About the International Institute of Tropical Agriculture (IITA) IITA is a major institute in a worldwide network of research and training centers. It was est abE shed as a nonprofit, autonomous corporation in July, 1967, on 1,000 hectares allotted by the Federal Republic of :-.1igeria. with initial funds from the Ford Foundation for buildings and development. Its governing Board of Trustees is an international group whose members include representatives ofIITA's target countries. UTA concentrates research and training in farm­ ing systems and crops of the humid and subhumid tropical zones of the world, with emphasis on maize, rice, cowpeas, cassava, soybeans, yams, sweet. pota­ toes, and cocoyams. It shares improved germplasm and varieties, pro· mising techniques, and other research findings through training programs, cooperating with scientists in many countries, and by posting scientists in national programs of other countries. Principal financing of UTA, like that of the other research and training centerS in the network, is arranged by the Consultative Group on International Agricultural Research (CGIAR)- an informal group of donor countries, development banks, foundations, and agencies. Support for UTA's research and train· ing core program in 1984 was provided by the Canadian International Development Agency (CIDA), Overseas Development Administ ration of the United Kingdom (aDA), U.S. Agency for International Development (USAID), World Bank, Interna tional Fund for Agricultural Development (IFAD), Ford Foundation, Rockefeller Foundation, and the governments of Australia, Belgium, France, India, ltaly, Japan, Netherlands, Nigeria, Norway, and Federal Republic of Germany. Other donors also provide funds to the Institute, particularly to support specific research or training programs. Correct citation: International Institute of Tropical Agriculture, 1985. Annual Report for 1984. Ibadan, Nigeria Published by the International Institute of Tropical Agriculture, Oyo Road. PMB 5320,lbadan, t\igeria IITA ANNUAL REPORT 1984 International Institute of Tropical Agriculture Contents Board of Trustees . ... . ... . .. . . ... .. . . . . . ... . . . . .... ... .. . VI Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. VII Rice Improvement Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Genetic Improvement . . . .. . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . 1 Entomology. . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . .. 12 Pathology .. . .. , , . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . .. 17 Virology .. ..... . ...•. . .... . .. .. .. . .. .... . " . .• . ...... , .... . 20 Agronomy . ... . .. , . . ..•..... . .. , . . . • . . . . . . . . . . . . . . . . . . . . . . . 21 Grain Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 25 NCRE Project . .. ...... . ... . . ' . . .. . .... ' .. .. , . . . . . . . . . . 26 Genetic Improvement .. ' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Agronomy ......... _ . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . .. 29 Maize Improvement Program ..... ..... ... ... ... .... . . , 33 Genetic Improvement ..... . ... .. ..... . .. .. .. ....... . .. . . _ . . . 33 Entomology ... ' . .......... . ...... . , , . . . , .. . . ' ..... .... ' . . .. 43 Pathology . .. , . . .. . ..... , . .. • . , . . •.. • .•. • . . •..... ' • .. . , . . . . 49 Virology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . .. 52 SAFGRAD Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 54 Genetic Improvement . .. . .. .. .. . ... ... ... , . .. . ' . . • . . . . . . • . . . 54 Agronomy . .. . , . . .. . , . . .... . . . , , . . . . . . . . . . . . . . . . . . . . . . . . . .. 56 Entomology ......... .. ....... .. ............ ... .... .. _ . . . . . . 60 NCRE-Cameroon Maize Improvement . .. ... .. .. . . , . . . .. . ... .. 62 Yiid-altitude Maize Improvement ................. , . . . . . . . . . . . 62 Low-alt itude Maize Improvement . .. _ .. ... . .. . . . _. .. _ . . . . . . . . . 63 Grain Legume Improvement Program . . . . . ... , , . . . . .. 67 Cowpeas . . .... . .... . ..... _ .. . . . _. .. _. ........... . . . ........ . 67 Genetic Improvement . . . . _ .. ... ... ...... . ....... . ......... ,. 67 Entomology .. . . _ . . . .. . . . . .. , . . . . . . . . . . . . . . . . . . . . . • . . . . . . . .. 76 Virology . . .. . . . .. . . . . . . .. .. . .. . . . . . . . . . . . . . . . .. . . .. • . . .. . . . 86 Pathology .... _. . , ... _ . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Agronomy ........ . . . . ... . . . . . . .... .. . . . .. . .. . . . , . . . . . . . . .. 87 IITA/SAFGRAD Project . . , . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . .. 87 Genetic Improvement . . . . . ... .. .. _. . .. _ . ..... . ....... _. . . . .. 88 Entomology .. , .. .. , , ..... . .. . .. .. .... .. ..... , ' .. ...• . .... " 90 Agronomy . . ............ , ..... . . .. . , .. .. , .. . . , . . . . . . . . . . . .. 93 EMBRAP AJIITA Legume Improvement in Brazil . . . . . . . . . . . . .. 97 IITAjICRISAT Program in Niger , .... , . .. . . . . . . . .. . . .. . . . .. .. 99 The ICRISAT Sahelian Center . . , .. . . ... . .. , .... , .. .. .... ..... 100 lIT AJIRRI Collaborative Research Program . _ . . . . . . . . . . . . . . . . 102 Soybeans _ , . ... .. ................. . . .. .. . .... .. . _. . . . . . . . . . . 104 Genetic Improvement , ....... . . .. ... ... .. .. _ . _. . .. ... .. , 105 International Trials .. .... . .. . . • , . . . . . . • . . . . . . . . . . . . . . . . . . . .. 105 Entomology . .. . . . _ . .... ..... .... . . .. . . ... . .. . .. . .. . . .. _ . . .. 109 Agronomy .. . .... ... .... .. ...•. ' . . ..... .. ' ... . .• . •. , . . • . . .. 112 Microbiology . .... . . . . • .. . .. . . .. .. . .. . . .• . . . .. .. ..• .... . .... 116 Root and Tuber Improvement Program. . . . . . . . . . . . . .. U9 Cassava ......... . ....... . .......... . ...............•. . ...... U9 Genetic Improvement . . . . . . . . . . . . . . . . . • . . . . • . . . . . . . . . . • . . . .. U9 Virology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 120 Biochemistry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . .. 121 Tissue Culture. . . . . . . . . . . . . . . . . . . . . . • . . • . . . . . . . . . . . . . . . . . . . . 124 Biological Control . . . . . . . . . . . . . . . . . . . . . • . . . . . • . . . . . . . . . . . . . .. 124 Zaire National Cassava Program .....•........ ... .. . ......... 139 Genetic Improvement ...... , . . . . . . . . . . . . . . . . • . . . . • . • . . . . . . .. 139 Entomology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . .. 139 Pathology ... ............ ........ .. .. ... • ......... ......... 141 Crop Management ........... .. . . . .. • .. . .•...... .... ..... . . . 142 Yams .................... .. ... ... ...... •.......•............ 143 Virology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 143 Pathology. . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 144 Biochemistry. . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 144 Sweet Potatoes .............. . . . ..... .. .. ..... . . ......... .. . . 145 Genetic Improvement ... . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . .. 145 Pathology. . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 146 Virology . . . . . .. ... . .. ................. . ... ... .. .......•.... 146 Biochemistry . . .. ........ .... ............. .. ................ 147 Nematology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 Root Crops Research in Cameroon. . . . . . . . . . . . . . . . . . . . . . . . . . .. 148 CNRCIP 1984. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . .. 148 Cassava. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 149 Genetic Improvement .. . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . .. 149 Cocoyams. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 149 Root Crops Research in Rwanda ... ........................... 152 Farming Systems Program. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 155 Soil and Land Management. . . . . . . . . . . . . . .. . . • . .. . . . . . . . • . . .. 156 Soil Physics and Conservation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 156 Soil Fertility ....................................' . . . . . . . . . .. 163 Soil Microhiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 168 Cropping Systems and Weed Management. . . . . . . . . . . . . . . . . . .. 172 Weed Management. .. ... .. ... .. . .... .... .. ... . . .... . ..... . . . 172 Cropping Systems ............ .. .... . .........•.. . .. . . ... . ... 175 Plantain and Cooking Bananas ................ . . . ... ... . ..... 182 Farm Mechanization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 185 On-Farm Research and Socioeconomics. . . . . . . • . . . . • . . . . . . . .. 188 OFR Activities of NCRE Project, Cameroon . .... .. .. ..... . .. .. 200 OFR Activities, Rwanda ...................... . ............ .. 206 International Cooperation and Training Program ... 209 Project Evaluations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 209 Training ....... .. ...................................•.... . . 209 Documentation, Information, and Library. ...... .. ... 217 Lihrary ....... ... .. ... .... ....... . .... ........ ........ . .... 218 Public Affairs . ... ......... .. ......•..........•.......• .... . 218 Publications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 219 Conference and Visitors' Center ............. • .. ........•..... 219 Interpretation and Translations . .... . ..•. ... •... .. .. ...• . .... 220 Research Support Units . ................................ 221 Farm Management . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . .. 221 Genetic Resources Unit . .............. . .... . . . . . ..... . .. . .... 221 Germplasm Distribution ... .. . .. .. ..... .. .... ... .... . ... . .... 221 Biometrics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . 223 Analytical Services Laboratory. . . . . .. . . . .. . . . . . .. . . .. . . . . . . .. 223 lITA Weather . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 225 Ibadan . .......... . .. ..... . ..•. .. ....... . .............. . ... 225 Onne .. ........ . .... . ..... . ........ .... ...... • . ... ...... ... 225 Personnel . ......................... ... .... .. ... .. . . ...... , 227 Collaborators and Trainees . ......... .. ... . ......... . ... 229 Publications . ............. ... ......... . ........... . .... ... 232 Conference and Seminar Papers . . . . . . . . . . . . . . . . . . . . . . . 234 Acronyms and Initials . . .. ...... . ...... .. .. .. . ........ ... 237 Board of Trustees Dr. JohnJ. McKelvey, Jr. (Chairman) Rockefeller Foundation (retired), Richfield Springs, New York, U.S.A. Mr. D.E. Iyamabo (Vice-Chairman) Director , Department of Agricultural Sciences , Federal Ministry of Education, Science and Technology, Lagos, Nigeria. Prof. Ango Abdullahi Vice·Chancellor, Ahmadu Bello University, Zaira, Kigeria. Dr. Leopold K. Fakambi Executive Secretary, International Federation of Agricultural Research Systems for Development/Africa, Cotonou. Republic of Benin. Dr. E.H. Hartmans Director General. IITA. Ibadan. Nigeria. Dr. Hidetsugu Ishikura Director General. Japan Plant Protection Association. Tokyo. Japan. Dr. Klaus J. Lampe Head, Department of Agriculture and Rural Development, German Agency for Technical Cooperation, Eschborn. West Germany. Dr. E. de Langhe Laboratory of Tropical Crop Husbandry. Catholic University, Heverlee. Belgium. Dr. R.C. McGinnis Dean of Agriculture. University of Manitoba. Winnipeg, Canada. Mr. M.S.O. Nicholas Director. Agricultural Services Division. Food and Agriculture Organization, Rome. Italy. Mr. N.O. Popoola Permanent Secretary, Federal Ministry of Agriculture, Water Resources and Rural Development. Lagos, Nigeria. Prof. L.A. Wilson Dean of Agriculture. Faculty of Agriculture. Uni versity of the West Indies. St. Augustine. Trinidad. West Indies. Prof. T.M. Wormer Senior Agronomist. Royal Tropical Institute, Amsterdam, :-letherlands. Kano state, Nigeria, an estimated 13,000 fanners are Foreword growing it. Other varieties that mature in only 60 days are being adopted widely , particularly in rice· based farming systems as soil moisture is critical for Nineteen eighty·four was the year the food crisis in them only about 40 days. Africa climaxed and caught the attention of people Soybean production in tropical Africa has been and leaders throughout the world. Thus, it was the hampered by introduced varieties having poor seed year to take stock of our achievements and our vigor and viability and requiring rhizobia that are challenges. Let me state here that this exercise has not in tropical soils. Through painstaking research, gi ven my colleagues and me every reason to be opti· UTA scientists ha ve developed high yielding soybean mistic about the future of food production in Africa. lines with superior seed longevity that nodulate Our improved cassava varieties out yield tradi· freely with indigenous rhizobia. tional varieties from 2 to 18 times, the highest under UTA's rice program has identified several high· severe disease pressure. The improved varieties, yielding, early·maturing, drought and blast resistant requiring neither fertilizers nor pesticides, are now varieties for the upland areas. Other UTA lines with grown on about 1.5 million hectares in various good seedling vigor and high·yield potential are African countries and are being multiplied in 24 becoming popular in the vast hydromorphic areas. African countries. Highest yields, of course, are in the low land irrigated A revolution in yam production is in the making as culture for which we have developed high·yielding a result of the development, in collaboration with the varieties with excellent grain. National Root Crops Research Institute, Umudike, of Among other exciting 1984 developments are pro· minisett and microsett techniques of seed·yam pro· mises of resistance in maize to stem borers and Striga, duction. Yam acreage has been limited by the scar· both of which cause large losses in Africa. Also city and cost of planting materials. The new tech· promising is alley cropping, a stable and productive nology will remove those constraints and increase alternative to shifting cultivation. yields substantially. A major challenge is how to get new varieties and IITA maize varieties resistant to lowland rust and technologies adopted more rapidly. Technology blight diseases are spreading in many West African transfer is particularly demanding where extension countries. In Nigeria alone, they are grown on about systems are extremely weak or nonexistent. Our 1 mi Ilion hectares with grain yields of 3 to 5 tons/ha, approaches include helping build African insti· compared with 1, tons from the local varieties they tut ions through training and collaborative research. displaced. Maize streak virus is perhaps the most We have trained more than 3,200 persons, 90% from devastating disease of maize in Africa. Newer reo African countries. We are taking our research closer sistant varieties are being distributed and their to potential beneficiaries through intensified on·farm potentia l impact is immeasurable. When maize streak research and decentralization . We opened a station virus epidemics in 1983 and 1984 devastated local in Benin Republic during 1984 for more efl'ecti ve work maize in many parts of Africa, resistant varieties in Francophone African countries, and moved a were remarkably productive and drew widespread cowpea breeder to Niger to work on problems posed farmer attention. by the Sahelian ecology. Hybrid maize, developed at UTA, has given on·farm We strengthened our socioeconomics unit and yields in the forest zone of 4 to 8 tons/ha compared began a major research thrust on the role of women in with 1t-2 tons of the best local maize . In t he savanna agriculture. During 1984, we organized many field zone, hybrid yields are even higher, up to 11 tons/ha. days for farmers , agribusi nessmen, and policy Farmers' enthusiastic reception of hybrid maize led makers. to the incorporation of two Nigerian commercial seed At midpoint of the 1980s, we are proud of past companies in 1984 that will begin producing and accomplishments, grateful for the generous support selling hybrid seed in 1985. of our donors, and looking forward confidently to Our new cowpea varieties have attributes that are the day when Africa will be self·sufficient in food critically important in various ecologies , including production. the Sahelian zone. Varieties with multiple disease resistance. insect resistance, and extra·early ma­ turity have been released in 31 countries in Africa, Latin America, Asia, and the Caribbean. One variety, ERMOND H. HARTMANS in particular, TVx 3236, has been widely adopted. [n Director General Rice Improvement Program The Ric e Improvement Program emphasizes breeding JOInt venture by three international institutions, superior rice lines that resist locally occurring IITA, IRRI, and WARDA. It will be the chief vehicle stresses and give stable, high yields. Three main fortesting genetically di verse materia l in sever al rice ecologies have been targeted to evolve and test new growing areas. rice breeding lines. They are the drought· and blast· Rapport with Nigeria's national progra m was prone uplands, the hydromorphic areas , and the strengthened during 1984. We were additionally lowlands. The program is multidisciplinary with involved in the Coordinated Rice Experimental Trial multilocationa l testing of genetically improved rice (CRET) for the National Cereals Research Institute lines. Management practices for rice under various (NCRI), and the institute's facilities at Badeggi, Bida, ecosystems also received attention. All new lines to multiply r ice seeds for testing programs. emanating from the breeding program were tested for superior grain quality and cooking characteristics. Genetic Improvement Several lines with high yield potentia l and reo The objective of the rice breeding program at lITA is sistance to drought and blast have been identified in to develop improved high yielding varieties with the uplands of Ibadan and in the acid uplands of the r esistance to physical and biological st resses for humid forest regions at Onne. growing under upland, hydromorphic, shallow ITA 305 and ITA 315 have been top yielders under swamp, and irrigated lowland conditions. Grain the harsh uplands of Ibadan, Ikenne, and Onne. ITA quali ty is also considered. 301 and ITA 303 showed excellent adaptation under Africa has many constraints to rice production. acid uplands of the humid forest zone. Blast is the major disease in all growing environ· When we used a toposequence t ransect to test rice ments. Rice yellow mottle virus (R YM V) is a potenti al lines for hydromorphic ecologies. ITA 306 showed threat to development and expansion of r ice in great promise under fluctuat.ing water conditions. hydromorphic, shallow swamp, and irrigated low· TOx 915·101·1 and TOx 960·42-1 are promising for lands. Diopsis and striped borers are a lso important. eventual use in the wetland project of the Farming Among adverse soil problems, iron toxicity is com· Systems Program. manly encountered, and rice in the mid·alti tude areas The breeding program to incorporate tolerance to suffers from low temperature. rice ye 1I0w mottle continues successfully; 33 new Drought, particularly in the dry zone of Africa, is crosses with dwarf and semi·ta ll parents have re­ t he major physical stress for upland rice, while acid sulted in testing large F , populations, and about 300 soils limit production in the high rainfa ll humid F, lines are poised for field tes ting in the RYMV· zones. Important diseases besides blast are grain endemic areas of the Republic of Niger. discoloration, sheath blight. and leaf scald. The pink The work on iron-toxicity continues as pot trials borer and grain·sucking bugs are the maj or insect and field testing techniques for screening. Entries pests in t he upland. with considerable tolerance have been identified in Research activities on breeding during 1984 were iron·toxic areas. Several lowland lines with high geared to developing better variet ies w here the yield potential and good grain characteristics have impact of stresses would be least . The breeding also advanced to large scale testing. program continued its 1983 activities in testing In Cameroon, where we support the National introduct ions, generating breeding populations Cerea ls and Extension Project (NCRE), the emphasis through hybridization, growing and evaluating pedi· is on vertical improvement and diseases, weeds, and gree materials , and conduct ing observational and insects of rice. We have one entry with considerable replicated yield trials. Screening for the stresses resistance to sheath rot disease that is to be released mentioned. and for quality remained major com­ soon in the Ndop Plain area. ponents of the breeding program. In 1984 added A new dimension of the rice improvement program emphasis was given to breeding for hydromorphic is the International Rice Testing Program for Africa. and shallow·swamp ecologies and tolerance to iron Seeds already have been multiplied for the 1985 tria ls toxicity and low temperature. We intensified breed· in several sub-saharan countries. This program is a ing for resistance to rice yellow mottle virus. 1 2 Rice Improvement Program Hybridization Program are particularly conducive to such foliar diseases as The 1984 crossing block was planted early with 240 blast, leaf scald, sheath blight, and sheath rot. Neck varieties and lines as parents in the hybridization blast and glume discoloration also are severe at program. Included were elite varieties and lines as Onne. donors for plant type, yield potential, and better grain The 1984 season at Ilorin was generally good except quality and donor varieties as sources of resistance or that the rainy season was too short for late-maturing tolerance- to provide adaptability to the various entries. Helminthosporium leaf spot was the most target environments. Of the crosses made in 1984, prevalent disease. At Ibadan, drought was severe 70 some 230 single, three-way, and four-way crosses were to 90 days after planting the 01\', PYT, and A YT. advanced to the F 2 generation for evaluation in 1985. Blast, grain discoloration, and leaf scald also were About one-third of all the 1984 crosses were for severe. The crop at Ikenne suffered from the same upland and two-thirds for hydromorphic, shallow stresses as at Ibadan although somewhat less. At swamp , and irrigated rice ecologies. anne, diseases were comparatively less severe in 1984 Major focus in upland crosses was resistance to than in previous years. Performance of test materials blast, drought tolerance, and improved grain quality. there reflected tolerance of acid soils. Very-early and late-maturing varieties also were considered in choosing parents for the upland cross. Pedigree Nursery For crosses targeted for hydromorphic and shallow­ The upland pedigree nursery was grown during the swamp ecologies, plant type and adaptability to 1984 wet season of 1984 at Ibadan, Onne, and Ikenne. rainfed lowland growing conditions were the major At each location, each line was sown in two-row, 5-m objectives with emphasis on resistance to rice yellow plots with seeds dibbled at hill spacing of25 X 20 cm. mottle virus and blast, iron toxicity tolerance, and A total of 930 entries, including early to advanced early maturity. generation materials derived from 53 crosses, were To develop cold-tolerant varieties for Africa's mid­ grown. Every 10th row in the nursery was a control a lti tude zones , we used cold-tolerant parents in (ITA 235, ITA 257, or OS 6). ITA 257 is an early generating segregating populations for the Cameroon maturing variety and ITA 235 is medium maturing. highlands. Both are improved plant types known for good yield, Using male-sterile lines from mRI, we also in­ stability, and adaptation to upland conditions. OS 6is itiated work on developing composite populations. commonly grown by fanners. From the male sterile lines that we pollinated with Many of the lines that looked promising and were selected upland, rainfed-lowland , and irrigat ed­ selected from the nursery after the growing season lowland materials, 45 crosses have been produced were derived from the following crosses: TOx 894, from which F 2 seeds were harvested. The first cycle of TOx 1752, TOx 1870, TOx 1889, TOx 1941, TOx 1945, intercrossing will be done in the dry season of 1985.­ TOx 1951. TOx 1854, TOx 1857, and TOx 1889. From T.M. Masajo and J. W. Gibbons Ibadan, Onne, and Ikenne about 650 selections were made from 430 lines. Seeds of the advanced selections Upland Rice will be multiplied in the irrigated lowland for testing The upland r ice trials conducted in the 1984 wet in the upland observational nursery and 1985 wet season consisted of Pedigree Nursery (PN), season. Seeds of late maturing selections will be Observational Nursery (ON), the Preliminary Yield shared with the lITA plant breeder in Liberia for Trial (PYT), and the Advanced Yield Trials (A YT). testing there.- T.M. Masajo, K . AI/uri, and J. W. Test locations were Ibadan, Ikenne, Ilorin, and Onne Gibbons in Nigeria. Ikenne represents the t ransitional zone between savanna and forest. Soils are fairly uniform Observational Nursery while rainfall follows a bimodal pattern. Average The 1984 upland Observational Nursery (ON) was precipitation is about 1,600 mm annually. Leaf scald grown at all four sites: Ibadan, Ikenne, Ilorin, and and neck blast are often severe. Onne. The Ibadan trial was drilled at 60 kg of seed/ha Ibadan (lITA) is also a transitional zone with in two-row plots, 3 m long with 25 cm between rows. bimodal rainfall (average of 1,300 mm annually) but At Ikenne, Onne, and Ilorin, rice was dibble planted soils are coarser, more heterogenous, and more prone with 25- X 20-cm spacing in the same plot size as in to drying. Drought and diseases, notably blast, leaf Ibadan. The t est was not replicated. The 107 entries in scald, and glume discoloration, are the major stresses the 0 N were selected from the 1983 pedigree nursery. at lITA. The savanna environment, characterized by lITA Observational Trials, and such international a short growing season, is represen ted by Ilorin nurseries as the WARDA, Initial Evaluation Trial where it is drier than at other sites but pressure from (lET), and mTP International Upland Rice diseases is relatively less. The Onne site is in the high Observationall\'ursery (IURON). rainfall. humid zone with monomodal rainfall pattern Among the test entries, 10 were highly susceptible of about 2,450 mmaverage annually. Soils at Onne are to blast but at least 15 carried high resistance. Among acidic, highly leached, and with poor CEC. High them were ITA 256, TOx 516-28-103-2-1, TOx 857-12-1- humidity and cloudiness during the growing season 1, TOx 939-107-2-101-2-1, TOx 1729-1-9-1, TOx 1729-1-9- Rice Improvement Program 3 2, TOx 1857-1-101-1-2, TOx 1857-2-101-1-3, and TOx TOx 1857-1-101, and TOx 906·2-6-204-1-1-1. Each yield· 1873-101-1. Some were also resistant to leaf scald and ed more tban 2,OOOkg/ha. At other locations, TOx 936- glume discoloration. Lines tolerant of drought at 81-3-3-101 did well at Horin, TOx 1780-7-1-201·1 at reproductive stage at Ikenne were derived mostly Ibadan, and TOx 1010-6-3-3 at Onne. Averaged over from crosses TOx 936, TOx 955, TOx 1010, and TOx the four locations, ITA 189 and TOx 1010-6-3-3 were 1857. the top ranking entries. ITA 189's height caused some Some of the most promising entries across the four lodging at Onne and Ikenne but it showed resistance tests are shown in Table 1.1. Four of the selected to drougbt, glume discoloration, and sheath rot. In entries, TOx 955-208-1-101, TOx 1010-14-1-3-1, TOx general, most of the PYT-E entries resisted blast. ITA lOW·24·2-6-4, and TOx 1739-101-4-1, are early matur­ 216, TOx 936-81-3·3·101, TOx 1757-1-7-201-1, and TOx ing and should be suitable for the savanna zone's 1780-7·1·201·1 were scored highly resistant. But TOx short growing season. Later-maturing selections like 1857·1·101, which performed well at Ikenne, was TOx 1729-1-9-2, TOx 1857-2-101-1-2, TOx 1857-2-101-1-2, susceptible, so it will be dropped from further trials. and TOx 1857-3-102-2-1 should fit the humid zone's At Ikenne and Onne no entry showed resistance to longer season. All promising entries selected from the leaf scald. 1984 observational nursery will be advanced to tbe replicated PYT in 1985.- T.M. Masajo, K. AI/uri, and Preliminary yield trial-medium. The highest J. W. Gibbons yielders in the 1984 PYT·M are shown in Table 1.2. Entries that did well at both Ikenne and Onne were Preliminary Yield Trial TOx 1857·3-102·1, TOx 1768-1·2·201·1, TOx 1779-3-1· 201·1 , and TOx 1757·1·28·201·1. Tbose four and TOx Two groups of Preliminary Yield Trial were grown 1768-1·2·201·1 also had highest overall average yields during the 1984 wet season, Preliminary Yield Trial­ across the three locations. Yield and maturity data in Early (PYT-E) and Preliminary Yield Trial·Medium Table 1.2 sbow later maturing lines yielding better (PYT·M). Each group had 22 entries from t he 1983 than the early at Ikenne. In addition to drought Observational Nursery (ON). PYT·E was grown at tolerance, they apparently had time to recover from Ibadan, Ikenne, Ilorin, and Onne. PYT·M was grown the drought that occurred 70 to 90 days after sowing. at all sites except Horin , which represents the sav· TOx 1857·3·102·1, TOx 1779·3·1·201·1, and TOx 1757·1· anna zone where early-maturing varieties are need­ 28-201·1 resisted blast and showed some resistance to ed. All trials used six·row, 6-m plots in three repli· glume discoloration, sheath rot, and leaf scald. Four cations in a randomized complete block design. At of the 22 test entries, TOx 1871·19·1, TOx 936·81·3-5- Ibadan and Ikenne, seeds were drilled at 70 kg/ha 201, TOx 936·87·10·1·1, and IR 4505·4-1·2, were highly while at Ilorin and Onne. the trials were dibble· susceptible to neck blast.-J. W. Gibbons, K. AI/uri, planted at 25· x 20·cm spacing 4 to 5 seeds per hill. R. Dobson, and T.M. Masajo. Fertilizer was applied at 90·60-60 kg/ha. Advanced Yield Trial Preliminary yield trial-early. Data on entries selected from the PYT·E are summarized in Table 1.2. The Advanced Yield Trial·Early (A YT·E) and The lines that did well at !kenne, where yields were Advanced Yield Trial·Medium (A YT·M) each had 22 highest , were TOx 1i80-7·3-102·1, TOx 1780-7·8-201·1, entries, composed of selections from the 1983 A YT Table 1.1. Some promising rice entries selected from the upland Observational Nursery (ON), 1984 wet season Phenotypic Plant acceptability height,l Maturity 2 Drought Entry score l cm days score3 TOx 1729-1-9-2 .. 2.7 94 127 5 TOx 957·109-2.1. ....... . 3.3 118 120 5 TOx 1010-6-9-3·201 ..... . 3.3 112 121 3 TOx 1010-14-1-3-1 ...... . 3.3 105 114 3 TOx 1739·101·4·1 ....... . 3.3 102 109 5 TOx 955·20B-I-101 ...... . 3.7 93 118 3 TOx 955·208-2-101-101·3-1·1. ........ . .. . .. . .... . .. . 3.7 83 123 1 TOx 1857·3-102-2-1 ..... . 3.7 102 127 5 ITA 235, Control. ...... . 3.7 112 120 3 TOx 939·107-2-101-2-1 ... . 4.0 94 124 1 TOx 955·201·2·101. . . . . . . . ...... . 4.0 97 121 1 TOx 1010-24-2·6-4 ................ . 4.0 109 118 3 I Average from Onne. Ikenne, and Ilorin . ~Average from Ibadan. Dnne, and Ikenne. :lIkenne data taken at reproductive stage. Now: Scoring for drought tolerance and phenotypic acceptability at maturity was based on Standard Evaluation System for Rice. ffiRI. 1980. Drought on 0 to 9 scale where 0 = no symptoms and 9 = all plants apparently dead. Phenotypic acceptability on 1 to 9scale where 1 = excellent and 9 = unacceptable. 4 Rice Improvement Program Table 1.2. Performance of selected rice entries in the Upland Preliminary Yield Trial (PYT) at locations in Nigeria, 1984 Plant Yield. kg/h. height. lvlaturity, Grain g,ualiti: Ent ries llarin Ibadan [ken ne Onne Mean em days Length' Shape' Amylose, % PYT-Early ITA 189 . .. ..... . ........ 1.871 1.396 1.899 1.942 1.777 137 123 3 5 20.7 TOx 101()'6-3-3 .. . .. 1.671 1.400 1.855 2.118 1.756 87 125 3 5 23.1 TOx 1780-7-1-20J.\_ . ...... 1,632 2.134 1.929 815 1.628 99 118 3 5 15.7 ITA 216 . ... . . . .... . ... ' . 1,467 1.568 1.939 1.347 1.580 107 122 3 5 13.9 TOx 936-81-3-3-10 1. ....... 2.011 1.006 1.906 1.191 1.529 99 128 3 5 14.4 TOx 1780-7-3-102-1. . .. ... . 1.156 1.066 2.596 1.116 1,484 100 139 5 5 20.7 ITA 257. Control . . . . . , . . . 1.583 1.651 1.729 1.578 1.635 96 94 3 5 19.3 ITA 235. Control . .. , ..... 1.558 1.657 1.504 1.689 1.605 102 116 3 5 14.8 OS 6. Control. .. ...... .. . 1.585 1.144 1.041 1.513 1.316 128 121 3 5 19.3 M ean ... . . .... .. ... .... 1.470 1.320 1.890 1.478 1.540 LSD. 5% . . . . . .. ... . . . .. . 625 776 872 568 C.V.,% .... . . .. ... 31 41 31 25 PYT-Medium TOx 1857-3· 102-1 . ....... . 610 2.602 3.054 2.089 89 141 3 5 21.3 11972 . .......... ., . . . . .. 1.377 2.208 2.508 2.031 70 140 3 5 23.0 TOx 1768-1-2-201-1 ...... .. 736 1.950 3.328 2.005 92 125 3 5 23.3 TOx 1779-3- 1-201-1. ..... . . 834 2,408 2.662 1.968 87 131 3 5 20.8 TOx 1757·1-28201·1. .... . . 782 2.883 2.036 1.900 91 135 5 5 19.9 TOx 936-1555·1·2·1 ....... 1.361 1.967 1.859 1.729 88 124 5 5 22.3 ITA 235. Control . . ..... . . 1.186 1.245 2.464 1.632 97 117 3 5 13.1 ITA 257. Control . . . .. . .. . 1.436 1.437 1.103 1.325 84 ll9 3 5 14.8 OS 6. Control. . . . . . . . . . . . 871 1.375 888 1.045 103 127 3 5 19.9 Mean .............. . 1.041 2.076 2.049 1.722 LSD, 5%. . . .. .. . . . . . ..... 605 736 821 C.V. . % ..... . .. ...... . . . 41 22 31 IScoring based on Standard Evaluation System fo r Rice. ffiR[, 1980. Grain length scale: 1 = extra long to 7 = short : shape sC8le: 1 = slender to 9 = bold . Kote : Sta t ist ica l computations used data from all entries. and preliminary trials. A YT-E and A YT·M were tion and shea th rot. grown at Ibadan. Ikenne. and Onne. AY T·E also at All entries in Table 1.3 will be kept for 1985 trials. Ilorin. The experiment used six·row. 6-m plots in four Others will be nominated for testing in national replications and arandomizedcomplete block design. programs and interna tional nurseries. Cultural practices were tbe same as those described Advanced yield trial-medium. Data on yield a nd for ON and PYT. major agronomic and grain characteristics of selec­ Advanced yield trial-early. Yields and major ted entries in the 1984 A YT·M are shown in Table 3. agronomic and grain characteristics of selected At Ibadan. the highest yielder was TOx 936-8-4-1-2; at entri es are shown in Table 1.3. This year's trial Ikenne. TOx 891-212-2·102·2·101·1 , and at Onne. ITA confirmed the potential of TOx 936-397·9·1·2, top 128. In similar 1983 multilocational trials. TOx 926-8- ranking entry across four locations in 1983. Figure 1.1 4-1·2 and TOx 891·212·2-102·101·1 performed satisfac­ shows varietal mean yields against stability index for torily. They have been numbered ITA 305 and ITA the A YT·E across locations in 1984. The most desir­ 303, respectively. ITA 128. another promising entry able varieties are those with higb mean yields and (Table 3) has been nominated for testing in nat ional stability index close to 1 (See TOx 936·397·9·1·2. no a nd international trials ; 11568 a lso yielded well but 18). Among the selected entries in Table 1.3 control appeared to be susceptible to drought at Ikenne and ITA 257 yielded highest at Horin ; IRAT 156. at glume discoloration at Ibadan. TOx 936-8-4·1·2 and lkenne; and TOx 936·397·9·1·2. at both Ibadan and TOx 981·212-1·201-1·103·1 appeared more drought Onne. TOx 936-397·9-1·2. IRAT 156, ITA 225, and TOx tolerant than other entries.- J. W. Gibbons. K. 937·87·10-4·2·1 resisted blast. ITA 132 also yielded A lluri. R. Dobson. and T.M. Masajo sa tisfactorily but seemed susceptible to blast at Ibadan. Of entries ranked in Table 1.3, ITA 225 and Rice Varieties for Acid Upland Conditions IRAT 156 were among tbe most drought tolerant. In 1984. attempts to develop superior varieties for better than TOx 936-87·10-4-2-1 ITA 182, and at acid upland conditions were continued and inten­ Ibadan. both showed resista nce to glume discolora- sified at IITA's high r ainfall station. Onne. Fifty.one Rice Improvement Program 5 Table 1.3. Performance of selected rice entries in the Upland Advanced Yield Trial-Early (AYT-E) at locations in Nigeria, 1984 Plant Yield, kgfha height, Maturity, Grain 9ualit~· Entries Ilorin Ibadan Ikenne Onne :Mean em days Length l Shape l AYT-Early TOx 936-397-9-1-2 ..... _ . _. ... .. , . . . . 1,562 1,934 2,299 2.470 2,066 86 111 3 5 !RAT 156 .... . .. . . . . , .. . . , . . . . .. ..... 1,182 1.292 2,765 1,874 1,778 103 116 3 5 ITA 132 . . . . . . . . . . . . . . . . . . . . ... 1,282 874 1,958 2,492 1,652 101 111 3 5 ITA 225 ..... . .. . . . ... ... . . .... .. . ... 1,393 826 2,118 1,937 1,567 94 113 5 5 TOx 936-87·10-4·2-1 . . .. . . .• ... . . . 1,133 1.095 2,257 1,779 1,566 85 120 3 5 ITA 182 ........... . . . . . . . . . . ..... .. . 1,297 1.128 1,659 2.054 1,535 102 112 3 5 !RAT 109 . . . ......... . .. . . . ....... ... 1,590 1,284 987 1,785 1,412 90 102 5 9 ITA 235, Control. ....•. . •............. 1,520 1,007 1,423 833 1,196 92 110 3 5 ITA 257, Control. ..... . . . . . . . . . . . 1,924 1,467 1,649 1,166 1,552 94 91 3 5 OS 6 Control ... . .......•... .. . . .. . ... 1,057 861 1,137 1,616 1,168 115 116 3 5 J..fean .. . . . .. . .. . . . • . · ... ... ... .. .... 1,419 1,166 1,851 1,647 1,521 LSD,5% . . ........... · ... . .. . ... . . . .. 342 651 719 524 C.V. % .. . .... . .. . .. . . . ... . .. . . . . . .. . 31 46 31 25 AYT-Medium' TOx 936-8-4-1·2 .............. .. .. .. .. . 2,138 2,598 2,172 2,303 91 120 3 5 TOx 891·212·2·102·2·101·1 ..... . . . . . . .. . 1,605 3,138 1,257 1,999 77 124 5 5 11568 . . .... . . . .. . .. . .... . - .. . . . .... . 1,622 3,080 1,023 1,908 71 141 3 5 ITA 128 . . .... . .......... . .. .. . . .... . 1,263 1,904 2,411 1,859 98 116 3 5 TOx 891·212·1·201·1·103-1 .. . .... . . . .. . . 998 2,854 1,669 1,840 80 136 5 5 TOx 1857·3·2·201·1 ... ..... .. ... . .. . . . . 863 2,216 2,200 ],766 91 133 3 5 15775 .... . .... . . . .. . . . .... . . .... .. . . 1,005 2,251 1,930 1,729 68 134 3 1 ITA 257, ControL .. . . . ... . . .. . .. . .... . 1,109 1,873 1,882 ] ,621 95 90 3 5 ITA 235, Control. . . ... . . . . . .. .. . . . . .. . 1,774 754 998 1,175 115 97 3 5 OS 6 Control ... . . . _. . · .. . . .. . . . . . . .. . 1,277 1.147 1,135 1,186 121 112 3 5 Mean . . .. .. . . . .. ... . . . .. , .. , . . . . . .. , 1,272 2,099 1,689 1,686 LSD.5% . . .. . . . ........ . . . __ . .. . .. . .. 433 669 558 C.V,, % .. . .. . . . ..... .. . ... _ . . .. . .. . . . 27 26 26 Note: Statistical computations used data from all entries . lScoring based on Standard Evaluation System for Rice. IRR[ , 1980. 2 A YT Medium was not grown at llorin. tolerant entries selected from a previous screening tions with high yield potential not adapted to Africa trial (UTA, 1983), 110 new lines, and 66 entries from upland stresses. Without severe upland stresses in the Acid Upland Soil Screening Set, 1984 (an 1983, UPLRi 5 yielded 2.7 t/ha in acid and 3.1 t/ha in IRTP/INSFER trial from IRRI) were all e\'aluated for plots with lime and P. performance in acid soil. Twenty·four were evaluated Susceptible cultivars like IR 5260·1 and IR lO198·67· in a replicated yield trial; the others were screened in 7, which do not do well with stresses absent, are acid soil. In addition to ITA 117, ITA 235. and TOx eliminated. All cultivars that are retained are used in 1753-1·24, each reported acid tolerant last year, six the breeding program or as varieties based on merit. more entries superior to local controls, LAC 23 and All entries in the acid, upland-soil screening set OS 6, were identified from the yield trial. (1984) from IRTP, planted 20 July, performed poorly. Varieties screened for acid tolerance differed in July- August plantings at Onne were severely response to added lime and P. Based on response to damaged by sucking insects, stem borers, sheath acid soils, we divided the lines into three groups: 1. blight, and glume discoloration, and no entry yielded Those judged tolerant with satisfactory yields. They satisfactorily. The only entries that produced some yielded about 2 tjha in acid plots and some responded filled grains were ITA ll7, IRAT 104, IRAT 112, IRAT positively when lime was added. Most of the superior 170, and local controls ITA 141, ITA 307, and ITA 315 entries in Table 1.4 qualified. All have African back­ (TOx 936·397·9·1-2).- K. Alluri grounds. 2. The tolerant, but low yielding lines. all traditional African varieties, yielded about 1 tjha with negligible effect from applied lime and P (LAC 23 Hydromorphic Rice and OS 6). 3. Those in the susceptible group that Development of suitable varieties for hydromorphic performed poorly in the acid plots but well when lime and rainfed shallow swamp is important to increasing and P were added and disease pressures were low (IR production and expanding the rice area in Africa. Our 5931-ll0-1, UPL Ri 5). They are improved introduc- breeding objectives for those environments included 6 Rice Improvement Program Stability index an upland with irrigated lowland below. The middle 2 .7,.----------------------, section, constantly wet from seepage and shallow 12 ground water, was typical ofhydromorphic soils. The entries were planted in two-row plots unre­ 2.61- plica ted along the slope in the transect starting from upland and down to the lowland. The upland and the 2 middle hydromorphic sections were directly seeded 2 .1 I- by dibbling seeds at 20-cm spacing along the row and planting four seeds per hill. Rows were 25 em apart. 1.8 I- 7 17 The lowland irrigated section was planted with 21- I. day·old seedlings the same day near the end of June 24 that the upland and hydromorphic portions were 15 I- seeded. Total fertilizer applied was 90-30·30 kg/ha. On the upper, middle, and bottom parts of the 16 toposequence, data were gathered on each plot: plant 1.2 I- 8 height, maturity, number of panicles, grain yield per sq m, and phenotypic acceptability score at maturity. 4 Rainfall was so favorable that the crop on the upper· 0 .7 I- 18 most slope did not experience stress. 13 Many entries, as expected, performed well under 0 .6 I- 20 lowland but not upland conditions and vice versa. But many vari eties that performed well in the irrigated 10 • lowland also did well in the hydromorphic area but 23 0 .3 I- not upland. That supports earlier UTA findings that 14 2221 15 in the absence of stress, lowland irrigated varieties may do well under hydromorphic soil conditions. 0 .0 I- Without drought stress, several entries performed fairly well across t he transect (Table 1.5). Se lections 3 with at least average phenotypic acceptability rating -0.3 I I I I 1000 1250 1500 1750 2000 of3.0 and betterwereTOx 1835-11·1, M 57B·I46·1, TOx Entry mean yield 1838-3·1, and TOx 902·5·103·3·101·2. M 57B·146·1 was introduced from Indonesia through International Figure 1.1 . Stability index and mean yield of r elated rice Rice Observational Nursery (IRON) of IRTP and ent ries in the Advanced Yield Trials Early (A YT·E) over tested at IITA in 1983. The selected entries in Table four locations in the 1985 wet season. 1.5 will be evaluated further in replicated trials in 1985. In general, plant height, tiller number, and yield Variety 1 = IRAT 109, 2 = IRAT 156, 3 = IR 10068-11·1, increased with descent from upland to irrigated 4 = ITA 130, 5 = ITA 132, 7 = ITA 175, 8 = ITA 182, lowland, or as moisture regimes became more 9 = ITA 184, 10 = ITA 187, 11 = ITA 225,12 = TOx 909-11· 201·3-1-1,13 = OS 6, 14 = ITA 235, 15 = ITA 257, 16 = TOx favorable- T.M. Masajo, A. Kumwenda, and N. V. 936-87·10·3·101, 17 = TOx 936-87·10-4.2.1 , 18 = TOx936-397· Nguu 9· [·2,19 = TOx955-202·2·101, 20 = TOxl0l0-6-3-5, 21 = TOx 1010·24·15·3, 22 = TOx 1768-3-1-1, 23 = TOx 1768-3·1·1·101· Lowland Rice 1.24 = TOxI780·9·2·201·1, 25 = TOxI780·18·2·201·1. Only about 11 % of sub-sahara Africa's rice acreage is irrigated lowland rice, but that percentage is expec­ ted to increase as irrigation facilities are developed high yield potential, short to intermediate stature, to create the most favorable environment for rice. resistance to blast, drought toleranc e, and ability Yields from irrigated lowland rice are often higher to perform relatively well under various moisture and more stable than on rainfed environments. The regimes. We also continue to screen breeding ma­ short stature, lodging resistant varieties with erect terials and introductions for adaptability. leaves, high tillering, and high yield potential are considered best for irrigated lowlands. Improving Observational nursery on toposequence. To varieties by incorporating resistance or tolerance of follow up the 1983 agronomic experiments comparing the stresses mentioned earlier is the major objective several varieties on toposequence, we used 130 en­ of our program for irrigated rice. tries in an observational nursery in a similar field Activities on lowland irrigated rice in 1984 in­ during the 1984 wet season. The objective was to cluded hybridization and selecting and evaluating identify materials that perform fairly well across introduced varieties. The following were grown and various moisture regimes that hydromorphic, rain­ evaluated: Pedigree Nursery (PN), the Observational fed, and shallow swamp rice often encounter. The Nursery (ON), the Preliminary Yield Trial (PYT), and area was a 60· X 5O·m block on a slope that included the Advanced Yield Trial (A YT). Rice Improvement Program 7 Table 1.4. Grain yield and related characteristics of superior rices under severe acid upland conditions, Onne, Nigeria, 1984 Grain Plant Days Visual score l yield, height to Panicles} Early Pan icle Leaf Sheath Leaf Pheno Variet~ t /ha cm ftw . m' vigor exstn. blast blight scald accept TOx 936·81·1·4 ... . .. .. . 0· · · ······ 2.9 [08 109 283 3 2 [ I [ 4 ITA 303 . .. . ..... .. ..... ...... . . 2.6 93 104 325 4 3 2 3 2 3 TOx 1010-23·1·7·5 .. .. . . . . . . .. . 2.5 llO 96 347 3 3 1 1 2 5 ITA 307 ... . .... . 2.3 79 88 336 4 3 2 3 5 4 TOx 1012·12·2·1. . . .. . . . . 2.3 88 91 331 2 2 1 4 3 4 TOx 854·101·3·20[·1·1·1 .. 2.1 L03 100 288 3 2 1 1 3 4 IRATI70 ....... 2.1 100 86 325 1 1 1 3 3 4 IRAT \04 .. . ..... . . . . . . . . 2.0 ll5 90 293 2 1 1 2 2 3 LAC 23. Control .. .......... ..... 1.2 149 105 192 1 1 2 2 1 4 OS 6, Control . . . . . . .. .. . . .. . . .. 1.0 134 84 213 1 1 2 2 2 4 Overall mean . . .... . ... .. ...... 1.54 S.D ... . .. .... . . ....... . .. .. . .. 389.0 LSD, 5%. •• • ••• •••••• • • • 0 • •• • •• 0.84 C.V.,"!. . . .. . .. . ....... . .. . 25.2 Entries tested . . . . . .... . . . . ...... 24.0 IScale or o to 9 based on Standard Evaluation System for Rice. Pedigree Nursery Observational Nursery [n 1984, the pedigree nursery was grown in bot h the The 1984 lowland observational nursery grown duro dry and wet seasons . Seedlings were grown in raised ing' the dry season was composed of 130 entries beds in the lowland field and transplanted when 21 to selected from the 1983 obse rvational nursery, 24 days old in two·row, 5·m plots with 25 em between International Rice Observat ional Nursery (mON) of rows and 20 em between hills, one seedling per hill. mTP, mRl, and [nitia l Evaluation Trial (lET) of Every 10th row was planted to one of the controls WARDA. Entries were grown in five·row, umepli· (ITA 212 , [TA 306, or m 42). The dry season nursery cated plots, 5 m long a t 20- X 20·cm spacing, one was composed of 189 entries and the wet season, 490 seedling per hill, and with 120-0·0 k g/ha of fertilizer. entries. The growing seasons were in general favor­ Data on days to flow ering, height, and tiller number able except for moderate·to·severe neck blast during per hill were taken with observations on diseases, the wet season . pests, and phenotypic acceptability rated at maturity. About 200 selections were obtained fo r further The most promising entries selected a re shown in evaluation in 1985 pedigree land observationa l nur· Table 1.6. All ofthem had been advanced to the 1984 series. Many selected lines were derived from these wet season lowland PYT. Most of the entries that crosses: TOx 978, TOx 981. TOx 3010, TOx 3015, TOx were discarded were eit her too tall , poor plant type, 3027, and TOx 3034.- T.M. Masajo suscept ible to lodging or blast, or h ad poor grain type.-T.M. Masajo Table 1.50. Characteristics of promising entries grown under upland , hydromorphic, and irrigated Table 1.6. Promising entries in lowland conditions in a toposequence, 1984 wet Observational Nursery (ON), 1984 wet season season Phenotypic Plant ~at· Grain Plant Mat· No. of acceptability height, urity , Panicle! yield/ Phenotypic height , urity, tillers Entry score ! em days m' m 2 , G: Entry acceptability:! em days per hilJl TOx 1835-11·1 . .... . 2.3 107 129 217 465 TOx 1835-11·1 . .... 3 122 125 11 !vi 57B-146-1. . ..... . 2.7 l OB 132 238 352 TOx 1838-3·1 ...... 3 120 125 10 TOx 1838·3·1 . . .... . 2.7 109 127 204 416 TOx 1857·3·109-1 . . .... . . . 3 103 123 10 TOx 902·5-103-3-101·2 3.0 105 132 243 420 TOx 1857·3·109-2 . . 3 123 123 8 B3894-22C·5!v1-5-1·1 . 3.3 103 124 223 355 lOx 902·5·102·2·103-101. . .. 3 115 127 9 IR52 ........... . . . 3.3 98 129 261 476 lOx 902·5·103·3-101·2 ... . .. 3 118 130 12 IR 21015-72·3·3·3-1 . . 3.3 111 123 273 378 TOx 960·58-1. . . ....... . . . 3 122 109 9 IR 25861·84-3-2 . .. . . 3.3 99 117 277 475 11377 ........ . .. ... .. 3 122 125 11 IR 28128-45-3-3·2 . . . . 3.3 98 110 273 251 16221 . .. . .... . .. . .. .. . .. 3 128 127 9 IR46 .. . .. . .. . .. . .. 3.3 114 128 384 358 ITA 212, Control ...... 3 114 125 10 ITA 212, Control .. . 3.7 106 127 295 568 1 Productive tillers counted at maturil). . 1 Scoring based on Standard Evaluation System fo r Rice. rRRI. :lScoring based on the Standard Evaluation System. IRRI. 1980. 1980. (1 = excellent to 9 = unacceptable). 