergraduate thesis, University ofNgaoundere, Ngaoundere, Cameroon. Purdue Universityllnstitut de la Recherche Agronomique (PurdueIIRA). 1987. Detailed Project Annual Report, Bean--Cowpea CRSP, Purdue UniversitylIRA-Cameroon Project. Purdue Universityllnstitut de la Recherche Agronomique pour Ie Developpement (PurduelIRA). 1992. Detailed ProjectAnnual Report, Bean--Cowpea CRSP, Purdue UniversitylIRA-Cameroon Project. Purdue Universityllnstitut de la Recherche Agronomique pour Ie Developpement (PurduelIRA). 1993. 1993 Detailed ProjectAnnual Report, Bean--Cowpea CRSP, Purdue UniversitylIRA-Cam- eroon Project. Purdue Universityllnstitut de la Recherche Agronomique pour Ie Developpement (PurduelIRA). 1997. 1996 Detailed Project Annual Report, Bean--Cowpea CRSP, Purdue UniversitylIRAD- Cameroon Project. Purdue Universityllnstitut de la RechercheAgronomique pour Ie Deveioppement(PurduelIRA) and the National Cereals Research and Extension Project Testing and Liason Unit (NCRE/TLU). 1990. On-farm tests of storage technologies: 1990-199I. In Supplement to the 1990 Detailed Project Annual Report, Bean--Cowpea CRSP, PurduelIRA-Cameroon Project. Rose, F. 1980. Supply shifts and the size of research benefits: comment. American Journal ofAgri- cultural Economics 62: 834-837. Schultz, MA. 1993. Economic assessment of cowpea grain storage techniques: a case study of North Cameroon. Unpublished MSc Thesis, Dept. of Agricultural Economics, Michigan State University, East Lansing, MI, USA. Sterns, J. and R Bernstein. 1993. Assessing the impact of cowpea and sorghum research and exten- sion in northern Cameroon. Michigan State University, International Development Working Paper No. 43. Michican State University, East Lansing, MI, USA. Vijverberg, W.P. 199 I. Profits from self-employment: the case of Cote d'l voire. World Development 19(6): 683-696. Wolfson, J. 1990. Analysis of cowpea production and storage methodologies used by small farmers in northern Cameron. PurduelIRA Cowpea Storage Project, January 1990. 423 Digitized by Google 5.6 Identifying cowpea characteristics which command price premiums in Senegalese markets M. Faye1, J.L. DeBoer2,A. Sene3, M. Ndiaye3 Abstract Since the 1980s cowpea has become an alternative cash crop in northern Senegal. However, in 1997, surveys showed that farmers often sold their crop at unprofitable prices. The hypothesis was that farmers could improve selling prices if they pro- duced cowpea with the characteristics demanded by consumers. To identity those characteristics, data were collected in six markets from January 1998 to December 1999. Regression analysis was used to estimate a linear relationship between price and grain characteristics. The results showed that consumers were willing to pay premiums for grain size (119 FCFA/IQO g) but discount price for number ofbruchid holes (0.62 FCFAlhole), red eye (40 FCFA/kg), red skin (30 FCFAlkg), and smooth skin (21 FCFA/kg). Introduction Cowpea (Vigna unguiculata [L.] Walp.) commonly referred to as black eyed peas in the US, is an important source of nutrition in West Africa. This crop serves to bridge the hunger gap between the planting and harvesting periods of the main food crops. With 10% of the area cultivated, cowpea is the third most important crop in Senegal after millet, the main food crop, and peanuts, a major cash crop. Cowpea was traditionally grown in Senegal for food (Tall 1991 ). However, since 1985, following several years of poor peanut harvests, cowpea has become increasingly viewed as an alternative cash crop. This is particularly true in the northern part of the country where a short rainy season and an annual rainfall of less than 300 mm does not favor peanut production. For this reason, the Senegalese Agricultural Research Institute (lSRA) is currently engaged in a research program focusing on the breeding and dissemination of early matur- ing (less than 45 days) and high yielding varieties of cowpea. The purpose of this program is to improve production and to promote cowpea marketing by providing earlier varieties, which could be marketed before the traditional varieties and receive a higher price. In 1997, as part of the research program, surveys were done to assess the impact of the new varieties on production and cowpea marketing. The results of the survey revealed that the area planted to cowpea had not expanded as expected. In addition, the farmers were found to still be selling their production surplus at prices below their cost of production. Problem statement Farmers face difficulty in selling their production surplus at profitable market prices. In order to address this problem, it is necessary to know how the buyers value the different 1-3. ISRAlCNRABP 53 Bambey, Senegal. 