8 Rice I mprouement Program Preliminary Yield Trials wet season. Identified as possible entries in Advanced The 1984 low land preliminary yield trial (PYT) en· Yield Trials for 1985 were BW 295·4 , P 1377·1·15·M·1· 2M·3, IR 19661·150·2·2·2·1, and Tax 725-1·8·201·1. BW tries were grouped by early (PYT·E) and medium 295·4 , Tax 725-4, and TOx 725-1·8·201·1 showed good (PYT·M) maturities. Entries were selected from the blast resistance. TOx 711·2·2·2·1 was dropped as 1983 Observational nursery, WARDA Initial susceptible to blast. Evaluation Trial (lET), and IRTP International Yield Nursery (IRYN). The PYTs were conducted in both Advanced Yield Trials the dry and wet seasons of 1984 with l.6- X 5·m plots in three replications in a randomized complete block Entries in 1984 lowland advanced yield trials were design. Seedlings ~ereraisedin beds'lnd transplan ted grouped by early and medium maturities, 24 entries in 21 days after seeding, 2 or 3 seedlings per hill in 20· x Advanced Yield Trial·Early (A YT·E) and 26 in 2O·cm spacing, with 120·0.0 kg/ha of fertilizer during Advanced Yield Trial·Medium (AYT·M). Entries the dry season and 90.60-60 kgJha in the wet season. were selected from 1983 trials and from international yield nurseries of IRTp·IRRI and WARDA and in· Preliminary yield trial-eady. The PYT·E during c1uded two controls. the dry season consis ted of 22 test entries. Six were The trials were conducted during both seasons. dropped for poor performance and seven new selee· Seedlings , raised in beds. were transplanted 21 days tions were added for the wet season trials. Controls in after seeding. to 2· X 5·m plots in hills spaced 20 x 20 both seasons were TOx 103 and IR 36. The dry season trial was grown in irrigated lowland; the wet season em in a randomized complete block in four repli· tria I , in both irrigated lowland and a hydromorphic cations with fertilizer applied at 120.0·0 (dry season) site at UTA. Table 1.7 gives the results. and 90-60·60 kgfha (wet season). Among the highest yielding ent ries from the dry Advanced yield trial-early. Table 1.9 gives per· season were Tax 915·101·1 a nd TOx 960.42·1 but they formances of selected entries in A YT·E compared yielded only slightly more than control TO. 103's with two controls, IR 36 and TO. 103. ITA 234 yielded 7,551 kg/ha. During the wet season under irrigated 7,035 kg/ha during the dry season, 6,338 kg/ha in the and hYdromorphic conditions, B 9894·22C·5M·5-1·1, wet season, and more than 6 t/ha in 1983. It resists Tax 960·42·1, and SiPi 692033 were consistently blast and has some tolerance ofRYMV. Nearly all the superior. TOx 960·42·1 was most resistant to blast and selections except IR 2042·178-1 and KAU 166, have it has long, slender grains. It will be in the Advanced high amylose content. Tax 894·28·201·1·2 and Tax Yield Trial for 1985. 894·28-201·1·5 combine high yields, earliness , and good grain type. Preliminary yield trial-medium. The dry season PYT·M was composed of 18 test entries and two Ad vanced yield trial-medium. Table 1.10 sum· controls, IR 42 and ITA 212. Five entries were dropped marizes data from A YT·M. Mean yield of the dry for poor performance and 10 were added to make 23 season crop was significantly higher than that of the for the 1984 wet season, with the same two controls. wet season crops, with the same varieties performing Table l.8 summarizes the performance of promising well both seasons. Averaged over seasons, several entries. Yields were higher during the dry than the also performed well in 1983. IR 19670·263-3·2·2·1 and Table 1.7. Performance oftbe most outstanding selected entries in tbe Preliminary Yield Trial-Early (PYT-E) under irrigated and hydromorphic soil conditions, 1984 dry and wet seasons, Ibadan, Nigeria Grain yield. kg/ha Grain characters Hydro· P lant Irrigated Irrigated morphic height,l Maturity, Amylose. Variety dry season dry season wet season Mean em days Length' Shape2 °/~ B 9894-22C·5M·f>.1·1 ..... 7,415 5,014 6,485 6,305 114 120 5 5 23.4 TOx 960-42·1 ....... . 7.679 5.062 5,951 6.231 113 119 3 1 24.4 TOx 91f>.lOl·1 . .... . . .. . 8,083 4.106 5,227 5,805 106 125 5 1 26.2 SiPi692033 . .. ... 6,865 4,926 5,400 5,730 113 116 3 5 25.4 32 Xuan f>.0 ............ 7,296 4,673 5,190 5,720 109 112 3 5 23.3 ill 18348-36·3·3 . . .. . ...• . 6,967 4,429 5,368 5,588 105 110 3 1 24.0 IR 36. Control ........•. 6,715 4,197 2,757 4,568 97 109 3 1 25.0 TO. 103, Control. . . 7,551 4,344 3,485 5,227 90 112 5 5 27.0 f\,1eaTI ... . ..........•.. 6,535 4,526 3,961 LSD,5% . ..... ......... 1,701 533 1.501 C.V.,% ...... ....... . . . 15.8 7.1 23.0 l Mean of irrigated trials only. 2Scoring based on Standard Evaluation System for Rice.lRRI. 1980. Note: Stat istical computations used data from a ll entries. Rice Improvement Program 9 Table 1.8. PerformaDce of .elected entries in the lowland Preliminary Yield Trial-Medium (PYT -M), 1984 dry and wet seasons, Ibadan, Nigeria Grain yield, kg/ha Plant Grain characters Dry Wet height, ' Maturity,· Amylose, Variety season season Mean em days Length' Shape' % BW29f>.4 .................. , . . .. .. 9,041 5.334 7,188 122 132 3 5 24.0 P 1377-1·If>.M·I·2M·3 .. .. ... . .. . .. . . 8,273 4.894 6,584 125 124 3 5 23.5 P 1496-7·7M .. ........ . . . ... , . ... . . 7,954 4,811 6,383 122 129 3 1 24.0 P 1274·S-8M·l·3M·!. . .. . ... .. .. .. . . . 7,633 4,789 6,211 115 123 3 5 23.6 ill 19661·150-2·2·2-1 . . .. . ... .. .. . .. . . 7,249 5, 117 6,183 lOS 128 3 1 24.1 P881·17·6-B·CR-6-7 .... . .... . .. . ... . 7,065 5,020 6,043 109 124 3 5 24.3 TOx 711·2·2·2·1 .. ... .. ... .. . ... ... . 6,999 5,057 6,028 117 140 3 5 24.5 ill 42, Control ... .. ... . ........ .. 7,135 4,766 5,951 113 130 5 5 28.1 ITA 212, Control ....... . , . ........ . 8,283 4,790 6,537 113 124 3 1 26.3 Mean .. . .... . ....... . . . .. . , . . . . . . 6,954 4,907 LSD, 5% . .. . . . . . . . . . . . .. . .. . . . . . . . 1,495 672 C.V., %. . . .... . ........ . . . ... ... . . 13.0 8.3 Statistical computations made on all entries t Mean of two seasons. 'lScoriog based on S t.a ndard Evaluation System for Rice,lRRI, 1980. Table 1.9. Performance oeselected entries in the lowland Advanced Yield Trial-Early (A YT-E), 1984 dry and wet seasons, Ibadan, Nigeria Dr~ ~eld, kg/ba Plant Grain characters Dry Wet height,' Maturity,l Amylose, Variety season season Mean em da~s Length' Shape ' 0,,0 ITA 234 .. . . . . . . . . . . . ............. . 7.035 6,338 6,666 115 122 3 5 29.6 6638 . . .. . .... .. .... . .. ... .. ... .... 7,071 6,205 6,638 116 122 3 1 28.6 ITA 222 ......... .. . . ... ........... 6,627 6,539 6,583 114 123 3 5 29.2 ffi2042·178-1 .. .................... 6,665 5,727 6,207 110 125 3 5 24.0 BR 161·2B·59 .. ............. , ..... . 6,414 5,995 6,205 102 120 5 5 29.5 TOx 894·28-201·1·2 ..... , . .. , ....... . 5.871 6,210 6,040 115 116 3 1 28.5 ITA 233 . . . . . ..... . .... . .. ... .. .... 6,630 5;246 5,938 114 124 3 5 28.4 TOx 894·28-201·1·5 . .. . . . . . . . . , ... .. . 5,845 5,666 5,755 115 118 3 1 28.9 C1321·9 . . .... . .. ... .. . .......... . 5,325 6,017 5.671 100 114 3 1 27.4 ill 21015·8Q.3-3-1·2 . . .. . . .... ... ... . . 5,412 5,829 5,620 111 115 3 1 27.9 ITA 230 . . . .... . .. . .......... . .. . .. 5,058 6,On 5,534 115 116 3 5 29.6 ill 36, Control ..... .. ... . .. ....... . 3,690 4,918 4,304 no lOS 5 5 28.2 TOs 103, Con tro!' . . . . . . .. . . ........ 5,318 5,771 5,544 99 112 5 5 30.3 Mean .. .. ..... ..... .. .. .. .. . ... .. 5,181 5,497 5,339 LSD,5% .................... ... ... 1,431 794 808 C.V.,% . . . .... . ..... . ............. 19.5 10.2 15.3 Statistical analysis used data from all entries I Mean of two seasons . 2Scoring based on Standard Evaluation System for Rice, JRRI. 1980. lET 6279 showed hlgh resistance to blast but IR 4422· screening trial. were used as donors. They are LAC 98-2·61, ITA 244, and ITA 249 were susceptible and 23, Moroberekan, and OS 6, which are tall, low will be dropped. Most of the selections ranked tillering, upland varieties highly adapted in Mrica, medium for grain shape and have high amylose ITA 235, an improved upland variety developed at content.- T.M. Masajo UTA, and CT 19, a semi-dwarf, early·maturing selec· tion recently introduced from India, Lowland elite Breeding for Resistance to RYMV varieties and lines used in crosses with the resistant donors included ITA 121, ITA 212, ITA 222, ITA 230, We launched an intensive program to introduce ITA 206, IR2042-178-I , IR21015-80-3-3·1·2, IR 9/29-67·3, resistance to rice yellow mottle virus (RYMV) into IR 9828-91·2·3, and Taichung Sen yu 285. They have elite breeding lines intended for lowland, hydromor· been superior in liTA and international trials. phic, and shallow swamp environments. Lines that Seventy.two crosses for RYMV were performed in have shown consistent hlgh tolerance of the virus in 1984, single crosses, backcroases, double crosses, and 10 Ri ce Improvement Program Table 1.10. Performance ofsolected entries in the lowland Advanced Yield Trial-Medium (AYT-M), 1984 dry and wet seasons, (badan, Nigeria Grain yield, kg/ha Plant Dry Wet height, I Maturity, l Grain characters Variety season season Mean em days Length' Shape' Amyl. % lR 4422-98-2·6-1 .. . .... .. ... . .. . .. . . 8,631 5,600 7.115 124 132 5 5 27.1 lR 28118-138-2·3 .. . .. . .... . . .. . . .. . . 8,335 5,822 7,079 116 128 3 5 28.3 lR 1967()'263-3·2·2·1 .. . . .. . . .. .. .. . . 8,386 5.450 6,918 111 127 3 I 26.3 IET6279 . ...... . . .. . ... . . .. .. .. . . . 9,On 4,540 6,807 119 133 5 5 25.6 lR 2928-7·3-1·1 . . . . .. . ........... .. 8,347 4.744 6,546 121 132 5 5 22.2 lR 42, Control .... . . . ............ . 7,503 3,835 5,669 110 131 5 5 26.5 ITA 212, Control. . ............ .. . . 8,361 5,049 6,706 108 122 3 1 28.6 Mean ........ . .. , . , .... . . . . .. . . . 7,511 4,510 6,0l! lBD, 5~~ ......... , . . , , .. . .. . .... . 1,329 824 771 C.V. , ~/o ....... . .. . . . .. . ... ... . ... . 12.5 12.9 13.0 I Mean of two seasons. ~ Based on Standard Evaluation SY3tem for Rice. IRRI , 1980. Note : Statistica l computations used data from a ll entries. three-way crosses. Some F I s of single crosses were Yields per hectare of the top eight very early used as parents for further crosses and some were entries were statistically similar (0.05). But line C advanced to the F 2 gen eration for screenhouse test­ 1158·7's 37% yield variability indicated instability. ing for r esistance t o RYMV and to the field for The instability factors need to be identified . The eight selection of agronomic types. Table 1.11 shows F 2 early entries may be used in advance yield trials. populations that produced at least 20 selections. Harvested area of only 1.92 sq meters may inflate Seeds from F, populations of the backcrosses, three­ yield estimates somewhat. way crosses, and double crosses have been obtained Twelve of the 30 early-maturing entries produced for 1985 evaluation.- T.M. Masajo, V. T . John, and G. high yields (Table 1.13). Yield variability was high Thottappilly among several, especially BG 380-2 and Taichung Sen 16. Again, instability factors need to be studied. KA U International Trials 1727, IR 36, and Si·Pi 692033 seem to possess yield During 1984, trials from the global International Rice stability but all promising lines should be critically Testing Program (ffiTP), the lRRI and African Rice tested against blast. Testing Program (ARTP), and IITA were conducted Twelve of the 19 medium maturing entries with at Ibadan to select adaptable lines to be genetically similar yields are shown in Table l.14. ffi 42, ffi 31917- improved for eventual release to farmers . 31-3·2, and ffi 22082·41-2 seem to have more stability Three International Rice Yield Nurseries, ffiYN-E than the others. The superior entries from the three (very early), ITYN-E (early) and ffiYN-M (medium), IRYN tests can be used for wider testing in Africa in were tested with the results shown in Tables 1.12, 1.13 the advanced yield trials.- Kaung Zan and 1.14. Table 1.11. F , populations that produced at least 20 selections in field or screenhouse when evaluated for rice yellow mottle virus Table 1.12. Pedigree, leaf and neck blast reactions, (RYMV) resistance, 1984 wet season and yields of the top eight entries in the IRYN-VE 1984 tests, IITA, Ibadan No. of plant selections Yield Leaf :- H I > L H > M,S followed by ROK 16 and m 2061·522-6·9. The trial Burkina Faso .... 10 I> H,L I>L H > M results indicate that under good management in the Cameroon .. , ... 20 H > I [ > L H > M > S favorabl e uplands of Mba Plain short·strawed va· Chad . .. . .. . .. . 10 H > I [ > L H > M.S rieties such as m 52 and m 2061·522-6·9 can be grown. Egypt .... .. . 5 H, I I , L H,M, S Under poor management by small·scale fa rmers, Gambia ..... • o. 19 I>H,L I > H H >S,M ROK 16, a taller, awned line has consistently per· Ghana ......... 9 I > H I>L H > M, S formed well, shown resistance to major diseases, and Guinea ......... 14 H > I I H > [ Guinea Bissau ... 15 I>H I M > S > H is a better choice for this location. Ivory Coast .. 24 I > L I.L M > H >S At )ldop Plain P 4023d·Tb·14, mAT 104, Seratus Liberia ...... 21 I L H > M >S Malam Muta nt, ITA 212 , ITA 222, and ROK 16 gave Madagascar 20 I > H > L I H > S > M high yields. But the two ITA lines, which have short Malawi ...... 10 L L > H, I S straw, may not perform well under farmers' con­ Mali . . . . .. . . 7 H >I I H > I ditions . The other entries will be further tested for Nigeria ...... 18 I > H >L L >I S > H, M yield stability (Table 1.41). Senegal ..... 20 I > H I > L H >S Sierra Leone. 24 I>H,L I > L H, M > S Irrigated Rice Tanzania . 12 L >I L > I S > H Zaire ........ 10 I > L L>HI, I H.M > S In various trials conducted at Mba Plain, ITA 222, Zambia. 0 • • •• 10 H,I > L 1>H1.L S > M ITA 121, ITA 212, CICA 8, X·3·DT, Cisadane, m 3273· 339·2·5, m 21015·80-3-3·1·2, and ITA 232 have per· 1 L = low, I = intermediate, HI = high intermediate, H = high. 28 = soft, medium, H = hard. formed well (Table 1.42). Of these ITA 222, CICA 8, ITA 212, and m 3273·339·2·5 have given stable yields Upland Rice and shown acceptable grain quality and moderate disease resistance over the years. These four lines are In the trial for early lines at Mba Plain, mAT 10 gave definitely superior to Tainan V, the currently recom­ the highest yields, followed by ITA 120, mIM209, and mended variety at N dop. Information about them has 28 Rice Improvement Program Table 1.40. Performance of selected entries in upland In the trials at Ndop Plain. B29838-SR·51·1·2·1. rice trials in Cameroon, 1984 IR 7167-33-2-3. IR 22082-41-2. B29838-SR-51-1·2-1. Mba Plain Cisadane. IR 2061·522·6·9. and B 2161·C·MR·57·1·3·1 Days to performed well in 1984. Because IR 7167·33·2·3 and B Grain, Plant 50% Blast 21Bl·C·MR-57-1-3·1 have consistently done well over yield, height, flower, Lear Neek the years and proven superior to Tainan V. infor­ Genotrpe kg/ha em DAS 0-9 0-9 mation about them has been given to the parastatal Advanced trial early maturity organization UNVDA (Upper Nun valley Develop­ IRAT 110. . . . . . . .. 3.471 100 79 3 3 ment Authority) for large scale testing in farmers' ITA 120. . . . . . . . . . 3,363 95 82 3 3 fields. Preliminary results indicate that the lines IRIM 209. . . . . . . . . 3,070 103 79 3 3 are performing well and gaining acceptance among IRAT 10 (Check).. 3.040 84 77 3 3 farmers. IRAT 107. . . . . . . . . 2,758 III 80 4 3 A total of 2.406 new entries were introduced for Average Yield. . .. 2A89 testing under upland. rainfed lowland. and irrigated LSD, 5% ......... . 902 C.V., %.. . . . .. . . 24.9 Advanced trial medium maturity ill 52 .. ...... .... 3,182 79 107 3 3 Table 1.4Z. Performance oCselected entries in ROK 16 . . . . . . . . . . 2,698 140 101 3 3 irrigated rice trial in Cameroon , 1984 IR 2061-522-6.9. . . . 2,601 87 96 2 5 Mbo Plain CICA 8 .. .. .. .. .. 2,403 86 III 4 7 IRAT 79, Control. . 1,767 110 99 3 3 Days to Average yield. , . . 1,948 Grain, Plant 50% Blast LSD, 5% . ......... 594 yield. height, flower, Leaf Neck C.V .• %. ...... . . 22.5 Genotype kglha em DAS 0-9 0-9 Disease sca le : 0 = no replication. Advanced trial 9 = very susceptible. ITA 222 .... ..... . 5.407 90 105 3 3 ITA 121. ......... 4.896 85 103 3 3 Table 1.41. Performance of upland rice lines at Ndop Cissdane . . ...... . 4.707 94 Itl 3 I Plain IR 3273-339·2·5 .... 4.705 90 107 4 3 ill 54 .. . ....... .. 4.641 89 108 3 3 Days to Tainsn V, Control. 3,540 96 93 5 3 Grain Plant 500/0 Leaf Brown Average yi.eld ... : 4,393 yield, height, Rower. scald, spot, LSD. 5% ......... 816 Genotype kg/ha em DAS 0-9 0-9 C.V .• % ....... .. 13.0 Preliminary trial African Low land Rice Yield Nursery P 4023d·Tb-14 . . . 5,671 93 119 2 2 ITA 232 ..... ..... 5.509 96 lOS 5 3 IRAT 104. . . . .. . 3,746 93 129 2 1 ill 21015·8()'3-3·1·2 . 5.306 96 107 3 3 Seraturs rnalam ITA 212 . . ..... ... 5.090 88 106 4 3 mutant . . . . . . 3.581 94 118 2 2 CICA 8, Control .. 4.829 82 no 4 3 ITA 212 .. . . . .. . 3.353 65 136 2 2 ITA 123 .......... 4.669 92 100 4 5 M 18.. . ....... . 3,289 96 115 2 1 IR46 ....... . . .. . 4,648 93 108 3 3 Salumpik it. . . . . 3.242 104 119 3 2 Average yield 4.513 ICA 25. Control . 1.857 85 96 3 2 LSD. 5% ......... 1.035 Average yield.. . 2.832 cv., Owo83D];m-LSRW . .... . •••• • •••••• •• • " • • 0 • •• ' •• 4.7 131 6.2 3.7 12 4.3 1.5 Akure83DMR·LSRY . ... . .......... . . . ... . 4.6 124 5.9 3.7 13 4.1 2.0 Ikenne 83 DMR·LSRW . . . . . ... . . . . ... . .... .. . 4.4 122 6.3 3.5 18 3.5 1.5 Ikenne 83 DMR·LSRY ..... . ...... . ... . ..... . •• 0 • • 0 " 4.4 119 5.8 3.6 17 3.8 1.3 Across83 DMR·LSRW .... ..... . . . .. . .. . ... . . . . .. . . . 4.3 119 5.6 3.2 8 4.2 2.8 Across 83 DMR·LSRY .... ' 0 • • ••••.•• , • •.•••• •• 0 •• 0 0 0 4.2 114 5.9 4.0 10 3.3 2.8 Owo 83 DMR-LSRY ..... . , •• • • 0 ••• 0 • , •• , •• • 0 • • • • 0 0 • 0 4.2 114 5.6 3.3 13 3.7 1.8 Akure 83 DMR·LSRW .... ... ..... . .................. 4.1 114 5.4 3.5 18 3.5 1.5 Controls Ikenne 81 TZSR· y.1. . • , 0 0 , • ••• • 0 • • • 0 , • 0 0 0 • 0 0 ' • 0 0 • 0 0 • 3.7 100 6.4 0.2 91 4.5 1.8 Across 81 TZSR-W·l '0' 0 0 0 •• • •••• • 0 . ' 00 • • • • • 0 •••• 0 •• 3.6 100 5.6 0. 3 91 4.9 1.8 TZB . .............. 0, • • ' •• • • 0 •• •• •• • • ••• • , 0 .0 •• • •• 3.2 89 7.1 1.4 73 1.1 5.0 Western Yellow. • •• 0 . 0 •• 0 •••• 0 •• • • • 0 • • • • • • • • • • • • 0 • • 2.5 68 6.1 0.4 93 0.9 5.0 LSD. 5% ....... . . . ....... . . . .. . . . .. . . .. ........... 1.2 16 1.2 0.9 CoV., ~~"""'" 0 ••• , •• • •••• , • • ••••••• • • • • , . , •• , •• 31 44 25 21 [Figures are computed within wbite and yeUow varieties Beparately. using appropriate control variety. 21 = resistant, 5= susceptible . Afaize Improvement Progra.m 39 under streak compared with yields of widely grown Table 2.8. Parent populations of experimental controls ranging from 0.4 to 1.4 tjha under mildew and varieties being converted to streak 0.9to 1.1 under streak environments. Streak resistant resistance varieties from TZSR·W·l and TZSR-Y-l show good Pop. Backcross yields under streak but very low yields under downy Parent population no. Description generation mildew attack. The DMR·LSR varieties gave good Tuxpeno-I . . . . . . .. . . . 21 Late white dent ... . ... BC, yields under severe attack by streak or downy MezcJa tropical blanca 22 Late white semi-dent .. Be, mildew. Blanco Cristalino-3 . .. 25 Late white flint ... . . . . Be. Populat ion improvement in the late (DMR-LSRW, Amarillo dentado . . . . . 28 Late yellow dent. ..... BC; DMR·LSRY) and early (Dl'.'!RESRW, DMR·ESRY) Tuxpeno Caribe . . . . .. 29 Late white dent. ...... Be, maturing populations continued during the year. Blanco Cristalino-2 . .. 30 Early white flint . . . . BC, Resistant plants in half·sib families under downy Amarillo Cristalino-2. 31 Early yellow flint ..... BC. mildew pressure at Kabba and Owo were selfed. The Antigua x Republica In termediate yellow S, progenies were further screened under streak Dominicana. . . . . . .. 35 dent . .. . .... .... .. BC, La Posta . . . . . . .. . .. .. 43 Late white dent. ...... BC. pressure and advanced to S, during the 1984/85 A.E.D. x Tuxpeno. . . 44 Late white dent . ... . .. BC" irrigation season. They will be recombined in 1985, Blanco dentado-2 . 49 Intermediate white then full·sibs will be generated for international dent .............. BC, testing in 1986 in areas infested by downy mildew and White flint QPM. . 62 Whi te qua lity protein streak.- J.M. Fajemisin maize . .... . .. . .... BC3 YeUowdentQPM .. . .. 66 Yellow quality protein COD version of experimental varieties to streak maize .. . ..... . .. . . BC ) resistance. The conversion of experimental va­ TZB .......... .... .. Late white Oint ....... BC) rieties to streak resistance continued, with the 13 source populations shown in Table 2.8. Varieties from each population have performed well in previous Table 2.9. Comparison under negligible streak variety trials in Africa, and they are converted to incidence of streak resistant conversions increase yield stability. Conversion is by backcross­ with normal counterparts at four locations ing, and for each backcross, the most promising in Nigeria, 1984 experimental variety from the respective population Grain Days Plant Ear Ear is used as the recurrent parent. yield , Yield to ht. , ht .. aspect, One or more ofthe conversions were included in all Variety kg!ha index l silk em em (1- 5)' the streak·resistant experimental variety trials con­ Poza Rica 7822 5,351 100 58 207 105 2.6 ducted in 1984. Those results are given in other EV. 8322·SR BC, 5,247 98 59 211 109 2.2 sections of this report. In some of the late maturing Across 7729 4,930 100 59 194 103 2.6 variety trials conducted in Nigeria, the streak sus· EV. 8329·SR BC, 5,268 107 57 209 105 2.8 ceptible counterparts (recurrent parent) were in· Poza Rica 7843 4,663 100 60 218 118 2.3 cluded as controls. Yield and other agronomic char­ EV. 8343-SR BC, 4,883 100 59 208 107 2.5 acters of the streak resistant conversions were similar to those of their susceptible counterparts Across 77283 4,890 100 60 225 126 2.1 (Table 2.9). Table 2.10 compares the conversions with EV 8328-SR BC? 4,772 98 57 219 125 2.1 their counterparts under heavy artificial streak ISusceptible variety = 100. infection at UTA. The streak susceptible varieties :21 = exce llent , 5 = very poor. yielded virtually nothing, whereas the resistant ;1 Data from three locations only. conversions yielded well and grew normally. The susceptible varieties were stunted and their flower· Table 2.10. Streak resistant conversions compared ing was delayed. with normal counterparts under streak We do not plan backcrossing beyond the BC, pressure at UTA, 1984 generation of each conversion; however, variability Grain Days Plant Ear Streak for yield and other agronomic characters in the yield, to ht ., ht., rating, converted materials seem to justify selection for Variety kg!ha Bilk em em (1-5) ' higher yield and increased uniformity. Scientists in Poza Rica 7822 ... 156 61 101 56 5.0 national programs are encouraged to do that work . EV. 8322-SR BC, ... 4,643 57 184 95 1.0 Some of this work has already been initiated. We Aeross 7729 ... 153 61 88 41 5.0 generated full·sib families while converting experi· EV. 8329-SR BCa. . 4,269 56 182 91 1.3 menta l varieties from population 43 after the Be3 generation and in the conversions from populations Poza Rica 78432 . • . .• . •. 30 and 31 after the BC. generation. Full-sib families EV. 8343-SR Be, .... .... 5,156 59 201 116 2.0 derived from population 43 were tested in a 13 x 13 Across 7728 ............ 507 61 136 74 4.5 simple lattice at !kenne, Samaru, and Mokwa, all in EV. 8228-SR BCa. ....... 5,632 56 163 81 1.0 Nigeria. The trial in Mokwa was planted in a field I] = excellent. 5 = very poor. heavily infested with Striga, a parasitic weed. 2No data obtained. 40 Maize Improvement Program Distinct differences were observed for tolerance to administrators, and extension staff the yield super­ Striga among the families. The most tolerant families iority of maize hybrids on farmers' fields. That was were recombined in a streak nursery at nTA in the achieved in both the forest and savanna areas. More 1984/85 dry season, to form a variety resistant to both than 3,000 farmers participated in 13 organized field Striga and streak. Another variety based on the days (3 forest and 10 savanna). Tables 2.11 and 2.12 combined data from Ikenne and Samaru also will be show the performance of maize hybrids under recom­ formed. Full·sib families fro m conversions of popu· mended management in the savanna and forest areas, lations 30 and 31 were evaluated in Ikenne during the respectively. The management involved 22- 25 x second season of 1984, and the best families were 75-90 cm, spacing 140 N, 60 P, and 60 K kglha selected to form uniform early maturing varieties. including top dressing with N, and good weed con tro l. Late, white, streak-resistant, backup pooL The best performing white hybrid under good This pool is asource of new germplasm for TZSR·W·l. management in the savanna was 8322·13. It yielded In 1984, 532 half·sib families from the late white 8.2 tlha (mean of six locations). N ext were 8321·21 and backup pool were planted at Ikenne in the forest zone 8321·18 with average yields of 7.9 and 7.3 t fha, r e­ and at Samaru in the savanna zone. The families were spectively (Table 2.11). The three hybrids gave uni­ grown in isolated, half·sib recombination blocks. formly high grain yields and showed desirable agro­ Selection was for plant type and yield potential nomic traits in all the locations. The best performing among and within families. At harvest, 550 ears were yellow hybrids were 8340·12 and 8341·12. 8322·13 selected from bot.h locations for the seventh cycle of yielded twice as much as Local TZB, one of two key selection.- M. Bjarnason varieties in Nigeria. In the forest zone, white hybrids 8326-17,8346-3, and 8322-13 yielded best while 8329-15 Hybrid Project was superior to any other yellow hybrid (Table 2.12). The white hybrid 8322-13 performed well in both The major objectives oflITA's hybrid maize project savanna and forest locations with an average grain were continued in 1984. We tested 24 hybrids under yield of 8.2 t fha in the savanna and 5.9 tlha in the on-farm conditions , offered hybrid maize trials and inbreds to national programs, worked to develop forest zone. The best farmers in the 1984 on·farm trials in the savanna averaged more than 8 tlha and in the high-yielding, stable hybrids for lowland for est and forest zone , the best averaged from 6.8 to 7.7 t/ha. In savanna areas, selected high·yielding, disease­ resistant hybrids adapted to mid·altitude ecology, the overall performance (yield, farm maintenance. and extension role), several farmers should be named conducted agronomic sturnes of selected hybrids and master hybrid maize growers for 1984. open-pollinated varieties, and helped to produce Performance of hybrid maize in on-farm trials hybrid maize seed. during the second rainy season in the forest area of On-farm demonstration trials. The special pro­ Nigeria , 1984, was also impressive (Table 2.13). Most ject supported by the Nigerian government in 1982 of the hybrids selected from the forest second season produced 24 high yielding hybrids with good agro· were similar to those selected from savanna 10' nomic traits that we tested in 1984 in O.I·ha plots in cations, and agreed with our 1983 experience. on·farm demonstration trials in the forest and sav­ Environmental conditions during the second season anna zones of Nigeria. Participating farmers were in forest seem to be simi1ar to savanna environment selected on their voluntary interest in hybrid maize, (good sunshine) except for a short rainy period. and their willingness to provide fertilizers and her­ Average yield of the selected hybrids was 5.0 t/ba, bicides, on r eadily accessible land. The National high enough to encourage future maize grain pro­ Cereals Research Institute (NCR!), the National duction in forest second season. Excessive rain in the Accelerated Food Production Project (NAFPP), the first season presents great difficulties for drying Federal Department of Agriculture (FDA), and the maize. Good stalk quality and relatively early matur­ Ovo State NAFPP collaborated with lITA in select­ ing hybrids or open·pollinated varieties may boost in'g farmers and monitoring the trials in all locations. malze production in forest second season In the savanna on-farm demonstration trials, the environment.- S.K. Kim, e.O. Alofe, L. T. Ogunremi, Kaduna State Integrated Rural Development Project Y. Efron, F.H. Khadr, J. Fajemisin, T. Adenola (KSlRDP) and the zonal lRDPs, the Agricultural (NAFPP), and AERLS·ABU. Extension and Rural Liaison Services (AERLS) of the Ahmadu Bello University, Federal Department of International testing of maize hybrids. Two Agriculture (FDA), Kaduna, and the lITA collabo· hybrid maize trials (white and yellow) were offered to rated. In the savanna area, on-farm trials were interested national programs. At least 17 African located in Kaduna, Kana, Sakoto, Niger, Bauchi, countries and other institutions outside Africa re­ Kwara North, and Plateau states. In the forest areas, quested the trials. The Nigerian zonal hybrid trials states covered were Oyo, Ogun. Lagos, and Ondo. conducted by the National Cereal Research Institute, In the savanna and forest zones, data were obta ined Moor Plantation, Ibadan tested some selected hy· from about 150 ha on mostly small-scale farms and a brids at 10 locat ions . few big farms. The primary objective of the on·farm Selected hybrids from 1984 on·farm demonstration trials is to convincingly demonstrate to farmers,local trials throughout forest and savanna locations in Maize Improvement Program 41 Table 2.11. Grain yield (t/ha) of selected maize hybrids from on-farm demonstrations, savanna, Nigeria 19M (all white exept for t,,·o yellows, 8340-12 and 8341-12) Grain yield, t/ha No. of Hybrid %1 Ra nge locations Remarks 8322·13. . ..... . . . .. .. . . . . . .... . .... .. . 8.8 157 6.9-11.8 6 Res. s tr eak , Striga, lodging 8321·21. . . .. . . . . . .. .... .. . ... ... . .. .. . . . 7.9 141 6.1- 11.1 12 Res. str eak, Striga. lodging 8321·18. . . .. . . . . . . . . . .. . . . . . .. . . ... . . . . . 7.3 130 4.9- 10.3 11 Res . streak , Striga, lodging 8321·11. . .. . . . . . . . . . .. ... . ... .... ... . . . 7.0 126 5.4- 9.4 18 Late. res. streak. S triga 8346-3. . .... .. ... . . . . . . .. ... . . . .. . . ... . i .O 126 5.2- 9.2 11 Short . r es. st reak, lodging 8338-1. . . . . ....... . .. . ...... . . .... . .. . 7.0 126 5.9- 8.0 6 Res. streak, lodging 8322-3. ..... .. ..... . .. . .. . .. . .... . 6.9 124 3.7 · 9.0 16 Short. res. s treak, Striga, lodging 834()'12 .. . .. ..... .... . ... . .. . .. .. . 6.9 124 5.8- 8.0 2 Early, res. to streak, Lodging 8341·12. . . . . . . . . . ..... . . .. ..... . 6.6 118 5. 1- 7.8 4 Early, r es. to streak, Striga TZSR·\V·l . ... . .. . .. ... .. . ..... . . 6.4 100 3.6- 7.7 15 Late r es . streak Local. .. . ... . . .. ... . . . .. . .. . .. .. . 4.1 74 1.9- 5.2 5 Late. good adaptation in savanna I Percentage of TZSR-W -1. Table 2.12. Grain yield (t/ha) of selected maize hybrids from on-farm demonstrations, forest, first season, Nigeria, 1984 (white except for two yellows, 8329-15 and 8341·6) Grain ~ield , t /ha No. of Hybrid ~/O I Range locations Rema rks 8326·17 . ..... . ... . . ..... . . . .. . ... .. ...... .. . .. . 6.0 126 4.0--8.4 10 Res. streak /lodgin g 8329-15 . .. . ....... . ..... . ... .. .. ... . .... . .. ... . 6.0 126 4.4- 7.5 24 Res. streak , e xcellent tip cover 8322-13 .. . . . .. ... . . .. . . . . ....... . .. . ..... .. . .. . 5.9 125 5.0--9,4 4 Res. streak/lodging 8341·6 . .. . .. .... ... . . . .. . . .... .. . .. .. .. . .. . . .. . 4.8 94 2.1 -7.1 16 Early res. streak/lodging TZSR·W·1 .. .... .. . . . ... . .. . . . .... . .. . .. ..... . . 4.8 100 2.2-6.7 23 Res. streak . sus. lodging Onne TZPB ............ ......... . . . . .. . ... .. .. . 4.7 98 4.4--6.1 3 Res. rust , blight Local . .. . . . ... . . . .. .. ... ... .. .. . . . . . . . .. . . . . . . 2.3 48 1.0--3.5 15 Adapted 1 Percentage ofTZSR.W.l . Table 2.13. Grain yield (t/ha) of selected maize (Table 2.45. last t able in t his r eport). In t he Republic hybrids from on-farm demonstrations, of Benin. two hybrids , 8322-13 (8.1 t) and8328·10(7.4t), forest , second season, Njgeria, 1984. (Two yielded significantly more than the best open· yellow, 8330-16 and 8329-15) pollinated varieties used as controls. No. of In the Internationa l hybrid maize trials planted at Grain yi eld. t/ha loea· IITA, Ibada n under artificial streak i nfestation Hybrid %1 Range tions Remarks2 (Table 2.15). selected hybrids under heavy streak 8330-16. 5.0 162 4.8- 5. 9 2 Res. lodging pressure yie lded significantly more than the top· 8322·3. . 4.8 155 4.6-4.9 2 Res. lodging, Striga yielding, streak·resistant, open·pollinated variety 8329-15 .. . 4.3 139 2.6-5.5 8 Res. lodging, Striga (EV 8343·SR) and the susceptible control (TZB). 8321-18 . . 3.9 126 2.1-5.9 9 Res. lodging, Striga UTA·developed maize inbred lines were requested TZSR·W·l 2.9 100 1.3-3.9 9 TZESR·W 2.9 94 2.2- 3.3 by at leas t 42 institutions worldwide in 1984; 11 3 African nationa l programs, many USA priva te seed I Percentage of TZSR· \V. l. companies, and others in Asia. Additionally, seeds of 2All s ix resist streak. more than 500 advanced lines were shar ed with CIMMYT, Mexico, and USAfD Projects in Honduras Ni geria performed equally well under experimental (Table 2.16).- S.K. Kim , F. Khadr, Y.Efron, and J . sta t ion conditions. In Nigeria best selected hybrids Mareck from the experimental fields remained 8322·13 (W), 8321·18 (W) for savanna, and 8326-17 (W) and 8322·13 New hybrid and line development. The hybrid (W) for forest zone (Table 2.14). Hybrid 8346·3 (W) is a maize project depends on continuous development of high yielding topcross (O.P. variety X inbred line). new hybrids and improved parental inbred lines. For Relatively low yield performances by hybrid maize t he lowland forest and savanna areas, high yielding were obtained under the experimental conditions in yellow hybrids are emphasized. More than 500 new 1984. hybrids developed from the advanced fixed lines P erformance dat a on hybrids from outside Nigeria (S. -S. stages) were tested at seven locations. Two are limited, but hybrid maize performed equally well high yielding, open.pollinated varieties selected in Cameroon and RepUblic of Benin. In Cameroon at previously were included as controls. Among the new four locations, four white and yellow hybrids out· hybrids tested , three modified single crosses, one yielded the best open·pollinated varieties 30 to 47% three· way cross, and three new single crosses were 42 Maize Improvement Program Table 2.14. Grain yields (t/ha) of white hybrid maize eties and private seed company (Pioneer and Funk) at experimental stations in Nigeria, 1984 tropical hybrids. ~id· All lines were ad vanced from 8 3 to S, during first Forest, Savanna altitude Across locations a nd second rainy seasons at UTA, Ibadan (forest) or Hybrid 2 10c. 410c. 110c. mean % control Samaru (savanna).- S.K. Kim, F. Khadr, Y. Efron, J. 8322-13 .. . . . 7.2 7.0 2.5 6.5 133 Fajemisin, and Z. T. Dabrowski 8326-17 .... . 6.1 6.5 8.2 6.5 133 8321 ·18 .... . 6.9 6.4 4.8 6.3 128 Mid-altitude line and hybrid development. 8346-3 ... . . . 6.4 6.3 4.7 5.9 120 Hybrid maize is widely grown in East and Southern Cantrall . . . 6.0 4.8 3.5 4.9 100 African countries. Major objective of this mid­ altitude, line·development project was disease reo 1 Control was open pollinated variety Poza Rica 7822·SR. sistant, inbred lines for national programs. Target diseases are maize streak virus I maize mottle/ Table 2.15. Grain yields (t/ha) of hybrid maize over chlorotic stunt, H. turcicum blight, P. Borghi rust, open-pollinated controls under streak and ear rot. Hybrids selected on 1983 test cross pressure at IlTA, Ibadan.l984 data were sent to Cameroon, Tanzania, Zambia. and Yield Yield Streak Zimbabwe. All 75 advanced hybrids and 250 test ":tbrid Type' k~:lha index2 index3 Tating~ crosses were pla nted at Vorn (National Root Crops 8322·13 .... Single cross 6.7 145 857 1.0 Research Institute), Plateau state, Nigeria (1 ,300 m 8328-10 .... Single cross 6.2 133 792 2.3 above sea level). All advanced hybrids were planted 8338-1. .... Single cross 5.~ 123 726 1.0 in four-row plots with four replications, while test 8321·18 .... Single cross 5.5 120 709 2.0 crosses were planted in single-row plots with two to 8326-17 . ... Single cross 5.3 115 680 1.8 three replications. Two open-pollinated, mid-altitude PR7843·SR a.p.SR 4.6 100 591 2.0 varieties, TZM8R-W and EV 8344·8R. were included 8346-3. Top cross 4.5 98 579 3.3 as controls in all trials. 8346-1. . _. _ Topcr06s 4_2 90 534 2.8 Grain yields and major traits of the selected hy· 9091 ...... Inbred 2.4 51 301 1.8 GusauTZB a .p. non·SR 0.8 18 106 5.0 brids from the advanced trials are summarized in PR 7843 ... a .P.non-SR 0.7 16 94 5.0 Table 2.18. Four selected hybrids significantly out· yielded (6G-77%) the top·yielding, open·pollinated LSD,5% .. .. ............. 1.0 0.6 TZMSR·W. All selected hybrids carry resistance to C.V.,% .. .............. . . 15.0 17 the major maize diseases. H. turcicum was artificially lType of germ plasm (D.P. = open pollinated). inoculated in all maize trials and line performance 2Yield index _ % yield of test materials over the open pollinated. tests. Relatively good selection was made against H. strea k resistant (SR) control. PR 7843·SR. turcicum. Striga aspera infestation was not uniform, 3Yie ld index = % yield of test materials over t.he average of two streak susceptible D.P. controls. but ,ome hybrid. and lines are more susceptible "Streak rating : 1 = resistant, 5 _ susceptib1e. (stunting and poor ear development) than others. One set each of internationa l hybrid t ria Is was also planted at Vom Station to observe reactions of hy­ Table 2.16. Countries that requested IlTA maize brids in mid·altitude ecology. Two lowland hybrids, inbred lines in 19841 8326-17 (Tzi 17 'x Tzi 18), and 8346-1 (Pop. 49 x Tzi 10) Africa: Ghana, Ivory Coast, Egypt. Nigeria, Senegal, had good yields in the mid-altitude ecology. Both Cameroon, Ethiopia. Kenya, Reunion, carry resistance to H. turcicum. Tanzania, Zambia, Zimbabwe. More than 200 new mid-altitude hybrids will be Asia: PalUstan, Philippines, Thailand (3), Korea. developed from the 1984 data. Additional seed of the America: USA (11). Columbia, Mexico (CIMMYT), selected advanced hybrids will be produced during Guatemala. Honduras, Venezuela, Brazil dry seasons for 1985 testing by the interested national (17). programs. Development of germplasm for the mid· altitude ecology will in future be carried out in a Europe: W. Germany. program jointly funded and staffed by CIMMYT and INumber of insti tutions requesting seed is given in parenthesis. UTA and based in Zirnbabwe.-S.K. Kim, F. Khadr, L. Everett. and J. Fajemisin promising (Table 2.17). We selected modified single Hybrid maize seed production. A major obstacle crosses primarily to increase hybrid seed. Two re· to hybrid maize development and cultivation in lated sister,l ine crosses (F 1) were used as seed parent. developing countries is that seed production is re­ About 15 to 30% seed yield increase is expected by latively complicated and difficult. A private seed this approach. Approximately 50 kg of seed of each company, Agricultural Seeds Ltd. , produced 85 tons selected hybrid will be produced during dry seRSon of hybrid maize seed from more than 100 hectares in for future tests. 1984 in savanna area, supervised by UTA and New line development has been vigorously pursued National Seed Service staff mainly in Kaduna state, with additional germplasm sources. Major materials and 30 ha in forest area in Oyo and Ondo states. This are derived from new CIMMYT experimental vari, was the first attempt to produce hybrid maize seeds in Maize Improvement Program 43 Table 2.17. New maize hybrids selected across locations in Nigeria, 1984 Yield, Yie ld. Number of Hybrid Pedigree t/ha index' locations Modified siogle cross 8435·10 . .... . . ....... .. . .. . ... . .. .. . .. .. . . . .. .. . . . . . ... . (9071 x 9091) x 1368 7.1 99% 7 8435·20 ....... .. . .... . .. ... .. .. .. . .. . .. ... . . .. .. . . .. . .. . (9071 x 9089) x 1393 7.1 99% 7 8435·8 . .... ... .. . .... . .. .. ... .. . .. ...... . .. . ....... .... . (9071 x 9030) x 1368 6.8 95% 7 3-waycross 8434·11 ... . . . . ......... .. . .. ... ... .. .. ...... .. ... .. .... . (4001 x 9848) x 9490 6.9 98% 3 New single cross 8428·19 . . . .. ..... .. .. . ...... . . ... ........ .. .. .... .. .... . 1368 x 5012 8.0 134% 6 8425-7 ... ... ......... . . . . .. . ... .. .. . .. ...... .. ... .. . .. .. 4001 x 9432 6.6 129% 7 8425-9 .. . ..................... .. . .. . .. . . ...... . .... . ... . 4001 x 9485 6.3 123% 7 : Yield index = Modified single crosses and 3-way crosses: % yield of new hybrid over control 8322-13 (single cross) and new single cr osses: % yie ld of new hybrid over two open-pollinated controls, G U 88 U TZB and PR 7843-SR. farmers' fie lds in Nigeria. A one·day training course rice varieties to stem borers. was offered to a ll contract seed growers before plant­ We had two 9- x 35-m screenhouses const ructed ing. IITA scientists and two t echnicians supervised where we developed new methods to screen ma ize for seed production in the sa vanna areas. The target was r esistance to the mottle virus disease by releasing C. 1 t /ha of seeds (i hybrid, t inbred seed inc.rease of triangula viruliferous leafhoppers. We reared C. male). Fifteen hybrids averaged approximately 750 triangula insects only on millet because raising t hem k g/ha of hybrid, and 500 k g of male seed increase. on maize would risk contaminating them with the Hybrid seed yields varied widely with management, Vlrus. site, and field conditions. Some plantings were de­ Rearing methods, infestation technique, and layed more than one month. Good site selection with eva luating resistance, however, should be based on fertile soil, good seed bed preparation, and adequate so lid information regarding the influence of biotic fertilizers are major items for success. Detasseling and a biotic factors on insect ecology, biology, and should be carefully supervised until seed growers behavior r elated to food habit, oviposition, or move· fully unders tand the importance of seed purity. ment. So we constantly continue experiments and Hybrid seed growers received 3,000 naira/ton com­ field observation s on maize-pest biology, ecology, and pared with approximately 600/ton for commercial host-plant r elationships. Data accumulated can be maize grain so most of the 1984 seed growers want to used later in developing new strategies for integrated continue producing hybrid seed. As expected, hybr id management of maize pests in Africa. maize seed production in the south (forest area) During 1984 we trained many groups to use t he proved more difficult than in the north (savanna). methods. techniques, and equipment needed to start Timely harvesting and drying should be studied for rearing insects artificially and to use them in evaluat­ future hybrid seed production in forest zone , Nigeria. ing maize resistance. We a lso taught them to identify Several modified methods of the single·cross. seed· major Cicadulina species occurring in Africa. As a production system were studied at UTA during the r esult, r esearch groups in Burundi, Zambia , Togo, dry season. By changing female and male row ar· Burkina Faso, and Zimbabwe have initi ated or plan· rangement from conventional 2 :1 ratio, seed yield ned insect·rearing projects and releasing Cicadulina may be increased 20 to 30%. Studies will continue leafhopper for resistance screening. during the 1985 cropping season.- S.K. Kim, L .A. An expedition was organized in Togo , a t the rep­ Oke, T. Bamidele. and F. Khadr quest of the Director of Agricultural Research, to identi fy the commonest Cicadulina species, which Entomology may be used later for mass rearing by the national maize program. Cicadulina spp. A constant, uniform insect popu lation is essentia1 to Table %.18. Selected disease resislant (all resist screen maize for resistance under artificial infes­ streakt H. turcicum, and P. sorghl), mid· tation . We are now rearing two groups of insects : altitude maize hybrids in Nigeria, 1984 (1) Cicadulina leafhoppers t o screen maize for r esistance Yield, Yield to maize streak virus (MSV) and the maize mottle/ Hybrid Pedigree t/ha index chlorotic s tunt diseases; and (2) stem borers to 8326-17 . . . ....•. . • . . .. •.. . . 2098 x 9091 8.3 177% screen new maize crosses and varieties for r esistance. 8445-6 .................. . 120 x 152 8.0 174% Addit ionally, Dr. M.S . Alam used many eggs and 8445-1 . . . . . . .. . . •. . . . . . . .. 129 x 150 7.6 184% larvae of the pink stalk borer, Sesamia calamistis , 8445-8 .... . ..... .. .. .. .. .. 154 x 32 7.4 160% provided by our Insect Rearing Laboratory to screen TZMSR-W. . . ... . . .. .. .. .. . O.P. 4.7 100% 44 Maize Improvement Program Species composition in West Africa. We Com· 1.25 m', with a range from 1 to 19. In northern and pared efficiencies of sweep nets and net cages for central Togo and in Plateau areas, averages were sampling Cicadulina leafhoppers from grasses and from 2 to 2.4 insects/sample unit. Sites with grass maize by collecting insects from selected s ites in species that the insects prefer yield many more various climatic zones in Nigeria each month of the insects tha n the randomly selected sites. Knowing growing season. Leafhoppers were also sampled in that should help one plan a sampling procedure to four areas in Togo in November 1984. Total collect the many insects needed to initiate a local Cicadulina leafhoppers. sex ratio. and species com· maize-resistance screening program. positions were recorded . Significant d ifferences were observed in the species Since Cicadulina individuals are used to monitor composition of local CicaduLina populations in var· their natural infection with MSV under field con· ious climat.ic zones. But CicaduLina mbiLa was ditions, an efficient method for co llecting live leaf· domin ant in most areas sampled in Nigeria (Fig. 2.2) hoppers is needed. A catcher we found efficient is a and Togo (Fig. 2.3), with marked differences among C. frame (or four corner supporters) of aluminium pipe mbiLa populations. Those from Jos Plateau were placed on the ground with fresh grass or cereal plants darker than those from the northern Guinea savanna and quickly covered with a dark green cotton cloth and the forest zone . The portions of C. arachidis and except for one side of fine, transparent n etting. C. simi/is never exceeded 15% of the total. C. mbila Leafhoppers (and other insects) responding to the populations occupied lowland and mid·altitude ecol· light source, move from the grasses onto the net. ogies, but not where daily t emperature exceeded where one may selecti vely collect them with an 2BoC. Then 49% were C. trianguLa species in northern aspirator. Nigeria and 57 to 61 % in central and northern Togo, We always collected large numbers of Cicadulina with their higher average temperatures. In all cases, insects at the end of the rainy season and from 65 to 80% of the Cicadulina were females. irrigated maize, wheat, and gr asses (Fig. 2.1). Dr. M.S.K. Ghauri of the Commonwealth Institute Leafhoppers are not randomly distributed in grass. of Entomology, London, U .K. confirmed the identity land . They aggregate on younger plants of Digitaria, of our CicaduLina species .- Z. T. Dabrowski and Eleusine, Pennisetum, Polystachion, and Brachiaria. VA.O. Okatlt In the Kadawa Irrigation Scheme (Kana, northern Nigeria), the number of leafhoppers from consecutive Comparative biology and virus transmission. samp Lings varied from 6 to 57 per 1.25 m' of area To determine the percentage of active MSV transmit· sampled (Fig. 2.1). t er s among native Cicadulina , we reared native In Togo, the coast area provided the most species separately and monitored them on young CicaduLina leafhopper s, an average of 10.5 adultsl maize seedlings. To improve mass r earing, we also Total rainfall (mm) 300 Moor Kadawa Jos 240 Plantation Kano 800 Ibadan 180 600 120 400 60 200 OL-----~----------~ O~~.L ____~ L-~ Average-number of Cicadulina spp/1.25 m2 na--------------~ 20 20 16 16 12 12 8 8 4 4 o o SONDJFMAMJJASOND SONDJFMAMJJASOND SOND J FMAMJ JASOND 1983 1 1984-- 1983 1 1984--- 1983 I 1984--- Figure 2.1. Populations of Cicadulina leafhoppers were high soon after rainy season ended. Maize Improvement Program 45 JOS PLATEAU NORTHERN GUINEA SAVANNA ( Kadawa, Kano) ()() TRANSITIONAL ZONE ( Moor Plantation, Ibadan ) ~O~(jO~~eJc; Sep Oct Nov Apr May Aug Sep Oct I Dec 1983 ----+-1--------1984---------1 mbfla g@m c. friangu/a m c. Slim/is • C. arachidis Figure 2.2. c. mbila was the dominant leafhopper in most Nigerian areas sampled. studied tbe biology of one population eacb of C. mbila previous list of six East African Cicadulina species as from Jos Plateau , Kadawa Irrigation Scheme (nor­ MSV vectors. thern Guinea savanna), and Moor Plantation, Ibadan Nymphal development periods of Cicadulina popu­ (transitional zone) and one population each of C. lations that originated from various climatic zones triangula from Moor Plantation and fiT A used for showed significant differences (Fig. 2.4). The C. mass rearing. Development periods of eggs and triangula IITA population that we used in mass nymphs. survival rates, and oviposition were com· rearing for resistance screening developed much pared under controlled and natural conditions. sooner than the wild C. triangula population from Suitabilities of four host plants, maize, millet Moor Plantation. Ibadan. Simi larly, the C. mbila (Pennisetum typha ide), Digitaria exci lis, a nd sorghum population from the Kadawa irigation scheme. Kano, (cultivars Rio and Atlas), were tested for five experi­ developed sooner than either population from Jos mental Cicadulina populations in the screenhouse. Plateau. All Cicadulina populations were three to We studied acquisition· access feeding periods and four times more prolific on miUet than on maize, efficiencies in transmitting streak virus of five sorghum, or Digitaria grass. Because large numbers Cicadulina populations by collecting nonviruli­ of leafhoppers always occupied Digitaria plants ferous insects from separate colonies and feeding under field conditions. we used them in the experi· them on infected plants for one oftbese six times : 30 ment. Some farmers use that host plant as a food crop. seconds. 15 minutes, 2 hours, 24 hours, 48 hours, a nd C. triangula IITA colony was significantly more 96 hours then releasing individual insects on f>. to 7· fecund than C. mbila. Both species had high ovi­ day-old susceptible maize seedlings for 24 hours at position rates the 2nd, 3rd, and 4th weeks after adults 26°C. emerged (Fig. 2.5). We concluded that species with Field catches and screen house bioassays showed high reproductive potential s hou ld be used in mass that, in general, the percentages transmitting MSV rearing and that old females should be replaced with were low at the beginning of the rainy season (from 0 young ones every fourth week. to 3%in thetransitionalzone and 10 to 13% in theJos Detailed studies on maize streak-virus trans­ Plateau zone), indicating a broad reservoir of the mission by various Cicadulina species and popu­ MSV in local grasses. Higher percentages of active lations showed that 46% of the C. triangula UTA transmitters in wild Cicadulina populations were colony transmit MSV after 24 hours of acquisition always observed in the middle and at the end of the feeding on MSV·infected maize plants. That com­ ra iny season, reaching 22.5% in Moor Plantation, pares with 32% of the wild C. triangula population lbadan, and 20% in Jos and Kano areas. from Moor Plantation (Ibadan), 25% of two C. mbila All four species of Cicadulina that were identified populations from Kana and Jos Plateau areas, and in West Africa transmitted MSV. Two additional 23% of C. mbila from Moor Plantation. species C. simi/is a nd C. arachidis, are added to the Virus transmission bioassay showed that only five 46 Maize Improvement Program locations and climatic zones, especially between relative moist areas in Jos Plateau and drier areas of the transitional zone and northern Guinea savanna.