2. Department of Agricultural Economics, Purdue University 47906 IN, USA. 424 Digitized by Google Identifying cowpea characteristics which command price premiums in Senegalese markets cowpea qualities. Nongovernmental organizations (NGOs), which are involved in new cowpea variety extension, ISRA, and cowpea producers and authorities in Senegal are all interested in this problem as they require more information on cowpea market price and grain characteristics. The specific objective of this paper is to measure the relationships between price and grain characteristics in order to identify priority areas for the breeding program. Literature review and theoretical model Based on the economic principle that product demand stems from the utility provided as a function of its quality characteristics (Berndt 1991), a hedonic pricing model was used to analyze the data. Since its introduction, numerous economists have employed hedonic pricing models as a tool for estimating the price-quality relationships of commodities over time or through cross-sectional data analysis. One of the earliest examples of this methodology dates back to 1974 when Sherwin Rosen (1974) first sketched on scratch paper a model of product differentiation based on the hedonic hypothesis that goods are valued for their utility-bearing attributes. In his model, Rosen used observed product prices and the specific number of characteristics associated with each good to define a set of implicit or hedonic prices. Brorsen et al. (1984) further contributed to the acceptance of this analytical tool by studying market acceptance of rough rice. The Brorsen study revealed that several fac- tors are involved in the distinction of rough rice. He evaluated the ability offederal grain inspectors to explain the factors that led to the grade classification and estimated the discount associated with each factor using a hedonic price model. Espinosa and Goodwin (1991), with the same motivation as those authors cited earlier, employed a profit maximization framework and hedonic pricing model to assess the impact of wheat characteristics on market price. This paper follows the framework outlined in the Espinosa study with one notable exception: the hedonic-pricing model used here on cowpea does not incorporate the pos- sibility of processing cowpea since data based on these characteristics are unavailable. Despite the absence of processing attributes, the general theory of hedonic pricing as developed by Espinosa closely relates to the current study in one important way: it follows a consumer goods approach and considers individual characteristics as utility providing attributes in a utility maximizing problem. Under this approach, an agricultural product is desired because of its particular quality characteristics. From this, it is assumed that cowpea consumers behave as utility maximizing agents. From the first order condition of the utility maximization problem can be derived the general form of the hedonic price model. This function would be expressed as a regres- sion of the following form: Pit = ao ~ k = 1 ~k Zitk Where Pit = Per unit price of cowpea ao = Intercept ~k = Marginal value of characteristic k Zitk = Amount of characteristic k in good i at time t Some authors use the semi-logarithm functional form or combine linear and quadratic time trends and dummy variables. In this paper, the linear model is used because of its 425 Digitized by Google Cowpea postharvest and socioeconomic studies theoretical interpretation (discoWlts orpremiwns) and also because it is easier to explain to market participants. Source of data Data were collected from six markets (Fig. 1) chosen according to their location and volume of cowpea sales. In each market, samples were bought from five different vendors every month from January 1998 to December 1999. The choice of the vendors at a given market was done randomly. For each sample, the following variables were noted: market price, skin texture, skin color, eye color, weight of 100 grains, length, width, and number of bruchid holes per 100 grains. Also observed were the locations (markets), the gender of the sellers, and the selling period (month). Econometric method Regression analysis was employed to estimate a linear relationship between price and grain characteristics. Generalized least square (GLS) was used to correct for temporal error correlation across the cross sectional observation. The dependent variable was price (P) in franc CF A per kg. The independent variables were: average weight (W) of 10 grains in mg, number of bruchid holes (NH), skin color, skin texture, eye color, and grain size (refers to average length of 10 grains multiplied by average width of 10 grains). Mauritania Sainl..ttllilll : • FII4 Bote. ClIp I~ Dattare Thlw Mali &.rD. "Gambi~ I .. Guinea Bissau Guinea Figure 1. Market locations. 