-V.A.D. Oko/h, Z. T. Dabrowski, and Y. Efron Stem Borer Research As in previous years, our stem borer research was directed primarily at Sesamia spp., which are preva­ lent in the West African forest zone and particularly severe in southeastern Nigeria. Eldana saccharina is the predominant borer in southwestern Nigeria. The 1984 Sesamia agronomy and breeding trials were conducted at four sites: Umuahia (Amakama Research Station, NCRI), Onne (UTA high rainfall substation near Port Harcourt), Sapele (farmer's field), and Ikenne (lITA research farm). The first three sites are in southeastern Nigeria, and Ikenne is in the southwest. Trials were planted bet ween late July and early August as a "second rainy season" maize crop, the season when stem borer infestations are highest. GHANA BENIN Trials were conducted to determine the effect of fertilizer, insecticide, and variety on plants subjected to na tural infestation of Sesamia stem borers. In PLATEAU DANYI addition, progenies from the TZBR cycle·2 population were screened as part of the ongoing population improvement program. Maize growth was good at Onne and Ikenne but poor and variable at Umuahia and Sapele. Data taken at all sites included plant stand at 3, 6, and 9 weeks and a damage rating at 9 weeks, using the 1- 9 scale shown in Table 2.19. Damage ratings were used to calculate incidence percentages-the percentage of plants present at 3 weeks that showed evidence (a damage rating between 2 and 9) of borer damage at 9 D weeks. Severity rating, the average damage rating of CmbllfJ plants that showed evidence of borer damage, was !Imli'!l! also computed. Percentage survival is the proportion C trtongu/fJ ~~;@,Jm of the plants present at 3 weeks that are still living at IIIJIB 9 weeks. Additional studies were conducted to de­ C. simlHs termine the prevalence and behavior of stem borers at _ C oroch/dis Onne. Each week 100 plants were sampled to de· termine the number and position of the eggs and Figure 2.3. In Togo C. mbila were dominant except where larvae. Larvae were placed in vials and reared to daily temperatures exceeded 28°C. C. triangula species adults for identification. accounted for 57 and 61% of the total in central and Sesamia infestation of young (3-S weeks old) plants northern Togo. was low at Sapele and Ikenne, moderate (30-40%) at Onne, and high (80%) at Umuahia. At harvest (15 weeks), Sesamia infestation levels were low at all C. triangula (Ibadan) and one C. mbila of six acquired sites except Sapele where 78% of the ears showed MSV within 30 seconds. None from any other popu­ borer damage. The decline in Sesamia infestation by lation acquired the virus that quickly. The acqui. harvest in Umuahia (also noted last year) indicates sition rate generally increased with time of feeding. that unknown control agent(s) may be active. High The order of efficiency was: C. triangula (Ibadan) > levels of Sesamia during harvest at Sapele show that C. mbila (Jos) > C. mbila (Ibadan) > C. mbila the timing of seasonal peaks varies at different sites. (Kadawa). At Ikenne, and especially at Onne, Eldanasaccharina A species or a population with the shortest develop­ levels became extremely high at harvest. Eldana is ment period, high fecundity, and efficient virus trans­ normally a pest of maize only during the post­ mission should be selected for mass rearing. flowering period. Eldana were first found at Onne Differences among species and populations may seven weeks after planting but by the 11th week partly explain the variation in MSV incidence among Eldana averaged 2.8 per plant and constituted 95% of Maize Improvement Program 47 Days 18·r--------.--------1r--------~------~r_------_, 01 st 16 ~ 2nd _ 3rd iiHi 4th 14 ~ 5th I III Total I 6 4 2 o U-a.:!. C.mblla (Jos) C. mblla(Kodowo) C. mblla( Ibadon) C.lrlOngu/a(lbadan)C. lriangula (IITA) Figure 2.4. Developmental periods of Cicadulina populations that origi"ne.ted from various climatic zones differed significantly. the borer population. The unusually high levels of variety trial to evaluate progress from selection and a Eldana at Onne show that its range of adaptation is progeny trial to identify superior half·sib families for wider than we had expected. continued improvement of the TZBR populatlOn Agronomy trials. The 1984 agronomy trials can· (Table 2.21). The variety trial consisted of Cycle 0 and sisted of fertilizer and insecticide (carbofuran) ex­ Cycle 2 of the TZBR popUlation, a hybrid control periments. Since the results pertaining to Sesamia at (8338-1), and two open pollinated controls (TZSR-Y:l sites other than Umuahia are either nonsignificant or and Bende White). The varieties were grown In similar, only the results from Umuahia are reported four-row plots. 5 meters long in 10 replications. (Table 2.20). As expected, applying carbofuran m­ Significant, but small, differences were observed for creased survival percentages and reduced the in­ survival and incidence percentages. Two cycles of cidence and severity of Sesamia attack. The purpose half-sib family selection (TZBR C. vs TZBR Co) did of this trial, however, was to find out what would not significantly improve the TZBR population. Both happen to the maize plants if they survived the initial TZBRentries were, however, significantly superior to Bende White, a local open poIJinated variety. Hybrid (3-6 weeks) infestation. As reported earlier, Sesamia infestation levels at Umuahia at maturity were low 8338-1 and the improved open pollinated variety, and essentially insignificant. TZSR-Y-l, were first used as controls this year. They The fertilizer trial showed that vigorously grow­ were chosen for uniformity and streak resistance. ing, fertilized maize had greater survival than un­ Streak resistant controls will be necessary when fertilized maize under heavy Sesamia infestation. large-sca le, artificial Sesamia infestation begins at Fertilizer did not affect the severity of attack but IlTA, since we do not want streak to confound the reduced the apparent incidence of st em borer. effect of borer infestation and we shall be unable to Reduced incidence of Sesamia on vigorously growing control streak with insecticides. maize has been noted frequen t ly in previous trials, The progeny trial consisted of 234 half·sib famili es w here maize growing in more fertile sections of the from Cycle 2 of the T:ilBR popUlation produced in field showed less damage. We assume this is toler­ 1983C season and 11 control plots each of TZSR-Y-I ance, however, it is also possible that vigorously and Bende White (Table 2.21). The entries were grown growing maize mechanically disrupts the eggs or in 16 X 16 simple lattice trials at Umuahia, Onne, and larvae. Ikenne. Only the Umuahia trial showed evidence of differences among the entries. The 43 selected fami­ Breeding trials. The breeding trials consisted of a lies were chosen solely on survival percentage and 48 Maize Improvement Program Number of eggs laid/week Table %.19. Damage rating scale used to evaluate Sesamia incidence percentages and severity ratings at nine weeks after planting 50 C fnonglJlo C mbdo Rating Description ( IITAI (JOS I 1 .. Escape. 2 ...... . Minor leaf feeding damage (pin holes). 40 3 . . ....... . Moderate leaf feeding damage. 4 ........ .. Severe leaf feeding damage. 5 .... .. . .. . Tassel killed . 6 .. Stalk killed above the ear . 35 7 .. Stalk broken below the ear. B. . Severely stunted but tassel or e-ar produced. 9 .. Dead (or missing) plant. 30 Table 2..20. Effects of insecticide aud fertilizer on 25 maize grown under heavy natural infestation of Sesamia stem borers in Umuahia, 1984 Survival, Incidence, Severity 20 " '. ',. Treatment 01'0 ~/~ rating Insecticide trial I 15 No carbofman , ... . 54.6 85.6 7.6 Seed treatment . ..... . .. 78. 3 62.9 7.0 Four applications . .. . . .. 86.5 46.4 6.2 LSD, 5'-0 . ... , ..... 7.4 to.7 0.57 10 Prob. ofF ... .. .... .001 .001 .001 .~. ::. Fertilizer tria1 2 ii· No fertilizer . ...... 52.4 92.4 7.5 5 Fertilizer . .. . ..... 59.3 81.7 7.6 LSD,5% . . .. ..... . 5.B 4.2 Prob. ofF . .. . .... . .05 .001 ns o I Each carbofuran application = 1 kg a.i.;ha. ~OO kg/ha 15·15-15 in two equal applications . at planting and six Months weeks later. Figure 2.5. Number of eggs laid weekly by 20 females of Cicadulina spp. reared on millet at 2SCC constant temperature. Table 2.21. Reaction of maize germplasm to natural infestation of Sesamia stemborers in severity rating at Umuahia. Selected families were Umuahis,l984 recombined in 1984C season to form TZBR Cycle 3. Survival. Incidence. Severity The recombinants will be self·pollinated in 1985A Treatment % 0,0, rating season, to form S , families for testing in 1985. We expect that the increased variation between and Variety trial TZBR Co. 64.8 83.0 7.2 uniformity within S, families will allow more effec­ TZBRC2 · .. · .. 85.3 7B.2 6.9 tive identification of Sesamia resistant families. Hybrid 8338· L . .. . .. .. . . 85.2 85.5 7.1 The 1984 trials at Umuahia were considerably TZSR-Y·l ..... ........ . 57.5 86.8 7.3 improved over the 1983 trials. Improvements included Bende White .. . . . . . . . . . 53.0 90.4 7.8 use of Mucuna cover crop during the first rainy LSD,5% ... . .. ... . . .. . . 9.13 5.74 0.62 season, liberal application of lime and fertilizer, seed Prob. of F .. . ..... .05 .01 .08 treatment with contact insecticides7 and frequent site Progeny trial I visits. Further site improvement is still possible. TZBR C2 (234) ... ..... .. 56.0 80.0 7.5 Based on soil samples collected in 1984, additional Sel. families (43) .. 69.9 74.6 6.9 lime and micronutrient. will be applied to our TZSR·Y·l (11) ..... 52.1 82.2 7.9 Umuahia field for the 1985 trials. All the trials Bende White (11) .. 55.4 80.3 7.7 conducted in 1984, except the insecticide trial, will be S.E.D ....... ..... 16.5 16.6 .54 repeated in 1985.- J.H. Mareck, Z. T. Dabrowski, A. Prob. ofF .. . .... . .. . .. . .05 ns .01 Carter, and K . Nwosu (NCRl) I Number of families 01" repJicates or each entr-y. Maize lmprouement Program 49 Pathology Striga Striga (witchweed, fireweed) is an angiosperm be­ longing to the family Scrophulariacea. There are 30 species described and 23, all of which are parasitic on other plants, occur in Africa. Striga hermonthica is the most prevalent species attacking maize in the West African savanna belt. Symptoms of Striga o infestation on maize may appear before the weed ... emerges during the latter stages of vegetative growth of the maize plant. Infected leaves show chlorotic blotches followed by scorching from the leaf tips and •o S ~"""'~ • s ~_"'''' • margins. The entire plant becomes scorched. resem­ s_~ ~ s"",. .• " $ _" • V._ ..i _~ bling signs of acute drought. even under adequate moisture. Severe infestation may cause stunting and Figure 2.6. Five Striga species were found infesting maize death before tasseling. in Nigeria besides an unidentified species in the Lake Chad basin. Survey of Striga. • In 1984. a survey of Striga on maize was conducted in Nigeria and parts of Benin experimental plots in 1983 ranged from 20% to 100%. Republic to study the distribution of the weed, its The middle part of Nigeria- Bauchi. Jos, Niger. and adaptation and agricultural importance, and host southern Kaduna- appears as a suitable hot spot to range. The survey was started in July and terminated screen maize for resistance to Striga. 30 September. At least two visits were made to Observations from mixed cropping of maize wi th farmers' fields, research trial plots, and farms belong­ root crops, other cereals, and legumes showed that ing to governments and institutions of higher educa­ the incidence of Striga was suppressed in a tion. In addition to assessing the level of infestation maize/soybean intercrop. It is planned to continue the survey in 1985 in and the effect on the crop. interviews were conducted collaboration with national programs in other count­ mainly to learn the history of Striga infestation in the area and the att itude of farmers about the problem. ries in Africa . The survey in Nigeri a and Benin Areas covered in Nigeria were within latitude 7 and Republic will be concluded.- V. Parkinson 15°:-.1 and in Benin Republic within latitude 6 and Controlled environment study. Striga is 7°30'N. A total of 12 Nigerian states were visited, of generally believed to be a problem in arid and semi­ which three were thoroughly surveyed. arid environments. It is essential t o determine the Five Striga species were found infesti ng maize in suitable soil texture for obtai ning satisfactory emer Nigeria. S. hermonthica and S. aspera were of econ­ gence of the weed in pot experiments. omic importance on maize and other cereals grown. 5-cm plastic pots were half-filled with the approp­ In the drier areas of Niger. Kaduna. Kano. and Sokoto ria te soil mix, ranging from 0 to 100% sand. Striga states, S. hermonthica was most severe whereas S . seeds (collected from Mokwa. 1983) were sprinkled on aspera was more important in the wetter Bauehl, Jos, the soil at 50 mg per pot and the pot was filled with t he Benue, and Kwara states (Fig. 2.6). Both species have soil mix. A basal fertilizer 15-15-15 was mixed wi th the pink flowers and are differentiated by the length of soil(l gjkg soil)_ The pots were watered and tbe maize the corolla tube extending above the calyx. variety TZB, known to be susceptible to Striga. was The other three species were described for the first planted. There were three planting dates: (a) maize time in Nigeria. S. forbesii . a broad leaf species with planted same day as Striga. (b) maize planted one salmon pink corolla . was found in Jos ; a yellow week later. and (c) maize planted t wo weeks later . flowered S. asiatica in Kwara and Niger states. and Two types of controls were used: the Striga without the third, a white-flowered unidentified species. was maize, and maize without Striga. All t he pots were found in the Lake Chad basin in Borno state. watered twice a week . S. hermonthica and the yellow flowered S. asiatica The experiment was conducted in growth chambers were found in t he Zou province in the Republ ic of with 12-hour photoperiods and at 300 e day and 23°e Benin. night temperatures. The plants were observed for Both S. hermonthica and S. aspera were found on Striga emergence and the date the first plant ap­ the following alternative hosts: sorghum. millet, peared was recorded. The experiment was terminated upland rice, lowland rice (flooded condition), ground­ at seven weeks when tbe plants were evaluated for nuts. sesame, and wild grasses. Thedamage caused by tota l number of emerged and underground Striga these two species was more severe on maize than on (Figs. 2.7& 2.8)_ sorghum or millet. The intensity of attack on maize Results (Table 2.22) showed no significant differ­ was quite variable~ ranging from severe infestation ence on Striga emergence from deferred planting of resulting in abandoned fields to slight attack . Yield maize. This indicated that preconditioning of the loss estimates based on visual survey and trials in seeds of the parasite was not necessary in pot experi- 50 Maize Improvement Program Figure 2.7. Emerged SI.riga plants on maize seven weeks after planting. ments. Striga emergence was delayed in heavy tex­ tured soil. There was a difference of 15 days between the emergence of the Striga in 100% sand and regu lar soi l (0% sand). There was. however, no significant difference on emergence between a 500/0 sa nd mix and , o i I textures containing more sand (Table 2.23). There were significa ntly more underground Striga pla nts in oa nd 25% sand mix, and on examination these plants were sma ller than Striga plants of co mparable age in Fi gure 2.8. Underground Striga plants attac hed to maize lighter soil texture. Based on these resu lts, we now roots (Photo five weeks after planting), use a 50:50 sa nd/soi l mix in the greenhouse to screen breeding materials for varietal responses to parasitism. a nd symptoms (leaf firing or scorching) were obser­ Limiting factors in this experiment were (1) the ved on maize about six weeks after planting. Most of plants could not be assessed for the effect of the the test materia ls were eva luated on a 1- 5 sca le, 1 parasite because of etiolation in the environment being highly resis tant, and 5 very susceptible. Striga plants t hat emer ged were a lso counted . All fixed und er whic h the work was conducted and (2) a corre lation between emerged plants and viable inbred lines, international and advanced hybrid Striga seeds could not be re lia bly estab lished .- V. tr ia ls, and experimenta l variety trials (EVTs) were Parkinson and J.M. Fajem isin Table 2.22. The effect of planting date and soil texture Striga resistance breeding. UTA Striga obser­ on Slriga e mergence in pot experiments at vations were initiated in 1982 at Gusau Station, in the UTA,1984 Sudan savanna zone. Even sowing St.riga seed in the Planting date of maize after maize field at t he Gusau station in 1983 and 1984 did Treatment, inoculating soil with Striga (weeks) not estab li sh Striga , so breeders could not screen germplasm against it. But Mokwa Station (National 0/0 sand o 1 2 Grain Production Company) in Guinea savan na had TO. emerged Striga plants high Striga infestation both years. In the Striga " hot o ......... , .. 39.5 40.0 -l0.5 spot" at Mokwa, 30 fixed lin es. 910 inbreeding lines 25 31.0 30.5 30.5 (S., stage), 309 hybrids, 693 top crosses, 58 open­ 50 ........... . 29.0 30.0 30.5 pollinated varieties, and 285 fami lies were tested. 75 27.0 27. 5 29.0 Plant establishment was very good (approximate ly 100 .......... ,. 25.5 27.0 28.5 95%) compared with an average of 75% in 1983. LSD, 5% ..... ". 3.1 3.6 3.9 The Striga hot-spot areas were severely infested C.V .. % .. 3.6 4.1 4.4 Maize Improvement Program 51 Table 2.23.Eft'ect of soil texture on Striga emergence Table 2.25. Grain yield and Striga hermonthica ratings in pot experiments at IITA, 1984 (1-{j) of maize hybrids and their parental Days to lines at Mokwa, Nigeria, 1984 Treatment, first Number of Striga plants Grain % sand emergence emerged underground yield, Stris,a rating I 0. ........ . .... . 40.0 8 60.5 Hybrid t/ha F , P, P, 25. . . . . . . . • . . . . . 31.5 31 22.5 50...... ....... . 26.5 46 ll.5 White grain 75 . . . . . . . . . . . . • . 27.0 43 12.5 8322·13 .. . . . . . . . .. ..•.. .•..... 7.2 1.5 3.4 2.0 100 . . . . . . . . . . . .• . 25.5 44 15.5 8321·18 ...................... . 6.5 1.9 3.4 1.9 TZB' . ............. . . . . .... . 5.2 3.3 LSD. 5% ........ . 1.6 10.2 11.3 8338-1 ..............•...• .. ... 4.3 4.3 2.0 4.6 C.V .. % ........•. . 2.0 10.7 16.2 LSD.5% ........ .. . ... .. .... . 1.7 0.7 C.V .. % . . . . ................. . 24.0 18.0 YeHow grain planted in four· row plots w,th four replications. All 8341·5............... .. .. .... 5.6 1.8 2.6 1.8 test cross testings. inbreeding lines. and family trials 8341·12. . .. ... . . . . . ... .... .•. . 5.6 1.8 1.8 4.0 were planted in single· row plots in two to three 8329-15.. . . .. . . . . . . .. ... . . .•. . 4.9 2.6 replications. Western Yellow 2 . .•• . . . • . . .• . . 3.4 3.8 Correlation coefficients among Striga and other LSD.5% ..................... 1.6 1.1 traits were estimated fro m the heavily infested four C.V .. % . ... . ...... . .......... 24.0 32.0 sets of hybrid trials (Table 2.24). Striga rating and IStriga fating : 1'"'" resistant, 5 = susceptible. grain yield were highly negatively correlated To compare families o75. 2'S 500 8 S~~216 0 of same stress level " 109 ]2" 18 >. 400 29 .!O ____ 203 0 ~ L .S.D. C> 300 (5%) ..... -- --------------..--=-- ..,...,--- 200 -' ....., --<:"'POPUkltIOfl Mean -..-­ 100 0 Sincje ,idges Tied ,klQes Sinole ridQes Tied 'idQes D ROUGHT STRESS Figure 2.9. Performance offamUies selected to form experimental varieties. Left, drought susceptible families; right. drought resistant families . Table 2.S1. Mean grain yield , days to silking, and plant heigh t of20 maize varieties, stress trials, Kamboinse. 1984 Grain Days Plant yie ld, to 500;, height. Entry' Origin kg/ha silking' em 6) Local Loumbila . .. ............ ... ..... . .... . Burkina 534 49.0 136 19) Pool 18 (SAFGRAD) ......... . ............. . . . SAFGRAD/II'I'A 524 53.7 112 13) Temp X Trop 42 .. . .... . . . .. ....• . . .... .. .... SAFGRAD/IITA 508 58.7 132 4 ) Local Raytiri ............................... . Burkina 505 47.2 132 1) Local Koudougou .. . . ........ . ......... . . ... . Burkina 491 48.0 118 10) Salita·104 . . . .. .. . ......... . .. . .... .. . .. . ... . SAFGRAD/II'I'A 475 51.0 120 20) EV81BB . ....... ....... .. . ....... . .. . .. .... . Tanzania 471 50.0 122 5) Local Pabre ... ... ..... .. . . .... . .. . . ........ . Burkina 459 53.2 121 14)DMR·Y ... . ..... . . . ....... . ........... . .. .. . II'I'A 422 58.6 132 17) TZE 16 Across·\V .................... .. ..... . SAFGRAD!IITA 422 53.2 U5 3) Local Diapaga ............•.. .• ......... . •.. Burkina 413 53.8 123 7) Jaune Flint de Saria ..... _ .... .. . . . ... . ..•... Burkina/IRAT 394 48.6 116 11) TZE·4 .... . ..................... . ...... . .. . . SAFGRAD/I1TA 389 53.8 122 16) Pirsabak (1)7930 . ...... . . . ................. . CIMMYT 355 57.5 122 12) Salita·2 ......... . .... . .• . . .•. .• . . .•. . . •. .• .. SAFGRAD/IITA 320 59.5 126 8) Pool 34 QPM ........................... ... . . CIMMYT 317 60.5 114 15) TZESRW ...... .. ...... . ........ ... ........ . I1TA 273 52.5 148 9) Early yellow ................. .... .. . . .. .... . Ghana 220 61.0 113 18) Composite 77 BD ........................... . Senega] 177 107 2) Local Kamboi nse .... . .....•..•... .... •. ..... Burkina 112 60.0 115 ~lean . ........... " ............... .. .. . . . ... ... ... . . ... ..... . " . ..... • . . 389 122 LSD, 5'}'. .... .... . . . .......... . .......... . . . .. ... .. ...... .. ...... ....... . . 154 12 C.V .• % ... . ... .. .... . ....... . ..... . ....... . .... . . .... .... . . ..... . .... ... . 49 12 'The entry number that precedes each entr y name is referred to in Figure 2.10. :.lData on flowering are given only for plots with lied ridges where 50% silking occurred to indicate the maturity cycle of each entry. 58 Maize Improvement Program looor Table 2.32. Planting date trial at Farako-Ba, Burkina Faso, 1984. Maize grain yield (kg/ha) at zero % moisture 900- Planting date June July July August Treatment 24 9 24 8 Mean 800 Low management Safita-2 1,680 1,320 1,060 590 1,165 Safita 102 1,260 1,115 650 235 815 700 High management Safita-2 2,075 1,815 1,230 545 1.415 •a 600 Safita 102 765 270 1.095 L-S.D. [5%) to compare varieties Means at same stress level Safita-2 1,880 1 570 1,145 570 1,290 Safita 102 1,440 1,415 710 250 955 •S 500 1,660 1,495 925 410 1,125 400 The rains ended in early October. As in 1983, part of the date-of-planting effect can be attributed to lack of 300 moisture at the end of the season and part to streak virus attack, which increased markedly in later plantings: 20, 43, 59, and 90% for 24 June, 9 and 24 200 July, and 8 August plantings. Because no plant showed "dead-heart" symptoms and lodging was not serious, we did not evaluate stem borer damage in 1984. IOO Density trial. This 1983 experiment was repeated in 1984 to estimate optimum density, responses to plant density, and to evaluate two promising maize More Less varieties (Safita-2 and Safita 102). Plant densities and Stress arrangements were as follow: Figure 2.10. Grain yields of 20 maize varieties under two drought stress levels, Kamboinse, Burkina Faso, 1984. Plants/ha Spacing Plants I hill Variety names are given in Table 2.31. 17,800 75 X 75cm2 1 40,000 75 x 50 cm2 1-2-1-2 and spacings (1 plant/hill) compared were 44,400 53,300 75 x 25 cm2 1 plants/ha (75 X 30 cm) and 59,300 plants/ha (75 X 22.5 80,000 75 X 25 cm2 1-2-1-2 cm). 106,700 75 X 12.5 cm2 1 The effect of planting date on maize grain yield (Table 2.32) was highly significant (0.01). The first (24 The experiment was planted 24 June in six-row June) planting gave highest yields (1,660 kg/ha) and plots, 5 m long, replicated five times. Soil preparation successive plantings gave lower and lower yields, to was by hand hoeing. Fertilizer applied was 200 kg/ha only 410 kg/ha for 8 August. The average decrease in of 14-23-15 at planting, and 150 kg of urea/ha in two yield was 28 kg/ha per day for plantings after 24 June. equal splits 30 and 55 DAP. Actual amounts were 97- It was nearly 50 kg/ha per day in 1983 for plantings 46-30 kg/ha. The experiment was carried out in the after 22 June. The management X date interaction same plot as in 1983, but varieties and densities were was significant in 1983 but not 1984; the yield trend rerandomized. Effective plant densities at harvest was the same with best response to improved manage- were near the intended 17,300, 36,100, 50,500, 76,700, ment at earlier planting dates. Safita-2 was signi- and 105,900 for Safita-2 and 17,500, 38,100, 53,300, ficantly better than Safita-102 at all planting dates 77,200, and 100,700 plants/ha for Safita 102. and at both management levels. The 1984 date X Highest grain yields in 1984 were with about 80,000 variety interaction was not statistically significant. plants/ha for Safita-2 and 53,000 plants/ha for Safita The 1984, results suggest that despite residual ferti- 102, which confirm 1983 results. Safita 102's mean lizer from 1983, the response to management in 1984 grain yield was significantly higher than Safita-2's would have been better if more fertilizer had been (3,120 vs 2,670 kg/ha). The rainfall distribution pat- applied for high management. tern in 1984 was better suited to medium maturing Total rainfall in Farako-Ba in 1984 was 815.1 mm, Safita 102 than in 1983 when rains ceasing too soon higher than in 1983 (775.3 mm) and distributed better. favored early maturing varieties. Maize Improvement Program 59 Ridging and earthing-up trial. The objective of this 6ODAP. Safita 102 maize variety was planted 24 June; trial is to evaluate ridging and earthing·up, either TVx 3236, cowpea variety, 24 June as monocrop and simple or tied, on maize growth, grain yield, lodging. 31 August as relay·cr op (68 DAP maize) at 118,500 and other variables under the soil and climatic plants/ha. Rotation started in 1983 will be continued conditions of the Farako·Ba Station, i.e., ferrallitic and varied through 1986. soils in the northern Guinea savanna ecology. Zero Furadan was compared with Furadan applied Planting was 25 June in the same plot as in 1983. both at planting (maize and cowpea) and 60 DAP The plot was hand·hoed and old ridges were leveled. (maize only). Cowpeas were protected weekly with The eight treatments were rerandomized in a Latin insecticide sprays starting 37 DAP. Only hand·hoeing square design; 14·23-15 fertilizer was broadcast be· preceded the 1984 plantings. fore hand·hoeing(200kg/ha) and urea (100 kg/hal was Maize grain yields showed significant differences equally split and incorporated at 35 and 60 DAP. between rotations (0.01), a significant (0. 01) response Safita 102 at 53,300 plants/ha (75 x 25 cm2 ) went into to Furadan (2,200 vs 3,120 kg/hal, and a significant six·row plots. (0.05) response to high management (2,210 vs 3,110 The 1984 results confirm those of 1983 concerning kg/hal. None of the interactions was statistically maize grain yield, with no significant differences (5% ) significant. between treatments (Table 2.33). There appears to be The rotation with maize following maize yielded no advantage to ridging (simple or tied) orof earthing significantly less (0.05) than maize planted after TVx up (simple or tied) over planting on the flat and 3236 cowpea or the local Logofrousso cowpea. Those keeping a flat bed in Farako·Ba's ferrallitic soils. The three rotations had mean maize yields of 2,640,3,390, 1984 data showed no significant differences in days to and 3,550 kg/ha, respectively (LSD, 5% = 424) . The 50% silking, ears with incomplete pollination, or positive rotation effect of cow peas on maize varies stem lodging despite statistical differences in 1983. with cowpea variety. We have nO indications yet on The 1984 data confirmed 1983 results that late earth· characteristics associated with the cowpea genotype ing up (70 DAP) significantly reduces root lodging, x rotation interaction. The positive cowpea rotation and early earthing·up (35 DAP) seems to increase it. effect was larger under high than low management (290 vs 580 kg/hal. That agrees with the rotation trial Maize-cowpea rotation and relay-cropping conducted in Saria (Sudan savanna) since 1979 and trial. This experiment's main objectives are to evalu­ indicates that symbiotic N fixation is not the ma in ate rotating maize and cowpeas in ferrallitic soils, factor involved (IITA/SAFGRAD in Annual Report , Furadan's effect on maize yields, maize-cowpea relay 1982). cropping on the succeeding maize yield , and cowpea Maize yielded significantly less (0.05) when it genotype x crop rotation interactions (see !ITA/ followed maize-cowpea in relay cropping than when SAFGRAD Annual Report, 1983). maize followed maize. Second·year yield reductions The experiment involves a factorial combination of should be considered when evaluating results the 2 management levels x 8 rotations x 2 Furadan first year from maize·cowpea relay cropping. The levels in a split·split·plot arrangement with 4 repli. yield loss may be related to more intensive exploi· cations. The plots had three rOws 5 m long and data tation of soil nutrients and water by relay cropping are from the center row. than by monocrop rotation. Results from other ex· Low management included 28·46·30 kg/ha and periments show that maize in these soils responds to 44,400 plants/ha; high management, 74·46·30 kgjha 75 or more kg P 20,/ha. In 1983 and 1984, we applied and 59,300 plants/ha. The 14·23·15 fertilizer was only 50-57.5 and 46 kg P,O,/ha, respectively, so the applied at planting and urea was sidedressed 35 and yield loss might. be eliminated by applying more P. Table 2.33. Ridging and earthing-up trial, Farako-Be, Burkina Faso, 1984. Maize grain yield (at 0% moisture), lodging (after the arcsine transformation) and other indicated factors Ears with Stem Root Grain Days incomplete lodg· lodg. yield, to 50% pollinatio silking .n,, mg, ing. Treatment kgjha I. ~/; % Planting on the fiat ... ... . .. ................................... . .... . . 3,380 64.4 9.9 3.4 30.4 Planting on the fiat , earthing.up 35 DAP .................... . ....... . ... 3,380 63.9 8.1 4.5 34.4 Planting on the fia t , earthing.up 70 DAP . ......... ..... . ............ . .. . 3, 140 64.9 9.2 6.5 13.3 Planting on the Aat, earthing·up + ridge· tying every other furrow 35 DAP .. . 3,840 64. 1 7.3 5.5 31.9 Planting on the fiat, earthing·up 35 DAP ridge·tying of every furrow 70 DAP . 3, 100 65.5 9.4 6.1 34.6 Planting on ridges .... . ... .. . .......... . . . . . ..... . .. . ........ , .... . .. , 3,350 63.8 7.6 5.8 27.7 Planting on ridges, earthing·up 35 DAP .................. .... ....... . .. . 2,960 66. 1 12.6 3.8 32.5 Planting on ridges, tied every other furrow. Earthing-up-35 DAP ... ... . • . . . , 3,220 62.9 9.0 4.1 31.4 Mean .......... . .......... . ............. , . . . . . . . . . . . . . . . . . .. . •.... 3,250 64.4 9.1 5.0 29.5 LSD. 5% . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520 2.1 5.1 5.8 9.1 C.V., % .. . ...... .. ..... . . , . .. . ..... . , . .. . .. . ... . ...... .. .... .... .. . . 15.8 3.2 55.3 116.5 30.6 60 Maize Improvement Program The positive monocrop cowpea rotation effect and the grain yield is considered, it appears that split appli· negative relayed. cowpea rotation effect on maize cations are better than a single application at plant­ yield need further study. ing. With no differences from timings ofN application Cowpea yields were not affected by management during the low r ainfall in 1983, more results are level (1,095 and 1,156 kg/hal Dr Furadan level (1,097 needed before general conclusions are drawn. Timing and 1,172 kgiha), but results from rotations varied of N application did not affect grain size, but the significantly (0.05). All the interactions were signi­ fewest grains per square meter came when all the ficant (0.05), except management x rotation . nitrogen was applied 12 DAP. The highest cowpea grain yield was with monocrop TVx 3236 (2,239 kg/hal. Relay·cropped with maize, it Entomology yielded significantly less (0.01), both when the pre· vious (1983) crop was monocrop maize (520 kg/hal and Maize production in the semi·arid African tropics is when it was a maize·cowpea relay (617 kg/hal. The greatly limited by low fertility and drought. Diseases 1984 cowpea yields were higher than those in 1983, and damage from millipedes, termites, stem borers, when rains stopped in mid-September. Although the and armyworms are additional yield reducers. 1984 rainfall pattern was better than in 1983, October Because they attack roots, stem, cob, and grain, was droughty, which may explain t he low cowpea termites are a major insect pest. Moreover, injury to yields under relay cropping. Cowpea pests were the roots is underground and often with no visible controlled by periodic insecticide applications. (termite) activity so it goes unnoticed and yield losses Timing N application trial. We used only half as are attributed to other causes. much nitrogen in the 1984 timing trial as in 1983. The Screening Maize for Resistance to Termites experiment consisted of a factorial combination of two nitrogen levels and five timed applications in a Work to date has included assessing yield loss as· split·plot arrangement with five replications. Two sociated with various pest species and preliminary additional treatments were one without added N and eva luation of cultivars for termite resistance. This one with 50 kg/ha. The N levels in 1984 were 25 and 75 experiment was undertaken to continue evaluating kg/ha . The timings of N applications are given in some improved and local cultivars with natural field Table 2.34. Planting was 25 June at 53,000 plants/ha infestation of termites at Kamhoinse. Twenty va· (75 x 25 em), with Safita·1()2 maize variety. Land rieties (12 improved, 8 local) of two maturity groups preparation was with the handhoe. No additional Por (15 early, 5 intermediate) were planted on three 5·m K was applied in 1984. long rows (density 53,333 plants/hal in a randomized In 1983, yield increments to N applications of 50, block design with six replications. Two treatments 100, and 150 kg/ha were significant but in 1984 the were used to determine intensity and effects of pestil· yield increase was significant (0.05) only for the first ence: no protection or two applications of increment (25 kg N/ha). Because we used the same Carbofuran; 1 kg a.i';ha at emergence, and 1.5 kg 40 main plots both years, we plotted total yield for both daYB after emergence. years against 1983 and 1984 totals. The potential Observations were taken on plant development residual effect from 1983 was overestimated and crop traits (stand establishment, lodging, tillering, silk· growth in 1984 was somewhat stunted by P deficiency. ing) , disease (streak), and insect pest incidence Both the amounts and dates of N applications (armyworms, stemborers. termites). Stand establish· resulted in statistically different yields (0.05 for ment was uniformly low in all varieties with response amounts, 0.06 for timing). The N level x timing similar to that from unfavorable growing conditions. interaction was not significant. Mean yield was But insecticide treatment strikingly improved (0.001) lowest when all the N was applied 12DAP and highest plant stand of all cultivars, presumably by suppress· from three equal applications (Table 2.34). If only ing seedling·rot and pre·emergence pestilence. Tiller production was moderate and much higher in Composite Jaune (Senegal), Safita 104, mAT 178, Table 2.34. Timing of nitrogen applications, Farako­ local Diapaga, TZSR-Y, and Safita·2 than in Jaune Ba, Burkina Faso. 1984. Maize grain yield Flint de Sari a (JFS), local Koudougou, local (kglba, at 0% moisture) Kamboinse, or TZE 4. Tillering in some crops (sor· ghum, millet) is a useful compensation mechanism to When applied, DAP Nitrogen , kg/ha offset stem injury but that is not the case with maize. 12 36 54 25 75 Between 60 and 70% silking had occurred at 47 DAP 1/1 1,315 1,645 in JFS, local Raytiri , and Composite J. followed by 28 1/2 112 1,415 2,025 to 46% in Safita 104, Safita 102, local Pabre , and 1/2 1/2 . ... . ... ... . . .. . 1,640 1,870 TZSR-Y, all substantially higher than in 1/2 1/2 . . . ...... . ..... . 1,810 2,140 1/3 113 1/3 . .......... . 1,995 2,100 Massayomba, TZESR·W, local Kamboinse, and Pirsabak (1) 7930 (20-26%). Mean. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,635 1.955 Lodging was assessed at 37, 45, 60, and 70 DAP, LSO·s. at 5%; Nitrogen : 240; timings: 407 ; timings at same before termite activity on roots became intense. nitrogen level : 575, timings at different nitrogen levels: 564. Safita 2, local Raytiri , and JFS suffered 25% lodging Maize Improvement Program 61 damage compared with 12 to 15% by local Tbe stem borers infesting stems and cobs, Sesamia Kamboinse, TZESR-W, and mAT 178 (Table 2.35). and Eldana, caused only mild damage (less tban 6% ). The incidence of streak and armyworms is reported in Termite injury was higher (27 to 35%) on roots than Table 2.36. Differences in varietal response to streak stems (10 to 24%) but in either case, cultivars did not were highly significant (0.01) and local cultivars differ significantly (Table 2.37). Termite injury on (Raytiri, Pabre , Kamboinse, etc.) appeared to be cobs and grain is given in Table 2.38. Grain dama ge in markedly suscept ible. Furadan effectively reduced genera l was higher than cob damage, especially in incidence. presumably suppressing the vectors early maturing cultivars. Injury was substan t ially (Cicadulina spp.). mAT 178, local Kamboinse, Safita higher in cobs of local Raytiri, JFS, Composite J. 102, and Massayomba suffered significantly more (Senega l) a nd local Koudougou than in TZSR·Y, damage (16 to 21% ) by the armyworm. Mythimnia TZESR·W, IRAT 178, or Pool 34 QFM. The pattern of unipuncta than did TZESR-W, JFS, Composite J. grain damage was similar to that of cobs ; it was more (Senegal) or local Fabre (8 to 11% ). Insecticide severe in L. Raytiri , JFS, local Koudougou , and treatment reduced the incidence 50% in all varieties. Composite J. than in Pool 34 QPM, TZSR· Y, TZESR· Y, or Safita 102. Cob and grain injury, particularly in Table 2.35. Lodging percentage of indicated maize the heavily infested cultivars, was significantly cultivars tested under field termite reduced by insecticide treatment. Conversely, pest infestation at Kamboinse, Burkina Faso, control had little or no effect on grain damage in less 1984 infested varieties. Early varieties seem to suffer more Cultivar Protected Unprotected termite damage (cobs, grain) than intermediate ones, presumably because they lodge and mature earlier so Safita 2 ...... . . .. .. .. ..... . . 23.4 27.4 1. Raytiri , .... .. . . the ears are more easily attacked by termites. An 22.7 26.5 JFS . ... . . .. . . . . .. . . . .. .. .. .. . . 26.6 22.5 early harvest of the early maturing varieties in sucha Pirsabak (1) 7930 . . ..• . . •. . .. . . .. 22.9 22.1 tria l, not done in 1984, should reduce the damage. Temp x Trop 27 . . . .. . .. . . . . . ... . 12.8 16.5 Grain yield was miserable, primarily because of L. Kamboinse . . . . . ..• .. . . _. . . .. . 13.9 15.2 poor rainfall (Table 2.39). TZESR·W, local Raytiri, TZESR·W . . . . ... . ..... • ... . . ... 12.1 15.2 and JFS produced about 725 kg/ba on insecticide· IRAT 178 . . .. . . . . . .. . . . •... • • . .. 13.2 11 .7 protected plots (compared with 200-400 kg/h a on Mean . . . .... ..... . . . . 18.720 untreated plots). Safita 102, TZSR-Y, lRAT 178, and LSD, 10% . . .. . . . .. .• . . 4.544 Temperate x Tropical No. 27 yielded around 100 C.V .• % . .. . .. . .. ... . . . 21 .200 kg/ha on insecticide protected plots. They are med· ium maturing culti\'ars that were just tasseling when Table 2.36. Incidence (% ) of armyworm and streak Table 2.37. Termite incidence (0/0) on roots and stems virus on selected maize cultivars tested at of selected culti vars of maize tested under Kamboinse, Burkina Faso, 1984 field termite infestation at Kamboinse, Cultivar Protected Unprotected Burkina Faso , 1984 Streak Cultivar Protected Unprotected L. Raytiri. . . . . . . . . . . . . . • . . . . . . . 8.37 16.12 Root damage 1. Pabre. . . . . . . . . .. . . . . . . . . . . . . 10.02 12.05 TZEBR·Y .. 35.8 40.2 L. Kamboinse . . . . .. .. .. . . . . . .. . 7.75 13.96 Temp x Trop 27 . . . . .. . . . . . . • •... 37.1 37.4 Composite Jaune . .. . . . . . . . . . . . . 10.17 11.53 L. Karnboinse .. . . . . . . .. .. .. . . . . . 34.9 36.6 Pool34QPM . .. . . .. . . . . .. . . . . . . 6.45 7.90 L. Koudougou ... .. . .. ... . . . ... . . 34.8 34.6 Temp x Trop 27 . .. . . . .. . . . .• ... 2.83 9.00 L. Raytiri . . . . . .. .. . .. . .. .. . ... . . 31.1 28.0 TZESR·W . . . . .. . .. . •. . •. .. . ... 5.88 5.74 TZE4 . . .. . ... . . . . . •.. . . . . . •. ... 24.7 34 .1 TZSR·Y ..... ... . .. . .. . .. . . . . . . 3.12 3.72 Safit. ·2 . . . . .... .. . • . .. . . . . . . .. . 26.6 28.3 Mean , ... 8.67 !RAT 178 ..... ......... . .. .... . . 26.4 27.9 LSD . 5% .. 4. 52 Mean . . . . . . . ........ . . ... . .. . 32.27 C.V ., '}';) ... 45.50 LSD, 5% . . . . . . .. . . . . • . . . . . . . . . . . N.S. Armyworm Stem da mage IRAT 178 .. . ..... .. . . . . ... •. .. . 14.01 26.55 Safita 102 ... . .. . .. . . . .. . . .... .. . 23.3 24.3 L. Kamboinse .. 10.88 22.13 JFS . . . .... . . . . . .. .... . . .. .... . . 19.0 24 .9 Safita 102 . . . .. . . . . . 12.46 20.35 L. Pabre .... . ..... . .. . .... . .... . 20.7 19.9 Massayomba .... . 10.55 19.82 TZSR·Y .... . . . . . . • . . . . .... . 19.1 21 .2 L. Pabre . .. 5.86 15.24 Safita 104 . . . . . . . . . . . ... . . 15.7 14 .0 Composite Jaune . 8.86 12.01 Pool 34 QPM . . .. . . . . . . . . .. . 17.6 11 .6 JFS . . .. . ... .... . 5.88 10.35 TZESR·W . . . . . . . . .. . .. . . 10 .5 12.9 TZESR·W .. .. . . . 4.11 11.51 TZESR·Y ... . • .. . . .. . . . . . . 11 .3 9.5 Mean ... . .. . . . .. .. . ... . .. . ... .. . . 13.100 Mean .. . .. . . . . _ . . . ... .• . . . . . . .. . . . 17.02 LSD, 5% . . ... . . .. .• . . • .... ... •. . .. 3.820 LSD, 5erg . . ' . . .. . . . .. ...• . . .. . . . •. .. N.S. C.V .. % . . . .. .. . .. .. . . . . . . . .. ... . . . 25.500 C.V., 'Yo. . ..... ••. . •. . .• •. . . ... ••. . . 91.90 62 l"faize Improvement Program Table 2.38. Maize cob and grain damage (%) by the rains stopped. Differences in yields among cul­ termites on selected maize cultivars at tivars were highly significant and strongly favorable Kamboinse, Burkina Faso, 1984 (0.001) to pest suppression. The results show that Cultivar Protected Unprotected cultivars differ in ability to cope with stress and Cob damage pestilence of the semi-arid environment. We shall L. Raytiri. 20.7 41.0 continue to harness the inherent qualities and, with JFS T14 K81 .... 21.2 31.2 sensible pest suppression, increase and sustain econ­ Composite Jaune . . . . . . 13.8 26.3 omic maize production.- J.B. Suh, M.S. Rodriguez, L. Koudougou. 7.0 23.0 and A.a, Diallo TZSR·Y ..... . 0.0 6.5 TZESR-W ..... . 0.0 5.1 !RAT 178..... .. .. .. . . .... . .. . 0.0 4.4 NCRE-Cameroon Maize Pool 43QPM. . . .. . . . ...... .. .. 0.0 2.7 Improvement Mean ... 9.83 LSD, 5% .. to.36 The maize section of the NCRE proj ect consists of a C.V.,% ... 91.90 breeding, an agronomy. and a testing and liaison Grain Damage unit. Activities ofthe last two sections are reported in L. Raytiri .......... . 42.1 43,5 the Farming Systems Program. Breeding activities JFS ............... . 28.2 37.8 are divided into mid· altitude and low·altitude pro· L. Koudougou ..... . 21.0 36.2 grams, In 1984, IITA added another breeder to streng· Composite Jaune .... . .... , ... . 14.2 30.7 then the mid·altitude breeding program, Pool 84 QPM ....... . ....... ... . . 4.4 0.0 TZESR·Y... .. ........ .. ..... . 0.0 0.0 TZSR·Y . ............... . . . .... . 0.0 0.0 Mid-altitude Maize Improvement Safita 102 ...... ... . . ... . . . . . . , .. 0.0 0.0 Variety Trials Mean ......... •. . • ....... • ' .... . . 12.70 LSD, 5%. 13.33 National variety trials. One early and one late set of C.V. ,% ........ .. , .............. . . . 91.50 this advanced variety trial were planted at 10 10· cations; seven early and nine late· maturing trials were harvested. The early set was composed ofBacoa and four CIMMYT subtropical experimental va· ri eties from populations 45, 34, 42, and 44. The experimen tal varieties were quite susceptible to H. turcicum. Table 2.39. Grain yield (kg/ha) 0(20 maize varieties Composition of the late set last year and this year under natural termite infestation at was similar (Table 2.40). Results presented are from Kamboinse, Burkina Faso, 1984 the seven sites most representative of tbe 1.000- to Cycle Pro· Unpro· 1,600·m. altitude area, As in 1983, t he best East Variety days tected ,ected Mean African hybrids substantially out yielded the best TZESR-W 95 744 402 573 a open·pollinated varieties. Hybrid SR 52, grown at five L. Raytiri 82 762 343 553 a of the seven locations, equaled H625 in yield and L.Koudougou 82 665 284 474 ab lodged less. As previously observed, differences JFS 86 725 193 459 abc among the five highest yielding, open·pollinated Composite Jaune 78 475 271 373 abed varieties were not significant across locations, except Safita 104 86 519 215 367 abed in stalk lodging. plant height (lower in Shaba), and H. L. Kamboinse 92 403 311 357 abcd turcicum susceptibility (higher in Shaba). We are L. Diapaga 90 540 118 329 bcde Temp )( Trap 42 90 343 216 280 bcder recommending that these varieties be consolidated. Safita·2 92 384 167 266 bcder EVT-MSR. Experimental varieties from the Pool 34 QPM 90 363 149 256 bcder TZMSR· W population were tested at four locations TZE·4 90 400 112 256 bcder (Foumbot, Dschang, Babungo. Wassande Exp.) to L. Pabre 90 260 232 246 cdef identify a suitable streak resistant variety. Babungo I TZESR·Y 95 296 174 235 cdef Pirsabak TZMSR had better plant type and H. turcicum re­ (1) 7930 90 256 162 209def Massayomba 110 270 107 189der sistance than the others. Its mean yield across the TZSR·Y 110 157 97 127 ef four locations was 8,242 kgJha. The trial will be Safita 102 110 112 84 98f repeated to confirm results and permit release of a !RAT 178 110 96 97 96f streak resistant variety for streak-prone areas. Temp x Trap 27 110 66 61 84r Mean .... 290.0 Introduction Screening LSD,5% ..... . . 190.6 A collection of hybrids and populations from East C.V. ,% ........ 57.3 Africa and CIMMYTwas screened at seven locations Duncans multiple range test. (0.05). from 1,000- to 2,OOO·m altitude across a wide range of Maize Improvement Program 63 Table 2.40. A summary across seven mid-altitude locations of indicated characters in National Variety Trial of late-maturing maize varieties, 1984 Days Plant Root Stalk Plant Ear H. Grain to height, lodging', lodging l , aspect, aspect, turcicum l , yield, Variety silk em 1- 5 1- 5 1- 5 1- 5 1- 5 kglha H625 .......... ... . .. . . ... 0·· · 0· ···· 87 285 2.2 1.9 2.3 1.8 1.2 9,883 Shaba ....... ···.·· · 0· · ·. 0 ........ 79 236 1.8 1.5 1.9 2.2 1.5 8,430 Composite 290 ... •••• • •• • •••• • • • 0 •• 83 276 2.0 2.5 2.7 2.1 1.4 8,300 Coca ........... ··. · · ·0 · ··.··. ··· · . 81 264 2.0 2.6 2.5 2.0 1.3 8,278 MLC .......... , ........ 82 277 1.9 2.5 2.6 2.2 1.4 8,096 SAW . . . . . . . . . . . . . . . . . . 82 268 1.9 2.2 2.6 2.3 1.6 8,071 Zambia·83 TZMSR-W (2) ... 81 241 1.9 1.8 2.1 2.1 1.9 7,637 Kasai ...... ............ 81 208 1.3 1.3 2.1 2.6 1.7 6,833 EV8344-SR. ........ 79 219 1.8 1.9 2.9 2.7 2.8 6,241 LSD, 5% .. . ........ 1.1 11.9 0.2 0.3 0.2 0.2 0.3 597 IFromsix ofthe seven locations. soil origins and fertilities. Results from the more Inbreds and hybrids_ More than 700 S, to S, adapted entries across six locations are shown in streak-resistant inbreds from IITA were screened at Table 2.41. The best yields were from Kenya 600 series three locations. Selection and selfing produced 489 and Zimbabwe late-maturing hybrids. The latter ears from 170 lines. In addition. 65 promising inbred lodged less and the former had less dented grain. The parents were identified in replicat ed trials where 424 Kenya 600 series, clearly superior above 1,500 m, will test crosses and 42 hybrids were evaluated at one form the basis of a new population for that zone. The location. highly related varieties, Coca, Compo 290, and Pool 9, all yielded well but were tall and prone to lodging. Population Improvement Acid-tolerant introduction screening. Twenty­ IPTT 34 was tested at Babungo with selections based five introductions from Brazil, CIMMYT, Thailand, on H. turcicum resistance. Families have been ob­ Zaire, and IITA were screened in a lime and no-lime tained for recombination and incorporation of streak (split block) experiment at three locations above resistance in efforts to form an early white popu­ 1,000 m and one below. Within-block variabiHty was lation. Ear-to-row selection was performed in Pool 9 high except at Wassande, so selections were made for plant type, and in Shaba and Kasai for H. turcicum visually. Entries selected to form a mid-altitude, acid­ resistance.- L. Everett tolerant population were ESALQYF3, ESAL Q5VF 1, CMS 36 (Brazil), Coca (Cameroon), HE 1066, HE 1049 Low-altitude Maize Improvement (Limagrain), Shaba (Zaire-CIMMYT), and MSR-F (UTA). Suwan 1 (Thailand) and population Variety Trials 28 (CIMMYT) should be further t ested at lower altitude. The lowland maize improvement unit conducted 66 Table 2.41. Data on indicated maize plant characteristics from the 1984 Introduction Screening of Mid·Altitude Varieties across six locations (Foumbot, Dschang, Santa (IRA), Babungot Wassande Ext., Was sande Farm), Cameroon 1984 Days Plant Root Stalk Plant Ear H. Grain to height, lodgingl, lodging I , aspect, aspect, turcicum yield, Line silk em 1- 5 1- 5 1- 5 1-·5 1- 5 kg/ha H614 . .... ... ... . ... . , .. , .. .. .. . .... 84 270 2.5 1.6 2.3 1.8 1.1 11.861 H8102 ....... . ...... ... . ... . .... . ... 89 285 2.1 1.5 2.3 2.3 1.2 11,579 H612 . .. . .. _. 86 274 2.8 1.7 2.6 1.5 1.1 11,380 H625 ... 86 266 2.7 1.9 2.3 1.7 1.0 11.337 Z5206 .. 81 252 1.8 1.3 1.7 1.1 1. 3 11.240 SR 52. 83 262 1.8 1.3 1.9 1.8 1.4 11,173 ROO .... 83 251 2.1 1.6 1.8 1.1 1.5 10,944 R613 .... 86 283 2.7 2.0 2.6 1.7 1.0 10,239 H622 .... ..... ... 83 250 2.4 2.6 2.3 1.7 1.1 10,212 H81M!. . . ... -...... . -. . .. ... 81 261 1.9 1.7 2.3 2.0 1.5 9,871 Coca ... ... , ... ... .... ..... ... .. . ... 81 258 2.0 1.8 2.4 2.2 1.4 9,688 Pool 9 .. ... , ... . ... .... .. 84 266 2.5 1.7 2.6 2.4 1.2 9,580 Shaba .. .................. . ... . .. ... 79 230 2.0 1.6 2.2 2.6 2.1 9,013 Composite 290 ............... .. .. . .. . 79 256 2.2 2.0 2.5 2.2 1.6 8.999 1 From fewer than six locations. 64 Maize Improvement Program trials in 14 locations during 1984. All savanna 10' C.V. resulted partly from root rot in Njombe and cations experienced drought during the grain·filling thieves in Nkolbisson. Kasai 1 and Ekona White, the period . Second season trials in the lowland forest controls. ranked 8 and 9 among the 10 entries. The top areas were a lso hit by drought, but maize streak virus three orfourwill be advanced to EEVT in 1985. was negligible. Three sets of EVT-LSRY also were tested at National Variety Trial (NVT). The trial con· Nkolbisson, Bertoua. and Ntui. Average grain yields sisted of 13 entries, including six from the 1983 NVT differed little among locations (Table 2.43). Suwan I and six from the 1983 experimental variety trial outperformed all entries , but differences across en· (EVT). The coefficient of variation ranged from vironments were not significant. Ranking of ge­ 11 to 49% . The higher C.V.'s were generally associated notypes differed among locations, indicating a signi· with highly varied number of plants harvested. When ficant genotype x environment interaction. One or C.V. of plants harvested exceeded 25%, the C.V. of two of the entries will be elevated to the EEVT in yield exceeded 29% except in Karewa where soil 1985. erosion caused the high C.V. This year tests with C. V. In three sets of EVT 12A from CIMMYT tested at exceeding 29% were not included in Table 2.42. Ntui , Njombe, and Nkolbisson during the first crop· Perhaps a covariance analysis with number of plants ping season, Across 8243 ranked first in yield and at harvest would reduce the high C.V.'s. out yielded Gusau-81 TZB, the best control, but not Gusau·81 TZB and Suakoko·81 TZPB were the top significantly. Rankings of the 18 varieties differed yielders in six of the 11 tests. Suakoko·81 TZPB widely by location, again indicat ing a high genotype tended to outperform Gusau-81 TZB hut not signi· x environment interaction. Across 8243 will be ficantly. The best streak resistant variety was Ikenne advanced to EEVT in 1985. (1) 8243. Grain yields in the sa vanna (Table 2.42) were Three sets of EVT·ESR were tested in the 1984 30% helow yields in the forest areas , opposite from cropping season at Kousseri, Bertoua, and Ntui. last year when the savanna was 28% above the forest Table 2.44 shows results from the second season in the zone. Lack of, and erratic distribution of, rain at all forest zone. Mayo Galke 82.TZESR·W, a selection savanna locations likely caused the reversed results. from Cameroon, had the highest yield (3,359 kg/hal In 1983-84, l\VT Gusau·81 TZB and Suakoko·81 and was significantly higher than the other 12 en· TZPB were decidedly better than Ekona White. Even tries. Seven entries, including one contro l (TZUT), though they lack streak resistance, they can replace yielded significantly more than Mexican 17 Early, the Ekona White because it also is susceptible. Where best available material for the semi·arid areas. The farmers plant late or during the second season, they top three or four entries will be ad vanced to EEVT in can use early maturing, streak·resistant varieties. It 1985. would, however, be desirable to have full·season , IRAT 178 and TZB Gusau outy ielded 10 other resistant varieties also. Ikenne (1) 8243 will be in· entries in the RUVT·2 trial from SAFGRAD under creased and evaluated further for that purpose. two Cameroonian environments in 1984 and will be advanced to EEVT in 1985. Experimental Variety Trials (EVTs). Three sets Four sets of the White Hybrid Trial (UTA) were of EVT·LSRW with 10 entries each were tested at tested twice at Sanguere and once at Mayo Galke the Bertoua. Njombe, and Nkolbisson during the first first season and once at Ntui t he second season. growing season. The highest average grain yield was Average grain yields in the savanna were at least 2 at Bertoua (8,520 kgfha), and the lowest at Njombe tlha higher than at Ntui t he second season, where (Table 2.43). The low grain yield and relatively high drought stressed maize during its filling period . Four Table 2.42. Yields and other indicated characteristics from the 1984 NVT across locations and across agro­ climatic zones Days Plant Ear Yield . to height. height. Yield. kg/ha Variety kg/ha silk em em Savanna Forest Suakok()·81 TZPB ................................ . 5,327 62 224 114 6,075 4,430 Gusau-81 TZB .. . . ... . ... . ............. ...... •... .. 5.307 62 208 101 6,059 4,405 Ikenne (I) 8243 .. . .. . .... . ..... .. . ...........•..... 5,275 62 218 109 6,002 4.403 PR 7822-SR Be, .... , ... . ........ . .. .... _. .. ...... . 5,247 60 209 101 6.035 4,302 Across 8043 . ....... . ............................ . . 5.120 62 211 109 5,891 4,195 Suwan 1 ....................... . ... ........ . ..... . 5,023 59 203 101 5,816 4,072 Kasai 1 . ............. . ... ...•. .. ... ... .. . .. ....... 4,772 60 198 87 5,358 4,068 PR 7843-SR ................. _ . .... .. . .. .. .... .... . 4,718 61 216 105 5,198 4,143 lkenne-81 TZSR-Y·l ................. . . . . .. .. ...... . 4.571 61 217 104 5,186 3.834 Bert()ua·81 TZSR·W·l .. . "." ... .. . .. " ... . .. ..... . 4,371 61 214 105 4,575 4,126 Aef()ss-81 TZSRW·l. . " ..... . . .... " .... " . .... ... . 4,291 61 208 104 4,563 3,965 Ekona White control. . . . . ........ . .............. . 4,093 62 247 128 4,580 3,509 Local varieties . .. . . .. . .. . ........................ . 3,921 60 227 116 3,874 3,977 Maize ImprDvement Program 65 Table 2.43. Location mean yields (kgfba) from the entri es, 8321-18. 8322-3, 8346-3, and 8322·13, out­ 1984 Experimental Variety trial of yielded the open-pollinated Poza Rica 7822·SR more indicated White (top) and Yellow (bottom) than 30% (Table 2.45). Pwani I and Pwani III, from maize varieties at indicated locations for Kenya, yie lded no more than Poza Rica 7822-SR, each indicating that they are not adapted 10 Cameroon 's Nkol· environment. Variety Bertoua ~jombe bisson Mean Four sets of yellow hybrids also were tested in the White seeded same three locations (Table 2.45)_ Five of the 11 EV. 8322-SR BC, 9.251 5.532 6,507 7,097 entries yielded a t least 30% more than the open­ Ejura-81 TZSR-W-l __. _. _ 9,144 5,917 5,897 6,987 pollinated Ikenne·81 TZSR·Y-l. They are 8341-5, 8329- EV. 8343-SR BC, . .. . .... 9,259 5,713 5,406 6,792 15,8341-12, 8340-1 2, and 8341·6. Some of them will be EV_8329-SRBC, . __ _ .. __ 8,194 5,944 6,480 6.480 advanced to EEVT in 1985_ Sekou-81 TZSR-W-I _. . . __ 8,158 5,963 5,125 6,415 Gusau-81 TZB . __. __ . _. _. 7,933 5,587 5.367 6.269 Population Improvement Across-81 TZSR-W-I . __ __ 7,942 5,521 5.578 6.247 Ekona White and Ekona Y eIlow. It was the first K.s.i 1 (control) . _. . ____ 8,546 4,634 5.200 6.127 season for Ekona White and Ekona Yellow in an Ekona White (control) . __ 8,662 4,400 4,767 5,943 intr a-population selection program. Previously r e­ Bertoua-81 TZSR-W-l . . . _ 7,942 5.221 5,577 6.247 leased by rnA, both are widely grown throughout the ~le8n . .. . .. . . . . .. , . . . .. 8,520 5,331 5,445 Cameroon lowland forest. The select ion program is to LSD. 5% . , ... . . . . . .. . .. 1,491 C.V ,% __ . __ . __ .. ... __ _ restore their yielding capacity and improve some 12 20 21 important agronomic characteristics, including stan· Nkol- dabili tyor lodging resist ance_H alf-sib families (95 for bisson Bertoua Ntui Mean Ekona White and 115 for Ekona Yellow) were tested Yellow seeded the second cropping season in three locations in 10 X Suwan L . . ,_. _ . .. 6,987 5.948 7,446 6,793 10 a nd 11 x 11 simple lattice designs , with two Across 7728 . .... . . 6,465 6,393 7,040 6.633 replications at each location. After results of the Ferke-81 TZSR-Y-l. 6,356 6.837 6,314 6,502 trials ar e computed, 10 to 15 families will be recom­ l\iaouli-81 TZSR-Y-l _. ___ 7,058 5.965 6,421 6,481 bined to make up cycle I populations _ EV_8 328-SR BC, .. . __. _. 6,231 6,437 6.281 6,313 Gene pool. During the first growing season in 1983, Across-81 TZSR-Y·l. __. _. 6.798 6.030 6,113 6, 314 a gene pool was created by crossing PR 7843-SR, Farako Ba-81 TZSR-Y-l _. 6.524 5.789 6.518 6,277 Ikenne·81 TZSR· W-1, the late, white, streak-resistant [kenne-81 TZSR-Y-l. . ___ . 6.699 5.682 5,457 5.946 backup pool, and Gusau·81 TZB as females with the Ekona Yellow (control) .. 4,880 6,498 6,392 5,946 common male PR 7822-SR in ha lf-sib isolation_ Mean . .. . .... . . ... . . . . . 6,440 6.175 6,442 6.352 During the second season, seed from 60 ears from each LSD. 5% . . ... . . . ... . . . , 1.569 1,401 1.034 1.335 fema le group were planted at Karewa. Equal num­ C.V., % ,_ .. . . . . .. . .. . . . 17 16 1] bers of seeds from each female group except Gusau-81 TZB were bulked and used a s male_ For each fema le group, 150 ear s were selected and the bulked seeds were randomlv mated at Nkolbisson the first season of 1984_ The ·procedure was repeated the second Table 2.44. Grain yield (kg/ha) of entries in EVT·ESR maize tests conducted during the second growing season of 1984 ; however . the r a ndom mating 1984 cropping season at Bertoua and Ntui, was done by hand-pollination because there was Cameroon neither space nor time for isolation procedures. About 490 pollinated ears were harvested a nd bulked Entries Bertoua Ntui Mean to consti tute popUlation NCRE-8401, which will be Mavo G. lke 82 TZESR-W 4.030 2.688 3,359 tested in 1985. Ike;'n. 82 TZESR-W. __ .. . . _. 3.374 2,70] 3,038 Gusau 81 Pool 16 _. ___. .. __ 3.564 2,474 3,019 Variety breeding. During the first cropping sea­ TZUT control . . _. _. __. . _. . ___. . 3,762 2,201 2,982 son of 1984, iO fa mi lies per female popUlat ion of the Gusau 82 TZESR-W __ .. _. ____ . __ 3.484 2.290 2.887 gene pool were test ed in a 17 x 17 si mple lattice at EV 8331 SR (BC2) . . __ . .. . _. __ _. . 3.291 2.474 2,883 four locations (NTUI, Bertoua, Sanguere, and EV 8335 SR (BC2)_ .. _ ... _ 3.240 2,490 2,865 Karewa). The experimenta l unit consisted of a single EV 8330 SR (BC2)_ . __ .. _. 3,045 2,297 2.869 r ow, 5 m long with 75 cm between rows. Data from the Mex 17 Control . ... . . . ..... . . . . . 2.598 2.288 2.448 two replications per location are being analyzed_ Five Across 82 TZESR-Y __ . . _. _. __ . . _ 3.588 !kenne 82 TZESR-Y __ .. _. _. __ .. _ experimental varieties wi ll be constituted from these 3.314 Bertoua 82 TZESR-Y _ . . __ _. _. . . _ 3,244 tests. [n the second 1984 seas.on , 250 families again Kamboinse 82 TZESR-Y ___ . ... . . 3.065 were selected from the gene pool that had been randomly mated one generation. The fami lies were Mean . . . . . . . .. . .. . ... ... _. .. . . 3.365 2,424 2,900 tested at Ntui and Berloua in a 16 X 16 simple lattice LSD,5% . . .. . .. .. . . . . . . .. . . . . . . 332 285 309 C.V .. % .. __ __. . .. . _. ... _. . __ . _. 18 18 to make up another set of experimental varieties_A ll the above developed experimental varieties will be 66 Maize Improvement Program Table 2.45. Partial results (Grain yield, kg/hal from white hybrid maize trials (top) and from trials with yellow hybrid maize (bottom) at indicated Cameroon ]ocatioDs, 1984 Sanguere, Sanguere. Mayo- Hybrid' 1st season 2nd season Galke Ntui Mean 0 ' 10 White grained 8321·18 .. ......... .... ... .. -. .... -... . .. , . . -. -.. 7,381(1) 5,240 (3) 7,487 (I) 2,644 (2) 5,688 147 8322-13 .. . . .. .. . .. . . . ... . .. -..... ... . . .... ... ... 6,317 (4) 5,728(1) 6.873 (3) 2,475 (8) 5,348 139 8322·3 .. . . -.- .. - .. .. .. ..... .. .. . .. . . . . -. . .. . - . .. 6.686 (2) 4,957 (4) 7.146 (2) 2,538 (6) 5,332 138 8346-3 .. . . ..... . ..... . .. . ....... . . . . . . . . . .. . .... 6,630 (3) 5,275 (2) 6,689 (5) 2,509 (7) 5,275 137 Mean ... .. .... .. ... .. ... ... ... .. ... ... ....... 5.434 4,043 5,940 4,451 2,388 LSD,5%. .... .. .... .. . - -.. .. ... ...... . ... 1,009 1,274 1,131 370 861 C.V. ,% . .. . - . . . . .. -... ... ....... 13 22 13 28 Yellow graIned 8341·5 ........... .. .. ..... .. - , . . . . . . . . . . . 6,527 (1) 4.942 (6) 8.207 (I) 3,723 (1) 5,850 144 8341-6 .......... . ....... . .. .... ...... . .......... 4,885 (5) 4.949 (5) 8,040 (2) 3.265 (2) 5,285 130 834()'12 ....... . . ....... . . .. . .. . ..... . . . . ... . 5,647 (3) 6,079 (I) 7,065 (3) 2,432 (7) 5.306 130 8341·12 ...... . . ... • ... . .... . . . . . . . . . . . . . . . . . 5,892 (2) 5,332 (4) 6,966 (4) 3,047 (5) 5,309 130 8329-15 .......•. . . . ... . . . . . . . . . . . 5,576 (4) 5.793 (2) 6,754 (5) 3,200 (4) 5,331 131 Mean '" . .. .. .... . ............ .. . 4,684 4.910 6,580 2,771 4,786 LSD,5O;" . . ....... ..... . . .. . . .. ........ . . ....... 1,884 873 971 200 932 C.V., % ............... . ... . . ..... . . .... ....... . 25 12 10 13 lHybrids that outy ielded the open· pollinated cont.rol more than SO"}i •. For the white hybrids. Poz.a Rica 7822 SR was the open-pollinated control ; fOT the ye llow hybrids, it was lkenne 8l TZSR· V.1. Ranking of each hybrid at each location is gh:en in parentheses. yield t ested and evaluated for streak virus resistance in 1985. Acid-tolerant line evaluation. Twenty varieties were tested at Mba Plain in 1984. Four of them were selected for their performances and will be used as parents in a cross with Gusau·81 TZB·8R in an effort to develop an experimental variety adapted to the acid Mbo Plain area. Hybrids, Inbreds, and Synthetics Diallel evaluation. During the first season of 1984, 45 variety crosses and their reciprocals were made in diallel fashion among 10 parents, to study t he breed· ing value and heterotic r elationships of lowland maize populations found adapted to Cameroon's environments. In t he second 1984 season, the 90 crosses and their lO parents were tested in four environments in a 10 X 10 lattice design. The results are being processed. Inbred line evaluation. Thirty inbred lines from lITA were tested at N'dock the first growing season of 1984 to identify adapted , productive lines to use as parents in developing synthetic varieti es. During the second season of 1984 at Ntui, Hawa iian inbred lines were tested to eva luate their adaptability to Cameroon lowland areas and resistance t o diseases in the lowlands. Resu lts of both experiments are being processed. New inbreds. During the first 1984 season, about 500 S, lines were generated. The second season about 500 S, and 1,500 8 2 plants were selfed . The objective is liDe development for hybrids and synthetic-variety development .- Jay Chung, Charles Tiw. and L. Empig Grain Legume Improvement Program The main objective of the program continues. It is to ing to grow an additional crop during the period land develop stable and high yielding varieties adapted to is normally left fallow. the different agro·ecological regions and suitable for To serve the Latin America region effectively, we the different cropping systems in the various tropical stationed a regional coordinator for cowpea and environments. soybean research with EMBRAPA in Brazil. He will To further strengthen the nationa l programs, we work on mono and mixed cropping with emphasis on added specialists to a network of regional institutes. acid soils with EMBRAPA (Empresa Brasileira de We stationed a cowpea breeder at the ICRISAT Pesquisa Agropecuaria·National Agricultural subcenter in the Sahel at Niamey, Niger, who will Research Organization of Brazil). concentrate on developing drought-resistant cowpea lines for grain, lea"es, and fodder for cattle. He also will do research on mixed cropping suitable for that Cowpeas region's millet·based cropping system. Genetic Improvement In Sudan savanna, at Ouagadougou, Burkina Faso. a small team is emphasizing development of Striga· International Trials and drought·resistant cowpea lines to fit into maize The·top performing lines selected from the multiIo­ and sorghum based cropping systems. Their work is cation variety tests conducted in Nigeria are multi­ 'through the Semi·Arid Food Grain Research and plied and formulated into Cowpea International Develop ment Project(SAFGRAD) under the auspices Trials (CIT). When materials and maturity groups of the Organization of African Unity 's Scientific, dictate, more than one trial is formulated. Technical, and Research Commission, with financial Formulating is in December-,lanuary and a circular support from the United States Agency for is sent to about 300 cooperators in more than 50 International Development supplemented by the countries explaining the nature and composition of Canadian International Development Research trials. Individual cooperators request one or more Cen tre (IDRC). sets of the various trials to meet their requirements. In northern Guinea savanna, at Zaria. northern We try to meet all requests but large demands often Nigeria, work funded by the European Economic make that impossible. The sets are despatched in Community (EEC) is on developing cowpea and March- April to fi ll requests. soybean lines for higher grain yields in that latitu­ Each Cowpea International Trial set includes a dinal belt. computerized field book, relevant information about In eastern Africa, Morogoro. Tanzania. funded by the trial, and two data sheets . One sheet is returned United Nations Development Program (UNDP), we by the cooperators to lITA when the trial is complete. are concentrating on mono and mixed cropping of The data are analyzed by IITA and a copy of the cowpeas and soybeans. analyses is immediately sent to the cooperators. After In southern Africa, Harare. Zimbabwe, our reo all data from most co-operators have been received. search is on cowpea/soybeans for the regions in both the results are compiled in a bulletin and sen t to each mono and mixed cropping that involves cowpeas and cooperator. National program scientists select the soybeans with emphasis on sorghum. best lines from Cowpea International Trials and For east and northern Africa, in addition to re­ formulate their own multi location trials for the search on cowpea/soybean cropping systems, we hope future. The most promising varieties are released to to identify leafy vegetable· type cowpeas for Kenya farmers or used as breeding stock by the national and cowpea lines with cold tolerance. We shall post a scientists. Thus far, cowpea varieties bred in the IITA scientist in Kenya with the International Centre of program have been released in about 30 countries in Insect Physiology and Ecology (ICIPE) at Mbita Asia, Africa, and Latin America. The annual cycle of Point. developing and distributing new improved cowpea In southeast Asia. IRRI, Philippines, the major lines to variqus national programs has created a emphasis remains on rice with cowpeas and soybeans global network of cowpea researchers. Their com· as alternate, relay, and mixed crops. We are attempt- bined data are used to select parents for crossing 67 68 Grain Legume Improvement Program programs that produce widely adapted, high-yielding kglha across 16 locations. TVx 1948-0l2F and TVx co\vpea varieties. 4677-01OE were nearly as good. All three carry mul­ tiple disease resistance and mature in 75 to 80 days in The 1984 international trials, Four Cowpea tropicaL environments. Of 19 new lines evaluated in International Trials (CIT) were constituted in 1984, this trial, the mean yield of17 exceeded means oflocal CIT-l had extra-early maturing lines (55 to 65 days); improved controls.- B.B. Singh and Remi Adeleke CIT-2, medium-maturing lines (75 to 85 days); CIT-3, bruchid-resistant lines mostly medium maturing, and Bruchid-resistant cowpea trial. This trial con­ CIT-4, vegetable-type cowpea lines, Requests for each sisted of nine bruchid resistant lines and a local set far exceeded our supply, so we attempted to control. The mean yield performance of different distribute a few sets in each country, Overall, 95 sets varieties at 19 locations are presented in Table 3.3. of CIT-I, 89 ofCIT-2 , 78 ofCIT-3, and 32 ofCIT-4 were Overall. IT 8ID-1007, IT 8ID-I032, IT 8ID-I064, and IT distributed to 54 countries, The most sets (216) went 81D-1020 constituted the high yielding group and IT to Africa, 43 to Asia, and 35 to Central and South 8ID-985, IT 8ID-988, and IT 8ID-994, the low yielding America including the Caribbean Islands, group. The top four carry multiple disease resistance Additionally 6; scientists and adaptive research and bruchid resistance. Also, IT 81D-1020 is resistant workers from 47 countries were supplied a few seeds to aphids. The low yields of IT 8ID·985, IT 8ID-988, each of promising varieties. and IT 81D-994 may stem from their late maturity and The results from 1984 have just begun to come in, photosensitivity. But they performed acceptably in They will be reported in the 1985 IITA Annual Report, West Africa. All three have large, rough-coated white seed, the type preferred in West Africa. Means over Results ofl983 international trials. As indicated locations ranged from 1,087 to 1,720 kg/ha. in the 1983 IITA Annual Report, three types of Cowpea International Trials were distributed in 1983. Phenotypic stability of cowpeas_ Overall mean CIT· 1 consisted of extra·early lines; CIT·2, of yields and regressions are considered phenotypically medium·maturing lines, and CIT·3, bruchid·resistant stable (Table 3.4). Varieties that showed stable per­ lines. In all, 212 sets were sent to 51 countries; 72 were formance were IT 82E-16, IT 82E·18, IT 82E-32, IT 82E- CIT·1, 70, CIT·2; and 70, CIT-3. Data were received for 13, and IT 82E-9 in CIT·1; TVx 4659-44E, TVx 1948- 30 extra-early maturing trials (CIT· I), 34 medium 01F, TVx 4677-01OE, TVx 4659-02E, TVx 3236, and IT maturity (CIT-2), and 33 bruchid-resistant trials (CIT- 82E-3 in CIT-2; and IT 8ID-lOO7, IT 8ID-1032, IT 8ID- 3). All were analyzed and summary results are being 1064, IT 8ID-I020, and IT 8ID-1l57 in CIT-3. All of published in a bulletin. Data from several locations them carry multiple disease resistance. Cowpea had high co·efficients of variability and low means, trials , always protected against insects, are not primarily from lack of insecticide sprays. Data from protected against disease. So varieties with multiple 16 locations each for CIT·1 and CIT-2 and from 19 disease resistance perform well over a wide range of locations for CIT-3 were fairly satisfactory and are environments with insecticide protection. presented here. Environmental factors other than disease have not affected the stability of these varieties, probably Extra-early maturity trial (CIT-I). The mean because they were initially selected for performance yield performance of nine extra-early maturing cow­ over a wide range of environments in Nigeria. The pea lines at 16 locations are presented in Table 3.1. data presented indicate that UTA's strategy for Means over the 16 locations varied from I t/ha for IT improving cowpea production is quite effective and 82E· 77 to 1. 7 t/ha for IT 82E-32 and IT 82E-9. The produces lines with wide adaptability.- B.B. Singh performance of each exceeded those of local controls. and Remi Adeleke All the high yielding lines have moderate to high resistance to all the major cowpea diseases, while the Hybridization and Early-generation low yielding ones are susceptible to brown blotch, Materials web blight, and anthracnose. The low yielding ]jnes The cowpea breeding program continued to emphas­ have erect growth habit and mature about one week ize multiple resistance to diseases and insect pests in earlier than the high yielding group. They did well in Cameroon, Ghana, Zambia, combination with higher yield, good plant type, desired maturity, acceptable seed quality, and wide Ethiopia, and Venezuela. The major advantage of adaptability. During the year, 461 crosses were made extra-early varieties is their ability to exploit short with parents selected from among improved lines that rainy seasons or residual moisture after rice or carry one or several desirable attributes. Most of the another major crop. Data from the internationaL crosses involved IT 82D-716 or its derivatives, which trials indicate that average yields of 1 to 1.7 tlha can carry multiple-disease resistance and resistance to be obtained in 60 to 70 days. Such yields compare well thrips and bruchids. The other parents were selected with higher yields in per-day productivity. for aphid and leafhopper resistance along with Medium maturity triaL Mean yields of different multiple-disease resistance, better seed quality, lines at various locations are presented in Table 3.2. higher yield potential, and desired maturity. Efforts TVx 4659-44E was best at most locations. ranging were made to develop hybrid populations involving from 900 kg/ha to 4,300 kg/ha with a mean of 2,220 multiple-disease and insect resistance in lines with Grain Legume Improvement Program 69 Tab1e 3.1. Performance of promising extra~arly varieties evaluated in Cowpea international trial 1 at indicated locations in 1983 Grain yield. kg/ha !ITA Mokwa Kamboinse Lafia Sekou Lome Garoua Kwadaso Variety Nigeria U. Volta Nigeria Benin Togo Cameroon Ghana IT B2E·16 ... ... .... . . . . .. .. . . ' . . - . . 1.473 1.828 2.252 1.251 2.937 791 2.456 1.664 IT B2E-18 ... .. . . ........ . . . . ... . ... 1.453 1.844 1.906 927 2.675 585 2.606 1.668 ITB2E-32 ...... .. .. ... ...... .. . .... 1.239 1.547 2.392 1.232 2.912 765 2.221 1.601 IT B2E-13 .. . ..... .. .. ....... .. .... . 1.616 1.297 2.428 1.106 2.294 779 2.428 1.523 ITB2E-9 . .. ................. .. .... 1,373 1,328 1.663 580 3.037 610 1,219 1.556 Control . . . . .. . ............. . ...... 1,183 1,687 1.477 144 2.034 751 548 1.697 Location mean. ... .. ...... . .. .... . . 1.199 1.428 1.806 735 2.11 3 600 1.908 1.522 LSD. 5% ... . ..... . . ....... . . 4B9 294 454 296 572 167 301 330 C.V .• % .... ... .... ...... . ..... . . . . 28.11 14.19 17.31 27.78 18.64 19.19 10.87 14.93 Table 3.2. Performance of medium-maturity varieties evaluated in Cowpea international trial 2 at indicated locatioos in 1983 Grain yield, kg/h. ITTA F/S IITA SIS Kamboinse Sekou Lome Dapaong Sotoubua Garova Variety Nigeria U . Volta Benin Togo Togo Togo Cameroon TVx 4654-44E . .. .... ..•. 1.157 1,031 1.221 2.735 2.367 1,752.00 2.202 3,344 TVx 1948-012E .............. 1.061 1,015 957 2.785 l,81i 1,601.60 4,212 2,494 TVx 4677-01OE ......... . .... 1.159 735 952 2.753 1,662 1,755.00 4,446 1.659 TVx 4659-02E .... ...... . ..... . . 1,264 700 917 2.265 1,659 1,271.00 3.848 2,484 TVx 3236-01G ....... ..... .. ... . 1.095 787 1.076 2.033 1,701 1,323.00 2,808 3.403 IT 82E-3 ..... ....... ... . . ... ... . 666 937 969 2.072 1.299 1,458.60 3.666 3.130 IT 82E-25 .. .. ...... 933 1.071 344 2.340 1.842 1,404 .00 3.328 2.888 TVx 466-07E ....... ............. 932 754 978 2.478 1,620 1.378.00 2,964 2.969 IT 82E-27 .......... ...... ....... 881 893 779 2.379 1.407 1,430.00 3.302 2.587 Location mean . .. .... ... . . . .. . . . 941 787 941 2.198 1,399 1,349.00 2.989 2.521 LSO,5% ..... .. , .......... 334 275 213 534 427 378.00 1.003 698 C.Y., ~~. . . . .. . . . ...... .. .... . .. . 25.93 24.60 15.97 17. 14 21.50 19.77 23.66 19.51 BtouaSS Kwadaso Njala Bambey Masumba Malkerns K 'Kaen Moroni Variety Cameroon Ghana S. Leone Senegal Zambia Swaziland Thail and Comores Mean TVx4654 ... 1,118 2.107 9,054 I ,B90 3,n7 1,359 3,079 4.371 2.220 TVx 1948-012F .. .. 988 1.749 972 2.004 2.067 1.971 2.392 3.989 2.067 TVx 4677-010E .... 1,180 2.065 1,081 1.593 2,494 1,606 3,105 3,645 1.995 TVx 4659-02E .... 1.139 1.587 1,092 1.432 2,166 1,450 2,506 3.030 1.801 TVx 3236-0lG .... 793 1.871 1.113 1,979 2,241 1,156 1,461 3,894 1.796 IT 72E·3 . .. . .... 972 1.773 1,004 1,785 1,426 2.528 2,450 3.571 1.794 IT82E·25 ....... . .. .. . 603 1.692 847 1,449 3,062 1,450 5.120 2,855 1.746 TVx 4661-07E ....... ... 1,232 1.563 617 1,557 1.903 1,494 2,126 3.323 1,742 IT82E-27 ............. 645 1.704 884 1,507 2,207 1,756 1,765 3,284 1,713 Control. . ....... . . .... 1.050 1.360 442 1,420 1,237 853 437 2,737 1,282 Location mean . . .... . . 836 1.599 3,001 1.55i 1,816 1,185 1.842 3.017 LSD. 5% . .... . 301 451 284 257 720 551 628 1.034 CV. . % . .. ......... . . . 25.39 19.89 25.81 11.62 27.95 32.77 24.05 24.19 white. brown, or red seed in both smooth and rough fourth , from December to February with irrigation. seed coat. The crosses were made in the greenhouse We created disease pressure in segregating popu­ where F I S also were raised. From the F, generation lations by artificially inoculating for diseases that on, hybrid materials were field tested. Using bimodal were not naturally severe. During the four growing rainfall and supplementary irrigation during the dry seasons, we evaluated 533 F 2 populations, 326 F, season, we grew four generations a year ! The first populations, 2,079 F 4 families , 2,186 F 5 families. 1,099 crop from February to April with irrigation; the F. families, and 202 F7 families, and selected many second. from May to July with rains ; the third, from desirable plants and progenies for further September to l\ovember also with rains; and the evaluation.-B.B. Singh 70 Grain Legume Improvement Program Table 3.3. Performance ofbruchid-resistant lines evaluated in Cowpea international trial 3 at indicated locations in 1983 Grain yield, kg/h. llTAS,S Mokwa Kamboinse Sekou Garous Sotouba Kwadaso Bouake Panama S,Cruz Line Nigeria Ni~eri a U. Volta Benin Cameroon 'fogo Ghana 1. Coast Panama Venezuela IT 8ID·I007 ....... 1,037 1,844 1,980 3,297 2.236 2,818 1,025 1.969 1,573 1,262 IT BID·1032 .... ... 914 2,359 1,861 2,894 1,371 2.609 1,161 1,544 1,639 1,822 IT 8ID·1064 ........ 646 2.312 2,218 1,774 2,362 2,630 1,254 1,437 1.606 2,142 IT BID·102O .. ...... 511 2,156 1,977 1,994 2.362 2.881 1.388 1,850 1,789 1,209 Control ...... . .. . . . 767 1,906 1,547 2,422 784 3,296 465 1,294 1,174 732 Location mean . ..... 741 1,887 2.024 1,789 2.097 2,390 301 1,644 1,463 1,375 LSD. 5% 640 451 405 618 428 765 380 622 277 544 C.V .. 5% .. ....... . . 24.23 16.47 13.80 23.81 14.06 22.05 32.71 26.05 13.04 27.29 Njala Njala 2 Bambey Makoholi Nsumba Chitala Nazareth Ubon Moroni Line S. Leone Senegal Zimbabwe Zambia Malawi Ethiopia Thailand Comores Mean IT 8ID·I007 . . . . .. ... 700 975 1.379 1.954 712 1,262 2.277 987 3,406 1,721 IT 8ID·I032 ......... 1,056 1,012 1,584 2.174 1.110 1.704 1.815 978 2.117 1,670 IT 8ID·I064 . . .. 994 1.062 1,432 2.243 587 1,973 2,089 1,122 1,789 1,634 IT 8ID·1020 .. . ... 1,044 894 1,856 2,039 698 726 1.542 878 2,387 1.569 Control. . ........ 1,1SO 862 728 2,469 1,751 2,461 3,151 337 2,389 1,563 Location mean 992 966 1,663 1,836 697 1,386.94 1.755 897 2.195 LSD. SOlo .. .... ...... 337 368 370 527 506 556 357 352 989 C.V .. % .. .. . . ....... 23.44 19.09 21.96 19.79 SO.08 28.07 14.02 27.04 31 .05 Table 3.4. Stability factors of cowpeas evaluated in international trials in 1983 at 16 locations lnt. Trial 1 Int. Trial 2 Int. Trial 3 Mean yield, Mean yield. Mean yield, Variety kg/ha b' Variety kg/ha b Varietz: kg/ha b IT 82E·16 1.666 1.20 TVx 4654·44E 2.220 1.30 IT 8ID·I007 1,721 1.30 IT 82E·18 1.621 1.20 TVx 1948·012F 2,067 1.30 IT8ID·1032 1,6iO .85 IT 82E·32 1,598 1.10 TVx 4677-010E 1.995 1.20 IT 8ID·I064 1.634 .92 IT 82E·13 1,540 1.00 TVx 4659'()2E 1.801 1.00 IT 8ID·1020 1,568 1.10 IT 82E·9 1.348 1.00 TVx 3236 1,796 1.10 IT 8ID·ll57 1,482 .90 IT 82E·41 1.173 .82 IT 82E·3 1,794 1.10 IT 8ID·ll37 1,469 1.35 IT 82E·56 1, 159 .84 IT 82E·70 96B .47 IT 8ID·985 1,373 .81 IT 82E·80 1,092 .83 IT 82E·49 950 .38 IT8ID·986 1,191 .90 IT 82E·97 1,041 .80 1T8LD·994 1,087 .71 I Regression of variety means over location means. Testing Advanced Breeding Lines Initial Evaluation Trials To select lines with wide adaptability, we evaluated A total of 251 n ewly developed breeding linea were the newly developed. advanced breeding lines in five evaluated in six initial yield trials. The first and to seven environments. The seven environments were second trials consisted of white seeded lines; the third two seasons at UTA and one at Mokwa, Samaru, trial bad cream colored lines; the fourth, brown seed Kano in ~igeria , Niamey in Niger, and Kamboinse in and rough coat; the fifth, brown seed and smooth Burkina Faso. The northern locations (Kano and coat; the sixth, primarily red seed but smooth and Niamey) and lITA's second season represented dry rough seed coats. Eight controls were used but not all environments; lITA's first season represented a were in each trial. The trials were distributed by seed· very wet environment, while Mokwa, Samaru, and color groups and planted in augmented design . In Kamboinse represented normal rainfall regions. each trial, the new lines were planted in single The breeding lines were evaluated in three sets of replication with control replicated three to five timea. trials: (1) initial evaluation of newly bulked F ./, The individual plots consisted of four rows 4 m long lines, (2) preliminary yield trials of breeding lines and 50 to 60 em apart. A 2O-cm plant·tQ-plant distance selected from 1983 initial yield trials and elite pro· was maintained in the rows, genies, and (3) ad vanced yield trials of the most Sixty lines performed as well as or better than promising breeding lines selected from 1983 initial controls. Several lines performed consistently better and preliminary yield trials. than others in all environments tested. IT saS.742-U , Grain Legume Improvement Program 71 IT 83S-742-13, IT 83D-339-5, and IT 83S-742-10 in Advanced Yield Trials Initial Trial 1; IT 83D-326-2, IT 83D-340-1, and IT 83D- 336-1 in Initial Trial 2; IT 835-728-3, IT 83S-728-3, IT Six advanced yield trials, each consisting of 20 entries, were conducted at IITA, Mokwa, Samaru, 835-728-13, and IT 835-782-5 in Initial Trial 3; IT 83D- 321-3 in Initial Trial 4; IT 835-682-7, IT 83D-317-1, and Kano, Niamey. and Kamboinse in randomized block IT 835-719-3 in Initial Trial 5; and IT 83S-687-15 in design with four replications. These trials consisted Initial Trial 6. Some low yields resulted from poor of the most promising lines selected from previous stands or early maturing rather than location. trials and represented different seed types. Trial 1 had Most of the top yielding lines with stable perfor­ cream colored seed; Trial 2, brown smooth seeds; Trial 3, red seeds; Trial 4, w hi te seeds, Trial 5, white mance resisted the major diseases. IT 838-742-10, IT rough seed, and Trial 6, brown and white seeds with 835-742-11, IT 835-742-13, IT 835-728-3, and IT 835-728- rough coats. Appropriate controls were used in each 5 combined resistance to aphids with multiple-disease resistance. Because they were evaluated in single trial. Yields at Niamey and IITA the second season replications, the lines' yields may heavily reflect were low because of short rainy seasons and drought environment. For that reason, when making selec­ stress. Performance varied widely from location to tions, we emphasized disease resistance, plant type, location, but a few varieties yielded consistently maturity, and consistent performance. The lines will better at several locations and also had less disease. be evaluated in r eplicated trials at several locations Lines that performed as well as or better than con· during the coming season. trols are listed in Tables 3.5 through 3.10. The most promising, considering yield potential and disease Preliminary Yield Trials reaction, were IT 82D-594·2, IT B2D-513-5, IT 82D·847, IT 82D-927. IT 82D-812, IT 82D-885, IT 82D·875, IT 82D· Ninety breeding lines were evaluated in five trials. 504·4,IT 82D·219, IT 82D·225, IT 82D·229, and IT 82D· Trial 1 had white seeded lines; Trial 2, cream colored; 216_ Reactions to diseases showed that most lines in Trial 3, brown ; Trial 4, tan and red seeds, and Trial 5, advanced variety trials showed moderate to high tan seeded lines. Controls varied in different trials. IT susceptibility to Septoria and scab diseases at 82D·716 and TVx 3236 were included in preliminary Samaru but this is not a major cowpea·growing area. trials 1 and 2; Ife Brown and IT 82D·789 in trials 4 and We are attempting to incorporate resistance to both 5. diseases even though they are prevalent only in The trials were conducted in randomized block restricted areas. Most of the new lines already com· design with three r eplications during IITA's first and bine resistance to these diseases. as indicated in second rainy seasons, at Mokwa, Samaru, Kano. Tables 3.5 through 3.lD.- B.B. Singh and S_A. Kamboinse, and Niamey. Yields of the best lines , in Shoyinka general, were much superior. The most promising by seed color were IT 83S·797, IT 83S·808, IT 83S·800, IT Extra Early Varieties 83S·818, and IT 83S·960, white; IT 83S·947 and IT 83S· We have emphasized disease and insect resistance 821, cream; IT 835·852. IT 835 853, and IT 835·871, while developing new 6Q·day cowpea lines and im· brown rough seed coat; IT 835·992 and IT 838·892, red proving existing 60-day varieties. We made many new and tan group; and IT 83S·873, IT 83D·873, and IT 83D· crosses and selected several progenies from segregat· 442, tan. They will be further evaluated in the coming ing populations, which will be evaluated in the season. coming season. Table 3.5. Performance of promising breeding lines evaluated in advanced yield trial! Grain yield, kg/ha Kam- Disease reaction I Line UTA Mokwa Samaru Kano Niamey boinse BF BB CYMV CAMV WB BBL SEPT SCAB IT 82D·513· 1. ... _ 2,143 1,638 1,729 1,469 438 1.552 1 1 3 3 2 1 3 5 IT 82D-594· 2 ..... 1,924 1,105 2,132 1,352 833 1 2 2 2 2 1 3 2 IT 82D·952 ...... 2,116 1,664 1,456 1,508 500 1.164 1 1 3 2 2 1 4 1 IT 82D-704 ...... 2,238 1,430 1,229 1,963 450 567 4 2 3 2 2 1 5 1 IT 82D·516·5 ..... 2,024 1,261 1,612 1,235 288 1.196 1 2 3 1 2 2 4 5 IT 82D-702 ...... 1,705 1,508 1,066 1,690 538 904 4 1 2 1 2 2 5 2 IT 82D-513-5 . .... 2,150 1,300 1,443 1,391 425 635 1 1 1 1 1 1 3 4 Control IT 82D-716 ...... 1,571 1,040 1,105 1,690 575 691 1 1 3 1 1 2 4 1 LSD,5% .. _ ..... 637 541 480 419 277 414 C.V .• % ......... 36 29 26 22 46 34 lReaction of all lines listed to CCr was 1, and all except IT 82D-516-5 with 2. had 1 for ANTH_ 72 Grain Legume Improvement rrogram Table 3.6. Performance of promising breeding lines evaluated in advanced variety trial 2 UTA Grain yie ld , kg lha Breeding First Second Kam- Disease reaction 1 line season season Mokwa Samaru Kano Niamey boinse CCR BP BB WB BBL SEPT SCAB IT 82D-847 1,502 778 2,457 2,002 1,651 530 551 1 1 1 2 2 3 4 IT 82D-752 .. 1,432 541 1,755 2,262 1,573 425 816 1 1 1 2 1 1 5 IT 82D-927 .. 1,723 572 2,080 1,846 1,378 238 800 1 1 1 2 1 4 3 IT 82D-786 " , 1,821 718 1,768 1,698 1,378 325 800 1 2 1 2 1 4 3 IT B2D-831 " 741 641 2,210 1,807 1,417 263 821 1 1 2 2 2 2 4 IT B2D-812 , , , 969 509 1,547 1,872 1,651 275 524 1 1 1 2 2 1 Control IfeBrown ... . ... . 1,100 591 1,937 1,807 572 366 643 2 1 3 3 3 4 5 LSD.5% .. . ... . . . 805 243 550 369 361 252 424 C.V.,% .......... 51 28 21 16 24 45 42 IThe reaction of all lines listed, except the Control with 3, was 1 for CYMV, CAMV, and ANTH. Table 3.7. Performance of promising breeding lines evaluated in advanced variety trial 3 UTA Grain yield, kg/ha Breeding First Second Kam- Disease reaction I line season season Mokwa Samaru Kano Niamey boinse CCR BB CYMV CAMV WB BBL SEPT SCAB IT 82D-865 2,029 830 2,171 2,327 1,157 325 984 1 2 1 1 3 1 4 3 IT 82D-861 l,459 531 2,197 2,041 1,144 500 1,478 2 1 2 2 3 2 1 5 IT 82D-880 1,653 644 2,210 2,015 1,131 450 1,246 3 1 1 1 3 2 3 5 IT 82D-871 1,349 610 2,002 2,197 1,157 663 1,142 1 2 3 3 3 1 1 5 IT 82D-875 1,284 632 1,898 2,288 845 425 1,281 1 1 1 1 1 1 1 2 Control IT 82D-716 630 1,625 1,781 1,092 588 1,137 1 1 2 2 1 2 4 1 LSD, 5~'O 319 206 559 474 343 242 611 C,V,,% 17 24 22 16 22 46 25 lThe reaction of all lines listed was 1 to both ANTH and BP, except for 4 to BP by IT 82D·875 Table 3.8. Performance of promising breeding lines evaluated in advanced variety trial 4 lITA Grain yield, kg/ha Breeding First Second Kam· Disease reaction 1 line season season Mokwa Samaru Kana Niamey boinse BP BB CYMV CAMV WB SEPT SCAB IT 81D-97D " ' , , , 2,070 571 2,054 1,469 1,417 586 1,326 1 2 4 4 2 4 2 IT 8ID, 1189-81, ' , 2,369 568 1,508 2,132 1,222 475 1,008 3 I 3 1 3 4 1 IT 82D-504-4, 2,158 640 1,599 1,742 1,183 375 1,263 1 1 1 1 2 3 4 IT 82D-699 __ 1,788 620 1,430 1.287 1,755 563 1,211 1 2 3 3 2 4 5 Control TVx3236 ,,_ 1,850 373 2,171 1,391 1,378 625 1,523 1 1 4 1 2 5 1 LSD, 5%, , , , , , , , 841 203 572 435 314 236 509 C.V., 0;0' . . . . .. . . 31 29 28 19 18 33 38 1 The reaction of all lines listed was 1 to CCR, AKTH , and BEL. Table 3.9. Performance of promising breeding lines evaluated in advanced trial 5 IITA Grain yield, kg/he Breeding First Second Kam- Disease reaction l line season season Mokwa Samaru Kano .N'iamey boinse BP BB CY:vlV CAMV WB SEPT SCAB IT 83D-225 2,203 491 1,989 1,963 1,118 600 1,646 1 3 3 3 2 3 4 IT 82D-703 2,340 404 1,703 1,820 1,209 769 1,133 5 1 3 3 2 4 2 IT 83D-219 , 2,285 662 1,573 1,716 845 525 1,591 3 2 3 1 2 2 2 Control TVx 3236 , , 1,979 456 2,223 1,729 1,456 726 1,124 1 2 3 3 2 4 1 TV x 4659-03E 1,534 140 1,066 1,768 1,235 406 1,505 1 2 1 1 1 2 2 LSD. 5"i~ ..... 576 201 585 501 332 240 531 C.V .. %, ....... 22 31 29 24 23 28 33 lThe reaction of all lines listed was 1 to CCR. ANTH, and BBL. Grain Legume lmprouement Program 73 Table 3.10. Performance of promising breeding lines evaluated in advanced variety trial 6 UTA Grain yield, kgJha Breeding First Second Kam- Disease reaction I line season season Mokwa Sarnaru Kano Kiamey boinse BB CYMV CAMV WB BBL . SEPT SCAB IT 83D·27 . . .. 2,066 543 2,080 1,937 1,469 650 1,337 2 4 4 2 1 4 2 IT83D·206 . . .. . 2,779 413 1,469 2 ,470 1,066 500 1,386 I 4 4 2 2 2 4 IT83D·235 . . .. . 1,929 726 1,547 2,015 1,079 838 1.628 3 4 3 2 1 2 4 TV x 3236-6- 1 . 2,286 553 1,521 1,859 1,170 781 1,194 I 3 3 2 2 4 I IT 83D·229. 1,979 467 1,235 2,405 1,053 938 1,238 1 3 1 2 3 2 4 IT 83D·216 . 2,137 380 1,586 1,950 1,274 600 1.334 2 2 2 2 I 4 1 Control TVx 3236 . 2,441 291 1,859 1,521 1,404 1.025 948 I 3 3 2 4 1 LSD,.5% 665 ]72 506 611 321 278 603 C.V., % ,_ 25 26 24 24 22 29 40 I Except for a 2 by IT 830 -216 to BP and a 2 by IT 830-206 to ANTH , Blli ines listed had 1 for CeR, BP. a nd ANTH . Among several new early· maturing lines evalu· high yielding groups in both wet and dry plots. On wet ated, IT 838·818, IT 838·962, IT 838·844, IT 838·849, IT plots, IT 83D·442, IT 82E·9, IT 82D-889, IT 84E-I24, and B38·B50, IT 838·945, IT B3D·37B, IT 830·379, IT 83D· IT 82D· 716 showed similar yield potentials, all signi­ 375, IT 838·883, IT 83D·442, IT B3D-200, and IT 84E·124 ficantly better than IT 82E-60's, which was reduced by appeared qui te promising. IT 83D·442 with brown a severe attack of Cercospora cruen/a. Threshing smooth seed and IT 84E·124 with brown rough seeds percentages in drier plots. were significantly less for have good yield potential, matured between 55 and 60 only three lines (IT B2E-9, IT 82E-60, and IT 84E·124), days , and have resistance to severa] major diseases. indicating poor pod filling. IT B4E·124 and IT 83D·442, They will be distributed to various national programs the newest 60·day lines, appear quite good for rice in 1985. fallows . The data indicate that 70· to 75-day varieties can be Extra Early Varieties in Rice Fallow equally successful in rice fallows if there is no Fi ve extra-early varieties were evaluated in rice pressure to vacate land within two months and family fallow on r esidual moisture with medium maturing IT labor is available for multiple pickings. 82D-716 as a control. As soon as rice harvest was The data on yield components and harvest indices are presented in Table 3.12. All five extra-early complete, early maturing cowpeas were planted (30 varieties had very high harvest indices (from 46 to November). The field's natural gradient r esulted in an upper portion with very low moisture and a lower 51 'Yo), not significantly different. IT 82D·716's harvest portion with extremely wet soil but no standing index (37%) was significantly below other varieties. Thus, it produced more biomass than other varieties water. The t rial was planted in a split plot design with so itmay be desirable where straw i. needed for fodder three replications. Varieties constituted the main or green manure.-B.B. Singh plot with moisture regimes as subplots (upper part of slope dry, lower pa rt of slope wet). Each subplot consisted of five rows 6 m long and 45 em apart with 20 em between hills within the row, two plants per hill. Table 3.11. Performance of 6O-day cowpea varieties in Oecis was sprayed 20, 30, and 40 days after planting to dry and wet plots on rice fallow control insects. Data on ma turity, threshing per· centage , and grain yi eld are presented in Table 3.1l. Days to Threshing Grain maturity % yield, kg/h. All variet ies in the drier part of the field matured earlier than those in the wet part ; extra early va· Variety Dry We t Dry Wet Dry Wet rieties, 4 to 5 days earlier ; medium·maturing IT 82D­ IT83D-442 . . 52 56 73 72 1.428 l ,570 itS, 12 days earlier. Yields of all the extra-early IT82E·9 .. ... .... 56 61 65 il 1,184 l,549 varieties were lower in the drier plot than the wet IT82D-889 ..... .. 52 56 69 68 l,06O 1,34] IT84E· ± 24 ..... .. 56 60 64 76 914 plot. 0 ifferences in yie ld were significant for IT 82E-9 1'l68 IT82E·60 .. .... . . 56 6l 64 60 90] l ,043 and IT 84E-124, but medium·maturing IT 82D-i16's IT 82D·7I6 , .... .. . 70 82 76 75 1,518 1,490 yields were similar in both dry and wet plots. The medium·maturing IT 82D-71S produced profuse vege­ ~ean ...... ... ... 57 62.3 69 74 1,168 1,394 tative growth and inter-plant competition in wet LSD,5% Between dry and wet. plots but not in drier plots. Also, IT 82D· 716 required within same variety ... l.0 5.1 327 two harvests but all the extra early lines were picked at one time. Between varieties Yield potentials of the lines differed significantly. across dry and wet . ... . 0.73 3.6 23] IT 83D-442, IT 82E·9, and IT 82D-716 were among the I A medium maturity line. 74 Grain Legume Improvement Program Table 3.12. Yield components and harvest indices Vegetable Cowpea Varieties indicated extra/early varieties evaluated in rice fallows Concerted efforts were made to develop high· Seed yielding, bush· type vegetable cowpeas of different Straw 100 yield weight seed Harvest maturity groups that carry disease resistance. Pod/ Seedsl plant, plant, weight, index,' Severa 1n ew crosses were made and many segregating Rice plant pod g g 0 ' ~ /0 progenies were evaluated. The better ones will be IT 820·889 7.0 15.5 7.0 7.0 10.0 50.0 further evaluated. The most promising F. lines em· IT 82E·60 8.5 10.5 8.5 8.0 18.5 51.0 erged from crosses IT 810·1228·15 X IT 820·716 and IT 84E-124 12.0 11.5 9.5 10.5 12.5 47.0 IT 810·1228-14 x IT 820·716. Both F.lines combined IT 830442 7.0 14.5 9.0 10.5 11 .5 46.0 multiple disease resistance, medium maturity, good IT 82E·9 8.0 13.5 1l .0 10.5 13.0 51.0 plant type, and improved pod characteristics. They IT 820·716' 12.6 12.3 9.5 16.5 12.0 37.0 will be evaluated in replicated yield trials in 1985. LS~. 5% 3.9 3.3 4.5 4.9 2.2 9.5 C.V .. % 16.6 9.0 18.0 19.0 6.0 6.9 Initial Yield Trial of New Vegetable Type Lines I Ratio of seed yield!plant and s traw weight/plant. ~Medium matu ring. Sixteen new breeding lines were evaluated in trials at lITA, Mokwa, and Samaru along with three controls. IT 810·1228·14, IT 820·38()'5, and IT 8ID·1228·10. Based on performance at different locations and disease reactions, IT 83S·344·4 and IT 83S·927 were most promising. Four others, IT 838·346·2, IT 838·683· Incorporating disease and insect resistance 2, IT 83S·917, and IT 83S·919, resisted web blight. a genes into extra-early cowpea varieties by major disease in vegetable cowpea areas. All the backcrossing. Highly yielding. extra·early (GO·day) controls were susceptible. cowpea varieties developed by UTA should increase cowpea production in the Sahel savanna (300 to 500 Preliminary Yield Trial of Vegetable mm of rainfall), where cowpeas can be grown in rotations with a limited fallow period and in rice Cowpeas fallows. Preliminary results indicate that with li· Eighteen n ew breeding lines were evaluated in a mited residual moisture after rice harvest, such extra· preliminary yield trial at UTA, Mokwa, Samaru, and early varieties as IT 82E·60 and IT 82E·889 should Niamey with IT81D·1228-14 and FARV·13 as controls . produce up to 1.5 t fha . The tria ls were planted in randomized block design A drawback of the 6O-day varieties is their suscepti· with three replications. Individual plots consisted of bility to major d iseases and insects, so we started four rows 4 meters long, and the middle two r ows were backcross breeding this year to incorporate disease harvested for green· pod yield. Row·to·row spacing and insect r esistance into four extra·early cowpea was 50 em with plant·to-plant distance 20 em. Data on lines. The new lines can be used by national programs green·pod yield and disease reaction of promising directly as varieties or indirectly as improved varieties are presented in Table 3.13. IT 83S·904 had germplasm. the highest mean yield, varying at individual lo­ cations from 6.5 to 15.7 t/ha. Other promising va­ Research Methods rieties were IT 83S·906, and IT 83S·911. Each yielded Four high. yielding, extra·early genotypes were selec· as well as or better than the controls. FARV·13, a ted for improvement. Each was the best performer in climbing type, yielded the leasL All tested varieties resisted several major diseases but the resistance of a its color group (white, brown, red , black)- to suit few to web blight and scab was weak . consumer preferences in various environments. Disease and insect resistance to bruchid, thrips. and cowpea aphid·mosaic-virus resistance. control­ Vegetable Type Cowpea International Trial led by recessive genes, will be incorporated by breed· Seven of the most promising bush·type vegetable ing scheme 1. Resistance to aphids and anthracnose, cowpea lines were included in International Trial controlled by dominant genes, wi ll be incorporated along with FARV·13 and Bush·Sitao as common using scheme 2. Three generations have been grown controls and a local control variety. Thirty·four sets since July 1984, the F , cross, BC, F,. and F , . of this trial were sent to co·operators in the tropics. In the International Yield Trials, the recurrent Results will be presented in IITA's 1985 Annual parents yielded well during 1983 in East Africa and Report. But yield data received from a few locations Southeast Asia. We expect the new genotypes to (Table 3.14) show high promise for some lines. Such maintain yield stability over a wide range of environ­ newly developed. bush-type varieties as IT 810·1228- ments when the genes for resistance to diseases and 10, IT 810-1228-13, and IT 810·1228-14 gave highest insect pests are incorporated into recurrent parents. green pod yields (6 to 16 t fha) at all locations, This project is funded by the Italian Department of significantly better than climbing type (FA RV-13) or Cooperation.