426 Digitized by Google Identifying cowpea characteristics which command price premiums in Senegalese markets The variables (month, market [location] and gender of sellers were handled as dwnmy variables as well as all the other qualitative variables. A base variable was defined for each group of dummy variables (Table 1). Expected signs for estimated parameters The common characteristics were those which could be taken into account in breeding programs, or were generally used to determine the value of cowpea grain. These variables included: number of holes per 100 grains, grain skin and eye color, grain skin texture, and grain size. The number of holes (NH) refers to the level of insect damage and is expected to have a negative sign. The signs for white skin color and rough skin texture are expected to be positive. Grain size (GSIZE) would have a positive sign because consumers prefer large seeds for their sauce or rice. Also because grain size refers to the quantity of flour, processors would be willing to pay a premium for it. Results This part will begin with an overview of the cowpea marketing and market structure in Senegal. Then it will discuss, respectively, the type of sellers and the distribution of skin color, eye color, and skin before reporting the results on hedonic price estimations. Market structure Compared to peanut for which the market is supervised by the government, cowpea has a competitive market without any government intervention. Relationships among market participants are based on informal agreement and on some ethics they define. In moving cowpea from farm gate to urban consumers, different linked steps are identified in terms of market participants (Fig. 2). Producers are the first cowpea suppliers on the market channel. They supply on aver- age 28000 tonnes per year, 71 % of which is supplied from northern Senegal (DSA 1998). These producers sell their produce directly to the collectors, wholesalers, processors, retailers, or to the consumers. Collectors, on the other hand, are individual entrepreneurs. They buy cowpea from market to market in the production area, and in return supply wholesalers, retailers, and processors. But to avoid any competition with the retailers (who buy from them) they don't sell to consumers. Most of the time, collectors use their own money to make their transactions. They can also get money from wholesalers based in towns depending on some specific relationships (relative or close friend). Collectors don't usually have a specific storage space except if their home village or town is close to the collecting area. In this Table 1. Base variables. Variables Skin color Skin texture Eye color Market Month Gender 427 Base White Rough White Sagatta October Female Digitized by Google Cowpea postharvest and socioeconomic studies Producers ----------, * ! r----- Collectors -----, 1----4.. Wholesalers ---+-..., ~ Retailers Processors Consumers .. Figure 2. Market participants and relationships. "The size of the arrows indicates the importance of the quantities of cowpea exchanged between two groups of participants. case they use one room in the house as a storage area and keep their cowpea in metallic drums or in plastic bags. The processors own small units where they process cowpea into flour or other cowpea- based products. Depending on their location, they buy their input from collectors, whole- salers, or sometimes directly from producers. Wholesalers and retailers are shopkeepers with the difference that wholesalers special- ize in one or two agricultural products. Consumers are the last users and account for all cowpea buyers whose objective is to consume cowpea as food. As shown in Figure 3, cowpea traders go from the production area to the rural or urban markets from where the product can be exported to neighboring countries like Gambia and Mauritania. Any movement in cowpea leads to financial charges, the most important being those for transportation and storage. In Senegal, loading cost is 25 FCFA per 50 kg bag in the rural market. Producers use carts and trucks to transport cowpea from their area to the rural market. They often pay between 50 and 100 FCFA per 50 kg bag. The same price is charged from rural to urban markets, no matter the distance. To transport cowpea outside Senegal, importers from Banjul and Mauritania use trucks. The transportation cost is fixed per load and varies between 50000 and 125000 FCFA depending on the volume of cowpea transported. The bags used to store the cowpea cost 428 Digitized by Google Identifying cowpea characteristics which command price premiums in Senegalese markets Production area (Sagatta and Mpal) Rural market ..... I--+--.u'''l m"ke' 1 t Foreign countries (Gambia, Mauritania) Figure 3. Spatial pattern of cowpea marketing. 