- F.A. Eisayed and B.B. Singh bush type (Bush·Sitao) and other local controls. IT Grain. Legume Improvement Program 75 Table 3.13. Indicated data from promising vegetable-type cowpeas evaluated in preliminary trial 6 IITA First Second Grain yield, kglha Disease reaction 1 Cowpea season season Mokwa Samaru Niamey BP BB CYYlV WB SEPT SCAB IT 83S-904 ... .. ... .. ... ..... 11,938 6.550 9,013 15,674 7,532 1 1 3 5 2 I IT 83S-908 ... .. . ........ · . 10,313 10,270 5.859 7,879 7,513 4 1 3 3 3 3 IT 83S-901. .... .. ... . .. . ... . 12,431 8,742 9,447 9,748 5,622 1 1 1 3 2 5 IT 83S-909 .... _. · . 11.875 9,767 9,395 9,745 3,690 1 I 1 2 2 1 IT 83S-899 ...... .. 7,198 6,346 10,244 11.258 5,875 1 1 1 3 3 3 IT 83S-906 ...... .. 9,683 4,423 7,003 14.262 2,693 1 1 3 2 2 1 IT 83S-911 ...... · . 11,703 7,662 8.355 5,134 2,288 1 2 1 3 .5 3 Control IT 8ID-1228-14 (Bush t)'pe) .. 8.694 4,464 6,500 15,988 2,808 I 1 3 3 3 3 F ARV-13 (Climbing type) . 6,086 6,083 9,221 8,383 3.408 4 4 4 2 3 3 LSD, 5% .................. 3,905 2,905 2.758 4.345 3.811 C.V .. °/~ ......... . .. . . .... . . 32 25 22 26 54 ' The reaction of all lines listed was 1 to CCR and ANTH and it was 1 to CAMV and BBL for all except a 3 to CAMV by IT 83S-908 and a 2 to BBL by IT 838-911. Table 3.14. Performance of improved bush-type vegetable cowpeas at indicated locations in Africa and Asia Yield, t/ha Ib.dan Mokwa Nyankpala Abomey Parawanipur Ganaruwa IRRI Cowpea Nigeria Ghana Rep. Benin NeEal Sri Lanka Phi 1i EEines Mean IT 8ID-1228-1O .. .... . ...... 14.3 10.7 9.4 12.1 7.7 6.8 16.0 11.0 IT 8ID-1228-14 . ... . .. . .... . 15.1 9.1 6.5 1004 8.5 9.7 13.7 lOA IT 8ID-1228-13 .. .. . .. . ...... 14.3 6.3 5.9 9.0 8.3 6.8 11.5 8.9 Bush·Sitao ... 6.5 8.5 4.4 12.1 5.6 5.8 7.3 FARV-13 ..... 8.7 8.5 3.0 6.2 6.2 7.0 6.6 Local Control. * * 2.0 7.9 3.4 4.5 12.3 6.0 LSD,5%. _. . . . . . .. . ... . . . . . 4.1 2.8 2.3 2.3 3.9 2.5 4.7 *Local control in Nigeria was FARV·13. 81D-1228·10 averaged 11 tlha over all locations ; IT length and root mass within four weeks after plant. 8ID·1228-14, 10.4 t/ha, and IT 8ID·1228-13, 8.9 tlha ing.IT 82E·60, IT 8ID·I020, and IT 82D-812 roots grew compared with 6 to 7 tlha by the controls. significantly faster than others, and IT 82E-60, TVx Green-pod picking started at most locations 45 to 5:; 3236, and IT 82D·716 produced more root mass than days after planting with five to eight pickings at 3· to others (Table 3.15). Trials in Sahelian environments 4-day intervals. In 75 to 80 days, 1 hectare produced have shown IT 82D·i16, TVx 3236, and VITA·3 to be more than 10 tons of green pods. These varieties can drought tolerant, and IT 82E-60to perform well in rice be grown under irrigation throughout the year in the fallows as well as in short, rainfall areas. Further tropics. They can thus provide fresh vegetables studies are in progress to elucidate relationships continuously and add to tropical farmers' income.­ between drought tolerance, root length, and root B.B. Singh mass.- A.M. Maponoko and B.B. Singh Breeding for Root Improvement Breeding for Disease Resistance Cowpeas normally are grown in the semi-arid regions Our progress continued in combining multiple dis­ of the tropics because of their known tolerance to ease resistance and high yield potential in different drought. The development of 6O--day varieties has maturity types and seed· quality groups. We screened widened the scope of cowpea cultivation. It is now all segregating progenies for resistance to web blight, possible to grow a crop of cowpeas with residual cercospora leaf spot , anthracnose, brown blotch, moisture in rice fallows or in areas with short rainy bacterial pustule, bacterial blight, cowpea yellow seasons. For all such ecologies, cowpeas with fast mosaic, and cowpea aphid·borne mosaic under nat­ growing, deep, dense root systems would be de~. . i.rable. ural infestation. The advanced breeding lines were to ensure quick establishment and better uptake of screened for all those diseases under both natural and moisture from the receding water table. With that in artificial infestations at IITA and for Septoria , scab, mind. we examined genetic variability for root'char­ brown blotch, and bacterial blight at Samaru. The acteristics in cowpeas in the greenhouse. Our pre­ disease scores for selected advanced breeding lines liminary results indicate wide variation in root are indicated in the evaluation section of this report. 76 Grain Legume improvement Program Ta hIe 3.15. Root characteristics of improved co"'peas once a week, from planting to harvest time. four weeks after planting Thirty-five lines t hat flowered and produced grains Root Root dry weight were selected for further evaluation, Characteristics length, per plant, considered important for selection include ability to Cowpea cm g germinate under severe cold and to complete physio­ IT 82E-60 ' 94,9 1.64 logica l life cycle under cold temperatures, Further IT 81O-lO20, 93,5 0,67 teStS are planned for 1985, Ten orthe lines selected for lT81O-812, , "'.'"'' 92,8 0,97 cold tolerance, all preceded by TVu, are 9,024, 9,753, VITA-7" _,, , '" .''' ,, '' '' ' 84,3 1.10 11,313, 1,925,8,216, 114,8,450,9,762, 132, and 11,309.­ IT 82D-il6 ' , " " ' " " , " , " ' , 71.8 1.14 S,A, Shoyinka. N, Ng, and B.B, Singh VITA-3.. ' " " " " , " _" , 70,5 1.23 TVx 3236 ' , , , , , , , .• ' , , . ' , , , , , 66,5 1.87 LSD,5% ' , ' , , .. ' .• , , , , , , ' . ' , 21.7 0,74 Entomology Breeding for Multiple Disease and Insect Resistance A number of new crosses were made and many Several selected lines combine moderate-to-high resistance to most of the diseases mentioned above, segregating progenies and advanced breeding lines were screened for multiple djsease resistance and Inheritance of Resistance to Brown Blotch, resistance to aphids, bruchids, and leafhoppers; 174 Septoria Leaf Spot, and Scab breeding lines were resistant to bruchids; 511, to aphids; 17, to leafhoppers, Many F.linesderivedfrom To plan an effective breeding strategy, we used the cross IT 810-1020 x IT 82D-716 combine multiple genetic studies to elucidate the inheritance of re­ disease resistance and resistance to thrips, aphids, sistance to brown blotch, Septoria leaf spot, and scab, and bruchids, These lines will be evaluated for yield Resistant and susceptible parents for those diseases and other agronomic characters during 1985, are indicated in Table 3,16, F " F2 , and backcross A numher of F 2 populations developed from early­ populations involving resistant and susceptible pa­ and medium-maturing parents carry multiple disease rents for each disease were screened under heavy resistance with resistance to bruchids and thrips, disease pressure ensured by artificial inoculation , bruchids plus aphids. or aphids plus leafhoppers, The results indicated Mendelian inheritance for each Thus, recombinants with resistance to several disease, Resistance to brown blotch is controlled by a diseases and to a combination of leafhoppers, thrips, single recessive gene pair; resistance to Septoria, by aphids, and bruchids are expected in both early and duplicate dominant genes; resistance to scab, by a medium maturing groups, single r ecessive gene pair, The data further indicated Through repeated backcrossing, we incorporated that IT 82E-16 has a single pair of recessive genes for aphid resistance into IT 82E-60, a 60-day cowpea. and resistance to brown blotch and duplicate dominant into TVx 3236, the thrips-resistant line, Aphid re­ genes for resistance to Septoria, VITA-4 and IT 82E- sistant versions of these varieties will be evaluated 77 have the same one dominant gene for resistance to for yield and other agronomic characters in the Septoria, TV x 3236 has a single pair of recessive genes coming season,-B.B. Singh and S,R, Singh for resistance to scab, The phenotypic expressiorrs of all the genes that condition resistance are quite Breeding for Pod Borer (Maruca testulalis) pronounced and easily discernible in segregating Resistance populations, so it, should be quite easy to genetically manipulate resistance characters in breeding Maruca pod borer (MPB) is a serious pest in most programs,- B.B, Singh, J,A, Abadassi, S,A, cowpea,growing areas of the world, Severe natural Shoyinka, and A.M, Emechebe infestations with no chemical control can cause high yield losses, There are several accessions that are Screening for Cold Tolerance much less susceptible to pod borer than the common control under heavy infestation, A tota I of 407 cowpeas accessions selected by the We studied the accessions (TVu 946, Nigeria; TVu Germplasm Resources Unit (GRU) were screened for tolerance to cold at the Kadawa Irrigation Research Station, Kano state between February and May 1984 Table 3.16. Reactions 1 of parental cowpea lines to during the harmattan season, The cowpea lines indicated diseases originated from the highlands of East. West, and Line Brown blotch Septaria Scab Central Africa. around the Mediterran ean coast, and TV. 3236 ' , '" ,"" , , , . . . . . . . . . R S R from South Africa, where temperatures are generally VITA-4, _"' " " ..... . ....... " S R MR lower than in the traditional cowpea-producing IT 82E,16, , " , " , .. . ... . .. R R S areas, Each line was planted in 4-m long, single,row IT 82E,60 ' , " , " , .... . .. . ..... . S S S plots, 8 February 1984, with 20 cm between plants, IT 82E,77 "', . .. . .. . , " . . .. , . . . S R S Irrigation was provided through a gravity system IR = resistant ; MR = Moderately resistant.; S = susceptible, Grain Legume Improvement Program 77 1, Nigeria; Kamboinse local, Burkina Faso; and IT first two are black·seeded; VITA 3 has dark red seeds 82E-9, Nigeria) to see if they have the same source of and a thicker pod wall than the others. TVu 1980 had genetic resis ta nce; how many genes control r e­ the sharpest drop in moisture content between days sistance to MPB; and to incorporate MPB resistance 12 and 19, followed by TVx 3343·03E in the first into extra·early IT 82E·60. Crosses among the five season. In the second season , TVu 1890, TVu 1, I'f82E- parents in t he greenhouse gave us 100 to 200 F , seeds 9, and Ife Brown had the lowest moisture contents per cross. Crosses among resistant parents will and t he sharpest drop between days 12 and 19. TVu accomplish the first objective, and crosses between 1890 also had the least seed damage. In a field the resistant and the susceptible control IT 82E·60 experiment in Mokwa, TVu 1890 and TVu 1 had the will achieve the second and third objectives. Between fewest insects per unit area and the least seed 50 and 70 F, seeds from each cross were advanced to damage, The results indicate that both TVu 1890 and the F 2 generation. And 10 to 20 F, seeds from each TVu 1 are ~omewhat resistant wit h resist ance gover­ cross were seeded in t he glasshouse to develop the ned perhaps by more than one gene. These studies will back crosses F ,P, and BC, F ,P 2' Four hundred to 600 be continued with in vestigation s added on the effects F 2 seeds/cross from each resistant X resistant and of pod·wall structure and such secondary plant resistant x susceptible population are being grown products as tannins and alkaloids.- L.E.N. Jackai in the field (January to April, 1985). The F 2 material will receive maximum protection against diseases Maruca pod borer (MPB). As a continuation of and insect pests. . our work on cowpeareslstance to l\1PB, \ve evaluated 430 selections from the previous screening of progeny During August, 1985, the parents, F" BC,F,P"and F 3 progeny r ows will be planted in Mokwa, a "hot linea in rvlokwa. Plot sizes were two 3-m rows with no spot" for MPB, and sprayed at 10·day intervals with replications. Susceptible cultivar IT 82E·716 and Nuvacron to contro l /lower thrips and pod bugs. resistant TVu 946 were planted as controls once in Plants will be scored for pod damage. Populations every 10 entries, and a susceptible spreader row, once in every two ranges as multiplication foci for the pest. with enough seeds will be tested at lIT A a lso. If the r esults of crossing among resistant parents All plants were sprayed with monocrotophos at 200 show transgressive segregation for pod borer r e­ g a.L /ha four times at weekly intervals starting at sistance, yield losses could be significantly reduced , flower-bud ini t iation . especially in "hot spots" by incorporating resistance At pbysiological pod maturity, the test lines were (from all genetic resistance sources) into common rated visually for damage as shown in Table 3.18. varieties. Even mild resistance may reduce the num­ ~one of the 430 lines (each with TVu 946, TVu 1. and ber of chemical sprays needed to control MPB. Kamboinse local background) was highly resistant. Results fro m the inheritance study will help us Two entries had 11 to 25% of their pods damaged; choose strategies for breeding for MPB resistance.­ others more, asshown in Figure 3.1. Mean score of the F.A. Elsayed and L.E.N. Jackai resistant control was 50% ; that of the susceptible control , 78%. Only 9.5% of the 41 lines will be Host Plant Resistance evaluated further.-L.E.N. Jackai Pod-sucking bugs (PSBs). In 1982 and 1983. seed of Fluctuations and yield assessment. Studies TVu 1890, TVu 1, TVx 3343·03E, and [T 82E·9 were have been conducted during the past three years on damaged much less by pod·sucking bugs (PSBs) than population fluctuations ofMPB. We hope to increase other varieties were. PSBs populations then exceeded our understanding of population pressures under 20 bugsj2-m row. which field experiments are carried out and to assess In the first of our 1984 studies, we caged first instar yield losses associated with various populations in nymphs with intact cowpea pods on potted plants different locations. using cloth· mesh sleeves, S nymphs pe r cage with 8 The experimental design was the same as described replications- 40 nymphs per line. The insects were for pod sucking bugs except that the unprotected plot transferred to fresh pods every other day. We reo corded development times from nymph to adult and bug mortality on each cowpea line. Ife Brown was Table 3.11. Development time (N,-Adult) of included as the control. The results are shown in Clavigralla lomentosicollis on indicated Figure 3.1. CDwpea lines and varieties In the second study, 12 pairs of adult Clavigralla Total developmental tomentosicollis were released in mesh cages measur­ Cowpea time (days) (+ se) ing about 1.5 m3 containing three cowpea Jines and Ife Ife Brown. . . . . .. . . . .. . .. . . . . .. . .. . . . 15.3 ( =0 .11) Brown as control. The insects were released when IT82E·9 .. . ....... . .. .. . ....... . 15.0 (=0.0) plants were at the early podding stage. After two TVu I .... . ............ . . 15.3( ;: 0.13) weeks, measurements were made on pod and seed TVu 1890 .. .. . . .. ...... .. 15.3 (±0.18) damage. Even though developmental time did not IT 82D·638 .......... .. .. .. 16,6 ( ± 0.37) vary significantly among most lines (Table 3. 17), TVx3343-03E. .... .. . . 15.7(±0.1l) mortality was significantly higher on TVu 1890, IT VITA·3 . . . . . . . . . . . . 16.0 ( ± 0.59) 82E-9, and VITA 3 (Fig. 3.1) than on other lines. The TVu 1985 ,... ... ... ... ..... .. . .. .. . 16.2 (±0.13) 78 Grain Legume Improvement Program Cumulafive rTIOrtollf)' (%) Number of Iorvoe per 20 flowers 60 IITA MOKWA 151 Sooson 2rd Seoso<' DOP AU(l 22,1984 50 70 [Xl" .o.pt'I I_ ,I984 Sept _ , 984 601- 1 40 t , , 501- / / TV>: 3343 -03 E 30 .a.----.a.----.a.----.a.----.a. I T'o'\I - 1 I 301- I 20 / -- Ite Brown I .----.----.-~-.----. 20 I- rfr" + lrOif ~ a l...l:~' L,05:l::-08,l:,12L.t5'-::::,9"2~2L..J.-:;:22-k;2.!-.2;;9~1IU 03 06 {) 13 17 20 24 27 I June Oct----+-Nov I Oc t Deys ofter coqinCiJ Figure 3,2. ~'daruca pod borer populations were muc,h Figure 3.1 Death percentages of pod-sucking-bug nymphs higher the second season at UTA than the first b~t thelr caged on indicated cowpea lines. reproo.uction period was shorter the second perlOd and shorter than at :Mokwa. Table 3.1S. A visual scale for rating l\1aruca damage to cowpea that question we initiated studies in 1983 to de· Score Pod damage % Score Pod load" termine the damage or infestation threshold for 1 = Il-5 5 Occasional bare peduncle control action. The experiment was set out in a 2 &--10 4 = Up to 15% peduncle, bare randomized block design with six pre·set, damage or 3 11- 25 3 = Up to 30% peduncle, bare infestation levels : 0, 20, 40, 60, SO, and 100%. Each 4 26-40 2 = Up to 6O(1~ peduncles bare level was replicated four times. The experiment 5 41--{;0 1 More than 60% peduncles received four weekly blanket sprays of monocroto­ 6 = 51--60 Bare phos (200 g a.i./ha) to suppress thrips and pod bugs. 7 = 61 - 70 "Pod load: the ratio of 8 = 71-85 peduncles without pods: From each lO·row x 15·m plot, we collected four 9 8&---100 peduncles with pod,_ sub-samples of 20 flowers each once a week for three or four weeks, starting with flower onset, and noted the number of infested or damaged flowers. When any pre·set infestation level was exceeded. it received received four weekly applications of man ocra top has deltamethrin spray the same day. The day the last (200 g a.i./ha), starting at flower·bud initiation. That treatment does not interfere with borer populatIOn flower samples were collected. 20 pod samples were collected and damage·infestation levels were simi­ build.up. Five flowers/plant were sampled from ~ve plants on pre-determined l·meter row samphng larly assessed. Yields were taken at harvest. The initial infestation or damage observed was SimIlar III points. Samples from each plant were held In a all plots during first and second seasons but varied separate vial with 30% alcohol for subsequent slightly more the second season (Fig. 3.3). examination. The MPB population in 1984, as in previous years, Fewer sprays were required during the second than was higher the second season at lITA. The repro­ the first season in 0 and 20% infestations, so the pre­ ductive period was shorter the second season at IITA set threshold was exceeded oftener in the first season so fewer samples were collected. In Mokwa the (Fig. 3.3). In no case did pod damaged or infestation population was high over a longer period than at lIT A estimates influence a decision to spray based on (Fig. 3.2). Yield losses were similar to those m earher flowers. studies, and almost twice as high in Mokwa as at Grain yields from each season are given in Table UTA. We do not understand why because larval 3.19. Significantly highest yields were from fully population was higher at UTA the second season. It is protected plots (zero infestation). possible that this is related to the stage of the crop at The profit margin a small farmer would expect by which the attack was most severe. In 1983 the losses fully protecting his crop does not justify spraying to at :v1okwa were about five times those at IITA. prevent all damage. But two well-timed applications We shall continue these studies at both locations during the second cropping season will keep dam,,;ge two more years.- L.E.N. Jackai or infestation between 20 and 40%. Further studies are needed for proper economic assessment and Damage or infestation threshold. When should before firm recommendations can be made.-L.E.N. cowpeas be sprayed against pod borer? To answer Jacka; Grain Legume Improvement Program 79 Damage/ infestation (%) ATh=O ATh=20 ATh=40 ATh=60 ATh=80 ATh=IOO 100 SI I 24 31 7 24 31 7 24 31 7 24 31 7 24 31 7 24 31 7 I---Oct ----+--Nov-i Sampling date Figure 3.3. Fluctuation in damage·infestation by Maruca pod borer on cowpea TVx 3236 flowers and pods (dark areas) and spray action (arrows) under pre-set infection levels, liT A second season, 1984, Broken lines represent damage- infestation thresholds (Alh). Storage beetle. Resistance to the bruch id. program to control bean flies. Callosobruchus maculatus, was tested in the labo· Cooperating with the Asian Vegetable Research ratory wit h 29 cowpea varieties includingTVu 2027, a and Development Center (A VRDC) in Taiwan, we seed resistant control. Ten seeds of each test variety screened cowpeas in field plots for bean fly resistance were infested witb 2 pairs ofO· to l-day·old insects for during October and December 1984- a total of 753 15 hours, and 5 pods of each variety were infested with cowpea lines. Based on plant- infestation percentages 5 pairs of the insect s for 24 hours. Temperature of the and the number of beanfiy larvae and pupa ria per laboratory ranged from 21 to 29"C; relative humidity, plant, we selected 15 lines for further evaluations in from 63 t o 84% . .Moisture contents of the seeds ranged Taiwan and other Asian and African countries where from i .3 to 8.1%. The hardness of the pod wall was bean flies persist. Data from 8 of the 15 entries are tested with a penetrometer that we designed. given in Table 3.21.- S.R. Singh, N.S. Talekar: H.S. The indices of insect susceptibility gave 12 va· Chiang, and L.E.N. Jackai (* Entomologist, A VRDC, rieties with r esistant seed or pod wall (Table 3.20). Taiwan) Apparently varieties resistant to seed damage were susceptible to pod-wall damage and vice versa . An Screening New Breeding Lines attempt will be made to combine both characters into resistant Iines.- S.R. Singh and M. Owusu-Akyaw Flower thrips. We screened 607 progeny lines (F 3) from the breeding program for resistance to thrips Beanfties. Agromyzid bean flies are a major pest of during the second season of 1983 through the second legume crops in t ropical and subtropical countries. season of 1984_ Progeny lines were planted in 5-m Total crop losses have been reported in parts of Asia, single rows, and test lines were visually rated for Australia, and Africa. Beanfly larvae feed inside the thrips damage by three evaluators using a 1 to 5 scale. stem, damage the phloem, and finally kill the plant, At the end of each season, individual plants in especia lly at the seedling stage. Its presence is not selected progeny rows were ha rvested and co nsti­ noticed until plants start to die. So host·plant re­ tuted the bulk for the next season's screening. sistance is important in an integrated management Thrips populations are higher and cow peas suffer 80 Grain Legume Improuement Program Table 3.19. Cowpea yields related to pre-set pod-borer more damage the second season than the first. Entries infestations, 1984 selected in the F. generation are scheduled to go into Pre.set Marketable grain yield. kg/ha replicated trials. infestation , 0/0 First season Second season In another trial advanced breeding hnes and germ· D .. . . ~1 1.601 plasm accessions that had gone through preliminary 20 . . . . . . . . . . . . . . . . . . .. 74D 1610 scr eenin g were evaluated for thrips resistance in a 4D ............ .. ......... 717 1,377 series of ra ndomized. complete block designs the 60 .... .. ... .... .... .... ...... 684 1.327 second season of 1983 and the first season of 1984_ No 60 ............. ......... . 608 1.116 insecticide was sprayed and the entries wer e natur­ 100............. . .. . . ...... ... 573 921 ally infested. with spreader rows planted to build up Mean . . 697 1,359 the infestation . TVx 3236 and VITA 7 were included LSO, 5% .. . ... ...... . . .... . . . . 79 156 as resistant and susceptible controls. Numbers of thrips per r aceme and per flower were recorded for each entry. Visual ratings were at 30 and 42 DAP. Table 3.20. Susceptibility of seeds and pods of Resistance levels of breeding lines IT 82D·713. IT indicated cowpea Jines to C. maculalus 82D-716. and IT82D·952 and germplasm accession TVu 4544 were comparable to that of r esistant control Eggs per unit TVx 3236. Except for IT 82D·851, however , no entry no. of seeds Index of Pod wan or pods differed significantly from the susceptible control but susceptibility : hardness ~ germplasm accessions TVu 7376 and TVu 8154 a lso Cowpea Seeds Pods Seeds Pods x 10' Kim' showed some resistance. They are semi·prostrate. IT820·4~·7 .. . 1.6 11.1 0.D48 0.020 17.1 Fie ld examinations indicated that their growth habit IT 820-4~· 1 .. . 1.7 7.2 0.D45 0,038 11.8 and raceme spacing causes nonuniform infestation , IT 820·453·2 .. . 1. 7 7.1 0.045 0.035 11.4 IT 820 ·486-4 .. . 1.7 7.3 0.045 0.038 16.6 so a few racemes escape infestation. IT 820-il6 .. 1.6 7.9 0.045 0_040 14.9 During the three seasons, we selected from 4 to 9% TVu 2027 .. ... . 2.6 14 .8 0.036 0.038 15.3 of 1,283 progeny lines for further study.- S.R. Singh VITA-7 ....... . 2.0 16.5 0.070 28.6 and A.B. Salifu IT 82D·516-3 .. . 1.9 10.2 0.066 21.8 IT 820·524-3 .. . 2.4 6.2 0.065 22.3 Screening Techniques IT 82D·504-1 .. . 2.2 15.9 0.053 21.5 IT 82D·504-4 .. . 2.1 15.8 0.053 22.3 Thrips. In screening for resistance to thrips under IT 820·542-2 . . . 2.2 8.5 0.050 24.4 field conditions. escapes are a lways possible. Additionally, r esistance must be measured at an S.E.. . . .. . 0.002 0.003 0.7 appropriate time to ensure detecting maximum differ· LSD. 5% .. 0.007 0.001 1.9 C.V. . % .. 8.841 29.933 12_1 ences. Also positions of eitherresistant or susceptible test lines influence their resistance ratings. We . .. F , adults emerging IIndex ofsusceptlbJIlty = log % . . The initiated a trial in 1984 to elucidate how such factors Developmental penod affect the expression of resistance. higher the index, the more susceptible the variety. The test included two r esistant and two susceptible lForce per un it area in Ne,..tons/m~ requi red to penetra te the pod wall. genotypes to a llow comparisons of both position Table 3.21. Evaluation onITA cowpea germplasm Table 3.22. Damage (mean rating) ofindividual lines for agromyzid beanfly resistance cowpea genotypes relative to their paired BF/damaged BF!IO Plant combinations. Cowpea plants, plants mortali ty Genotypes paired with lines % (L + ., P) /0 Combination F-898 .. . . . . . . . . . . . . . . 20 0 0.0 83 F-824-4 . . -- .... . . 30 1 6.1 Controls I TVuR-1554 .... . . ...... 40 1 0.0 RJR 1 ••••• • • 1.64 (0.13)' 83 F-814-1. ..... 40 0 2.2 R, R, . .... . . 1.86 (0.21) TVu 3858 . ..... . . 0··.0 . 0·0 0. 40 1 3.6 S,S, . .... . . 4.29 (0.27) F-895 ......... ..... .... .... . 40 1 7.5 S2 S, .... _. . 3.07 (0.31) 83 F-668-2 .. . ..... . .. .. ... .. 40 1 7.7 IT 82D· 702 . ... . ... .. .. Mixed pairs I 50 1 0.0 TVu 156' R,R , 1.93(0.20) 2.21 (0.21) . ..... 0· ·.··· 100 18 22_5 TVu 2328' . .. . . . .... . . - 100 17 27.7 RIS 1 • • • 2.00(0.18) 4.21 (0.33) 83 r·723-1' ............. 100 21 33.3 R, S, 2.00 (0.21) 3.93 (0.29) Virginia Cowpea R I ... .. 0··· ··· 100 17 37.9 2 S 1 2.29 (0.19) IT 82D-952' . . . . . . . . R, S, 2.29 (0.22) 3.93 (0.29) . 100 20 44.7 S,S, 4.14 (0.31) 3.50 (0.33) Mean . .. .. . . . 0. . . . . 90 7 16.4 IR J = TVu 1509; R2 = TVx 3236; 8 1 = V1TA 7; S :l = TVu 76. 'Susceptible con trols . :lNumbers in parentheses are standard errors. Grain Legume Improvement Program 81 effects: A resistant (R) row planted beside a suscep­ Humber, Boyce Thompson Institute.- S.R. Singh, tible (S) row or two resistant and two susceptible K.M. McGin nis, and M .K. &ars, entomology pro· rows. and t.ime of rating. fessor, University of Guelph, Canada We paired the genotypes with each other in a complete factor ia l combination planted in that pat· Insecticide Screening tern. Data were collected at 30, 35, 37, 39, 42, 44, and 46 DAP. Genotypes were rated on a scale of 1 to 5, Cowpea insect pests. Ten insectiddes were tested representing slight to heavy damage. for efficacy as fo liar sprays on t he major cowpea pests Table 3.22 presents relative damage to test geno· in Ibadan . Nigeria, during the second season of 1984. types in different combinations. A significant de. Some were tested at more than one dosage. This was crease in resistance of R genotypes was observed m the second year of testing for most of the chemicals, RS combinations. Only the susceptibility of S, signi. some having been tested three years. Two, Sherpa ficantly increased when grown beside a resistant Super and Cymbush Super ED, were new to the test genotype. Higher thrips populations.on S In relatIOn this yea r. The latter was included because cowpea to R in RS combinations mav have tnggered denslty. growers in severa] African countries are interested in dependent movements fro~ S to R genotypes. Or it and Electrodvn spraying techn ique. factors other than crowding may have been reo As in 1983, th~ cowpeas wer e planted in randomized sponsible, fo r example, fast deterioration of the block design with four replications . Plot sizes were initially favored S genotypes. " six rows each 5 m long, and TVx 3236 was used. The RS interact ions observed a re consldered 1m· TW1:!nty flower bud clusters (racemes) and 20 flowers portant factors fordecisions on resistance to thrips. A were sampled per plot at fou r times at weekly in· screening trial's precision depends largely on reo tervals starting at flower· bud formation (for racemes sponses of the resistant and susceptible cont rols. The only), and examined for thrips and Maruca pod borers preliminary results presented here indicate that we (MPBs). At onset of pod forma tion , we took weekly need to pay more attention to positioni ng controls In samples of pod sucking bugs (PSBs) by the drop. cloth resistance screening studies. Further studies are in technique, three such samples per plot each samphng progress.- S.R. Singh and A.B. Sali{u date. At pod maturation, 50 pods were randomly sampled for seed-damage esti mates. Some of the results are shown in Tables 3.24 and 3.25. Screening Methodology Control of thrips was generally good (Table 3.25). Aphids. The current screening method at UTA to Generally counts of two or fewer thrips per fl ower or identify aphid·resistan t cowpea hnes ldentlfies reo raceme ar e considered economically acceptable. sistant lines but not the level of resistance. Current Most treatments were effective on the MPB but with experiments are to determine differences within and significant differences , which is not surprising ~s between resistant and susceptible lines. most of the formulations contain Cymbush or Decls, We used the following cowpea lines : TVu 36, TVu bot h known to be effecti ve on MPB. The organophos. 300 TVu 801 and TVu 2896 as aphid resistant, and phates, however, were less e ff'ecti ve in uncombined Vl'fA.7 as su'sceptible. They were planted in 5·inch formulations. Except for Hostaquick , each formu. pots in the screen house and thinned to one s~edling lation gave excellent control of PSBs, with Cymbush per pot. Seedlings were used when the first tnfohate Super ED giving the best resu lt. leaf was full y expanded, 10 to 14 days after em­ Grain yields resu lting from the first 13 treatments ergence . One fourth·instar nymph was confined to the did not differ statistica lly but were highest with center trifoliate leaf with a 'clip·cage' consisting of endosulfan + deltamethrin at 500 + 10 g a.i. /ha , plexiglass rings mounted to a hairclip. The nymph was allowed to molt to an adult, then nymphs pro· duced in 24 hours were left on the leaf and the adult was removed. When they r eached the 4th instar, they Table 3.23. Fecundity of Aphis craccit.'ora Koch (F I were confined individually to a fresh leaf and the females) reared on five cowpea accessions number of nymphs each adult produced was counted . under screenhouse conditions; temp = 23 Table 3.23 shows the percentage of aphids surviv· to 33°e, RH = *95% ing to t he adult stage and the six-day fecundity of F I ~/o of nymphs reaching Nymphsl females. No nymphs survived to the adult stage on adulthood and day Resistance TVu 801 but in subsequent screen house tests a few Accession producing nymphs (x • se) level aphids survived on this line. AU nymphs reached the TVu 801. . .. . .... . .. . . . .. 0 o HR adult stage on VlTA·7. TVu 300 .. .. . . . . . . .. . 22.0 1.17+0.17 R Whi le rearing aphids under screen house con­ TVu 36.. . .. . ... .... .. . .. 78.5 1.56 =0.11 R TVu 2896 . . . .. . . . . .. .. . . . 53.3 ditions we found an entomophagous fungus that 1.23 ± 0.04 R effecti~ely VITA· 7. . . . . . . . . . . .. . . . .. tOO.O 3.33+ 0.19 killed aphids in all instars , especially high S densities of aphids under high humidity. The fungus HR = highly resistant ; R = resistant; S = susceptible. was identified as Neozygil£s {resenii (NowakowskI) I female lived only 5 days. Number of nymphs!day was ca lculated over 6 days for Remaudier e & Kellet (Entomophthoraceae) by R.A. 5/14 females. 82 Grain Legume Improvement Program Table 3. 24. Counts o(thrips in racemes and flowers of The cost of sampling was calculated as man-hours cowpea TVx 3236 under indicated required to collect and process a sample under each insecticide treatments, UTA, second method. Sampling precision was determined on the season, 1984 basis of the relative net precision statistic (RNP), Number of calculated as RNP = [l/(Cs) (RV)] 100. where Cs = insects per cost in man-hours of sampling and processing one Dosage 20 20 sampl e. and RV is the mean relative variation. The Insecticide g a.i.jha racemes flowers larger the RNP, the greater the precision for the effort expended. The degree that relative density estimates 'NURELLE-D' Cypermethrin + r efl ected actual population trends was determined by chlorpyrifos ... . . . . ... . . . . 50 +500 0.6 • 83.2 g correlation and regression analysis. SHERPA PLUS ' All selection criteria indicated that shakes and Cypermethrin + dimethoatc 30+250 0_8. 00.7 g water traps were the most desirable for intensive SHERPA PLUS ' research (Tables 3.27 & 3.28). They retur ned the Cypermetbrin + dimethoate 30+400 1.1. 79.0 g h igher precision for unit-effort expended (see RNP SHERPA PLUS" values), but plant shakes were preferable to water Cypermethrin + phosalone . 30+500 1.5. 70_1 g traps as the latter did not correlate satisfactorily with Oftanol •. . _. _. .... . . . .. . . . 500 2.6 d. 52.3g absolute populations. and are not easily ca librated_ Tamaron M- • • .. . . , .. . ... ... 500 2.7 d. 97.2 fg DECIS K. + ROGOR " Table 3.25. Counts of ,Maruca larvae in Rowers of Deltamethrin + dirnethoate 12.5+400 2.8 d. 163.3 ef cowpea TV:I. 3236 under indicated THIODAN' + DECIS R insecticide treatments, IlTA , second Endosulfan + deltamethrin . 700+10 2.9 de 163.3 ef season, 1984. NURELLE·D' Dosage Number of Cypermethrin + applied. insects per chlorphrifos .. . . ... .. .. 25+250 3.1 de 166.1 de Insecticide g a.i ./ha 20 flowers CYMBUSH SUPER ED' Cypermethrin + dimethoate 15+20 5.2 cd. 102.5 efg ·CYBOLT·'- Flucthrinate . ..... . .... . . . .. . . .. 60 0.3g CYBOLT ' Flucythrinate ..... _. . .... . 60 8.3 cd 410.7 d NURELLE-D > Cypermethrin + chlorpyrifos .. . 50+500 O.Ug DECIS ~ + ROGOR ~ Deltamethrin + dimethoate 6.2+200 11.4 bc 259.0 cd SHERPA SUPER' THIODAN" Cypermethrin + phosalone . . . . . 30+500 0.gefg Endosulfan + deltamethrin. 500+10 11.7 c 257.4 cd THIODAN" + DECIS" CYBOLT' Endosulfan + deltamethrin . . . .... 700+10 1.0 fg F1ucythrinate. . . . . . . . . . . . . 50 14.0 bc 342.8 be THIODAN" + DECJS~ Host.quick ' . .. . .. . . . . _. •. 550 28.1 a 586.0 s Endosulfan + deltamethrin .... _. 500 +10 1.0 ef H,O, Control. . _. . . . . . . . . . . 2,501 23.4 ab 267.1 c CYBOLP' Fl ucythrina te . . . ... . . .... . ..... 50 1.2 erg SD. . . . . . . .. . . . . . . .. . .. . . . . . . .. . . .. 6.4 77.4 NURELLE-D " Figures followed by the same letter do not durer s ignifiCAntly Cypermethrin + chlorpyrifos ... . . 25 + 250 L2 efg (0.05 DMRT) SHERPA PLUS' Sprayed 35. 45. SO. & 55 days after planting on 10.9-84. Cypermethrin + dimethoate . . . . . . 30+400 L9d. -Trade names used only for clarity ; no endorsement of product is intended. SHERPA PLUS ' Cyperrnethrin + dimethoate ...... 30+250 2.2cd followed by cypermethrin + dimethoate at 30 + 250 g CYMBUSH SUPER ED · Cypermethrin + dimethoate .... . . 15+20 2Ad a.i./ha (Table 3.26). Those formulations also gave good r esults in 1982 and 1983 trials. The n ew DECIS ' + ROGOR' Deltamethrin + dimethoate . . . .. . 12.5+400 3.3 bcd Electrodyn formulation gave comparable r esults. Oftanol !!" . ..... . ... ..... .. . . . . . . 500 3.9 he The focus in 1985 will be on how t hese products affect natural enemies of cowpea pests, and efficacy Hostaquick,!5 . .. . . .... . . . ... . . . .. 550 4.5ab with longer intervals between applications .- L.E.N. Tamaron Ji: . . ... . .. . . . . . . . . . . . .. . 600 6.1 a Jackai and S.R. S ingh DECIS" + TOHOTt> Deltamethrin + dimethoate 6_2+200 6.2a Sampling techniques for flower thrips. We H20, Control . ..... . . . . .. __ . 2,501 4.2 ab continued to evaluate sampling t echniques for flower SD ..... . .. __ . .... . . . . . .. _. . . . . . . . . . . . 1.3 thrips. Previously sampling efficiency was deter­ mined by r elative variation (RV). This year we Figures followed by the same leLter do not di ffer sign ificantly expanded selection criteria to include sampling cost, (0.05 DMRT). Planted 10 September 1984, s prayed 35, 40. 45. & 55 OAP. sampling precision , and absolute-relative density *Trade names used only for clarity ; no endorsement of product is relationships. intended. Grain Legume Improvement Program 83 Table 3.26. Flower and pod production and yield (kgfha) of cowpea (var. TVx 3236) under different insecticide treatments, UTA, second season, 1984 Dosage Flowers Pods applied, per per Yield, Insecticide g a.i./ha plant plant kg/ha *THIODAN ' + DECIS ' Endosulfan + deltamethrin . .. .. . . . . ... . - .. ...... 500+10 3.8 ab 7.7 ab 1.758.5 a SHERPA PLUS' Cyperrnethrin + dimethoate .. ... ......... 30+250 4.1 ab 4.8 bc 1,681.8 ab SHERPASUPER~ Cypermethrin + phosalone ...... . .. . . . .. .. ...... 0+500 5.2 ab 9.6ab 1,628.0 abc SHERPA PLUS ' Cyperrnethrin + dimethoate ····· · · . · · · .· · . 0' ·· . . 30+400 7.2 a l1.4 e 1,563.3 abc CYBOLT" Fl ucythrina te . ......... . ..... . ........ ....... . 60 4.6 ab 9.4 ab 1,560.3 be NURELLE-D ~ Cypermethrin + chlorpyrifos . .. .... ... 25+250 5.8 ab 13.2 a 1,551.0 abc THIODAN" + DECIS ' Endosulfan + deltamethrin .. .. . ... . . . 700+10 4.9ab 8.1 ab 1,490.0 abc DECIS ' + ROGOR" Deltamethrin + dimethoate . ..... . .. .. .... , . .. ... 12.5+400 5.3ab 5.1 bc 1,436.3 abc NURELLE-D" Cypermethrin + ehlorpyrifos . .. . ..... . . . _. .. .. ... 50+500 3.6ab 4.9bc 1,425.3 abc CYMBUSH SUPER ED" Cypermethrin + dimethoate 0 •• • • ••• , • • • • • • • , • • • • 15+20 5.5ab 7.3ab 1,384.0 abc CYBOLT" Flucythrinate .................... .. .. . .. . ... .. . 50 4.2 ab 7.6ab 1,290.5 abc DECIS~ + ROGOR " Deltamethrin + dimethoate .... .. , . ..... . .. .. 6.2 +200 4.3ab 7.2 ab 1.217.8 abc Oftanol " ....... . .. . .. . .... .. . .. . ...... .. ... .. .. 500 3.7 ab 6.7bc 1.207.0 abc Tamaron !C . ... . , .. ...... . . ..... ...... . .. , . . . . .. 600 4.9ab 4.8 ab 1,134.5 be Hostaquick 'J!.' . , . . . . , .. ... .. , .. .. . . . . .. . . . , . . , . , . 550 4.1 ab 5.3 ab 1.098.5 c H20, Control . .. . . .. . ... _. .............. .... . . . . .. ... 2.0b 1.4 c 410.8 d SD ... ..... .............. . .... . .. ............................ 1.7 2.8 335.9 Figures follow ed by the same letter do not differ significantly at (0.05 DMRT). Planted 10 September. sprayed at 35. 45. 50. & 55 OAP. *Trade names used only for clarity ; no endorsement of product is intended. Hence, water traps may have little utility except for To improve our understanding of the population general surveys and populat ion monitoring. The cost behavior of this pest complex, we continued monitor­ and ease of construction make water traps the easiest ing population fluctuations and assessing their im­ form of sampling technology. Plant shakes can be pact on cowpea yields. Cowpea TVx 3236, was planted useful on the breeder's plots where rapid population in unreplicated 625-m 2 plots. One plot was completely assessment is often desirable. Several adhesive­ protected with weekly sprays of Decis (12.5 g a.i./ha) coated plywood boards may be easily carried to the and monocrotophos (400 g a.i.jha). The other recei ved field for successive samples. And the alcohol method only two sprays of decis. one at flower-bud initiation is still acceptable for research.- S.R. Singh and A.B. and one 10 days later, to control thrips and pod borers. Salifu The experiment was conducted at the I1TA research farm both cropping seasons and at Mokwa during the Population Studies single cropping season. Populations of PSBs were sampled each third or fourth day by drop-cloth (D-C) Pod sucking bugs continue to be a major problem on at predetermined sampling points, and visually (V-S) cowpeas in Africa's tropical savanna and forest on five rows of 25 m each. Counts were expressed as zones. A complex of about nine common species bugs per 2-m row (area sampled by DC method) (Fig. usually includes three or four major species. The 3.4 & 3.5). When the crop matured, yields were highest population has been reported from Mokwa, in estimated on four groups of 25-m, five rows each. the Guinea savanna zone of Nigeria where tempera· As in the past, populations at Mokwa were much tures generally exceed 30°C during the August to higher than at I1TA. But populations at both lo­ November cropping season. cations were much lower this year than in previous 84 Grain Legume Improvement Program Table 3.27. Cost (Cs), mean relative variation RV, and Mean number of bugs per 2.m row l'elative net precision (RNP) of sampling 20r-------------------------------~ methods for thrips on cowpeas Spray Sampling method Cs ' RV RNP Absolute C and B' . . . . . ....... ...... . 0.20 9.73 51 16 j Relative methods:! Unp 120 days). Maturity is the day after planting that of two breeders , an entomologist, a physiologist, a 95% of the pods have changed color and few leaves microbiologist, and a pathologist at!ITA, a breeder in remain. Days to maturity may vary in different northern Nigeria, and regional coordinators in the environments. Philippines and Brazil. We shall station regional The 1984 trials are not complete; some in the coordinators in Tanzania and Zimbabwe in 1985. southern hemisphere will not be harvested until During 1984, the soybean group increased varietal April 1985. Even though 1983 was dry in most of West testing in Nigeria to include sites in Ibadan in the Africa, the soybean lines performed well (Table 3.59), rain forest transition zone; Ilorin and Mokwa, both in with lower yields of late-maturing lines reflecting the Guinea savanna; Zaria, in the northern Guinea effects of rains ceasing early. Experimental lines in savanna; Kadawa, in the Sudan savanna; BukunI, at both trials maintained seed viability during eight 1400-m elevation; and Samaru·Kataf in the soybean­ months' storage. growing area of Nigeria (Fig. 3.7). Adding the new In 1985, we expect to offer cooperators for the first 106 Grain Legume Improvement Program yond the t r aditional soybean areas in Nigeria. For the first t ime farmers grew improved soybean va· r ieties in Oyo, Kwara, and Bauchi states. They were able to sell their products at the local ma rk ets. Eight farmer s who tried the earlier line TGx 536-02D in • ZARIA southern Kaduna state, a traditional soybean grow­ ( "or.r..m &._ 5ovcn"G I ing area, reported an average yield of 1.8 t/ha com­ • ..IOS (1400 m.'Iwo' 1OO I • N\)f('tfWI ZON.I(WI. pared with the O.6-t/ha average expected from the 5awrFMt" a.",1'1«1 s.....~ ...... ) traditional Malayan , 'ariety. z • Uniform Yield Trials • The uniform yield trial comprised lines previously tested at multiple locat ions a nd screened for profuse (without i nocu lation) nodulati on, improved seed longevity, disease resistance, and agronomic charac­ ters. The best lines will be included in 1985 in· ternational t rials. The medium maturing and the late Figure 3.7. IITA conducted multilocat ional soybean va· maturing, uniform trials were grown at six and three r iety t ria ls at the Nigerian sites shown and vs rietal tests in sites, respectively, in 1984. All yield tria ls received addition at Bukuru and Samaru-Kataf. 150 kg/ha 15·15-15 and 200 kg/ha of 0-18·0. The entries were sown in four replications in six-row plots 6 m time a trial of earl ier maturing lines « 105 days) as long with 75 em between rows and a n optimum plant well as trials of medium and late maturing lines. We population of 270,000 plants/ha. Other trials reported anticipate requests for more than 400 trials. in this section had the same row length, spacing, and populations but varied in number of rows . Mean Nigerian National Yield Trials yield, disease, and agronomic performances of some The !ITA soybean program continued to cooperate in promising lines in uniform trials (Ta ble 3.61) indicate 1984 with researchers from the Institute of progress in resistance to bacterial pustule Agricultural Resear ch and Training, the Institute far (Xanthomonas campestris), shattering, and lodging in Agricultural Research, and the National Cerea ls lines with good yield, height, and a range of Research Institu te in planning and growing trials at maturities . 14 sites in Nigeria. Some results from the trials ar e shown in Table 3.60. Advanced Yield Trials The development of improved earlier maturing Lines included in the advanced yield t rials of early (8 lines has aroused interest in growin g soybeans be· sites) medium (two trials at 4 s ites), and late (3 sites) Sites of 1984 Internotional Soybean Trials Figure 3.8. Cooperators in the 44 countries indicated used 199 international80ybean trials in 1984. Grain Legume Improvement Program 107 Table 3.59. Results from the 1983 international (except for one trial of late·maturing lines tested at soybean trials at sites in West Africa two si tes). Plots were two rows and there were three Yield Mean Emer· replications per location. Outstanding experimental range,l yield. gence, lines from these trials (Talbe 3.64) were superior to Entry kg/h. kg/ha Of " the controls in yield. disease, and agronomic ratings. Lines in the second group of PYTs (11 trials) Medi.um maturity TGx 297-l92C . ....... . .. . 508-3125 originated from a single-plant selection made in 1983 1,637 79 TGx 342-37 5D . .... . ....... . .538-3264 1,572 59 and sown as progeny rows in the 1983/1984 dry season. TGx 536-100C. .. ....... .. 803-2847 1.555 73 We screened the lines for bacterial pustule. lodging. Bossier . ............... . . . 661-1917 1,197 41 shattering, and seed longevity in the dry season. and selected 280 lines for trials. The most promising ones Grand mean .... . . .. . .. . . . 1.383 74 Standard error , ......... , are listed in Table 3.65. Lines in this series of trials 380 11 C.V. , O;~ .. . .. . , ...... • • . .. 26 16 including the most recently produced adva nced breeding material, had less pustule and shattering Late maturity and lodging and included good agronomic lines that TGx i09-06D ... .... .. .. . . . 553·2569 1,321 65 mature in fewer than 105 days. After the methodology TGx 330-03E . ...... ..... . . . 506-2361 1,294 75 TGx 30S-036C ... . .. . ...... . for incorporating profuse nodulation and seed long­ 472·2500 1,200 86 lAC 8. control .... . ..... . . . 32 eVlty had been developed and merged into agro· nomically acceptable lines, we then emphasized Grand mean ... .. .. .. .... . .. .. . . .... . 1.122 77 incorporating better agronomic characters and dis· Standard error , ...................... . 456 5 ease resista nce and put more emphasis on breeding C.V .• % . ....... . .... . .. . . .. ......... . 41 7 early maturing lines. Lines in these trials are the first lRange and means from seven locations for medium maturity and from this effort. from four locations for the late maturity trials. 2After 8 months ambient storage at fou~ locations for medium maturity and at two locations for late maturity trials. Segregating Generations were tested in replicated yield trials in 1983 and evaluated for nodulation and improved seed long. During 1984 we made about 150 crosses. Each cross evity. The early trials included lines that matured in had the potential of uniting profuse nodulation. seed as few as 95 days and gave adequate yield when longevity, and resistance to bacterial pustule shat· planted 12 July at Zaria. But no entry was outstand· tering. and lodging . Parents included high.pe~form. ing for all desirable traits. Further progress in ance lines developed at [ITA and lines from Latin selecting early maturing lines is reported in another America, Asia. and diverse African environments . section. We make crosses throughout the year and advance Medium maturing. advanced trials included 63 from F 2 at rbadan through a modified single·pod short and tall entries in two trials. Some lines were descent (SPD) method where one pod is harvested superior to the high·yielding control (Samsoy I). had from each plant and populations are maintained good resistance to bacterial pustule. did not shatter. without selecting for environmental adaptation until and remained erect (Table 3.62). Data are not in· the F4 or F . generation. In the last generat ion we c1uded from Zaria as some lines failed because of soil made three or more SPD selections to produce nearly crusting and a dry spell immediately after sowing. identical populations of each cross. One set of each Late·maturing lines are maintained in the soybean population is grown at Mokwa in the southern breeding program because their maturities match Guinea savanna, one at Zaria in the northern Guinea those of traditional varieties grown in Nigeria and savanna. and one population is stored for later dis· some other countries and will fit directly into tradi· patch to cooperators if it is found to have a high tional farming systems. Some experimental lines proportion of desirable plants. Chosen populations. whose mean yields were double that of the traditional in volving parents with special adaptations, are plan· variety Malayan had resistance to bacterial pustule. ted during the rainy season at Kadawa in the Sudan remained more erect, and in most cases shattered less sa vanna and near Jos at l,4()()..m elevation. (Table 3.63). Experience from the past three years in One reason to maintain populations by SPD until Nigeria and from recent international trials in· they are fixed is that plants selected in West African dicates , however. that medium-maturity lines have lowlands seldom are adapted to high elevations or higher yield potential and more promise for new latitudes exceeding 15°. We think we serve research­ farming systems. ers better in those environments by providing un· selected, fixed populations from which they can select Preliminary Yield Trials adapted plants, grow progeny rows, and 'soon begin replicated testing. The regional coordinator's Preliminary yield trials (PYT) were divided into two (Philippines, Brazil. Zimbabwe. Tanzania) role will groups. Seven trials, grouped according to maturity, be important to each region. as he selects lines comprised entries previously tested in replicated appropriate to that region. The coordinators will also yield trials and re--evaluated in 1984 at three sites recommend thatiiTA make specific crosses that have 108 Grain Legume Improvement Program Table 3.60. Performance of soybean lines in Nigeria national yield trials, 1984 Mean yield, kgjha' Reaction 3 to Entry Forest Savanna P lateau Nodulation 2 Pustule Shattering Lodging Medium maturity 'fGx 297· 192C ........... . • . . , .. . 3,672 1,660 1,342 2.3 MR Heavy Slight TGx 297·35C ....... .... . .. . ... ....... .. 3,435 1,805 670 2.1 MR Moderate Slight 'fGx 536·02D ............... ... . .. . .. .. 3,081 1,866 1,616 2.4 MR Moderate Slight Samsoy 1. Cont rol. ......•...•. . . .. .. ... 3,370 1.900 1,188 3.9 MR Moderate S light Samsoy2 . ... . . 3,042 1,739 1.084 3.3 S Kone None Grand mean ... . 3,104 1,711 1,168 2.8 Standard error 588 296 265 0.6 C.V.,% ... .. . . . 19 17 23 22.1 Late maturity TGx 306·036C . . ... ... ...... . .. . 3,052 2,019 3.2 MR Moderate Slight TGx 306-036D. . .... .. . 2,816 1.684 3.0 MR Moderate Slight TGx 442·012D . . ... . .. . 2,625 1.489 1.8 MR Moderate Slight TGx 713·011D ......... •...... .• ... 2,281 1,667 2.9 S Moderate None Malayan, Contro l . . . . . 2,573 1.387 3.0 S Heavy Heavy Grand mean ...... . .. . . ...... . 2,654 1,570 2.9 Standard error .. . ........ . .. .. . 375 315 0.7 C.V .. % ......... .. . . . ............... . . 14 20 23.2 I Mean of four locations in the forest zone, six loca tions in the Guinea savanna zone, a nd two loca tions on the J os Plateau. ~ 1 = No nodu les seen : 3 = Nodulation comparable to inoculation. :lR = Resistant ; MR = mooerately resistant ; S = susceptibl e. Table 3.61. Mean yield, nodulation, and some disease and agronomic data from the 1984 soybean uniform trials. Yield, Nodu lation, Height, ~.turity Entry kg/ha 1-5 ' Pustule' Shattering Lodging em days Medium maturity TGx 814·23D .. . . . . . . . . . . . . . 1,983 3.4 R Slight Moderate 78 121 TGx 814·26D ..... .. ...... . 1,946 3.5 MR None None 67 118 TGx 539-5E ... .. . .... . . ...... ... . 1,940 2.9 MR Slight None 83 118 TGx 536-02D . . .. . .. ...... .... . ... 1.907 2.5 MR Slight Moderate 71 114 Samsoy 1, Control ...... . ... . ... . .. .... 1,893 2.9 S Moderate Slight 67 119 Bossier, Control . ..... .. . .. .. .. .... .... 931 2.0 R None None 47 112 Grand mean . ......... .. ...... . ... . .. ~ 1,649 2.5 71 117 Standard error ........... .. . .. . .... . .. 313 0.4 9 3 C.V.,% ...... . ....... ....... . .... .. .. 19 15.2 12 2 Late maturity TGx 814-39D . .. .... . . ............... 2,554 3.2 R Moderate Slight 73 121 TGx 306-036C . ... .. ... .. ... ... . 2,313 2.4 MR Modera te Moderate 89 125 TGx 855·78D .. . ... .. ........ .... 2,220 3.0 r.m Moderate Slight 86 129 TGx 725-0110 . . . .. .. ...... 2,204 2.5 r.m None Slight 75 136 Malayan, Control .... . . . . . . . . . . . . 1,396 3.1 S Moderate Heavy 122 137 Grand mean ........ . . .... .. . .. 1,956 2.7 83 126 Standard error ............ ....... 377 0.4 11 3 C.V.,% .......... . .. . . .. . ..... 19 15.0 13 3 '} = l\o nodules seen; 3 = Nodulation comparable to inoculation . 2 Xanthomonas campestris. R = Resistant : MR = moderately resistant ; S = susceptible. the potential to develop good varieties for their 1984/1985 dry season at lbadan and [kenne. They will regions. be screened for shattering, profuse nodulation, seed In 1984,388F " and42F• populations were grown in longevity, and lodging. Then lines remaining will Mokwa and in Zaria. In addition, 37 chosen F . enter multilocational trials in 1985. populations were grown in Kadawa and Jos. Seed from selection of single plants began as early as 85 Screening Method for Shattering days after planting (DAP) and continued at approxi· A laboratory method is now being used to screen F 5 mately 10000ay intervals until as late as 135 DAP. and F 6 progeny rows for shattering resistance. When Selected plants were sown in progeny rows in the pods become light hrown, 25 are carefully removed Grain Legume Improvement Program 109 Table 3.62. Mean yields, disease and agronomic performance of some promising lines in advanced trials of medium.maturing lines at four locations in Nigeria in 1984 Yield, Kodulation, Height , Maturity, Entry kg/ha 1- 51 Pustule~ Shattering Lodging em days Trial 1 TGx814·36D .... . - . .. . .... . . -. -.. -. .. 2,4iO 3.0 R None ~one 60 118 TGx814·27D ........... .. . . .... .. -.... 2,186 3.4 R None Slight 67 116 TGx855-29D · . . . . . . . . . . . . . . 2,093 2.8 R :Moderate Moderate 73 115 TGx 725-010 · . . . . . . . . . . . . . .... . - . . . ... 1,972 3.1 :MR :Moderate Moderate 67 117 Sarnsoy 1. Control ........ . .... .. .. ... . 1,846 3.0 S Slight Sli ght 66 116 Bossier, Control .. . . . ... . .. . ..... . . ... 843 1.8 R Non e None 50 107 Grand mean .... .... ... . .. . . . . 1,724 2.7 64 ][4 Standard error . 397 0.4 10 3 C.V .. % .. . ...... ....... ...... 23 15.5 15 3 Trial 2 TGx 725-05D ... . - ....... . . . . . . . . . . . . 2,072 2.7 MR :Moderate Moderate i5 116 TGx855-68D · .. .. ...... . ........ 2,085 2.8 MR Moderate Slight i6 116 TGx 832-13D .... .. . . . . . . . . . . . . . . . . . 1,924 3.2 MR None None 66 116 TGx813-6D .. .. . ...... .. . ..... ... 1,806 2.9 MR None Moderate 66 113 Samsoy I, Control . . . . . .. . .. . . ... .. - .. 2,017 3.0 S Moderate Moderate 62 116 Bossier. Control .. . _. ..•. . ....... ... .. 646 2.2 R None None 45 109 Grand mean . .. . .. . ... . .. . . . ............ 1,658 2.7 70 115 Standard error . . .......... . . . . .. . .. . . . . 378 0.4 10 3 C.\' .. C}~ . . ... . .... . ... ...... ...... . ..... 23 15.0 14 3 11 = No nodules seen ; 3 = Nodulation comparable to inocu lation. 'l Xonthomonas campestris . R = Resist.ant: MR = moderately resistant ; S = suscept ible . Table 3.63. Mean yields, disease, and agronomic performance of some promising lin es sown in an advanced yield trial oflate maturing lines at three locations in Nigeria in 1984 Yield, Nodulation , Height, Maturity. Entry kgjha 1_6 1 Pustu]e2 Shattering Lodging em days TGx 855-610 ... .. ........•... 2,193 2.4 MR Moderate Moderate 77 128 TGx 56().3D ......... . ... .... . . 2,166 2.6 R Heavy Slight 81 128 TGx 855-64D ......... . .... . . . 2,016 2.6 MR Moderate Moderate 91 128 TGx854-4D ............. . . . ..... . 1,940 2.6 R Moderate Moderate 87 117 Samsoy 2, Contro l . . .. ....... . ... . 2,010 2.3 S Moderate None 50 117 ~alayan , Control. ..... ... . ... . .. . 1.077 2.7 S Moderate Heavy 127 138 Grand mean . . ..... . .... . . ...•.. • . 1,855 2.3 80 125 Standard error .................. . 399 0.5 10 4 C .V, , ~~, .... . ..... ... . .......... . 22 20.9 12 3 I I = No nodules seen : 3 = Nodules comparable to inocu lation . 2Xanthomonas campestris. R = Resistant ; MR = mooerately resistant ; S = susceptible. ~ Eva l uated at Mokwa Cattle Ranch onl y. from each plot and held at room temperature for 10 to grown have no serious insect pest problems. But the 20 days. All samples collected on the same day are southern green stink bug (SGSB), Nezara viridula, is then placed into an oven set at room temperature and a serious problem in some locations, reducing yields the temperature is increased about l°e dai ly. When up to 60% (UTA Annual Reports, 1982 and 1983). We, 50% or mo re of the pods shatter , the sample is therefore, evaluated a number of lines from our removed and the date recorded. Samples that remain breeding program in Mokwa, Nigeria, in 1984 for in the oven longest without shattering have good resistance to the SGSB. The parent of a few of the shattering resistance.- K.E. Dashiell. W.R. Root, lines had'been identified a s having some resist ance to and L.L. Bello this pest . Planting was 14 June and 28 June in plots of two 6- m rows each. No insecticide was applied , and stink Entomology hugs exceeded 15 per 2·m row length. Host Plant Resistance Field scoring for stink ·bug damage was by t hree scientists. Averages of their three scores are the Stink hugs. Most parts of Africa where soybeans are entries ' r atings (Table 3.66). 110 Grain Legume Improvement Program Table 3.64. Mean yields, disease and agronomic performances of some lines grown in preliminary yield trials in Nigeria in 1984 Yield, Nodulation, Height, Maturity, Entry kg(ha 1-5' Pustule2 Shattering Lodging em days Early maturing TGx 81l-lOD . .... . . .... --. .. . .. . ..... 1,372 2,9 R Moderate Moderate 73 99-108 Medium maturing TGx 855-lOID ' , , , . · -.. - ... 2,012 2.8 MR Moderate Moderate 79 107- 123 TGx814-30D ........ 1,976 3.2 MR None None 70 118-123 TGx559-ID .. - . . . . . . . . . . . · -. - - , .. 1.939 2.9 lvffi Moderate None 63 107- 115 TGx814-53D ....... - .. . . . · -. . - . ..... 1,931 2.6 R Moderate None 69 112- 135 TGx539-3F ... . . . . . . . . . . .. · -. .. . - ... -. 1,926 2.8 MR Slight None 56 107- 112 TGx 814-33D ... . .. ... . .. . -... . . -. 1,879 2.8 lvffi None Slight 65 113- 135 TGx 814-32D . ... .. . .. . .... . . . . . . . - 1,863 2.6 lvffi Slight None 63 112- 135 TGx 539-8E . ... .. . - . -. . ... - . 1,860 2.9 lvffi Slight Slight 59 107- 117 TGx311-4ID . -. . ... -. 1,812 2.9 MR Slight Moderate 75 105- 122 TGx855-59D 1,805 2.7 MR Moderate Moderate 83 107- 124 TGx 302A·4ID ... ......... 1,772 2.B R Moderate Slight 89 109-120 TGx 814-54D ......... 1,732 2.B lvffi Moderate Slight 65 !l3- 134 TGx 297-11F .. ........ . 1,623 2.7 R Slight None 80 !lG-lIB TGx573-2E .. .......... . 1,583 2.5 R Slight None 70 105--10B TGx 23!>-22D . . . , . . ... . ......... . 1,416 2.B R Moderate None 54 99-117 Late maturing TGx 855-063D ... . ............ . 2,449 2.4 R Moderate Moderate 84 125-138 TGx 814-28D .. , . . ..... , .. ... 1,905 3.4 ~ffi None None 50 118-129 TGx 814-38D .. . . . . . . , ... . ... . .. , .. , .... 1,873 3.0 R Slight Slight 72 118-180 Controls Samaoy 1 . ... ............ . .. . ......... . 2.271 3.1 MR Moderate Slight 69 118-120 Jupiter . ..... .. .... ...... ............ . . 2,211 2.0 S None Moderate 101 11G-1I7 Samsoy 2 . . . .... .. . . . .. . .. .. . .. . .... 2.110 3.2 S None None 53 11&--120 Bossier . ............... , .. , . ...... ..... 1,326 2.1 R None None 4B 100-107 Malayan .............................. 1.100 3.1. S Heavy Heavy 121 130-140 11 = No nodules; 3 = nodulation comparable to inoculation . 2R = Resistant: MR = Moderately resistant: S = suscepuble. The lowest scores were from resistant parents. transmission of this disease by B. tabaci, we used 8 h Most of them mature early and, thus, escape heavy as the minimum period for acquisition of the virus. damage, even though the scores represent two plant­ Then transmission occurred within six hours. Even ing dates. Nineteen progeny lines scored hetween 5 wi th fi ve w hi teflies per plant, transmission proved to and 6.5, which is promising under the heavy stink· hug be quite inefficient.- H. W. Rossel infestations. More than half the lines were highly suscepbble with heavy damage along most of the stem Soybean Mild Mottle Disease length but more so in the upper half of the plant where most of the SGSB feeding takes place. 14 June plant­ Cowpea mild mottle virus (CMMV) causes a soybean ings suffer ed more damage than 28 June plantings. disease that is cbaracterized by mild mottle and Only entries with scores of6.5 or lower (sigma = 19 mosaic symptoms. It has been commonly observed at or 18%) will be evaluated in 1985. Several had reason· lIT A as well as in farmers' fields in Nigeria during the ably high yields under the experimental conditions.­ past several years (lIT A Annual Reports, 1979 and L.E.N. Jackai, K.E. Dashiell, and L. Bello 1983). It is difficult to identify the presence ofthe virus solely on plant symptoms because infected plants of African Soybean Dwarf many culti vars recover or do not show symptoms at all even when the virus is still present. And the virus Studies on transmission of African soybean dwarf by often occurs in mixed infections with soybean mosaic its vector, Bemisia tabaci, indicated that the causal virus (SMV) and African soybean dwarf (ASD). SDS agent undergoes a short period of "latency" in the agar-gel diffusion tests to identify the virus in crude vector. This may indicate that the virus is of a 50- sap from soybeans have given inconsistent results, called ucirculative" type that! upon acquisition, although in the electron microscope the characteris­ becomes available for infection only when it re­ tic carla-like particles of CMMV can usually be appears in the saliva of the insect after passing the observed. gu t wall of the intestine. To provide a reliable diagnostic tool, the direct Based on the results of earlier experiments on double-antibody ELISA method was developed and Grain Legume Improvement Program 111 Table 3.65. Mean yields, disease, and agronomic performances of some lines grown in preliminary yield trials in Nigeria in 1984 Yield, ~odulation , Height, Maturity, Entry kg/h. 1-5 1 Pustule2 Shatte ring Lodging ern days Early maturing TGx 573-2090 . ......... .. . .... .. .... 2,406 2.4 R Yloderate ~oderate 65 102- 105 TGx 849-2940 . . . . . . . . . . . - . . . . . . .. ..... . 2,363 2.8 MR Slight Slight 65 105-113 TGx 854·39E ........ .. .... . ... .. . .. 2,149 3.0 R Moderate None 69 100-110 TGx888-49C ..... . .... . .. . .. . . . . . .... 2,137 3.0 R None Moderate 66 98- 100 TGx93O·9C · . , . . . . . . . . . . ...... ... 2,129 2.3 R Modera te Moderate 93 99-101 TGx854·33E . ....... .. . .. . . - ... ... . . 2,083 2.9 R Slight Modera te 78 99-103 TGx 849-2970 . ... , .. .... ...... .. .... 2,004 3.2 MR Moderate Slight 69 105- 109 TGx 848-3130 . . . . . . . . . . . . . . . . . . . . . 1,780 3.1 MR Moderat e Slight 80 104 111 TGx 824·4E · . . . . . . . . . . . . . . . . . . . . . . . . . 1,509 3.4 R Moderate Moderate 64 100-111 Medium maturing TGx 814-39E .. . . . . ... . . . . . . . . . . . . . . . . . . 2,872 3.6 MR >Jone Slight 76 117- 133 TGx 814-36E ..... ....... ..... ... .. ... .. 2,770 3.2 MR None Slight 71 103- 127 TGx814·34E .. ... . .. .. .... .. .... .. ..... 2,690 3.5 MR None None 74 117-135 TGx855·14E ..... . . , ... .. ..... .. ... .. .. 2,677 2.8 R Moderate Moderate 100 111- 127 TGx 849-2470 ... . .. .. . . .... ... , . . . .. 2,684 3.0 R Yloderate Moderate 86 111- 120 TGx 814·33E .. .. . . . . ... ..... .. ... 2,583 3.3 R None Slight 72 118-130 TGx 814·44E .. , .. , .. . . . . , . . . . . . . . 2,553 3.2 MR None None 73 117-134 TGx 814·50E . ...... . . . . . ... ... . . .. . . . 2,552 3.2 MR None Slight 79 119--135 TGx 816-4E · . . . . . . . . . . . . .... ..... . ... 2,515 3. 1 R None Moderate 80 111}-135 TGx 814·49E . . . . . . . . . . .. . . .. .. .. .. . . . . . 2,445 3.0 R None None 70 118- 131 TGx 814·45E .. ....... .. ... .. ... -.. . .... 2,441 3.4 MR None None 66 112- 133 TGx 814·46E .. ... ... .. .. ... ...... .. . - .. 2,441 3.4 MR None None 63 110-133 TGx 849-47D . . .. . .. . . . .. .. . . .... .. . . ... 2,4 16 2.9 MR None Moderate 78 103-119 TGx849-8D .. . ...... .. . . .. . .. . . .. . .. 2, 333 2.6 R Slight Moderate 80 no-1I8 TGx 849-70 ... .... .... . ... . . .. ....... 2,1 38 2.7 R Moderate Moderate 78 111).. 123 TGx 814·43E .. . . -. .. . .. 2,125 3.4 R Moderate Slight 80 99-128 TGx 849-73D . . . . .. ...... . .. . . 2,100 2.6 R Slight Slight 77 110-120 Late maturing TGx 293·63E . .... . .. .. . .. . .. . . . . . . 2,411 2.9 MR Moderate Moderate 82 119--131 TGx 302A·I02E ..... , ... .. . .. . . . .. . 1.964 2.6 MR Moderate Moderate 70 119- 128 Cantrols SamBay I . .. .. ... .... .. . .. .... .. .. ... .. 2,270 2.7 MR Moderate Slight 71 118-120 J upiter ... .. .. .. ..... ... . ..... .. .. ... .. 2,211 2.0 S None Moderate 101 110-117 Samsoy 2 .. . .. ... ... .... .... .... .. .. . .. 2,110 3.2 S None None 53 118-120 Bossier . .. . . .. . . . ..... . .. . .. . .... . . . . .. 1,155 2.2 R None None 51 100-107 Malayan .. . .. ......... . _ . . . .. . .. .. . . .. 1,100 3.1 S Heavy Heavy 121 130-140 I} = No nodu les; 3 = nodulation comparable Lo inocula tion. 2R = Resistant ; MR = moderately resistant ; S = susceptible. tested. An antiserum against CMMV had a t iter of microtiter plates, we added 1 ml of healt hy soybean 1/256 in microprecipitin tests. To remove any anti· juice to every 20 ml of conjugate to cross·absorb any bodies against healthy plant proteins, the antiserum antibodies against healthy plant proteins. was repeatedly absorbed with sap from healthy soy· When an extraction buffer consisting of 0.25 JJ bean leaves until no further precipitate was formed. potassium phosphate (pH 7.5) with 0.1 M EDTA and The gamma globulin fraction from the ahsorbed 0.25% sodium sulfite was used, positive reactions antiserum was purified by ammonium sulfate pre· were obtained from the sap of infected soybean plants cipitation and by filtering through a column ofDE22 up to a dilution of 1/250. ELISA proved to be a reliable cellulose. The best results were given by a gamma method of diagnosing CMMV in soybeans.- G. globulin coating of 10 Ilg protein/ml, with an in· Thottappilly cubation period of about 14 h at 4°C, followed by a second coating of O.5°/~ ovalbumin, with an incu­ New Golden Yellow Mosaic Disease bation period of about 12 h . A second coating of 0.5% In 1984 during our annual survey of soybeans in the ovalbumin prevented the background nonspecific main growing areas of Nigeria, we observed a new reaction that. was caused by healthy preparations. A and prominent disease causing a bright. conspicuous conjugate dilution of 1/250 gave the best results wi t h yellow mosaic. The disease proved to be readily infected soybeans. Before adding t he conjugate t.o the transmissible by sap. In electron microscope s tudies 112 Grain Leg ume improvement Program Table 3.66. Damage scores and grain yields of that, as with CMMV, the carla virus in diseased selected soybean cultivars eva luated for plants is readily transmitted by thi s insect vector and r esistance to the southern green stink causes symptoms in inoculated plants t hat are in­ bug, Mokwa , 1984 distinguishabl e from those we observed in the field Soybean Damage Yield. during our survey. but not like those usually as­ cultivar score 1 kg/ha sociated with CMMV. TGm 1909 .. ..... . . .. . . . .... . . . 3.5 472.8 It thus appears that a new type of CMMV is TGm 1367. .............. 4.0 582.5 involved thatdiEers greatly in symptomatology from TGm 1361. . . . . . . . . . . . . . . 4.0 134.8 t he common type found in soybeans at IITA and TGm 737 P. ...... . . 4.0 383.3 e lsewhere in Nigeria and causes only very mild TGx 536·ll3C ..... .. . 5.0 1,238.9 sy mptoms. In serologica l tests , strongly pos itive TGx 539-5E 5.5 1,718.9 reactions were observed when the new isolate was TGx 713·09D 6.0 1.543.9 TGx 814·40D .. . . ... . . . ... . . 6.0 1,291.7 tested with a CMMV antiserum. In the fi e ld and on TGx 299-7E . . ... .. .. . .. ... .. .. 6.0 577.8 plants grown and inoculated in the greenhouse , TGx 307·048D ... -. .. .. .......... . 6.0 1.040.0 symptoms of the disease were most prominent in the TGx 573·1 E .. . .... . . .. . . . ..... . . 6.0 1.196.1 local improved soybean varieties M·79 and M·216 Ma layan . . . . . . . . . . . . 8.5 433.3 (released as SAMSOY 1 and SAMSOY 2) and in Sarnsoy 1 ..... 8.5 866.7 re lated breeding lines such as M·90 and M·98. AU Jupiter . . . .... .... 8.5 441.1 these breeding lines have a common genetic back­ TGx 293·03D 9.0 222.2 ground , being progenies from a c ross between the TGx 709·01E 9.0 385.6 loca l varie ty " Malayan " and "Clemson Nonshatter· LSD, 50/0 . 2. 1 759.3 ing" whic h was obtained from the U.S.A .- H. W. 1 = 0- 10% damage (no pod absciss ion). 3 = > 10 < 25% damage Rossel an.d C. Thottappilly (s light pod absciss ion ). 5 = > 25 < 50% damage (moderate pod absciss ion). 7 = > 50 < 75% damage (loca lized high pod damage. several pods on grou nd ). 9 = > 75% damage (severe pod Soybean Mosaic Virus abscission. pods on ground ). During routine surveys of lIT A's experimenta l fi e lds at lbadan, samples wer e taken from different legu· of artificia lly inoculated plants , a high concentration minous pl a nts s howing virus- lik e sympto ms. The of particles was observed that strongly resembl e samples were examined in the e lectron mjcroscope those of CMMV, which has carlavirus·lik e part icles and inoculated to cowpea , soybean , and Nicotiana that a re stra ight, rigid , and a pproximately 650 mm benthamiana plants . Cassia occidentalis plants long. consistently s how ed mosa ic symptoms, including Cowpea mi ld mottle virus (CMM V), a common deform ed lea ves. Examination of crude juice pre­ virus of soybeans in Nigeria , does not cause symp­ parations with the e lectron microscope revea led toms like those (Fig. 3.9). Since it results only in very filamentous particles typica l of t he potyvirus group. mild symptoms or none at a ll , we assumed at first that When inoculated to soybeans, sa mples from C. occi­ possibly another unidentified virus (in mixed in· dentalis consistently s how ed sy mptoms that are fection with CMMV) was respons ible for t he disease. typica l of soybean mosaic virus (SMV). In SDS agar· WhiteRy-trans mission s tudies revea led , however , gel diffusion test s with SMV antiserum (iso late " micro·A "), confluent bands formed with the SMV isolates and iso lates from C. occidentalis. In ELISA tests with Hmicro-AI! antiserum, sa mpl es from C. occidentalis reacted with the same intensity as the soybean samples infected with SMV. We concluded that C. occidentalis was infected with SMV. This is the first report in Nigeria of the natural occurrence of SMV in a host other than soybeans.- C. Thottappilly Agronomy Varietal Response to Soil Texture and Fertilizer In previous years, soil t exture and soil fertility were identified as major factors affecting soybea n growth on Guinea savanna soi ls (Annual Report, 1981). The effects of texture on soybean growth may stem from Mn toxicity. which appears most severe on deep, coarse-textured soi ls. Concentration of Mn in upper­ Figure 3.9. Leaves showing symptoms of a new go lden most, fully expanded leaf blades of soybeans at full ye llow mosa ic disease of soybeans. bloom averaged 470 ppm on coarse-textured soil at Grain Legume Improvement Program 113 Yandev, Nigeria, where plants were severely stunted. plots. In Set A, growth in unfertilized plots was On two moderately coarse-textured soils in Mokwa in generally poorer than in fertilized plots; still growth 1984, leaf-blade Mn concentrations in stunted plants was luxuriant in two replicates on unfertilized, fine­ at the pod-filling stage averaged 980 ppm. textured soil. Soybeans that received no fertilizer on In 1983, two fields at Mokwa were sampled exten­ the coarse-textured soil were stunted, probably by si vely for subsurface soil texture and areas were Mn toxicity. Manganese concentrations in upper­ categorized as fine-, medium-, and coarse-textured. most, fully expanded leaf blades of three soybean The coarse-textured areas were in a field that pro­ lines at pod elongation and pod-filling stages are duced stunted soybeans in 1980 and 1981. The presented in Table 3.68. Tissue concentrations ofMn medium- and fine-textured areas were in a pasture were similar at the coarse-textured sites, but plants in where plant growth was excellent. Permanent plots the fertilized plots were not stunted, though less were established in selected areas of each textural luxuriant than those in medium- and fine-textured category. Half the replicates were fertilized. Fine­ sites. and medium-textured sites received 36 kg N, 45 P, and In Set A (Table 3.68) , three IITA breeding lines, 30 kg K/ha as 15-15-15 supplemented with boronated TGx 539-5E, TGx 536-02D, and TGx 297-192C, ranked single superphosphate. On coarse-textured sites the highest in yield. The high yield ofTGx 72o-0UD at the superphosphate rate was reduced to give 40 kg P/ha. coarse-textured, unfertilized site, where plants were Ten kg Mglha, 2.5 kg Znlha, 1 kg Culha, and 0.1 kg badly stunted, was notable. IAC·8 from Brazil was the Mo/ha were applied to fertilized plots as magnesium highest· yielding line on fine- and medium·textured sulfate, zinc sulfate, copper sulfate, and ammonium soils receiving high rates of fertilizer. IAC·8 was at molybdate. Fourteen soybean lines were planted, least somewhat resistant to lodging, which may late· maturing lines 25 June and medium-maturing explain part of its yield advantage under good grow­ lines 4 and 5 July. ing conditions. TGx 306·036C's relatively poor perfor· Yields obtained in 1983 tri a ls are presented in mance in 1984 may be attributed to poor crop estab· Table 3.67, together with soil data differentiating the lishment early in season due to erratic rainfal l. The sites. The 1983 season was characterized by below· local Malayan performed better on coarse·textured average rainfall , drought during the season, and than on fine-textured soil where growth and lodging early cessation of rains. Little evidence of severe were excessi ve. toxicity referred to above was seen in 1983, but plants In the coarse· textured sites in Set B (Table 3.69). grew much more vigorously on the fine·textured sites growth in unfertilized plots was similar to that in than on the coarse ones. Stunting a nd toxicity symp­ fertilized plots. The highest·yielding lines under high toms on the coarse-textured site were less because of fertility were TGx 297-35C and TGx 252-71C on fine­ low rainfall. textured soil and TGx 297-192C and TGx 742-00D on The highest-yielding breeding lines were TGx 306- coarse-textured soil. The highest· yielding lines in 036C and TGx 297-192C and the newly released unfertilized plots were TGx 71S-01 E and TGx 533-65C variety Samsoy 1 from Nigeria (Table 3.67). The line on fine-textured and TGx 744·01E and TGx 742·05D on that performed best in coarse-textured unfertilized coarse-textured soil. TGx 342-356D averaged highest plots was M 351, a sister to Samsoy 1. of all. Yields were higher on fertilized than on un­ The growth of soybeans in unfertilized plots in 1983 fertilized plots, but the relatively low mean yield on indicated high residual fertility despite moderately the fine-textured, fertilized plots reflects early lodg­ low values of available P at the fine- and medium­ ing from excessive vegetative growth. Lines appear­ textured sites (Table 3.67). Before the 1984 growing ing most resistant to lodging were TGx 716-0IE, TGx season, vegetation on unfertilized plots was transfer­ 307-048D , and TGx 713-011D. red to fertilized plots in the same field to reduce One objective for including early-, medium- and nutrient contributions from organic residues in late-maturing lines in the same trials has been to unfertilized plots and to increase contributions to determine if late-maturing lines provide a yield fertilized plots. High fertility plots on fine· and advantage over early-maturing lines under low fer­ medium-textured soils were treated with 36 kg N, 30 tility or other stressful soil conditions. Data from kg K(ha; on coarse-text ured soils, the P rate was 30 1983 and 1984 indicate no such advantage. The kg!ha. Secondary and micronutrients were applied at highest-yielding lines in coarse-textured, unfertilized half the 1983 rate. soils in 1983 were M 351 (harvested after 108 days), Two trials were conducted in 1984. Set A consisted TGx 30S-036C (122 days), and M 90 (107 days); in 1984 of 14 lines including the best performing ones from Set A, they were TGx 725-011D (140 days), TGx 536- the 1983 trial, controls. and the most promising new 02D (99 days) and M 90 (110 days); in Set B, TGx 744- breeding lines. Set B was on coarse- and fine-textured OlE (113 days), TGx 742-05D (l38days) , and M 351 (105 sites established in 1983 for a similar trial. days). On the other hand, late Malayan (138 days), Planting dates were staggered from 14 June to 8 TGx 307-043D (137 days), and TGx 713·0l1D (138 days) July to permit more uniform maturation conditions ranked and even yielded higher at this site on un· for lines of different maturity. Generally rainfall was fertilized than on fertilized, fine-textured soil, while adequate, with some drought stress after planting on TGx 742-02D (138 days) ranked the same on both soils. late-maturing lines and poor initial stands in some It thus appears that some late-maturing lines are 114 Grain Legume Improvement Program Table 3.67. Yields (kg/ha) of soybean varieties grown on soils of indicated texture and fertility. Mokwa, 1983 Site ch aracteristics Fine Fine Medium Medium Coarse Coarse fertilized unfertilized fertilized unfer ti lized ferti lized unfertilized Clay , o/~ ()-15 em 8 8 4 7 5 4 45-60 em 35 42 23 29 20 19 Available P+ (PPM .)' 12 6 17 8 22 17 Soybean Mean TGx 306-036 . . . . .. .. . . . . .. . .. .. . 2.530.' 1.920• 2.8OOa 2.360. 1.940. 2.060. 2.270 Samsoy 1. . .. . . ... • . . . . . .. .. . . . . 2.6100 2.000. 2.33Oa 2.31Oa 2.170. 1.9100 2.220 TGx29H92C . . ...• .. . . . .. . .. . .. 2.370. 1.8808 2.55Oa 2.100a 2.340. 1.850. 2,180 MOO .............. . . .. ..... . .. 2.500. 2,I80s 2,SOOa 1.9208 1,770 1.910. 2,130 TGx 307-047D ......... .... .. .... 2.5500 2,050. 2.56Oa I.BBOa 1,720 1.560 2,050 M35 ± ......... ... ....... . . . ... 2.080 1,8200 2,360a 1.460 1,690 2.080. 1,910 TGx 330-01020 .... . .. .. ... . .. . .. 2.000 1,6500 1,940 1.730 2,060. 1,7806 1.860 TGx573·6C .. . ., ..... . . .... 1,840 1,770a 1,960 1.540 1.660 1.350 1,680 TGx 533-OID . . ....... . . . . . ...... 1.880 1,6008 1,780 1.680 1,690 1.420 1.590 LAC-8 ...... . ....... . ... . . .... 1.660 1.660. 1,550 1.640 1.500 1.520 1.590 TGx 307-0430 . . . . . . . . . . . . . . . . . . . 1.840 1.540 1,910 1.640 1.190 1.480 1.570 Jupiter . . . .. .. ..... .. ...... . .. . 1.850 1.440 1,410 1.190 1.680 1.240 1.470 TGx 304·0590 . .... ... . ........ .. 1.710 1.310 1,610 1.250 1.380 1.180 1.410 Malayan .. . . . . . . . . ... ... . . . .. 1.080 1,070 1,600 1,430 730 910 1.130 Mean .. ,., .... . .. . ....... , .... . 2.020 1.710 2,060 1.720 1.680 1.590 LSD. 5% ., .. .. , .... . . . . .. , .... , 450 580 510 510 500 330 I = Determined by Bray 1 method. 2Yields in same column followed by "s" do not differ fiignificant ly from those of highest yielding lines. Table 3.68. Yields (kgfha) of soyb ean lines and varieties grown on soils of indicated texture and fertility set At Mokwa,1984 Site characteristics Fine Fine Medium Medium Coarse Coar se fertilized unfertilized fertilized unfertilized fertilized unfertilized Clay, ~/~ ()-15 em 8 8 4 7 5 4 4!H'iOem 35 42 23 29 20 19 Avail.ble P+ (ppm.)' 26 8 26 21 29 18 Mn cone (PPM.)' ± 2 370 270 370 340 580 5BO Line or variety Mean TGx 725-0110 . . .. ............ . .. . . NA3 NA NA NA NA 1.8500' TGx539-5E ..... ... .. . ....... . .... 2.819.' l ,500a 2.79Oa 2.030. 2,8408 1.240 2,170 TGx 536-02D . . ............ . ....... 2.360 1,690 2,510 1.760. 2,300. 1,5800 2.030 TGx 297·192C ............ ...... .. . 2. 330 1.8008 2,49Oa 1.690. 2.160. 1,370 1.970 IAC·8 . .... .. . ....... ' .... . .. 2.920. 1,580. 2,820. 1.470 1.8BO 1,130 1.970 MOO ......... .. ................. . 2.400 1,540. 2,300 1.790a 2.100. 1.490. 1.940 S8IDSOy2 ....... . ...... . .. . ....... 2,090 1,480. 2,430 2.040a 2,040 1.320 1.900 TGx 711·01D ......... .. ... . ....... 2.180 1,530. 2.380 1.660. 2,4400 1.140 1.890 S.msoy 1. . .. . ... . . . .. . •.. . .. . . . .. 2.040 1.710. 2.350 1.420 2.3000 1.404. 1.880 TGx 713-090 .. . .. . . ... .. .... . ... .. 261a 1.500. 2.370 1.570. 1.840 990 1.810 M351. .... .. . ....... .. .. .. .. .. ... 1.780 1,430. 2,260 1.550a 2.2908 1.380 1.780 TGx 306-036C . .. . .... ... ..... ..... 2.030 1,490. 2,020 1.700. 1.950 1.180 1,730 TGx 307-047D ... . .. ............... 1.920 1,310 1,990 1.520 1.580 1.000 1.550 M81.yan ........ _. . . .•...... . .... 980 840 1,330 950 1.680 1.4908 1.210 TGx 725-0110 ...... .. . .... . NA NA NA NA NA 1.850.' Mean .. ,., . . . . ... , .. , ...• ' ....... 2.190 1,490 2.310 1.630 2.130 1.330 1.860 LSO,5% .... . . . . .... . ............ 360 430 300 490 570 410 I = Bray 1 method. 2 = Average leaf blade concentration in three varieties at pod elongation and pod fi lling stages. 3NA = Data not avai lable becaui;e seed was stolen . • = Missing plot estimates. based on four of six reps. !>Yields in same column followed by "8" do not differ significantly from those of highest yie lding lines. more suited to soils with poor nutrient status than to large effects on soybean growth. Effects of fertility highly fertile soils. management the first year were not pronounced. but With these trials. we attempt to rationa lize variety they became more clearly evident the second year. testing according to soil factors (nutrient status. soil Residual P in some unfertilized sites remained high texture. a nd nutrient toxicity) recognized as having (Table 3.68) so a testing site off the research station Grain Legume Improvement Program 115 Table a.69. Yields (kgfha) of soybean lines and varieties on soils differing in texture and fertility, Set B, Mokwa, 1984 Site characterlst.ies Fine texture Fine texture Coarse texture Coarse texture fertilized unfer t ilized fertilized unfertilized Clay, % (}-15cm 14 15 10 9 45-60 em 42 42 21 24 Avoiloble P (PPM.) 24 11 22 11 Mean Breeding lines TGx 342-356D . . . .... .. , ....... .. .. .. 2,3900' 1,750a 2,450 • 2.0500 2,160 TGx 716-0IE . . .... . .. . . • •.... . .. . . • . 2,3300 1,760a 2,450. 1.9900 2,130 TGx 744-01E ...... . .... . . ... . ....... 2,2900 1,700a 2,3800 2,1700 2.130 TGx 297-35C .. . .. . . .. .. . . .. • . .•..... 2,500a 1,380 2, l lOa 1,7400 1,930 TGx 742-0 12D' . .. ............ 2,340a 1,070 2,260. 2,030a 1,920 TGx 713-011D . .. . .. . .. . . . . . .. ....... 1,760 1,5608 2,400a 1,840a 1,890 TGx 252-71C . . .... .. .......... . .... . 2,4500 1,560a 1,920 1,610 1.890 TGx 533-65C ....... . .. . .. . • . . .. . .. . . 2,000 1,75Oa 2,110. 1,470 1,830 TGx573-6C . . . . .. . .. . .. .. ... . . ... . .. 1,800 1,540. 2,110. 1,7600 1.800 TGx 742-05D ' ........ ... ............ 1,410 1,130 I ,SOOa 2,1600 1.800 TGx 307-0430' ..... . . .. ... .. . .. . ... . 1,280 1,310 1,920 1,8700 1,590 TGx 442-02D ....... . . .. ... .. . . .. .. .. 1,290 1,280 1,960 1.410 1,480 TGx 304-0590. ..... .. ... .. . . . . _ ..... 1,590 1,300 1,380 1.180 1,360 Controls TGx 306-036C ...... . .. .. .. ........ .. 1,990 1,5200 2.060 1,6900 1,820 Samsa), 1 ....... .. . ...... ... ... .... 2,4100 2,440a MOO . .. .......... . .. .. .. .. ........ M351 ... . .. . .... . •. .. . .. . ...... . . . 2,1l0a Malayan . . .. . .. ... . .... .. ..... , .... 1,600 TGx 297-192C . . ... .• . .• . . . . . . . . . . . . . 2,560a TGx 307-048D . . ......... . . . . . . . . . . . . 1.790 1.6900 Mean .. . . .. .. . ....... . . . , .... ... , . 1.980 1,480 2,190 1,790 LSD. 5% . .. . .. . .. . . .. . .. . ... ..... .. 370 340 490 460 IYieJds in same column followed by " a" do not difl'er signifi cantly from highest yielding ,,'ari ety. ~ln clude8 missing data estimates at fine- textured locat ions. would have been preferable. In future years, it will site and 181 ppm at the fine-textured site, while be possible to adjust management factors to meet in­ cowpeas sampled in a similar manner averaged 385 dividual site requirements by considering P status . and 224 ppm, respectively. secondary and micronutrient status, and M toxicity.­ Good growth of soybeans on t he coarse-textured D.A . Shannon site and relatively low concentrations of Mn con­ trasted with the poor growth observed in parts of this Effects of liming on Mn uptake field in 1980, 1981, and 1984 and with the high Mn Stunting and chlorosis in soybeans, indicating Mn concentrations in 1984 (Table 3.68). The drier con­ toxicity. were observed in a moderately coarse­ ditions of 1983 may have contributed to the textured field at Mokwa in 1980 and 1981. In 1983, a differences.- D.A. Shannon portion of this field was selected to see ifliming would increase yields and decrease Mn uptake. The surface Soybean nutrition soil was loamy sand, with sandy clay loam at 60 t o 75 In 1980 and 1981, trials at five locations in the cm depth. Finely ground, dolomitic limestone was southern Guinea savanna showed large yield in­ applied at 0, 1, and 3 tfha. Soybean line M 351 was creases from applying combination s of secondary and planted in half of each plot and cowpea TVx 3236, in micronutrients (UTA Annual Reports, 1980 and 1981). the othe r half. The trial was repeated in a fine· Where available P in soil was low, large yield in­ textured soil (sandy loam surface, sandy clay.to-clay creases were also observed after P was applied. In at 6(}-75 cm), where soybeans had grown well 1983, the trial was repeated at M ok wa on a de­ previously. generated Stylosanthes pasture (PH = 5.7, CEC = 2_3 Liming. which raised soil pH from 5.8 to 6.3 at the meq/ IOO g soil, Bray 1, available P = 9.2 ppm, organic coarse·textured site, and from 5.6 to 5.9 at the fine­ C = 0.6%). Phosphorus was applied at 0, 12.5,25. and textured site did not significantly affect yield or Mn 50 kg/ha as triple superphosphate. and K at 0 and 30 uptake in either crop. Manganese concentrations in kg/ha as KCI. Halfthe plots received a secondary and uppermost, fully expanded leaf blades of soybeans at micronutrien t combination of 10 kg Mg, 13 kg S, 2.5 full bloom averaged 205 ppm at the coarse·textured kg Zn, 1 kgCu, 1 kg B,and 0.1 kgMo/ha as magnesium 116 Grain Legume Improvement Program sulfate, zinc sulfate, copper sulfate, boric acid, and while inoculation increased the number of nodules ammonium molybdate . respectively. The plotsreceiv. and nodu le dry weight (Table 3.71). Adding rhizobium ing secondary and micronutrients, but not P plots, strains reduced yi eld and shoot fresh weight slight ly received gypsum to equal the Ca content in the 12.5 kg but the differences were not significant. For yields , N P/ha treatment. All plots received SO kg N/ha. times variety interaction was not significant (20~/o Soybean line M 351 was plant ed. level) but was highly sign ificant 0 %) for the other Applying 12.5 kg P/ha significantly increased t ra its. yields but higher application rates produced no In Figure 3.10, we have graphed the overall perfor­ additional yield increases (Table 3.70). P up to 50 mance of the highest yielding lines and two controls. kg/ha increased plant height, node number, and Bossier responded to both N and inoculant as did TGx lodging, indicating that vegetative growth responded 536·6C to a lesser degree. Lines, such as TGx 252-71 C more than yields to higher rates of p, Added K or and TGx 53&-02D, maintained stable and relatively second ary and micronutrients did not affect any factor measured. Growth was vigorous wi th consider­ a ble lodging throughout the trial. The low yield Table 3.71. Mean yield , s hoot fresh weight, number of nodules, and nodule dry weight with three response to P and lack of response to applied second­ fertilizer and inoculant treatments for a ary and micronut rients may be a ttributed to high trial grown at three sites in Nigeria in 1984 plant residues in the soiL- D.A . Shannon Shoot Nodule fresh dry Microbiology Yield ,' weight, Nodule weight, Treatment 1 kg/ha g number g Evaluation of soybean breeding lines with and +N ......... .... 1.624a 458a 16.7b 0.15b without inoculum and nitrogen fertilizer . A 0 ...... .... .. . . .. 1,360b 364b 18.5b O.ISb major goal oft he soybean breeding program has been + 1 .... ....... 1.298b 336b 63.2a O.60a to incorporate free nodulation (promiscuity) into St. error .. . .. . 15l 50.9 10.6 0.13 adapted soybean lines so farmers in the tropics can LSD, % ...... 265 89 18.8 0.23 grow soybeans successfu lly even if inoculant is not C.V .. % ...... 10.6 13.2 32.4 41.70 available. In 1984, we conducted an experiment to 1 +N = 120 kg/he N applied at 40 kg/ha N at 25. 50. and 75 DAP; 0 test 14 elite soybean breeding lines and two non­ = no appli ~d N, n o inocula nt; + I = R. japonicum applied. promiscuous controls in a split plot design with ~ Means fol lowed by the same letter do not differ signi ficantly variet ies as SUb-plots in whole plots of: 1. No applied (0.05). nitrogen. un inoculated ; 2. 120 kg/ha N applied as urea topdressed as 40 kg/ha N 25,50, and 75 days after sowing, uninoculated ; and S. no a pplied nitrogen, Yield (kg/ho) Rhizobium japonicum inoculant strains ffij 2124, ffij 2133, and ffij 2144 applied on the seed at sowing. The tr ia l was sown in four replications at Ilorin, 2000 Mokwa, and Zaria, Nigeria. Plots were 4.5 X 6m wi tb an intended population of250,00Q plants/ha. All plots received 400 kg/ha single superphosphate incor­ porated before sowing. G.297-35 C At a ll locations and in cross·site analysis, adding N significantly increased yields and shoot fresh weights TG. 539-5 E 1500 Table 3,70. Effects of applied P on soil-available P , yield, lodging, and height of soybean M 351 at Mokwa, Nigeria, 1983 Available Pin soiP Lodg· Papplied, (Bray I) Yield, ing 2 Ht., Node 1000 kg/h. ppm kg/ h. score em number 0 ...... 7 2,050 3.1 87 17.2 12.5 ..... 9 2,280 2.9 99 17.8 25 . ..... 10 2.290 3.3 105 17.9 Bossier 50 ...... 18 2,250 3.7 109 18.2 LSD,5% ... 3**'- 180' 0.4*'" 9*** 0.5*** OL-----'------'-----'-----' +N +1 o ns . ...... = not significant , significant at the 0.05 and 0.005 levels of probability, respective ly. Figure 3.10. The +N is for nitrogen ; +1 for R. japonicum I = Sampled at flowering. inoculant. and 0 for with neither . Yields are of highest. ~ 1 = no lodging, ;) = completely lodged. yielding lines. Results from three sites in Nigeria in 1984. Grain Legume Improvement Program 117 high yields over all treatments ; and other lines shown The growth·pouch technique a Iso was used to test a had higher yield with K and a slightly decreased yield rew lines' ability to nodulate with severa l rhizobial wi th inoculant. isolates (Table 3.73), as a possible method to evaluate We were surprised by the higher average yield from a variety's relative (high, medium, low) promiscuity. uninoculated, unfertilized plots than from the in· In this experiment, TGx 304·059D was the most oculated. The threefold increase in both nodule promiscuous because it formed nodules with 85% of number and dry weight indicate that the inoculation t he strains tested. Studies will continue to try to was satisfactory. Work will continue to test more improve this screening procedure.-J . Hohenberg· effective inoculant strains, but we can tentatively Gibbons, W. Root, K . Dashiell. L. Bello, and K. conclude that promiscuous soybean lines such as Mulongoy TGx 252· 71e and TGx 536-02D can give relatively high and stable yields even where an inoculant is not available. Table 3.73. Nodulation ofsoybeans by rhizobia I isolates from West Africa. Presence or Screening for freely nodulating soybean lines. absence of nodules tested in plastic Studies were continued to develop a laboratory pouches under laboratory conditions screening procedure to identify freely nodulating Rhizobium s train (promiscuous) lines and thereby reduce the number of ffij strains IRe strains breeding lines for field evaluation in the soybean breeding program. The procedure, developed in 1983, l'um­ Num­ Strains is based on the principle that some rhizobium strains ber ber nodu­ Soybean Number nodu- Kumber nodu­ lating. tend to form nodules only on promiscuous varieties of line tested lating tested lating 0' 10 soybeans. This year trials were conducted to identify a rhizobium strain or a set of strains that would Promiscuous nodulate with promiscuous lines but fail to form TGx 304·059D ... 20 17 5 5 85 nodules on nonpromiscuous soybeans. TGx 604·027C ... 19 9 4 4 72 A mixture of three strains, rRj 2154A, rRj 2161B, TGx 330·0540 ... 19 7 4 0 39 and IRc 430A, has given the best results. They are Samsoy 1 .. _ .. . . 20 7 5 5 60 Orba ..... . . .. . inoculated in plastic pouches on two pregerminated 19 11 4 3 74 seeds per test line. Four replicates were tested. After Nonpromiscuous three weeks , plants were rated for presence (+) or TGx 17·2GE .... 20 3 5 0 15 absence (-) of nodules. Bossier . ....... 20 1 5 0 5 Nodulation scores have not always been consistent Jupiter . ....... 20 5 5 0 25 across replicate pouches so a final + score is given if Willi ams . . . .... 20 3 5 0 15 more than half of the replicates are positive. A total of 126 lines were tested by this method. The results were compared with nodulation ratings for the same lines from field observations in which plants were up· rooted and scored for nodule mass in soil never inoculated with R. japonicum. Field results are assessed from trials in three replications and at three locations. The comparison of the two methods to evaluate promiscuity is presented in Table 3.72. There was only one false negative and 50% of the lines poor in the field were also poor in the laboratory. Studies will continue to try to improve this screening procedure; however. it can now be used to screen breeding lines before they are planted into yield trials. The advantage of the procedure is that many more lines can be screened in the laboratory than can be sown in replicated yield trials. Table 3.72. Correspondence between promiscuity ratings in the laboratory and in the field with three strains of rhizobium inoculant Lab Field rating rating Good Medium Poor Good. ...... . ..... .. .. .. ... 21 33 28 Poor.. .. ... . ............. .. .. 1 15 28 Root and Tuber Improvement Program The UTA Root and Tuber Improvement Program tinuous and Kinuani shows consistently superior consists of the core program and three international performance. The variety has shown particularly cooperative programs: (1) the Cameroon National good tolerance to excess water strain (moisture Root Crop Improvement Program (CNRCIP), (2) the stress) under inundated cond itions. Rwandan National Root Crop Improvement Program In Rwanda, the Rl\:RCIP has identified cassava and (R:-.IRCIP), and (3) the National Manioc Program sweet potato lines that are well adapted to the (PRONAM) in Zaire. The core program has received country's highlands (1,750 m above sea level). substantial financial support from the International Fund for Agricultural Development (lFAD); CNRCIP receives funding from the International Cassava Development Research Centre (IDRC), Canada, and the General Administrat ion for Cooperation and Genetic Improvement Development (AGCD), Belgium; RNRCIP is sup· The cassava breeding populations have been signi· ported by IDRC and PRONAM by t he United States ficantly improved for important agronomic charac­ Agency for International Development (USAID). teristics through continuous recombinations and Research on biological control of cassava pests has selections at UTA. Many good clones have been been supported with specia l grants from IFAD, the identified from the populations in Nigeria, Sierra German Agency for Technical Cooperation (GTZ), Leone, Liberia, Cameroon, Gabon, Zaire, Rwanda, and the Swiss and Austrian governments. The pr<>­ Burllndi, Uganda, Seychelles, Tanzania, Malawi , a nd gram cooperates with the International Center of Zimbabwe. Tropical Agriculture (CIA T), Colombia, in the in­ A total of 16 UTA improved clones in vitro have troduction of cassava germ plasm in seed form and in been distributed to 32 countries in Africa and India work with natural enemies of cassava green spider for evaluation under their conditions. Some clones mites (CGM) through the Quarantine Office of have shown promise in several countries , parti· Nigeria. The biological control staff also works cularly TMS 30572, which has been most popular in closely with the National Agricultural Resear ch Nigeria and has performed wel1 in Zaire, Ghana, Organ ization of Brazi l (EMBRAPA) and the Cameroon, Rwanda, and Uganda . It also has pr<>­ Commonwealth Institute of Biological Control duced good progenies in other countr ies. TMS 30572 (CIBC). seems to have excel1ent genetic potential in wide Good progress was made by the core program in adaptation and to maintain stability in a wide range popularizing among farmers and researchers the new of environmental conditions in Africa. system for producing seed·yam in Nigeria, Ghana, Resistance to cassava green spider mite (CGYf) has and Cameroon . Farmers who adopted the seed-yam been significantly improved since breeding emphasis production system were highly satisfied wit h it. The was focused on resistance to it. system generated enormous interest among farmers, Several promising cassava clones adapted to high particu larly in Nigeria. a ltitude conditions (1,680 m above sea level) have Epidinocars is lopezi, the most promising natural been identified. enemy of the cassava mealybug (CM), continues its The variety Kinuani, which has been released by natura l spread with established populations in !ITA's cooperation program (PRONAM) in Zaire, Nigeria. Togo, Caribbean, Ghana, Zai re, Congo, consistently out yielded loca l var ieties at 52 on-farm Zambia, and Rwanda where it has been released. It locations in Bas-Zaire. It a lso consistently performed a lso has been observed in Benin and Cameroon. we l1 after standing in water two to five days. Progress continues in improving genotypes with E valuation of breeding material in two environ­ resistance to CM and CGM. ments in Cameroon, Mujuka, with elevation of 80 m. PRONAYf's improved variety, Kinuani, has been and Babungo, with elevation of 1,130 m showed that multiplied on a large scale and widely distributed to clones produced substantial1y higher yields in lo­ farmers in Zaire, where Kinuani plantings now cations where their initial selections were made. exceed 800 hectares . Spread among farmers is con- Yields of materials selected at Babungo were halved 119 120 Root and Tuber Improvement Program when evaluated at Mujuka. Yields were also halved cassava vegetative materials across international when Mujuka·selected material was evaluated at boundaries. Babungo, which indicates that to minimize adap· tation problems, selection should be made under the Etiology ofCMD environmental conditions where the selected va rl· eties wil l be grown. In 1983, for the first time, a gemini virus isolated from CMD·affected plants was tr ansmitted back to cas­ V ariet y Trials sava, though only at very low incidence (IITA Annu a l Report, 1983). In 1984 the virus etiology of Uniform yield trials of 13 lIT A improved varieties CMD was confirmed beyond doubt by transmitting were conducted without fertilization at three 10' back to cassava the gemini virus associat ed with the cations in ~igeria, lIT A, Mokwa, and Onne, that disease. Transmission was achieved through sap represent a wide range of environments. Eight of t he inoculati on of true seedlings of cassava with crude 13 varieties were tested during both wet and dry juice of infected Nicotiana benthamiana. The cassava seasons at lITA. Mokwa is in the dry savanna; Onne, seed lings were raised from seed obta ined from Latin in the high rainfall zone on sandy poor soil; IITA. in America and at IITA from a local susceptible cu l­ the region of moderate r ainfall and soil fertility. The tivar, 60506. Of 57 seedlings inoculated that h ad been results are summarized in Table 4.l. raised from Latin American seed , seven were in­ TMS 4(2)1425, which has been reported to be fected . In one experiment in which 20 seedlings from resistant to cassava mealybug (CM) and somewhat to seed of the local culti var were inoculated, none was CGM, performed well over the locations and seasons infected, and in another experiment with 50 seedlings tested. Quality characters such as dry matter per· from local seed, only one was infected. centage, garification rate, and clone consistency a ll These latest attempts to transmit the gemin i virus rated high. Th is variety is being multiplied by the back to cassava show that it is very difficult a nd Nigerian Seed Service for distribution to farmers. It is depends on the genotype of the cultivar. The experi­ cultured in vitro and tests have shown it to be free of ments a lso underscor e t he difficulty of infecting cassava mosaic virus.