100 FCFA per Wlit and can contain 50 kg. As shown in Figure 4, farmers face variable prices (from 46 to 780 FCFA). Prices are lowest between October and December, which corresponds to the harvest period. Data overview Types of sellers Male sellers play an important role in selling cowpea in Dakar and Nioro. More than 91 % of the sellers in Dakar are men against 100% for Nioro (Table 2). In Bambey and Mpal on the other hand, females represent most of the sellers, with 76 in Bambey and 74% in Mpal of the population interviewed. It is only in Sagatta that equity between men and women selling cowpea was observed. Wholesalers are present only in Dakar (Tilene) and in Mpal. Skin texture Two types of skin texture, smooth and rough, were observed. Except for Castors where the percentage of cowpea with rough skin was 39%, Figure 5 shows that 50-65% of the grains had rough skin. Skin color In Mpal, 52% of the cowpea sold was white while in Bambey, Nioro, and Sagatta, the pro- portion of cowpea with mixed color was above 50% of the samples observed (Fig. 6). That the white-fleshed variety is the dominant in Mpal seems to confirm that buyers from Mauritania prefer the variety. This assumption will be tested in later studies. Eye color Black and red eye colors were observed with the black -eyed color representing more than 50% of the sample. In Bambey, Castors, and Tilene, 70-72% of the cowpea sold had black eyes (Fig. 7). According to the sellers, some consumers in Dakar prefer the black cowpea and particularly one local variety called Baye Ngagne because of its taste. Table 3 shows that price varies between 46 and 780 FCFA, with a mean of 260 FCF A. The highest level of damage observed was 100 holes for a sample of 100 grains due to the high rate of use of metallic drums to store cowpea. Throughout the survey, this level was observed only once in Dakar. The factors weight (W) and grain size refers to the quantity of flour of a seed sample. They, respectively, have a mean of 18 g and 54 mm2• 429 Digitized by Google Cowpea postharvest and socioeconomic studies 250 200 i:i!: 150 -U Q,j IJ .;: CI. 100 50 ~--__ --------------------------------------------~ Jan Feb Mar Apr May Jun Months Figure 4. Cowpea price variations in 1999. Table 2. Types of sellers interviewed (%). Castors Tilene Female producer 0 0 Female retailer 0 0 Total female sellers 0 0 Male producer 0 0 Male retailer 100 83 Total male sellers 100 83 Wholesaler 0 17 o Figure 5. Skin texture. 430 Jul Aug Sep Oct Nov Dec Bam bey 48 28 76 24 0 24 0 Nioro Sagatta 0 13.5 0 36.5 0 50 3 36.5 97 13.5 100 50 0 0 Mpal 38 26 64 13 19 32 4 Dsmooth ILIRough Digitized by Google Identifying cowpea characteristics which command price premiums in Senegalese markets Bambey Nioro Castors Figure 6. Skin color. 90 80 70 60 50 F'l;tlre 7. Eye color. Tilene Markets Markets Sagatta Mpal mOthers III Black ..:IRed 1:1 White [.'SIRed I:'l§IBlack The regression model: This model measures the explanatory power associated with all the variables listed as grain characteristics. The hypothesis tested is whether or not the information conveyed by dummy variables and all the quantitative variables jointly can explain the observed price variation. The output of the regression model is in Table 4. These results suggest that the model explains the weighted variation in price. The F -test rejects the hypothesis that there is no relationship between price and 431 Digitized by Google Cowpea postharvest and socioeconomic studies variables in the model. The model shows that the standard characteristics including two quantitative (NH, GSIZE) and five dummy variables representing the main grain characteristics (REDSKIN, BLSKIN, OTHERSKINC, SMOOTH, and REDEYE) are meaningful for buyers and therefore should be important for cowpea breeders. However, from the individual T-test, the relationship between number of holes (NH) and price is not significant. The model also shows that the selling period has an impact on the market price. Octo- ber (MO 1 0), which corresponds to harvesting time, witnesses a price collapse. Selling between January and September would lead to an increase in price