-S.K. Hahn, E. Chukwuma. seedlings from seeds of a local, susceptible cultivar, andJ.Otoo which probably explains why no infections were Virology observed in earlier attempts, with seedlings from local IITA sources. Another r eason was that nor­ As in past years, virology research on cassava fo­ mally only 10 to 25 seedlings were inoculated in those cused in 1984 on the etiology and indexing of cassava experiments. mosaic disease (CMD), one of the most seri ous con­ No transmission was obtained to Laportea aestuans straints of African cassava production. The a im of (L.) Chew (formerly Fluerya aestuans(L.) when it was this work is to facilitate the transfer of improved inocu lated by sap with the isolate fro m cassava. From Table 4.1. Results of cassava trials at three locations in Nigeria, 1983-84- Yield. tlh. Improved IITA Average Dry Dry cassava Season yield, matter, yield , Resistance to Gari clones 1 Season 2 Onne Mokwa tiha % tiha CMD CBB CGM quality TMS 50207 .. . .. . . .. 23.4 25.7 14.5 21.2 26.8 5.7 1.7 1.4 P TMS30337 .... ... .. . 23.7 24 .5 ll .6 19.9 284 5.7 1.9 1.3 m TMS -1(2)1425. .... . . .. . 30.9 17 .2 21.0 to.8 18.6 33.3 6.2 !.7 1.4 vg 'ThIS 50395 . . .. . . . . . . . . 23.0 20.9 26.9 6.9 18.6 27.9 5.2 1.6 1.8 4.0 g TMS 30572 ... . 22.4 to. 7 23.4 15 .3 18.4 31.3 5.8 1.6 1.3 3.3 vg TMS63397 ... 16.4 20.0 21.6 14 .0 17.9 34.2 6.1 1.6 1.1 4.3 g TMS 302ll ... 15.6 20.0 to.S 15.5 27.2 4.2 1.5 1.2 vp TMS 50081 ... 22.0 14.0 lOA 15.5 32.0 5.0 1.6 1.2 g TMS 30001 .. 17.3 18.2 11.0 13.4 29.9 4.0 1.5 1.4 4.S vg T MS 30.%5 .... 2l.! 12.4 14.2 8.7 13.2 31.4 4.1 1.7 1.3 4.1> g TMS42025 . . . . .. . 22.1 13.0 15.5 6.6 13.2 36.4 4.8 2.0 1.1 3.0 vg TMS 4(2)0257 ....... . 15.2 S.O 11.5 10.2 11.1 32.5 3.6 2.0 1.2 3.0 g TMS 60506. control ...... 20.9 9.9 15.2 7.0 12.5 28.1 3.5 2.4 1.9 3.5 g LSD . 5% . . .... . .. . ... 7.3 4.2 7.0 4.2 C.V., 1% ............... 23.1 23.1 26.; 31.9 S.E ... ... .. .. ... .. . . ... 2.56 1.44 2.49 1.48 lResistance to cassava mosaic disease (CMD), bacterial blight (CBB), and green spider mite (CGM) wa.s rated on n 1 to 5 scale , where 1 = no damage and 5 = severe damage. 2Ga ri qualit y was rated as follows: VG = .... ery good, M = moderate . and P = poor. Root and Tuber Improvement Program 121 this common weed species, isolates had earlier been Biochemistry obtained t hat strongly resemble the geminivirus isolated from cassava (UTA Annual Report, 1979). Cyanide Analysis of Cassa va Nor was transmission accomplished in Manihot To help breeders select low cyanide clones, we de· glaziouii by means ofwhitefty inoculation from CMD termined cyanide concentrations in cassava tubers affected cassava.- H. W. Rossel from the different yield trials using Cooke's auto· mated enzymatic method on 841 samples, 508 from Indexing CMD un iform yield trials, 196 from advanced yield trials, Cassava is currently indexed for CMD by inoculation and 137 from LCN samples and local cultivars. We to a susceptibl e alternate host, N . benthamiana. also analyzed 162 different plant parts , silage, and Methods of virus detection and indexing have been tuber extracts from Texagri a nd 12 feed samples from considerably improved by the enzyme·linked immu· NCRL- A.M. Almazan nosorbent assay (EUSA). This highly sensitive method permits rapid detection of extremely low Drying Cassava Tubers concentratIOns of virus particles in crude plant extracts. Reported below are the results of a study to Incidence of goiter in certain areas of Zaire (Id",i determine whether the direct double-antibody EUSA Island, Kivu. and Ubangi , Equateur) is high, perhaps method may be u sed to index CMD. from the antithyroid action of cyanide in cassava. An antiserum against cassava mosaic virus was Dried cassava tubers sold as chips, quarters, or halves produced by injecting a rabbit witb purified virus are sometimes contaminated with fungi . We analyzed preparations from N. benthamiana infected with cyanide concentrations in different sizes of dried CMD. The antiserum had a titer ofl/256 in an agar gel tuber to determine acid levels and the size with the diffus ion test. To remove any a ntibodies against least . h ea lthy plant proteins. the antiserum was repeatedly Five clones of medium-s ized cassava tubers were absorbed with sap from healthy cassava seedlings used: TMS 50395, TMS 30555, TMS 30572, TMS 60506, and h ealthy N. benthamiana until no further pre· a nd TMS 30001. Their cyanide concentrati ons in mg cipitate was formed. HCN/100 g fresh tuber were 12.2, 5.6, 7.3. 5.0, and 6.0 The gamma globulin fraction from the absorbed respectively. We peeled them and cut them into four antiserum was purified by ammonium sulfate pre· sizes: (1) chips, t inc h thick; (2) chips! incb thick; (3) cipitation and by filtering through a column of DE 22 quarters; and (4) halves. Each of the last two had a 6· cellulose. A gamma globulin coating of 1 Jig inch ma ximum length. Half of the material was sun· prote in/ml and a second coating of 0.5% ova lbumin dried, and the other half, oven-dried at 45°C for gave the best r esults; the second coating prevented comparisons. the background nonspecific reacti on t hat healtby Figures 4.1 and 4.2 show the cvanide concentration preparations cause. A conjugate dilution of 1/100 was on wet basis after oven or sun drying. The su n·dried used . Before adding the conjugate to the microliter materia ls had concentrations lowe r th a n 10 mg plates, we added 1 ml of healthy N. benthamiana sap HCN/lOOgdry weight basis, whi le the oven·dried had to every 10 ml of conjugate in order to cross·absorb va lues bigher, which indicates that sun-drying is any antibodies against healthy plan t proteins. more effective to reduce HCN conce ntration. Since tbe ex traction buffer proved particularly Fermentation before sun-drying. which is usua lly important, several buffers were tested. Three showed included in processing, wi]] decrease concentration the most promise: (1) 0.1 M sodium citrate, 0.004 M furt her. EDTA, 0.1 % thioglycollic acid, and 0.1% sodium . Drying more than t wo days in the oven essenti ally sulfite (pH 7.0); (2) 0.05 M Tris·HCI, 0.005 M EDTA did not reduce HCN except in the ~·inc h·thick chips. (pH 8.0); and (3) 0.05 M sodium citrate buffer, 0.005 M Sun-drying nine days instead of six davs reduced EDTA, 0.1% thioglycollic acid, and 0.1% sodium HCN. In both methods of drying, quarter~d materi a l sulfite (pH 6.5). had the highest HCN concentration. Depending upon the extraction buffer and the age The fungi observed on the quartered and halved of the plants, virus was detectable in N. bentlzamiana pieces were isolated a nd identified as combinations of extracts diluted from 1/3, 125 to 1/15,625 , and in cas· Aspergillus flavus , Aspergillus niger, Fusarium oxy· sava extracts diluted from 1/625 to 113,125. Therefore, sporum , Penicillium sp., Monilia sp .. a nd Sclerotinia ELISA detected the virus in cassava leaf extract even rolfsii. Toxins produced by the first three species are though the concentrations of VHUS vari ed potential carcinogens. Whether these toxins contri· considerably. bute to the observed endemic goiter is not known and Although its concentration seems to be low in may warrant investigation .- A.M. Almazan and cassava, the virus can still be detected by ELISA, R .L. Theberge especia ll y when the leaves are still young. This Fermentation of Pounded Cassava Leaves method should have important applications in epide. miological studies and in cassava indexing. Further Pondu , a popular Zairois cassava leaf dish, is pre­ work is needed to perfect the method so as to produce pared by pounding the fresh leaves to small pieces and consistent and r eliable results.-G. Thottappilly boiling a longtime in a small quantity of water before 122 Root and Tuber Improvement Program Cyanide concentration ( rng HCN/IOO g ) Leaves from five cassava clones, TMS 50395, TMS 20r---------------------------------~ 4(2)1425, TYIS 30572, TMS 30555, and TMS 30001, were used. Average changes in cyanide contents and in pH of the clones' leaves are shown in Figure 4.3. Pounding reduced HCN concentration at least 50%. 2 days Even at prolonged periods. fermentation had little IIIIII! 3 da" effect on cyanide concentration. The change from the 15 fresh green leaf to the cooked green leaf color. and the change in aroma increased with incubation time. The pH decreased continuously with increased fermen­ tation. Boiling for one hour after fermentation re­ moved some of the acids. hence. the slight increase in pH. The more significant effect of boiling though was that it reduced HCN to an undetectable level. - A.M. 10 Almazan Paste Viscosity Related to Cassava Bread Quality While determining the best baking formulas for cassava bread with pentosan, we noted that bread 5 quality depends on maximum paste viscosity of the cassava variety or clone number. Bread crumb of loaves from cassava flour with pentosan is softer and more moist than crumb of wheat bread, primarily because more water is needed for the right "liquid-dough" consistency. TMS 30572, o TMS 30001. and 60447 have high maximum pasce Fresh Chips Quartered Halved 14," 1/ viscosities and produced bread with normal crumb 2" structure (Fig. 4.4). TMS 30395 had gelled portions at Figure 4.1. Cyanide concentration in cassava tubers (wet the side and bottom of the loaf. TMS 518 and TMS basis) oven-dried at45°C two and three days. Percentages of 91142. both with low maximum paste viscosities, had dry matter are shown on tops of bars. Cyanide concentrotion ( mg HeN/IOOg) 35 IOr---------~----~----------------_. 78,6 lim • 6 days 30 .9 days .•8 i 25 5 ~ 8 520 I ~ E 620 .~ " a. .. • P-'Q. 6.00 E ~ 0 ~ • : ...~ 5.80 .!? o Frest'i Chips 1/4" 'hOi Ouoriered Halved 8 10 . '~ 5£0 1 ~ c Figure 4.2. Cyanide concentration in cassava tubers (wet 5.40 ~ ~ u ... basis) sun-dried six and nine days. Percentages of dry 5 5.20 li matter shown on tops of bars. 500 adding other ingredients. To ensure that its cyanide 0 0 2 4 6 concentration is safe, fermentation at 37°C in sealed Ste~ in prOCessinQ casseva leav-es plastic bags after pounding was introduced. The Figure 4.3. Changes in cyanide concentration Co .) and in effect ofthe added fermentation step was determined PH (-.-) of cassava leaves during processing. Steps involved by analyzing the cyanide concentration in the pro­ are (1) fresh leaves, (2) pounding, (3) fermentation: 2 hours, cessed leaves and measuring the pH of the leaves (4) hours, (5) 8 hours, (6) 24 hours. (7) boiling after fermen­ suspended in distilled water. tation: 2 hours, (8) 4 hours. (9) 8 hours, (10) 24 hours. Root and Tuber Improvement Program 123 Figure 4.4. Suitability of cassava varieties for bread of 75 to 100% cassava flour and pentosa n can be determined by paste viscosity values. Cassava lines used ror these loaves have low paste viscosities that pl·oduce loaves with gelled. clumped center. s imilar ge ls with centers so wet that they form ed a 25% wheat flour. 90% TMS 30001 + 10% soybean clump when s liced . TMS 50395 produced the worst Rour. 90% TMS 30001 + 10% cowpea flour . 100°.0 loaf- a ll ge l and an air space below the top crust. 60447, 100% TMS 30001. 100% 30572.- J . Casier. C. When wheat flour was added at 250;0 concentration , Depa.epe, R . Dewit, alld L. Van de Ven only cassava varieties with high max imum paste viscos iti es formed bread with normal structure. Tannic Acid's Role in Gari Processing and Wheat flour's lower absorption required less water. Quality Adding wheat flour improved crumb appearance. particularly that from TMS 60447. TMS 30395 still Gari. a traditiona l cassava product. is wid e l ~l con· had gelled portions. TMS 518 loaf was compact with a sumed in West Afri ca. Processing the tuber to ga ri collapsed center. TMS 91142 loaf still had a wet involves material loss and hours of labor. Removing center and ge ls at the s ides and bottom. The entire the peel and damaged portions can redu ce cassava crumb ofTMS 50395 was extreme ly wet. we ight by 45%. Certain steps in process ing =:;uc h as These results indicate that the suitability of cas­ grating, dehydration. and dryin g have been mecha· sava lines for bread of 75 to 100% cassava Rour and nized. but pee lin g is stil l done man ually. even in pentosa n can be determin ed by maximum paste commercia l plants. viscosity values .- A .M. Almazan We in vestigated t he possibility of us in g unpee led tubers for gari to minimize labor and cassava losses. Consume .. Evaluation of Cassava B .. ead Among IITA improved cassava lin es a re some. such as with Pentosan TMS 30001. TMS 4(2)1425. and TMS 60 142. with a white pee l. Gari from peeled and lInpee led tubers of Pure cassava flour cannot be used for baking bread TMS 30001 , TMS 30572. TM S4(2)1425, a nd TMS60142 because it lacks proteins with wheat·glu ten proper­ were prepa red. Unpee led tubers a lways produced ties. Adding an improver. like pentosan. is necessary grayish or brownish gari that darkened as storage to produce loaves with good baking quality (Fig. 4.4). increased. Nei ther grayish 01' brownish gari is accept­ Pentosans are polysaccharides consisting of the able to consumers who prefer ga ri that is white- or sugars arabinose and xylose, and are del'i ved from rye yell ow from added palm oil. or wh eat endosperm. Thinking the di sco loration likely resulted fro m The best baking formula for cassava bread with tannic ac id . we determin ed tanni c acid levels of pentosan has been 3 to 50/0 yeast, 3(% sUCrose. 2 to 30/0 pee led and unpeeled tubers. us ing the Polin · Dennis pentosan. 1% oil or fat, 70 to 85% water. 1% salt. reagent. Cari with peel a lways had higher tannic acid Using that formula , we prepared flours to use in concentrations. at least fi ve times higher in samples taste eva luations from oven·dried chips by milling on 15 mon t hs old and two times higher in the 15-day-o ld adisk mill or by pounding in a mortar with pestl e. and materials (Fig. 4.4a). Cyanide concentration in either then pass ing through a 180-1' s ieve. type ofgari was undetecta ble. The bread was evaluated at the IlTA canteen A survey of tannic ac id concentrations in the peel during lunch a nd in the Internation a l House lobby. and fl esh of various UTA improved cassava clones The orde.·ofpreference at the canteen was: 75% TMS and of local cultival's a Iways showed more ac id in the 30001 + 25% wheat flour. 90% TMS 30001 + 10% pee l (Tables 4.2 and 4.3). In commercial plants. a cowpea flour, 100% 60447, 90% TMS 30001 + 10% bleaching agent after fermentation mi ght prevent soybean Aour, 100% TMS 30001. 100% TMS 30572. At di scoloration du e to high tannic ac id in the pee l. the I-House lobby_ the order was 75% TMS 30001 + A.M. Almazall 124 Root and Tuber Improvement Program TCI'ric acid ().I 0//000 ) Table 4.2. Tannic acid concentration (ug/lOO g) in 120 r---------, IITA improved cassava clones Peeli ~ p. .l ed Clone no. Peel Flesh flesh TMS 30001 .. . .. .. . ... ........ . 100 20 5.0 100 • Unpeeled TMS30040 .. . .. . .... .. .... . ... 100 20 5.0 TMS 30555 ....... .... ..... .. .. 100 20 5.0 TMS 30572 ...... . .... ...... ... 80 30 2.7 TMS40150 ..... ... ....... .. '" 170 30 5.7 eo TMS 2025 . .. . .. ..... .... ..... . 170 20 8.5 TMS 50395 .. . .. . .... ...... . ... 60 20 3.0 TMS50142 ... · . . . . . . . . . . . . . 80 20 4.0 60506 .... ... . .... ... .. . . 70 20 3.5 60 TMS61677 ... · .... .. , ...... 140 30 4.7 TMS63397 ... · .. . ...... ... . 60 20 4.0 TMS 1525 . .. . ........... .. 110 20 5.5 TMS801O .. . . .... . ... ... .. .... 110 30 3.7 TMS8022 .... .. .. . , . ... . .. .... 90 30 3.0 40 TMS8034 .... ..... .. ... ... .... 100 20 5.0 TMS8042 . ... .... ... , .. ....... 70 20 3.5 TMS 4(2)0156 .. .. . .. , .. 100 30 3.3 TMS 4(2)0267 . .......... . .. . . . 100 20 5.0 20 TMS 4(2)0378 _ .......... .. .. . . 60 10 6.0 TMS 4(2)0780 ..... . .... . ...... 160 20 8.0 TMS 4(2)1425 .. ...... .. . ..... . 200 30 3.3 o TMS 30001 4(2)1425 :30572 TMS4(2) 60142 1425 Table 4.3. Tannic acid concentration (ug/100 g) in 15-Monlh-ofd QClfi 15-Iloy-old QOfi cassava local cultivars Figure 4.48. Concentrations of tannic acid in peeled and Peel! un peeled gari. Clone Peel Flesh flesh Oyarugba funfun . ...•.. . .... .. 50 20 2.5 Arubielu pupa ...... . • ...... 60 10 6.0 lsunikankiyan ..... . • . . .... 60 20 3.0 Ogunjobi . . . .... . 100 10 10.0 Tissue Culture Ojunkanye ........••......... 50 10 10.0 Oyarugba dudu. . . . .. . . . . . . 50 10 5.0 With 17 lines of cassava available for distribution in 5093 Gb .. . . . . . ....... . 80 10 8.0 1984, 37 packages were sent to 16 requesting count­ 53101 .. . . . . . . . ... . . . .. . .. . . 70 20 3.5 ries in Africa and two in Europe_ Three additional 56308.. .. .. .. .. .. ....... . . 150 30 5.0 varieties were certified by the Plant Quarantine 60447 .... . . . .. . . . . •. .. .. . ... . 80 10 8.0 Services (PQS), Nigeria. Information from collaborators indicated that yield potentia ls of IITA clones ranged from 25 to 45 mixture gave better results for the cultivars that were tlha. Reports from Zaire (PRONAM) on the cassava mosaic disease (CMD) score of UTA clones showed kept under ordinary light. In another experiment, we used peat moss a lone, the same level of resistance there as at UTA. Because receiving countries have problems estab­ one mixture of 2 parts peat moss and 1 part vermi· lishing cassava plantlets, we tried to increase the culite. and one of2 parts soil and 1 part vermiculite, as planting media. survi val rate by using different planting media and High light intensity (25,500 lux) gave high survival , light treatments. Using two varieties. TMS 30786 and from 80 to 100%. with the highest in the vermiculi te TMS 30001, we divided cultures into two sets. The and soil mixture. With ordinary cu lture, survival first was exposed to higher light intensity (5,000 lux) four days and the screw caps ofthe cu lture tubes were rates ranged from 30 to 80%, the highest in the mixture of peat moss and vermiculite.- S. Y.C. Ng loosened one day before transplanting. The second set was placed under ordinary culture condition (light intensity, 3,000 lux) and the caps were loosened Biological Control one day before transplanting. The three planting media were sawdust, peat moss, and a mixture of 2 Epidinocarsis lopezi in Africa parts of soil and 1 part vermiculite. Dissemination. E. lopezi was introduced from South Sawdust was not a suitable medium ; peat moss America and established first at lITA in 1981 and then alone gave best results when the plantlets were near Abeokuta (lITA Annual Reports, 1982, 1983). It exposed to higher light intensity; the vermiculite soil has since been released on an experimental basis at Root and Tuber Improuement Program 125 about 30 sites in 10 African countries (Table 4.4, Fig. released from the ground with the assistance of 4.5): Congo, Gambia, Ghana, Guinea·Bissau, Nigeria, national program personnel. Rwanda, Senegal, Togo, Zaire, and Zambia. Except E stabUshment. A natural enemy is considered for three release sites in Ghana, all insects were established when it has survi ved a full rainy season- Table 4.4. Release of Epidinocarsis lopezi in Africa from 1981 to 1985 Date Country Locahty Release Approximate Status carried number after outby released : release Xl8lto I 82 .. Nigeria lIT A, Ibada n IITA 1,150 Established VII 82 ....... Zaire Kinshasa IITAjPRONAM 200 Established VII 82 ....... Zaire Bas- Zaire IITA jPRONAM 800 Established VII/IX 82 and X 83 . Congo Brazzaville- Mantsournba ORSTOM' 3,000 3 Established IX 82, X 83 and IX 84 Congo Mbesrea ORSTOM 3 Not yet recovered XI82 ........ . . ... Nigeria Abeokuta (Ogun) Texagri IITA 2,050 Established V83 ...... .... Nigeria Onne (Rivers) IITA 420 Established VII 83 and V 84. Zaire Likasi area (Shaba) IlTAjPRONAM 325 Established XII 83 .... Nigeria Enugu (An.mbra) ASUT, Enugu' <2,000 Established X to XII 83 ..... Nigeria OIokoro (Imo) NRCRI' 5,000 Established Etiti (Imo) NRCRI' 5,200 Ibere (Imo) NRCRJ2 1,200 184 .... . . . . , . . . . . . Guinea· Bissau IITA 1,500 Established Bissau II 84 .. . .... . .. . .. . Nigeria Igbariam (Anambra) "ffiCRI' 500 Established [[84 ... . . . . . . . . . . . Senegal Thies lITA 250 Kot yet recovered II 84 .. . . . ... . . . .. . Senegal Tivaouane IITA 1,125 1\'ot yet recovered III 84 . . .. . .... Gambia Jambanjali IITA 250 Established III 84 ... . .. . ... Ghana Pokoase (Greater Accra) Plant protection 1,400 Established Ghana III 84 . . .. . . . . . . .. . Ghana Sege (Greater Accra) Plant protection 400 Established Ghana III 84 . . ... .. .. , . .. Zaire Lubumbashi (Shaba) IITA!PRONAM 400 Established III 84 ...... .. Nigeria Otobi (Benue) NRCRI 300 Established lJ[ 84 .. ... . . . Togo Glidji ferme, Bokokope lITA 1,500 Established Togoville [[[84 .... .. .. Togo Kamina lITA 500 Established III 84 .. . ... .. .. ... Togo Sokode lITA 500 Established V84 .. ......... Zaire Kikwit (Bandundu) IITA /PRONAM 250 Established' V84 .. .. . . .. . -,. Zaire Mosango (Bandundu) llTA /PRONAM 250 Established X84 .. . . . , . . . . . . Zaire Kazenze (Shaba) IITA/PRONAM Not yet recovered;j X84 .. . . . , . . , . . . Zambia Mansa UTA 800 Recovery not confirmed X84 Zambia Mwense IITA 2,560 Recovery not confirmed II 85. Rwanda Gisenyi CIBC 3,200 Recovery not confirmed II 85. Senegal Tivaouane Senegal 1,;00 Not recovered plant protection II 85. . . . . . . . . . . . Guinea- Bissau Guinea 1,;00 Established from Bissau plan t protection earlier release III 85 ..... ,. . ••. Ghana Bimbila UTA , aerial Recovered releases III 85 ... ........ Ghana Kumasi UTA, aerial Not yet releases recovered [[[ 85 ... Ghana Agbosome UTA, aerial Not yet releases recovered III 85 ... Gambia Jambanjali UTA 4,000 Established from earlier release lUnless stated otherwise all insects came as adults from IITA. ~Single specimen recovered; later no recovery. 2Reared from cultures supplied by nTA. "Release of field-collected material. 3Total for Congo 1982: 1983-84 figures not known. 126 Root and Tuber Improvement Program Cassava belt ----- CM distribution I1 1 11 1111 ·· ·· .. ............ .. . . .., . :-. :. . . . , ', ·· ....... .. ~ ~ distr ibution e­ S. lopezi recovered o- S. ~ released but not yet recovered ?• - Questionable distribution of eM IOOOkm. Figure 4.5. Distribution of CM and E. lopez; in Africa. the period of low eM populations- and has been covers about 450,000 km 2.-H.R. Herren, P. recovered in rearings again in the next dry season. Neuenschwander, WN.O. Hammond, R. Hennessey, Establishment has been recorded with certainty from and K.M. Lema releases in seven countries and possibly from two others (Rwanda, Zambia) but we have not yet seen Epidinocarsis lopezi attacks on cassa va m ealy­ specimens from them. No establishment has been bugs (eM). [n the 1983 Annual Report, we had linked reported from Senegal so far. successfu l parasitization of e M by E. lopezi with additional host mortality represented by dead eM Dispersal. The spread of E.lopezi has been compre­ that yielded no parasitoids. That phenomenon was hensively documented in a large cassava growing investigated in more detail by observing what hap· area in Nigeria. Samples were taken in a ll directions pened to mealybugs attacked by an E. lopezi female. fro m the re lease sites in I1TA (release November 1981) We observed wasps in the laboratory stinging and near Abeokuta (release November 1982) at the different stages of eM. As soon as a wasp fin ished beginning and end of the dry seasons in 1983 and 1984. stinging and lost interest, we carefully removed the A total of 481 fi elds were surveyed . Results from 231 eM and reared it in a petri dish on a leaf of Talinum fie lds sampled in December 1984 are shown in Figure triangulare. Mortality of the CM on day 6 and pro· 4.6. South and west of the Niger river, E. lopez; occurs du ction of mummies t hat yielded Ii ve parasi toids regularly in t he rain forest, transition zone, and were noted, and the results compared with the sur· Guinea savanna. North of the Niger river, eM den· viva l of un attacked mealybugs, kept under the same sities were very low and only a few E. lopezi were conditions (= control e). Resul ts of three experi· found. It spread over a dry season from 50 km in t he ments are given in Table 4.5. In the first experiment rain forest to 170 km in the Guinea savanna. The (A 1) a leaf with numerous eM of different stages parasitoid has dispersed from several release sites w hen they a re susceptible to attack (second to fourth into neighboring cou ntries (Benin, Angola, stage, pre-oviposition females) was offered to wasp Cameroon). In summary, by the end of the dry season , females of undetermined ages. The experiment in­ 1984-85, it is confirmed in 10 African countries and volving many females lasted several weeks. In A2, the Root and Tuber Improvement Program 127 that drew our interest in the next section.-P. Neuenschwander and E. Madojemu Melanization: the Mealybug Defends Itself Both in the field and insectary, cassava mealybugs that had been parasitized by the encyrtid wasp, Epidinocarsis lopezi, had black particles in their body cavities visible through the exoskeleton. Dissecting the mealybugs showed the black particles to be dead eggs and larvae of the parasitic wasp (Fig. 4.7), a significant finding because the female wasp lays eggs inside the mealybug. The egg normally hatches into a wasp larva that feeds internally for 8 to 10 days on the mealybug, eventually killing it. The wasp larva then metamorphoses into a pupa and emerges from the dead mealybug as an adult wasp ready to attack another mealybug. Dead, blackened eggs and larvae of the wasp inside live mealybugs indicated a defense reaction of the host against the parasitoid. That phenomenon, known in many insect species attacked by parasitic wasps, was reported in mealybugs as early as 1939. The defense mechanism involves the mea lybug's hemocytes Or blood cells first surrounding and en· Figure 4.6. Spread of E. lop"i from release sites at IITA capsulating the wasp's egg or larva. Then, by a (broken arrow) and Abeokuta (arrow), surveyed December process called melanization, the mealybug blackens 1984 .• = E. lopezi recovered; = released but not yet and hardens the innermost cells of the capsule, which recovered. likely kills the parasi toid by suffocation. Such a defense reaction is an example of the continuing co· evolution between a parasitoid and host. The para· wasp could choose different stages but each female sitic wasp attacks the mealybug. which, in turn, uses was offered exactly 150 eM, 50 of each stage, for two its defense mechanism to kill the egg a nd larva of the hours only, replicated four times. Only eM that had wasp. been stung are reported. In the third experiment (B) To determine melanization frequency and effect, the female wasps were offered only one eM stage. we performed the following experiment. A female E. The results (to right of I) indicate that E. lopez; lopezi was permitted to attack a 3rd or early 4th instar slightly, but not significantly, prefer eM in the third mealybug on a small cassava leafin a petri dish. After stage. In some instances, a female wasp, after retriev· observing the insertion and removal of her ovi· ing its sting, turned around and fed on the wound positor, we separated the mealybug from the wasp made by her ovipositor, a behavior called host· and placed the mealybug on a small cassava plant feeding. That enlarged the wound and partially enclosed in a plastic cover with mesh·screened open· drained hemolymph from the eM. All eM thus fed ings. After the mealybug fed on the plant seven days, upon died immediately. we removed and dissected it in 70% ethyl alcohol. The The results indicate that, at least where the female time interval assured that eggs normally would have had a choice, she significantly preferred the younger hatched and larvae would be readily visible. Of 345 host stages, which offer less resistance, for host such mealybugs, 44.3% were parasitized; and 98.0% feeding. We infrequently observed host·feeding on of the 153 parasitized mealybugs had some eggs first stages. By contrast, the percentage of eM that and/or larvae at least partially melanized. yielded a live parasitoid increased on the older host \Ve used a second series of experiments to see if a stages but generally remained relatively low. Even in second oviposition opportunity by a different female the no-choice experiment and excluding the host·fed E. lopezi would change the statistics. The second CM, successful parasitization was only 37.8%. A attack was within two hours ofthe first and followed large percentage of the eM died from mutilation by the same procedures, with these results: 68.6% ofthe the parasitoid, most right after being stung, es­ 347 mealybugs were parasitized; while 87.4% of the pecially when Qviposi tion lasted more than seven 238 parasitized mealybugs had some eggs and/or minutes. Death was delayed for about one third of the larvae at least partially melanized. CM, and mortality was significantly higher on the "At least partially melanized" refers to the fact younger stages- much higher than in the unstung that while all the eggs and many larvae of the control (C). Such mortality must be attributed to the parasitoid seemed to be completely melanized, other wasp. Finally, a fairly high percentage of CM (about larvae showed various degrees of melanization: only 40~/~) survived a wasp attack. It is that proportion the tip of the tail, or 20, 50, or 80% melanized (Fig. 4.7). 128 Root a.nd Tuber improvement Program Ta ble 4.5. Fate of Ph enococclis manihoti attacked by Epidinoca rsis lopez; in the laboratory, each eM stung only once. Four expe riments, A I' A :.o. B . C 2nd to 2nd 3rd 4th 4th I nstal' A B C D 211 X Number of mea lybugs attacked. .................. . ........ A, 100.0 123.0 102.0 325.0 A, 18.0 27.0 16.0 61.0 B 123.0 102.0 100.0 325.0 Contro l. not attacked ... . . . . . . . . . . . . . . . . . . .... . .. . . C 230.0 130. 0 130.0 490.0 I. (J 0 mea lybugs attacked. accord in g to stage A, 30.8 37.8 31.4 3.00 (1000 '0 ~ A + B + C ~ D) ............ . ........ .. A, 29.5 ~ 4.3 26.2 2.89 n. Fate of attacked eM( 100% = A. r esp. B. res». C. resp. D) A I 20.0 10.6 8.8 12.9 8.48* a) 0 '0 mea lybug host-fed upon. .. .. . .. - .... . .. - .......... .. ... . . A, 22.2 11. 1 6.3 13. 1 B 13.0 11.8 9.0 11.4 1.77 A , 18.0 27.6 2 .4 24.9 b) 4.82 0 ' 0 producing l ive parasiLOids . .. . . . ... ....... - A, 11.1 22.2 25.0 19.7 B 17.9 33.3 29.0 26.2 7.49* A, 20.0 31. 7 10.8 21.5 11.61 * c) °0 dead on 6th day. but not host-fed .................. . .. . A ";!. 33.3 11.1 12.5 18.0 B 34. 1 23.5 1 ~ .0 24.6 12.12* A, 42.0 30. 1 52.0 40.6 d) 0 0 C~ 1 surviving sting more than s ix days. ...... . .. ... . .... . .. . A";!. 33,4 55.6 56.2 49.2 B 35.0 3 1.4 48.0 37.8 III. 0 '0 morla lity of un attacked mealybugs (contro l) . . C 4.3 8.5 6.2 5.9 a Test orhomogeneilY : X:.! 0 .05. 2d.r. = 5.99' ;\ 1 and A ";!. pooled in ordel'lo rulfill requi rements or test. partia lly melanized. some mea lybugs wi th me lanized eggs a lso contai ned a completely unmelanized larva. Other researchers, us ing different spec ies of para­ sito ids a nd hosts. have stated lhat o nly pa r t ial me la nization probably is not suffic ien t to kill t he larva. Escape is possible for a larva may wri ggle from the me lanized capsule o r shed the me lanization at the next molt. So melanization must be complete or the larva will survi ve. The fa ilure to completely melanize a la r va may support the hypothesis tha t the host's hemocytes have onl y limi ted capacity to me lanize, especia lly if early instal'S a re a t tacked. In our second ex periments when mea lybugs were attacked a second t ime by a diffe ren t wasp. t he decrease in even partial me lanization from 98.0 to Figure 4.7. Part ia lly me lanized la rva of t he paras itic wasp. 87.4 0/0 may have resulted from a limi t to the Epi(/inocorsis lopez;. with anterior end visible (100 x). mea lybug 's a bili ty to mela nize. In terestingly. mea ly­ bugs stung once never had more than one larva per host. But mea lybugs stung twi ce usua lly had two or Our rate of pa l' as it ism wasonly 44. 3% • even though t lu'ee la rvae per host- most par t ia lly me lani zed bu t in each case the female E. /opezi inserted her ovi­ many notata ll me la nized . Although a mea lybug may positor in to the mea lybu g. Eggs t hat may have di ed contain more than one larva. only one will survive to before melanization would not have been detected in form a mummy and emerge as an adul t. So partia l our dissections, or perh aps the wasp does not lay an me lanization and superparasitism in both the fie ld egg every time. In our experiments or a second attack and the insectary explain why E. /opezi can overcome on the same mea lybug by a different wasp. parasit ism melanization and remain a successful paras itoid . in creased bu t o nly to 68.6% . An addi t iona l explana ti on may be t ha t t he wasp Alt hough 98.0% of the paras itized mealybugs a t­ takes adva ntage of the mea lybug's limi ted capac ity to tacked only once had some eggs a nd /or la rvae at least melani ze. and the wasp's ovipos iti ona l behavi or Root and Tuber Improvement Program 129 evo lved to keep it one step ahead of the mealybug's were harvested and a ll e M (2nd to 4th instar) defense reaction . Although she may not lay an egg counted . every time she inserts ber ovipositor into the mealy­ At the same t ime a random sample of 100 tips was bug, when E. lopezi does oviposit, s he usua lly lays taken from the field and dissected_ Results of the 34 more than one egg. Since the mea lybug's hemocytes pairs of sleeves that lasted the whole experiment (16 cannot melanize a ll the eggs, one will hatch into a pairs were lost) a nd of the 31 tips of the random larva that survives even partial melanization, con­ sample that contained eM are presented in Figure tin ues to develop, and eventually kills t he mealybug. 4.8. In the open sleeves the natural enemies reduced In a third series of experiments, we used early 4th t he e M population sevenfold compared with popu­ instar mealybugs directly from the insectary rearing lations in the closed s leeves (F = 38_3*). In the open cages containing co lonies of E. lopezi and placed t ips with eM, populations were sti ll substantially those with black particles visible through the exo­ lower, presumably because eM enemies were not skeleton on water leaves, Talinurn trianguLare, in petri impeded_ The natura l enemies responsible for the dis hes. The controls were mealybugs not exposed to reduction late in the dry season were E. lopezi a nd the wasp, therefore, n ever attacked or parasitized. loca l coccinellids_ We could not eva luate eac h 's Host feeding and/or excessive stinging was elim­ re lative importance. inated as factors in these experiments. First, we Early in the next dry season we did a similar s leeve­ wanted to ascertain if melanization caused the mealy­ cage experiment, with 50 pairs of open and closed bugs to die prematurely before the parasitic larva had sleeves and each tip infested with two eM egg masses. a chance to develop normally before forming the At the same time an open, clean cassava tip was mummy. Such premature morta lity was extremely infested with eM, and an uninfested tip was marked ; low with no s ignificant difIer ence between mealybugs 25 of the double pa irs (with and without sleeves) were with black particles (4.8%, N = 84) and the con trols harvested after one month , the other ha lf after two (4_1 %, N = 122). So parasitism, a t least as revealed by months. The field at the beginning of the experiment black particles, is not a mortality factor in mealybugs was absolutely clean of e M a nd, consequently, of E. before mummies form. lopezi. We re leased 1,000 wasps in the field the second Mummy formation and norma l adul t emergence week, a nd another 1,000 in week three, too few to stop occurred in 66.7% of the mealybugs with black the pest population build-up. In both types of sleeves, particles. This demonstr ates t hat E. lopezi overcame e M increased until some were c r aw ling on the or evaded melanization by its host. But 33.3% of the sleeves. There they could be stung from the outside by mea lybugs killed the parasitoid and thus prevented E. lopezi, so exclusion in this experiment was not mummification and a new generation of wasp complete. After one month, mean eM populations per progeny. tip were substantially t he same in both types of To summarize, melanization as a defense reaction sleeves, 199.4 in closed, a nd 186.3 in open sleeves by mealybugs is an evo lutionary fact. Through co­ [critical interval according to Tukey-A test = l.96x , evolution, however, the introduced parasitic wasp, calcu lations as log (x + 1)] . Overall parasitization Epidinocarsis lopezi, continues to be an effective rates were only 5.3 and 6.8%, r espectively. On the natural enemy in our biological contro l program.­ open infested tips, mean eM density was, however, D_S ullivan and P. Neuenschwander significantly lower , 4_8 with a parasitization rate of 34. 1 % . On tips that had not been infested experi­ mentally, eM density again was s ignificantly lower, Measuring Epidinocarsis lopezi's Efficiency 1.0 and with parasitization of22.5 % . E.lopezi was t he The efficiency of an establis hed natural enemy can be only natural enemy present, so it k ept e M popu­ esti mated in the field by excluding the enemy from its lations low on unprotected cassava. In both types of host_ The undisturbed development of the pest insect sleeves, however, parasitoids were largely, if not protected from its enemy is then compared with the totally excluded. Thus e M populations esca ped the s itua tion where the pest's natural enemy could wasp a nd reached high levels. In the second sample attack. We excluded naturally occurring E_ lopezi one month later , there were significantly fewer eM from eM at UTA by two methods , usi ng ga uze s leeves, on the tips with open s leeves than on those with and spraying an insectic ide that destr oys most para­ c losed s leeves (607 vs. 1,420; crit. interva I 2.29x), and s itoids without da ngerously harming e M. sti ll fewer on unprotected infested tips (39). In the second ha lf of the dry season 1983/84,50 pairs In a third exper iment (15-10-84) a 0. 3-ha field of of sleeves made of a fin e plast ic mesh (40 x 20 cm) cassava with fresh regrowth was homogeneous ly were installed on clean tips of cassava plants (TMS infested with one egg mass of eM per t ip, and E. lopezi 30572), which were then artificially infested with one was excluded by insecticide. Half of the fi e ld was egg mass of eM. One s leeve of each pair was tightly sprayed with carbaryl at I kg a. i. / ha ; the other half closed to exclude a ll predators and parasitoids; the was a n untreated cont ro l. Insecticide was applied other was left open to allow natural enemies access to weekly, 11 times, throughout the experiment, 25-10-84 e M. The opening at the bottom of the s leeve was so through 11-1-85_ We recorded eM populations by sma ll that the micro-climate in both types of sleeves weekly samples of 10 ti ps from the center portion of was essentia lly t he same. After 2~ months t he t ips each half of the field . The tips were dissected in the 130 Root and Tuber Improvement Program X CM density per sleeve (logX+ll Cassavo mealybugs per tip 200 200 .-- --, 100 I ± II , SE 2.0 I , 160 I,'S p4'oyed " ',I 50 . ., ,r W;lh 120 in• •e tietde I , I , 80 f , 1, .. - , , 40 I ' , • 10 1.0 I ,~ ~ ..... ., __, Without inMCticidt " 5 O~~~~~~~~~~~-~ Porositi~otion (%) 24 ::; ISp4'oyed ~i;1 Control 20 o L-...:Lt.rL.o..--:c-....,..-2.0/~ __- -.J.~L---1 0 Number of infested tips ::,:: 16 iH" :: .:.,.,,. 34 34 31 !: . !:" Closed Open No 12 sleeves sleeves sleeves Pairs of sleeves Figure 4.8. eM population 2! months after infestation, in closed and open sleeves, and from a random sample of infested tips without sleeves, IITA 1984: weeks ofter infestation laboratory and all stages of CM and mummies found were counted. Parasitism percentages were calcu* Figure 4.9. eM population development in insecticide­ treated and untreated plots (a) together with corresponding lated by relating the number of mummies to the parasitization rates by E. wpezi, (b) UTA 1985. number of living CM (2nd to 4th stage) found the previous week, presumably when the CM yielding the mumnues were stung by the wasps. Figure 4.9 shows mean CM (1st to 4th stage) cassava were grasshoppers. As the CM spread, how­ numbers for the two treatments. Up to 25-12-84, CM ever, arthropod fauna on cassava rapidly increased in populations increased markedly in the insecticide­ abundance and complexity. Indigenous coccinellid treated plot, and remained low in the untreated predators of other mealybugs, chiefly Phenacoccus control. As expected, the parasitism percentage was madeirensi8 Green and Ferri8ia virgata (Cockerell), low in the Held where E. lopezi adults were selectively adopted the newly arrived species as an alternate kIlled by msectIcides. The percentage was substan­ host. A few parasitoids also became adapted, par­ tially higher in the control. As soon as insecticide tially so far, to the new pest. The indigenous natural treatm~nts stopped, parasitization rates picked up, enemies moving into the cassava ecosystem were mdICatmg that wasps were immigrating into the followed by their own complex of parasitoids and field. Overall, the exclusion experiment successfully hyperparasitoids. Simultaneously, insects of other demonstrated the efficiency of E. lopezi in keeping species occupied the new, complex microhabitat of CM low up to 25-12-84. After that, CM populations cassava shoot apexes that were damaged and distor­ crashed in the treated field , but probably not from E. ted by CM feeding. The polyphagous predators at­ lopezi. We speculated that the heavily attacked tracted to the enriched insect fauna added still more plants became unsuitable for CM or hot weather and links to the developing food web. harmattan destroyed the CM.-P. Neuenschwander, A new stage in the development of the cassava F. Schulthess, and E. Madojemu arthropod fauna began in 1981 with the introduction and establishment of Epidinocarsi8 lopezi (DeSantis). Food Web ofInsects Associated With the A new complex of insects soon began to exploit E. eM lopezi as an alternate host. The primary hosts of the hyperparasitoids had been mainly parasitoids attack­ Before cassava mealybugs (CM) arrived in Africa the ing P. madeirensi8 and F. virgata. The arthropod only insects commonly found in large number~ on fauna now associated directly and indirectly with the Root and Tuber Improvement Program 131 CM number about 100 species but only about 20 are Hyperaspis delicatula. This species is common and common in one part or another of the pest's range. locally abundant on high CM popUlations in West Data in this report are based on surveys in 10 Africa. African countries with eM, where E. lopezi was H. senegalensis. In Zaire this species was common released but not necessarily established. on cassava only where eM were abundant. Searches for insects associated with CM were concentrated on, but not strictly limited to, cassava Hyperaspis pumila. Locally abundant on P. plants. Other plants, e.g., Acalypha and Hibiscus, manihoti. that the polyphagous P. madeirensis and F. virgata Hyperaspis vinciquerrae. This is the common small prefer, also were searched. Insects were collected by coccinellid on CM in Senegal and Gambia. three methods: directly in the field, rearing from Parasitoids of coccinellids and their hyper­ infested shoot tips, and rearing separately in gelatin parasitoids. Various parasitoids were reared oc­ capsules. casionally from immature coccinellids (Cl, C2 in Fig. 4.10 and Table 4.7). Only Metastenus at times became Coccinellid Predators of Mealybugs abundant; others have reported that it significantly Many species of coccinellids have been found on reduced coccinellid populations. cassava plants in Africa. A few in the genera Other predators of mealybugs on cassava_ Exochomus and Hyperaspis have become common Severa I predators, apart from the coccinellids listed predators (Bl in Fig. 410 and Table 4.6). Still most of under Bl, are fairly Common on cassava where they them are common only on homopteran prey on other develop and reproduce on CM (D). A few parasitoids plants (B2). Mite-feeding beetles found commonly on were found attacking cecidomyiids of this guild (F). cassava seldom, if ever, feed on CM. Similarly, N'umerous general predators also are occasionally diaspine scale predators, sometimes found on cas­ encountered on CM-infested cassava (E); their exact sava, do not form a part of the CM food web. place in the CM food web is not known, but the spider Exochomus {lavipes group_ In Africa this group Argyrodes often fed on E. lopezi. includes several species distinguishable mainly by Dicrodiplosis manihoti. So far ail host records of genitalic characters. Whether some should have this species are from P. manihoti in Africa. If it is in specific or subspecific rank is not clear. Often the fact host specific; it must have been introduced from most important predators of P. manikoti, this species South America. Ifit is not specific, its original host in group sometimes greatly reduces mealybug popu­ Africa remains undiscovered. In Nigeria, it is the lations late in the dry season. But they seem to prefer second commonest predator ofCM after coccinellids. aphids to CM. Spa/gis lemolea. Even when abundant, it has a patchy distribution in cassava. It is recorded from P. manihoti, P. madeirensis, and F . virgata. Te~ ............, ,,.., Facultative predators. their parasitoids. and <, saprophages_S ome of the insects most abundant on cassava arrive after mealybugs have damaged t he I~JI '''PM po · ...t ~,. ,g. cassava. Numerous small spaces between distorted, p,f<_~' o.~.,;u <, crowded apical cassava leaves and stem furnish them a protected microhabitat. In the protected spaces one finds living and dead mealybugs, large quantities of mealybug wax, honeydew, dead or necrotic plant t issues, and saprophytic microorganisms that use the accumulated organic matter as a food substrate. It is a rich environment for saprophagous beetles (Il, together with Psocoptera. In the same microhabitat, certain common microlepidoptera are facultative predators of CM, and others are suspected of the facultative habit (G). They, in turn, often draw their own complex of parasitoids and predators (H). Parasitoids of mealybugs on cassava_ Apart Po, ••_ from the introduced host-specific CM parasitoid E. o,~,,_ t_ lopezi, about 12 species of parasitoids indigenous to Africa can be reared from mealybug-infested cassava tips (J). Most attack P. manihoti only occasionally, if at all, and are adapted to P_ madeirensis and F. virgata. Figure 4.10. Coccinellid predators of mealybugs, parasitoids of coccinellids. and other predators of mealybugs on cassava. Epidinocarsis lopezi. It never has been reared from 132 Root and Tuber Improvement Program Table 4.6. A list of predators and sapropbages associated with cassava mealybug in Africa Guild Distribution according Kame in Africa Fig. \4.12. Araneida Theridiidae A rgyrodes sp. Zaire* E Hemiptera Anthocoridae Cardiastethus exiguus Poppius Congo, Zaire· E C. pygmaeus Poppius .. . . .. . . . . . . ...• . . .. .... Nigeria* E Xytocoris afer ! day Mononycne//lls progresivlIs 7.---------------~------------------------------------------~ 6 5 4 3 2 , 25 50 75 100 150 175 200 225 PhysiolQ9icol time ( degree-days) 14.4°C ) Figure 4.13. Egg production by M. progresivU8 over physiological time is drastically reduced by temperature extremes. Root and Tuber Improvement Program 137 reduce possible edge effects. No damaged or obviously I - No mll' dcn;gqe 2 - <5 0/. c;t ~o' diseased plants should be included in the census. Vor"'~1 Aqe _ Fiola -======== chlorotIC 3- ) !5". bYt<50~ol '-at c./'Il~lw; 4. rvlite abundance is determined by examining the 0.. . 4 - )~.,.~ ~t ;eel chlorotIC first fully developed leaf of each censused plant. This 5 - LecfO$6t1ld leaf can be distinguished from the shoot leaves by its Pt7't Miles ptr legl [hnogI.! ~ Ph;..,! 114,1'5 pet Ie~ ~ sc;o,. darker color and from the older leaves by its petiole, ~p. (2!1 '" ;>2OCI (I - 51 sample <2:)" }200(1-5) ) 25,( 200 ) 2!>,(200 which joins the stem at a more acute angle ( < 90°). The northernmost terminal branch is selected when , • 2 more than one branch is present. The observed mite " (active stages only) density is recorded as one of four " density classes: 1 = no mites , 2 = < 25 mites per leaf, 3 = > 25 but < 200 mites per leaf, or 4 = > 200 mites per leaf. 5. Ifa leafless terminal branch is encountered while •,• estimating mite abundance, record 'Ileaf drop" when -- mite feeding activity is suspected as the cause. The next nearest plant should then be censused. This Figure 4.14. Cassava mite (CGM) population monitoring procedure should be repeated until 30 plants with data sheet. leaves are evaluated ifpossible. 6. The mite damage on each leaf evaluated for mite abundance is then scored on a scale of 1 to 5 as fields at the Texaco farm near Abeokuta, Nigeria follows: 1 = no mite damage. 