with respect to the reference period. The best periods to sell cowpea appears to be from June to September. This can be explained by the fact that June and July correspond to the planting time when the demand for seeds becomes very high while in August and September most of the reserves would have been consumed and farmers are yet to harvest. Selling in November (MOll) or December (M012) would, respectively, decrease price received by about 25 and 40 FCF A. However from the individual t-test, November (MO 11) does not have a significant relationship with price. The impact of factor location (market) on price is significant and positive for all markets. For example, in Bambey, selling price increased by 20 FCFA compared to the price in Sagatta market. This is expected because when you move from the production area to the consumption area, price should increase. The positive effects of location on market price would likely reflect the difference in cost of transportation from the base market, Sagatta in the production area, to the other locations. Conclusion This analysis has considered hedonic price models using the main cowpea physical charac- teristics and other variables that can influence the market price of cowpea such as location (market) and selling period (month). The results show that buyers are willing to pay a premium for grain size and white skin color but discount price for any other skin color and number of holes. It shows also that selling cowpea between January and September Table 3. Univariate statistics. Variable p W NH GSIZE Mean 260.7978 18.6207 6.3260 54.0288 P = Price/kg in FCFA. W = Average weight of 10 grains. NH = Number of holes/lOa grains. GSIZE = Grain size. Std Dev 141.7557 4.3331 9.5609 9.7536 Minimum 46.0000 10.0000 0.0000 4.7000 Maximum 780.0000 94.0000 100.0000 93.8000 would increase the price received by farmers but that November and December would be bad periods to market cowpea. Bambey had the greatest positive effect on price even though the impact of the variable market was not significant. These results are useful for breeders, policymakers, and farmers for addressing cowpea price variation issues. The implications for cowpea breeders would be to focus on a new breeding program incorporating white skin and large grain size as main characteristics since buyers are 432 Digitized by Google Identifying cowpea characteristics which command price premiums in Senegalese markets Table 4. Parameter estimates. Variable OF Estimate Std Error T Stat Prob >ITI INTERCEPT 224.7531 25.0001 8.9901 0.0001 NH -0.6222 0.3484 -1.7856 0.0747 GSIZE 1.1910 0.3382 3.5213 0.0005 MSELLER 74.8358 11.3778 6.5774 0.0001 REDSKIN -31.8342 9.1192 -3.4909 0.0005 BLSKIN -4.7575 10.6163 -0.4481 0.6542 OTHERSKC -17.7121 9.7742 -1.8121 0.0705 SMOOTH -21.2304 8.4291 -2.5187 0.0120 REDEYE -40.5880 29.8127 -8.0700 0.0001 BLEYE -3.2755 9.4980 -0.3449 0.7303 MOl 51.7848 14.5992 3.5471 0.0004 M02 55.0610 14.6356 3.7621 0.0002 M03 68.1134 16.1236 4.2245 0.0001 M04 52.0172 16.7257 3.1100 0.0020 M05 91.3192 16.3468 5.5864 0.0001 M06 118.4365 15.9788 7.4121 0.0001 M07 119.8292 15.8663 7.5524 0.0001 M08 115.3758 16.5178 6.9849 0.0001 M09 127.7584 19.0388 6.7104 0.0001 MOll -22.3698 15.7593 -1.4195 0.1563 M012 -38.7204 16.1256 -2.4012 0.0166 CASTOR 10.1063 11.3919 0.8872 0.3753 TILENE 1.7355 11.1591 0.1555 0.8765 BAMBEY 20.0246 11.3710 1.7610 0.0787 NIORO 3.9886 12.2334 0.3260 0.7445 MPAL 3.2007 10.7045 0.2990 0.7650 F-test = 56.947 AdjR2 = 69% willing to pay premiums for these characteristics. Also, the breeder would need to consider insect resistance in order to reduce the price discounts due to insect damage. In order to enable producers to sell their cowpea at the best period (May to July), a policy that will lead to metallic drums price subsidy should be put in place. References Berndt, E.R. 1991. The practice of econometrics. Classics and contemporary. Addison-Wesley Journal of Economics: 29: 239-249. Brorsen, W., R.G. Grant, and E.M. Rister. 1984. A hedonic price model for rough rice. Bid/accep- tance markets. American Journal of Agricultural Economics 66: 156-163. Direction des statistiques agricoles, Rapport annuel Repuhlique du Senegal. 1998. Espinosa, IA. and B.K. Goodwin. 1991. Hedonic prices estimation for Kansas wheat characteristics. Journal of Agricultural Economics 16: 72-85. Rosen, S. 1974. Hedonic prices and implicit markets: product differentiation in pure competition. Journal of Political Economics 82: 34-55. Tall, S.G. 1991. Evaluation socio-economique des essais mini-kits. ISRAIDRCSPlProjet CRSP Niehe 1991. 51p. 433 Digitized by Google Digitized by Google