2 = chlorotic (white or (Fig. 4.15). Even though we had not used the third yellowish) spots present on < 5% of total leaf surface, revision, the results corresponded closely for both 3 = chlorosis more severe, between 5% and 50% of fields (L plot : r = 0.82, p < 0.01, d.f. = 15; Igbo-ora the leaf area affected; occasional stunting of leaf plot: r = 0.52, P <0.05, dJ. = 15). growth, 4 = chlorosis very severe , > 50% of the leaf The results show that the censusing t echnique can surface affected, the entire leaf appears yellowish be used to accurately monitor CGM populations, and from loss of chlorophyll; stunted leaf growth, or 5 = that new census· takers can be trained with a rew leaf is dead or has dropped from mite feeding activity. hours of carefu I instruction. Once proficient with the 7. Calculate census statistics as follows: methodology, they can determine the incidence, Infested plants (%) = abundance, and damage symptoms of CGM for a number of plants with CGM single field in 45 minutes compared with four man­ days when actual counts are taken. number of plants censused (with leaves) Some examples of how this procedure was used are Damage score = sum of individual damage scores Mites per leo' 200,-----------------------------------, number of plants censused ABEOKUTA, NIGERIA [GOO - ORA PLOT Abundance = [(0 x A) + (10 x B) + (110 x C) + (350 x D)j Estimated counts 100 ,./ Actual COUt\'tS N = mites/leaf \ , ' ....' ~.. - -'" -- ...... ' I .... \ ... ~ where : o L-~L-~"~ __~ ~ ~~~ A is the number of zero scores ~ r-----------------------------__, B is the number of < 25 scores ABEOKUTA , NIGERIA C is the number of > 25, but < 200 scores L- PLOT D is the number of > 200 scores (note that A+B+C+D should = N) 200 8. To accurately track mite population fluc­ tuations, monitor fields every two weeks. We revised the procedure three times during the year. First in June when we increased the number of 100 plants censused from 10 to 20 per plot. Then in October we added a zero class to improve the estimate of CGM densities during the rainy season. In O ~~~~~-L~-L~~~~~~:s-~-~-~-~~~oJ December for the last revision we again increased the Jon Feb loAor Apr M~ Jun J~ A~ Sep Oct NoV' number of plants censused from 20 to 30. 1984 The census procedure was tested by comparing Figure 4.15. Counted and estimated CGM densities from actual mite counts with estimated counts from two two fields near Abeokuta. Nigeria. 138 Root and Tuber Improvement Program shown in Figures 4.16 and 4.17. A comparison of CG M C G M Populalion moniloring populations on different cassava culti vars planted at Number per leaf Score the same time in the same location reveal a common 400 ,-'----------.-------,4 trend in CGM infestation, while the magnitude of A infestation depends on the variety (Fig. 4.16). Figure 4.17 shows that the relationship between CGM abun­ 300 3 dance and CG M damage-symptom scores also is Leaf damage variety specific, thus symptom scores should not be used to estimate CGM abundance wit hout pnor 1', I , knowledge of the relationship.- J.S. Yaninek 200 _.. .. " Mitas I \ 2 / '\ :I \ Natural Enemies ofCGM '~, \ 100 , I \ Three species of phytoseiids were in culture at UTA during 1984, including Neoseiulus idaeus (received in '\...:.. '" ' ... ; May. 1983), Phytoseiuluspersimilis (received in May, 1984), and Neoseiulus anonymus (received in August, 1984). Cultures were maintained by modified stan­ dard methods. The operation involved coordinating B VARIETY ' 63397 three trophic levels : plant hosts for the spider mites, spider mite colonies as hosts for the predators, and 300 3 leaf damaQe the predators themselves. Alternative host plants and spider mite species , were screened for possible use in the production V 200 Mites I 2 cycle. Cassava. mucuna, and groundnut were selec­ ted as host·plant materia l while CGM, a red mite, , / Oligonychus gossypii, and the ubiquitous two· \, ,, ....." 100 , ., ,. ' ,,,... ' spotted, Tetranychus urticae (= telarius) were reared , ... '" ',.,', ~ .... ~ as hosts for the phytoseiids. Cultures of phytoseiids were maintained in o L-_--L_ _. ...L_ _. .L_.......J'--_--L_ _~ 0 temperature-controlled rearing rooms on dark water· Jon F. . Mor Apr Jun proof substrates, isolated in a petri dish by a water· 1984 saturated matt of cotton wool. Leaves from t he mite· Figure 4.17. Relationship between CGM abundance and infested plants were collected then brushed with a CGM damage symptoms on (a) TMS 30555 and (b) TMS specially designed mite· brushing machine to remove 63397. the spider mites. The mites were then added daily to the phytoseiid cultures as food for the predators. Cultures during the dry season after four weeks Biotic potential studies similar to those of CGM aver aged 2,500 individuals starting with 50 adult were initiated on N. idaeus. The developmental females . threshold was determined to be 15.2°C. Overall, this phytoseiid developed to adulthood faster and lived longer as an adult than its CGM host at each tempera· C GM Population monitOring ture (Fig. 4.18), but total fecundity (as determined for Mites per leaf CGM) for the predator was 39, 45, 48, 48, and 26 eggs per female at 20",24°,27",31°, and 34°C. IITA,MGERlA We initiated mass-rearing studies to develop tech· :lOO niques to expand the small cultures for large-scale releases. The first study was to examine the feasibility of developing colonies of phytoseiids directly on 63397 plants infested with tetranychid mites in open screen­ 200 house rooms. Two-week-old plants were infested with spider mites, then transferred to the phytoseiid rooms after two to four weeks. Phytoseiids were then added from the petri-dish cultures directly to the infested 100 plants. Two to three generations of phytoseiids were reared at ambient temperature before being col­ 30572 lected, plant and all, for local releases. Maximum oL-__~ ____- L_ ___- L_ ___~ ~~L-~ production during the dry season varied between Jon Mar Apr May Jun 50.000 and 100,000 phytoseiids every four weeks. 1984 Experimental releases of Neoseiulus idaeus were Figure 4.16. Dry season abundance ofCGM populations on made immediately after permission was granted by five cassava cultivars. the Nigerian Plant Quarantine Service at the Texaco Root and Tuber Improvement Program 139 Developmental periods, Neoseiulus liiows tested at. Kinsatu and Menkao. At M'Vuazi. TMS Temperature 30572, and TMS 30555 (introduced in tissue culture from IITA) were included. TMS 30572's yield is [[JEw noteworthy (Table 4.8) and its value as a parent, as u..o demonstrated in the advanced yield trial, is now Protonymph evident. Bulk II j47, 30441612, and Bulkl122 gave yields Dout""~ two to four times that of the local control and showed ---------------------=;~" good resistance to CMD and CBB. Advanced Yield Trial Two advanced yield trials , one with 15 clones and another with 14, were conducted without fertilizer at sites representing a range of soils commonly en­ countered in Bas-Zaire ; M'Vuazi, an alluvial, valley 20 30 bottom, sandy clay loam with good fertility and soil Oayo moisture; Kimpese, a clay on a highly eroded sa vanna Figure 4.18. Developmental periods of N. idaeus.like that of plateau; Kisantu, an upland sandy loam of moderate cassava green mites. were related to temperature but it fertility. Performances of the six clones selected for develops faster and Ii ves longer than the mites. further testing are shown in Table 4.9. Four ofthe six clones are offspring of the UTA improved variety TMS 30572. In trial 1, clone 30572/172 gave the farm near Abeokuta, t\igeria, after rains started in highest yield across locations. It also was the highest April, then later in July. Phytoseiids were recovered yielding in the 1983 preliminary yield trial.- N.M. for only a few weeks following releases. Seasonally Mahungu and F.E. Brockman appropriate releases were planned for Nigeria during the 198411985 dry season. Releases of N. idaeus and N. Demonstration oflmproved Varieties anonymus were also made during the year in Zambia In 1984, PRONAM continued workin g in collabo­ (October) and in Zaire (November). Results of those ration with agricultural development agencies to releases were not available by the end of the year.­ demonstrate improved varieties to farmers in Bas­ J.S. Yaninek and H.R. Herren Zaire and Bandundu regions. Demonstration plots were established at 120 locations in Bas-Zai re and 21 in Bandundu region. Results of 46 demonstrations Zaire National Cassava conducted with the National Fertilizer Program in Bas-Zaire are shown in Figure 4.19. At all fertility Program levels and at all positions in the landscape, the improved variety Kinuani outy ielded the local va· UTA is providing technical assistance to the Zaire riety an a verage of 37%. National Cassava Program (PRONAM) whose objec­ tives are to develop improved varieties and tech­ Multiplication and Distribution of nology, make them available to farmers , and provide Improved Varieties in-service and degree training to Zairean research and extension workers . Significant progress was Approximately 600,000 meters of planting material of made toward those objectives in 1984. the improved variety Kinuani were distributed in Bas-Zaire this year. A new multiplication farm of 100 ha was established near the M'Vuazi research sta­ Genetic Improvement tion. It is estimated that Kinuani, released in 1983, is The PRONAM breeding program is developing va­ now being grown on 850 ha in Bas-Zaire. In Bandundu rieties with high yield potential, disease resistance, region, approximately 75,000 meters of cuttings of and good consumer acceptance. One variety has been improved varieties were distributed.- S.J. Pandey, officially released; monitoring its performance in C. Burr, and F.E. Brockman farmers ' fields confirms its superiority. New promis­ ing varieties have been produced and are in advanced Entomology tests. Effective biological control of the cassava mealybug in Zaire began in 1982 with releases of a parasite, Uniform Yield Trial Epidinocarsis lopezi, imported from UTA. This year Uniform yield trials were conducted at M'Vuazi and additional releases were made and the parasite's Kisantu in Bas·Zaire and Menkao in Kinshasa re­ effectiveness was evaluated over large areas of Bas­ gion. The latter site, on the Plateau des Bateke, is on Zaire. Surveys also were made across vast areas of poor sandy soil (Kalahari sand), which is extensive in Zaire to determine the distribution and severity of the Kinshasa and Bandundu regions. Twelve clones were pest and direction of spread. 140 Root and Tuber Improvement Program Table 4.8. Performance of cassava clones in uniform yield tri.al at three locations in Bas Zaire and Kinshasa regions, Zaire, 1983--84 Fresh root yield, t/ha Resistance to: 1 Clones M'vuazi Kisantu Menako .Mean CMD CBB 30572 .. ............ , ........ ......... . 17.0 3.0 2.7 Bulk II147 ................... . ........... 15.4 25.9 5.1 15.5 2.7 2.4 30555 .............. . ....... .. . .. . .. . .... 14.1 3.0 2.5 30344}6}2 ........... . ........ . ........... 14.4 13.1 9.3 12.2 2.6 2.5 BulkI122 ..... . ............ . . .......... .. 14.7 14.0 8.0 12.2 2.8 2.0 Mpelolongi}4 .... . ..... ...... , ...... . . ... 10.1 18.6 4.0 10.9 3.1 2.2 30005;28. improved control .. . . •.. . , .. .•... 11.1 8.5 6.0 8.5 2.8 2.2 B 4911 , ............ ..... , .. ... ..... .. ... 13.0 8.5 3.9 8.5 2.3 3.9 Mpelolongi.local control . ... . . ........ ... 6.4 6.4 2.4 5.1 3.1 3.0 LSD,5% .............. . .... . . . . . . . . . . . . . 6.1 5.8 2.3 C.V .• % . .................... . .. .... . 34.6 37.8 29.7 'Resistance to cassava mosaic disease (CMD) and bacterial blight (CBB) was rated on a 1- 5scale. where I = no damage and 5 = severe damage. Table 4.9. Performance of cassava clones selected from advanced yield trials at three locations in Bas Zaire region, Zaire, 1983--84 Fresh root yield, t/ha Resistance to: I Clones M'yuazi Kisantu K impese Mean CMD CBB Advanced yield trial 1 30572}172 ................ . . .. . . ... . ... , . 22.7 12.5 20.5 18.5 2.3 2.6 30572}175 ....... ..... . .. . .. . .... . ...... . 23.4 9.7 14.1 15.7 2.9 3.5 30572}4 ....... ... .. . .. . . . .. ............ . 16.8 8.9 13.5 13.1 2.0 2.7 3OO85}28, improved .. . ............... .... . 16.8 7.2 9.8 11.3 2.7 2.4 Mpelolongi, local control .......... .• . .. .. 15.9 5.5 15.0 12.1 3.0 2.4 Trial mean .. . ....... . ....... ... . , .• . . . .. 14.0 8.9 13.0 2.6 2.7 LSD,5O;' ... . .... .......... . . . . .. . 7.2 3.5 4.9 C.V.,% ............. .. ....... . ......... . 36.1 27.7 25.9 Advanced yield trial 2 &~n ............. . ... . . . . . .. ... .. , .. 20.8 10.2 15.5 1.5 2.9 A 56/2 .. ..................... .... .. . .. . . 17.0 8.8 12.9 3.0 2.4 30572}79 ....... . ....... . ..... . .. ... . ... . 19.0 5.4 12.2 2.5 1.8 3OO85}28, improved control . . .. . , . .. , ... , . . 13.9 8.6 11.2 3.0 2.2 Mpeloiongi, local control ......... . ... . .. . 12.9 7.8 10.3 3.0 2.4 Tria I mean . . . ......... , . . .. . .. .. , . .. . 15.5 7.9 2.5 2.7 LSD,5% . .............................. . 6.1 4.0 C.V., % ............. .. . ...... . ......... . 27.5 36.2 'Resistance to cassava mosaic disease (CMD) a nd bacteria l blight. (CBB) was rated on a 1-5 scale. wher e I = no damage a nd 5 = severe damage. Distribution and Damage by Cassava Effectiveness of the Introduced Mealybug Mealybug in Zaire Parasite Until 1984, the country's vast size and poor transpor­ Intensive surveys in Bas-Zaire and Kinshasa in 1984 tation infrastructure prevented surveys of large revealed t hat the introduced parasite E. lopezi, interior areas. This year, regions of Zaire were made released by an IITA-PRONAM team in July 1982, had accessible with an airplane leased by the UTA become widespread. In contrast , its invasion of Bas· Biologica l Control Unit. Fleuve (eastern Cataractes and eastern Kinshasa) Information was combined with that from previous h as been slow, probably becau se cassava plantings year s to construct a distribution map. The n e w are not dense there. information extends the mealybug's known western E. lopezi's effectiveness as a biologica 1 con troJ distribution as far nort h as Mbandaka, a lthough agent was studied by comparing mealybu g popu­ infestations there probably are not contagious with lations in areas wher e E. lopezi is absent (Bas-Fleuve those in Bandundu. and East Cataractes) with t hose where it is present Root and Tuber Improvement Program 141 Fresh root ),ield (t/ha) Lukenke (damage l.I, infestation 10%) and Dikula _Locol (damage l.2. infestation 17%). At Ntumbwe, where ~ Kinuani ( Pranam) I LSD (0.05) damage and infestations were high in September, mealybug populations were already high when E. /opezi were released. Control by E. lopezi could not be expected under those circumstances. In summary. our 1984 evidence strongly indicates that E. /opezi is substantially suppressing cassava mealybug throughout the parasite's wide area of distribution in Bas-Zaire. Results of initial releases in Shaba are equally encouraging. A concentrated effort will be made in 1985 to bring this successful biological control agent to regions of Zaire still threatened by cassava mealybug-Bandundu and Kiwu.- R.D. Hennessey and Muaka Toko Pathology Stem-tip Dieback Fertilizer rate {kQ / ha N,P.Os-,K10) Position I--Volley ---t I-- Slope ----4 r- PtateoJ---i This cassava abnormality was reported at M'Vuazi No. or locations 10 7 29 for the firs t tiroe in 1980. Death of tertiary branches Figure 4.1 9. Yield of improved variety Kinuani and a local during the dry season and decay of tuberous roots are variety at t.hree fertilizer r ates in three landscape positions symptoms. Although the causal agent is not yet (Bas·Zaire 1983-84) established , a fungus belonging to the genus Sclerotium was repeatedly isolated from decayed roots. (West Cataractes). An average of about six tips/field were examined, although the number varied from 1 to Effect ofPotassiurn Fertilization 20. Mealybugs of the 2nd, 3rd, and 4th instars were A previous trial indicated that potassium lowered the counted. incidence of stem·tip dieback. To further study tha t Data on mealybug populations in Shaba, where E. effect, we applied potassium sulfate to cassava plots /opezi were released in 1983 and 1984, are presented in at 0, 50, 100, 150, and 200 kg/ha, and a basal appli. Table 4.10. In the area of releases between cation of 75 kg N/ha and 50 kg P ,03/ha. We then Lubumbashi and Mulungwishi, mealybug infes­ recorded the incidence (percentage of affected plants) tations have been severe since the first years of and intensity (percentage of affected branches) every infestation (1978- 1980). Late dry season samples 15 days for two months after symptoms first appeared . generally are high (damage> 3.5, infestation> 80,%). At harvest, we recorded tuberous roots and yield of In other fields damage is low to moderate (2- 3), and marketable roots. infestations are high. High infestations occur after As K,O increased up to 150 kg/ha, dieback in­ mealybug populations build up and shoots collapse. cidence and percentages of decayed tuberous roots Soon after the beginning of regrowth young shoot.s decreased significantly (Table 4.11). Increasing K,O suffer li ttle damage, but a large proportion of them to 200 kg/ha seemed to increase fungus and decayed are infested, so fields with both low damage and low tuberous roots, which reduced the yield of market· infestation would be evidence of biological control. able tuberous roots. Two of the E. lopezi release sites fit those criteria- Using cutt ings from healthy and diseased plants, we found no difference in stem-tip dieback or per· centage of decayed tuberous roots between plants Table 4.10. Status of mealybug at Epidinocarsi{; /opezi from cuttings from affected plants and those from release sites in Shaba region , and in cuttings from healthy plants. Symptoms were not neighboring field (Lukenke, 10 km E.) expressed on newly developed shoots from recipro­ where no releases were made cally grafted cuttings from diseased or healthy pla­ Damage Incidence of nts. The plants will be further observed in the field. Date of in Sept. infestation in Sta tus of Effect of time of planting. Susceptible clone Locality release 198·1 Sept. 1984, ~b E. lopezi 02864 was planted monthly from November 1983 to Lukenke . 12-07·83 1.1 10 abundant April 1984. Incidence of stem·tip dieback, recorded Ntumbwe . 25-05-84 3.5 97 abundant during the 1984 dry season , was lower in later plant­ Kime .... 25-05-84 2.2 83 scarce ings. No symptoms were observed in the April plant­ Dikula .. . 25-05-84 1.2 17 abundant ing (Table 4.12). Stem-tip dieback was observed IOkm E. of Lukenke . no release 3.9 83 absent primarily in the dry season with no increase after rains began in October. 142 Root and Tuber Improvement Program Table 4.11 . Effects of potassium fertilizer on stem-tip dieback, decay of tuberous roots, and yield, M'Vuazi, Zaire, 1984 Yield , marketable Decayed tuberous tuberous roots. Alfected plants at day roots , % tlha Yield loss , % Healthy Alfected Healthy Alfected hea1thy vs . K,O rate, kgjha 15 30 45 60 plants plants plants plants a ffected, % 0 . .... .. . .... .. .. . .. . .. 10.4 46.6 36.2 67.6 27.0 40.1 18.9 10.1 47 50 .. . . ......... .. . . .... . .. 12.0 41.3 61.2 68.2 15.7 34.3 19.3 12.3 36 100 .. . · ·. 0.· ••. · .. ·· • • · ·· •· . 2.3 18.0 40.3 48.6 13.6 24.0 20.4 15.5 24 150 . . . ...... .. ........ ...... 1.2 11.9 23.0 27.7 12.8 25.7 25.7 18.0 30 200 . . . ... .. . . - . . .. . , ,_ . 2.1 17.0 35.5 43.4 20.5 25.8 li.3 14.8 17 LSD.S% ........ . . . . . . . . . . . . . 2.8 4.0 7.3 7.2 8.9 6.9 NS 3.4 Table 4.12. Effects of planting date on stem-tip Uninoculated plants served as the control. We coun· dieback (% or plants) observed at ted the edible leaves on the uppermost 25 cm of stem indicated July-8eptember 1984 dry-season tips at three-week intervals for three months. dates Production of edible leaves by uninoculated plants Planting 17 7 22 6 of both clones was similar. Inoculated plants of the date July August August September tolerant clone, Kinuani , and by the uninoculated Nov 83 .. . ... . ... . 6.0 40.0 49.4 15.2 control did not differ inedible leafproduction, but the Dec 83 . . . . . . . . . . . 13.0 40.4 46.6 48.5 susceptible clone, A56, 12 weeks after inoculation Jan 34 . ... ... . . .. . 12.0 26.9 36.4 38.7 produced 62% fewer leaves than the uninoculated Feb 34 . ... 4.7 25.4 34.9 37.8 ones. PRONAM variety Kinuani obviously resists ~ar34 ... .... . .. . 1.8 7.2 9.3 10.1 eBB so that, even under high eBB disease pressure, Apr 34 . . . ....... . 0.0 0.0 0.0 0.0 its production of edible leaves is not appreciably LSD,S"/., ......... NS' 24.4 29.0 31.5 reduced.- K.A . Muimba INS = >Jot s ign ificantly different (0.05). Crop Management Anthracnose Effect of Agronomic Factors on Improved A study was conducted to determine the importance Varieties of cassava plant debris as a source of inoculum of In developing improved varieties, we tested their Colletotrichum gloeosporioides f. sp. manihotis Henn., performance under more or less uniform conditions of causal agent of cassava anthracnose disease (CAD). crop management for standardization and practi­ Stems of clone 30122/2 and Mpelolongi with CAD cality. Crop management practices used by farmers , cankers were held at 25.3 to 26.5°e and 77 to 84% RH. however, may differ in one or more important factors Samples were taken monthly for isolations on potato from those used in tests. For that reason we compared dextrose agar to determine if the pathogen was the variety Kinuani with the local variety viable. C. g/oeosporioides f. sp. manihotis was isolated Mpelolongi under two cropping systems (monocrop­ from the cassava debris up to six months of storage. ped or intercropped with groundnuts) with and with· After seven months, saprophytic fungi made it im· out applied fertilizer (600 kg/ha of 17·17-17) on a possible to further isolate the pathogen. It thus relatively impoverished soil at 10,000 or 20,000 appears tha t cassava plant debris left in the field after plants/ha with or without leaf harvesting. A con· harvest may be a n important inoculum source. founded design with two replications was used. Root yields of the two varieties did not differ and Bacterial Blight root yields were not affected by leaf harvesting or by Cassava lea,'es constitute an important source of plant density. Intercropping with groundnuts reo protein for much of Zaire's population, but cassava duced the yield of both varieties, but Mpelolongi was bacterial blight (CBB) reduces the number and reduced 33%, Kinuani 64%. Kinuan.i has slow early quality of leaves. We studied the effect of eBB growth. Kinuani yielded more tban Mpelolongi only on production of edible leaves by a CBB-susceptible as a monocrop under high fertility . clone (A56) and a tolerant variety (Kinuani) by inoculating plants three months after planting with Response to NPK two stem injections of a bacterial suspension between The trial, initiated in 1982 at the request of the the third and fifth internodes of each branch and by National Fertilizer Program, to determine response abrading the five uppermost leaves and stem tips in of cassava to the three major nutrients was con· the presence of the bacterial suspension. tinued. Nitrogen at 0,75, and 150 kg/ha; phosphorus Root and Tuber Improvement Program 143 at 0, 50, and 100 kg P 20,/ha; and potassium at 0,100, what results could be expected from herbicide treat­ and 200 kg/ha were again applied in a 3 x 3 factorial ments. We used glyphosate + NSMA to control l. experiment using a confounded design with two cylindrica, and fiuometuron and metolachlor for replications. residual control of other weeds. This year, in contrast to the previous year, ferti­ Although we followed the standard recommen­ lizer significantly affected root yield. The first incre­ dations for glyphosate (250 liters spray solution/ ment of potassium increased yield 39% (Fig. 4.20). But hectare on 75 cm regrowth after slashing; no rain for there was no significant response to nitrogen or 24 hours after application; plowing three weeks after phosphorus. A response to potassium was also obser­ application), control of Imperata was poor. At the two ved in the yield of planting material; 100 kg K 20/ha lower rates of glyphosate, Imperata biomass at har­ increased multiplication 76%. Again this year, phos­ vest was greater than in the untreated plots. phorus did not affect yield of planting material. Apparently early suppression of other weeds by Unlike last year, there was no response to nitrogen. fiuometuron and metolachlor allowed Imperata to make more growth than without glyphosate. NSMA Plant Residue Management in combination with glyphosate at the low rate Plant residues often are burned before land prepara­ greatly increased Imperata control. Fluometuron and tion. Alternatively, they may be buried when making metolachlor did not provide residual control of other ridges. A common method of handling residues in weeds long enough to allow cassava to become Bas·Zaire, called mafuku, is to burn the residue under competitive with them. As Imperata control in· a covering of soil where combustion is not complete . creased, growth of other weeds increased. With five A trial was carried out to compare cassava yields hand weedings, there was no Imperata top growth at under the three methods of residue management. cassava harvest, and other weeds were adequately Superimposed were various fertilizer treatments. The controlled.-K. Landu, N.B. Lutaladio, and YE. residue was a grass fallow composed primarily of Brockman Hyparrhenia diplandra. Withou"t fertilizer, the mafuku method gave the Yams highest yield (Table 4.13). With that method only N increased yields. With residue burning, both Nand Virology PK increased yields. The response to PK in burned plots but not mafuku plots may be because ash was Yam Mosaic Virus washed away by the first rains . Burial of plant residues gave lowest yields with all fertilizer treat· Dioscorea rotundata and D. cayenensis are apparently always infected with a potyvirus, which can be ments, probably because nitrogen is immobilized, as phosphorus and potassium were not released by transmitted mechanically by sap inoculation and by combustion. aphids in a nonpersistent manner. The virus was t ransmitted to Nicotiana benthamiana, which showed Weed Control initial vein clearing followed by systemic mottling. To provide a faster, more reliable indexing method PRONAM cultivates about 100 ha of cassava in Bas­ for yam mosaic virus, direct double-antibody ElJSA Zaire to produce planting material of improved was developed. A gamma globulin coating of 1.0 )lg varieties. Adequate control of weeds, especially protein/ml and a conjugate dilution of 1/250 gave the Imperata cylindrica, by hand labor has been expen· best results. In tests of several buffers for extraction si ve and difficult. We conducted a trial to determine or dilution of the antigens, the best results were Fresh root yield (tlha) Plontinq material (km/ho) 16,--------, Table 4.13. Effect of plant residue management and fertilizer on cassava fresh root yield 14 M' vUBzi, Zaire, 1984 Plant residue Fertilizer treatment, kg,lha N-P 20~-K20 management 0·0-0 60-0·0 0·45·45 60·45-45 Mean Buried .... . 11.4 14.1 13.1 16.6 13.9 Burned .... . 13.9 16.7 15.4 19.2 16.3 10 Ylafuku .... . 17.4 20.0 16.8 19.2 18.3 Mean ...... . 14.2 16.9 15.2 18.3 16.2 LSD. 50;, Residue management means .......... . 3.6 100 200 100 200 Fertilizer means ..................... . 1.1 K2 0 rat. ( Kg! ho ) Fertilizer means with same residue management. 1.9 Residue management means with same fertilizer Figure 4.20. Effects of potassium on cassava root yield and treatment ............................... . 4.0 production or planting material, M'Vuazi , Zaire. 1984 144 Root and Tuber Improvement Program obtained with 0.25 111 potassium phosphate buffer (pH block design with four replications was set up to test 7.5}containing 0.1 MEDTA and 0.25% sodium sulfite. effects of 0, 50, and 150 kg/ha of 50·50-50 on three yam When serial dilut ions of the sap from infected D. lines. The fertilizer was applied equally three times, rotundata leaves were tested by ELISA. it was pos­ at planting and two and four months later. Planting sible to detect the virus up to a dilution of 11250, and harvesting dates were in April and October. With depending upon the age of the leaves. Concentrations increasing amounts of fertilizer . disease severity of virus in leaves vary considerably and seem to be increased in all three lines and yields were reduced highest in very young leaves. Older leaves collected (Table 4.15). In this test, using fertiliz er on water­ from diseased plants ga ve inconsisten t resu Its. yams infected with water·yam chlorosis increased Samples from tubers were also tested by ELISA, disease severity and consequently reduced yields. which detected the virus in extracts. Further work is Water·yam chlorosis likely results from a virus being done to perfect the detection method so it will belonging to the potyvirus group of filamentous produce consistent, reHable results and permit rapid rods.-R. Theberge indexing ofplants.-G. Thottapp illy Biochemistry Pathology Bitter Compounds in White Yam Chips Sclerotinial Rot Chips from eight varieties of D. rotundata (Omi , A major constraint to yam "seed piece" or minisett Jioku, Ekpe, [yawo, Iyo, Ukom, Asoko, and Aga) were production in 1984 was Sclerotinia rot caused by the prepared by frying slices 1.5 mm thick in vegetable or fungus Sclerotium rolfsii Sacco The fungus survives ground nut oil at 165°C for three minutes. All chips well in soi Is as a saprophyte on plant debris. The were bitter, unacceptable to the taste panel. pathogen readily invaded both water (D. ala to L.) and Of four methods tried to remove the bitterness from white (D. rotundata Poir.) yam species. the chips, two succeeded: blanching slices in steam A randomized complete block design was estab­ three minutes, drying them, then frying , or boiling lished consisting of 16 treatments with five repli. slices in distilled water three minutes, drying, and cations to determine which combination of four frying. The two successful methods indicate that the fungicides (includ ing a control) most effectively bitter compounds were hydrolyzed in high tempera· controlled Sclerotinia r ot. The fungicides used and ture water. They may be glycoalkaloids that are not their concentrations were: Aldrex T (formulation soluble in cold water but easily hydrolyzed in hot unknown) 1 g!120seed pieces, Demosan (Chloroneb)6 water .- Nell Mondy. Divisionof Nutritional Sciences. g{iiter, Difolotan (Captafol) 10 g/liter, and TectO 5% Cornell University dust (Thiabendazole) 1 gl120 seed pieces. Several fungicide combinations effectively con· Yams (D. rotundata. and D. alata) trolled Sclerotinia yam rot. Aldrex T and Demosan gave the best results. Two curing methods of treated Meristem culture media for t he two yam species (D. yam seed pieces were also studied. They included rotundata and D. alata) were developed in 1983. It is possible to index D. rotundata for yam mosaic virus shading the treated seed pieces in a conventional yam using the inoculation method and ELISA test , so barn and placing them on the soil surface andshading international distribution of D. rotundata could be with palm fronds. Fewer rots resulted fro m seed feasible in the near future. pieces cured in a traditionally shaded yam barn Multiplication of yam using model segments was (Table 4.14). best in Murashige and Skoog's medium with 20 gi l Water Yam Chlorosis sucrose, 20 mg/l L·cystein, 0.5 mgll kinetin with or without 0.5 mgll NAA. To study photoperiod effects An experiment conducted during the 1984 growing on multiplication rate, we test ed three selected yam season tested whether or not increasing amounts of clones, Abi and Nwapoko for D. rotundata and TDa fertilizer affected water yam chlorosis (yam scorch). 251 for D. alata . Highest multiplication rate was which often causes severe scorching of leaves and vines of Dioscorea alata L. A randomized complete Table 4.15 Mean severity (% ) afwater yam chlorosis (scorch) by yam line and fertilizer treatment. I Table 4.14. Percentage of rotted seed piece based on Ferti lizer, kgiha curing method Yam line o 50 150 Curing Method Rots. % TDa204 ..... .... ............ . 94.3 97.9 99.8 Yam barn .... .... .. .. .. .. .. .. .. . 18.0 TDa 297 . . ..... . . .... ....... . . 8.4 13.8 29.1 Seed bed. . . . . . . . . . . . . . . • . . . . . 49.6 TDa 310 ..... .... ... . ....... . . 66.1 89.6 92.8 S.E .. . . . . . . . . . .• . . . . . . .. . . . . . . . . . . . . . . 2.3 LSD.5% .. . . . . • . . . . . . •• . . . . . . . . . . . . . . . • . 6.4 1 Yam line, fertilizer , and their interaction were h ighly significant. Root and Tuber Impro vement Program 145 about 24 hours light by Abi, 16 hours for Nwapoko. Sweet Potatoes and 12 hours fo r TDa 251. We a lso studi ed e ffects of physical status of ell Iture media on the multiplication rate, using a solid med· Genetic Improvement ium . a stationary liquid med ium . and a liquid medium A lotal of 28 sweet potato clones in vitro have been on rotal"Y shaker at 70 rpm . Both Abi and Nwapoko's sent to 52 countries throughout the world. TlS 2544 multiplication rates were highest when grown on consistently performed well in Nigeria. Cameroon. liquid medium on rotary s haker. Rwanda . Sudan. and Kenya. It also performed we ll at A white yam. TDr 820. and a water ya m. TD a 25l. high altitude. and it seems to be adaptab le to a wide were transp lanted from in vit ro materi als to the field. range of environmental comli tions. They first were transp lanted to jiffy peat pelle ts in a Variety TIS 2498 performs better at low altilUde humidity c hamber. then in potting soil in plastic bags than high altitude conditions. TIS 70357 has shown kept in t he s hade. a nd finally in April planted in the consistently high and stable y ie lds in Nigeria. along fie ld with plastic mul ch. Harvest was in November with res ista nce to weevi l and the virus complex. Its 1984 with TDa 251's yield promising. Three or four dry matter percentages range from 35 to 40. TIS 8266 tubers of seed ya m size were harvested from each also has performed well in Nigeria. plant (Fig. 4.21). TDr 820's tubers were LOO s mall for Improved sweet potato c lones from !ITA (Tlb I and seed yam (Fig. 4.22). 527034) used in 430 farm-managed trials in Cameroon Cocoya m ( X sagittifolilllll ). More than 100 clones yie lded two to five times more than the best available of Xanthosoma sagittifolium were maintained in vitro loca l varieties and the improved clones' taste rat in g using meri stcm culture. S. Y.C. Ng was "good" to "very good." Variety Trials Ten sweet potato c lones were tested at three lo­ cations in Nigeria: II TA. in a region of moderate rainfall and soil fertility: Onne. in the high rainfall zone with sandy. poor soil; and Mokwa. in the dry sava nna zone. The results are summarized in Table 4.16. The wealher did not favor sweet potatoes in 1984 so yie lds were lower than in previous years. However. severa I c lones outy ielded TIS 2448 under the unfavor· ab le conditions. TIS 8250 s howed hi ghest fresh yie ld with 10.2 tlha but its low dry ma tter percentage gave it a low average dry y ie ld of2.lt/ha. TIS 70357 consistently performed well , particularly in terms of dry yield with its highest dr~· matter percentage 0[35.5. Also its weevil and virus damage scores were low. TIS 8266 a lso performed we ll over Figure 4.21. Tubers from D, alata. TDr 820: three exceed 9 locations and with high dry y ie lds and a low weevil­ inches damage score. S.K. Hahn, E. Parh. and E. Chuhwuma In tegr a t ed Contro l of Sweet Potat o W eevils Sweet potato weevils (Cylas spp.) are a major con­ straint to sweet potato production in Africa .. evera l species (C. formicarius. C. punc/ical/is. C. brunlleus) damage leaves. vines . and tubers in the field and in storage. Early planting and harvesting and earthing­ up tubers s ignificantly reduced weevil damage. Hi gh y ie lding. good qua lity clones with mod erate re­ sistance ha ve been deve loped and their resistance is be ing improved. But weevil damage to stored tubers remains serious. A few insects multiply rapidly and cause tremendous dam age. so a technique is needed to preven t. weevi I damage to stored tu bers. Tn an ex periment with four replications to study Figure -l.2:2. Tubers from D. rotll"dala. TDr 820: five exceed how time of planting and harvest ing afrects weevil 7 inche~ damage. half of each replication was harvested four 146 Root and Tuber Improvement Program Table 4.16. Results of sweet potato uniform yield trials, 1983--84 Sweet potato Fresh yield, t/ha Dry matter, Dry yield, Weevil Virus c lone !ITA Onne Mokwa Mean % t/ha score l score l TIS 8250 . .. . . . . .. .. . .. ............ .. 20.3 5.5 4.9 LO.2 21.0 2.1 0.5 1.3 TIS 8441 . . .. .. .. .. ...... .. .. .... ... 17.9 7.3 4.6 9.9 28.9 2.9 0.6 0.6 TIS 8266 .. .. ...... ..... .. .. ........ . 15.2 7.3 4.6 9.0 32.2 2.9 0.3 0.5 TIS 8504 .. .. ........ .. ....... .. .... . .. 18.0 4.8 4.2 9.0 25.5 2.3 0.6 1.3 TIS 8524 ...... . ............ ... ..... . .. 15.7 5.7 5.5 9.0 28.0 2.5 0.7 0.3 TIS70357 .. ... .. ...... .... . .. ......... 12.3 7.9 5.9 8.7 35.5 3.1 0.2 0.2 TIS 9465 . ...... . .. .... ...... . .. . .. . ... 14.1 5.1 5.8 8.3 30.0 2.5 0.6 0.0 TIS 8509 . . .. . . . .. . .... .. . . . . ....... . 14.3 3.4 7.0 8.2 30.0 2.5 0.6 0.1 TIS 2498 ..... . .. . . .. • ... .. . 14.3 4.4 3.2 7.3 33.0 2.4 0.6 0.4 Tlb 4 (check) ...... .. . . . .. , . . ... . . . . . 9.5 1.8 2.0 4.4 29.0 1.3 1.5 1.2 S.E. ..... . .. . . . . . . ....... . . . . . . . . . . . .. 1.3 0.8 0.8 0.6 I.1)D. 50/0 ... . ...... . ..... . . . . . .. .. . . . . . 3.6 2.2 2.2 1.7 'Weevil and virus damages were evaluated subjectively on a 0 to 5 scale where 0 = no visible symptoms and 5 = severe symptoms. months after planting and half six months after to healthy plants of the SPVD test clone, TIb 8 s.c. 9- planting. We used one susceptible clone, TIb 4, and H, and maintained for further studies. When the five three tolerant clones, TIS 21>32, TIS 2534, and TIS plants (designated Tlb 8s.c. 9-W.F. 2 through·W.F. 6) 70357, and recorded the number of feeding holes in were grafted to cuttings of TIb 8 s.c. 9-A. t he SPVD tubers selected randomly at harvest. Results (Table test clone preinfected with the aphid· transmitted 4.17) indicate that early planting and early harvest· component, three of five series showed prominent ing reduced weevil damage. Sweet potatoes planted systemic vein chlorosis and distortion symptoms. But as early as possible during the rainy season and only "fan· leaf" symptoms developed in two of the five. harvested before the dry season (when weevil popu· Characteristic systemic necrotic spotting did ' not lations build up) will be at least somewhat protected occur in any of the five series. It thus seems that when from severe weevil damage. The r esu Its on late compared in combination with one fixed aphid· planting and storage are still being co\lected.- K.M. transmitted component, there are various types of the uma and S.K. Hahn whitefly·transmitted component of SPVD and they may occur together in one plant.-H. W. Rossel Pathology A new fungal problem appeared in several sweet The Whitefly-Transmitted Component of potato plots at lIT A during the 1984 growing season. SPVD Foliar disease symptoms of various severities in· The nature of the whitefly·transmitted component of itially appeared on young leaves as chlorotic areas sweet potato virus disease complex is still not known between veins. In time, the chlorotic areas became with certainty, although at IITA long, fragile filam· brown, lesions coalesced , necrotic areas became entous particles have frequently been observed with detached and left tattered leaves. The causal agent the electron microscope. In an effort to confirm this was identified as a fungal species of Curuularia, and finding and reveal the true nature of the agent, the soil found deficient in manganese (Mn). When Mn attempts were made to obtain purified virus from I . was applied to the soil, chlorotic areas no longer setosa plants that were inoculated by viruliferous developed. But where Mn was not added to the soil, whiteflies. But since these plants did not show clear the symptoms and the disease persisted. The fungus symptoms of virus infection, even after whitefly Curvularia sp. in vaded sweet potato plants weakened transmission, all inoculated 1. setosa plants were used by Mn deficiency.- R. Theberge for purification of the agent. Virology Table •. 17. Weevil damage in early and late harvested sweet potato tubers compared at harvest Sweet Potato Virus Disease Average number weevil Studies of the whitefly·transmitted component of feeding holes/tuber sweet potato virus disease (SPVD) in Nigeria have Clone Early harvest Late harvest shown it can be transmitted by Bemisia tabaci only TIS 2532.. . .. .. . .. . . . .. . . . . . . . 1.95 17.80 with difficulty and apparently only from sweet potato TIS 2534... . . ... .. . . . .. . . .. .. . l.05 15.00 plants that also contain the second component of the TIS 70357 . . .. .. . .. . . . .. . .. . . . . 1.97 15.20 disease, which is transmitted by aphids. Tlb4 .. . .. .. ...... . . . . .. 1.89 34.60 Five Ipomoea setosa plants of six that became 14 months after planting. infected after inoculation by whiteflies were grafted 26 months after planting. Root and Tuber Improvement Program 147 Leaf and stem tissues were ground in 0.25 M Biochemistry potassium phosphate buffer (pH 7.5) containing 0.25% sodium sulfite, 0.01 M EDTA, and 0.2% poly· Total Sugars in Fresh Sweet Potato Tubers vinyl polypyrrolidone, and passed through a double Sweetness in sweet potato tubers depends on soluble layer of cheesecloth: 6.5% n·butanol was added to the sugar contents. In Africa, clones that are not sweet sap, which was stirred for 2 h and then centrifuged at are preferred. To assist breeders in detecting clones 5,000 rpm for 10 min: 8% PEG was added to the with low total sugar concentration, we extracted supernatant. The pellet was collected by centri­ soluble sugars with 80% ethanol and analyzed sugar fugation and resuspended in 0.02 M phosphate buffer contents by the phenol sulfuric acid method. Figure (pH 8.2) containing 0.1% mercaptoethanol (PB). 4.23 shows total sugar percentages in fresh sweet After low-speed centrifugation, the supernatant was potato tubers from 1983 and 1984 yield trials. In 1984 centrifuged at 25,000 rpm for 2 h in a R-35 Beckman concentrations ranged from 2.2 to 5.4% with a 3.9% rotor. The pellet was resuspended in PB as before. median. IR 1983 concentrations ranged from 2.0 to Cesium chloride was added to this preparation to a 8.8% and a 5.0% median .-A.M. Almazan concentration of38% (w{v) and centrifuged for 18 h at 30,000 rpm in a SW·41 Beckman rotor. A light but Tissue Culture clearly visible band in the middle of the tube was collected and examined by electron microscope. Twenty-four lines of sweet potatoes were available Filamentous particles slightly longer than those of for distribution during 1984; 47 packages of tissue the aphid-transmitted component, a potyvirus, were culture materials were distributed to 22 countries. observed. Further work is being done to confirm this Six additional lines were certified by the Nigerian finding and to elucidate the etiology of the whitefly­ Plant Quarantine Service. transmitted component.-G. Thottappilly and H. W. Yield performances ofl ITA clones in several count· Rossel ries are reported to be two to four times higher than A Possible New Potyvirus in Sweet local variety yields. TIS 2544 was the best yielder in Potatoes several countries. A potyvirus was isolated (by inoculation to N. ben­ An additional 400 sweet potato lines were added to thamiana) from a diseased sweet potato that showed our in vitro germplasm collections. Most came to lITA mild vein yellowing symptoms and had originally in seed form from elsewhere.-S. Y.c. Ng been collected from a farmer's field near Yandev, Nigeria. This potyvirus seems to differ from isolates of the aphid-transmitted poty·component of SPVD that had earlier been obtained by inoculation of N. benthamiana at IITA. Since the newly isolated virus Number of clones does not seem to be transmissible to the SPVD test clones, TIb8 s.c.9-A and .W.F. , we do not know whether it is a new and unrelated virus or a new type 20 of the poty-component of SPVD . In 1. setosa the isolate appears to produce symptoms identical to those of isolates obtained earlier by inoculation of N. 'E"~;'H' 1983 benthamiana. When plants of this recently obtained 11 1984 diseased clone from Y andev. JI,i igeria, were grafted to 15 healthy test plants ofthe SPVD test clone (TIb 8 s.c. 9· H), severe systemic chlorosis and necrosis symptoms developed. followed by extensive leaf abscission, a symptom that is known to occur with a certain " ! combination of components. The severity of symp­ 10 toms in the test clone suggests that the newly col­ lected. diseased sweet potato clone contains anum· ber of different types of the potyvirus component. one of the two components ofSPVD. Further characterization work is being done to 5 elucidate the relationship of the newly isolated potyvirus to the previously obtained isolates of the aphid·transmitted potyvirus component of SPVD. The new isolate is being maintained and propagated in N. benthamiana. The symptoms in this plant. being o very mild and hardly visible, differ greatly from 2.0 4 .0 6.0 8 .0 isolates of the potyvirus component obtained. earlier Percenl in this test plant species.- H. W. Rossel and G. Figure 4.23. Total sugar in sweet potatoes. 1983 and 1984 Thottappilly crops. 148 Root and Tuber Improvement Program Nematology Table 4.1B. Plant-parasitic nematode soil populations under soil regeneration fallow of Leucaena leucocephala Leucaena leucocephala Fallow to Control Years fallow Nematodes l Nematodes sleight Nonpara- Root· Root· IITA sites sitic Z Spirai3 lesion4 knotS Total Plant-parasit ic nematodes are common in soils of farms in the tropics. Except for what they do as plant 12 .. . ... . . . .. 12,500 47.0 47.0 parasites. nematodes are se ldom noticed because 8 . . . ... . . . . . 1,318 26.0 26.0 8 .. . .. . .... . . 2,458 3.0 3.0 they are small and colorless . Virtually every plant 4 .. . . . . . . . -.. 3,625 3.0 3.0 and crop has nematode parasites and some nematodes 3 .. . ...• . .. . . 5,420 0.4 0.4 are parasitic on many crops. So plant-parasitic n e­ 3 .. . ... • . .. • . 4,990 12.0 0.4 12.4 matode populations may increase year after year to 3 .. . . . .. . .. .. 4,775 1.2 0.8 0.4 2.4 the point of damaging crops. Nematodes are affected 3 .. . .... . . .. . 4,853 0.4 29.0 29.4 by the interplay of such environmental factors as soil ~lean .... . .. . . 4,992 10.1 5.2 0.1 15.5 texture, fertil ity, moisture, temperature, and the crop's resistance or susceptibility. I Number of nema todes/liter of soil. based on 30 lOO-cm:'l soil We are intensively evaluating alley cropping and samples. 2Number of nonparasi tic nematcxles in soil is an indication of soil regeneration with leguminous shrub species at biotic activity. They typically ranged from 10,000 to 3D,COO/liter of llTA. Alley cropping is growing a crop between shrub soil , (Prefallow means of plant· parasitic nematode so il borders that are trimmed and kept pruned during the populations under rice. maize. and cowpea ranged from 3,000 to food crop·growing period . The trimmings, left on the 24,OOOfli ter of soi L) :IS pira l nematodes = Helicotylenchuspseudorobustus, H. soil surface, serve as mulch and nutrient sources. The erythrinae. leguminous tree, Leucaena leucocephala, is being "Root-lesion nematod es = Pratylenchus se{aensis . P. brachyurus . used in alley cropping trials and for soil r egeneration. eRoot-k not nematodes = ,'I,leloidogynespp. Juveniles. Observations were made on plant· parasitic nematode soil populations after at least three years' growth at Table 4.19. Plant.parasitic nematode soil populations eight IITA sites. The sites, previously planted to under a Leucaena leucocephala-based alley cowpeas, rice, or maize, contained soil populations of cropping system 3,000 to 24,000 root·knot, root·lesion, and spiral Treatment Nonparasitic Spiral Root·lesion nematodes per liter of soil. Soil sample analyses showed drastic reductions in Leucaena ratooned ...... 9,000 123 4 nematode populations under L. leucocephala fallow Leucaena·rnaize alley cropped . . . .... .. ... . . 7,120 186 49 (Table 4.18). Where the fallow was converted to alley Maize . ....... . . . . . . . . . . 8,438 1,170 4,551 cropping, the parasitic nematode population reo mained low in th e crop year (Table 4.19), which indicates that cropping systems based on a planned L. leucocephala fallow will control plant· parasitic nemat.odes.- F.E. Cave ness Root Crops Research in Cameroon Nematode-destroying Fungi A study during the 1984 growing season was designed CNRCIP 1984 to isolate and identify nematode-destroying fungi. The Cameroon National Root Crops Improvement Soil samples and/or litter was collected from several Program (CNRCIP) aims to increase yields of root IITA locations. Several recovery t echniques were and tuber crops in production systems suitable for tried and the method whereby corn meal agar (CMA) low·resource farmers in Cameroon. The job involves petri plates-overrun with either Pythium or identifying and incorporating di sease and insect Phytophthora spp.-were baited , proved to be the resistance into root crops that yield well, have high most successful. nutritive quality that consumers accept, and that ar e For convenience. nematode-destroying fungi were adapted to the ecology. Also needed are cultural classified either as predatory or nonpredatory. Fungi practices suitable for low management farmers to belonging to the former group develop extensive maximize returns of improved or local root crop hyphal network systems on which at various lengths, varieties with minimal inputs. trapping devices are produced, while the latter group Another important CNRCIP task is training produces spores on fruiting structures emerging Cameroonian researchers and technicians for a con· directly from infected nematodes. tinuous crop improvement program. Genera of nematode·destroying fungi isolated, During 1984, the most advanced cassava selections including the location from which they were ob· reached the uniform yield trial stage in three major tained, are listed in Table 4.20.- R. Theberge and F . agro·ecological zones, and a multilocational evalua· Caveness tion of the best clones is underway. Root and Tuber Improvement Program 149 Table 4.ZO. Nematode-destroying fungi isolated from indicated locations at IITA Paddy Hydromorphic Forest Nematode Vegetable Fungus Type] water-line area litter microplots farm litter Verticillium sp.. . .... .. ...... . .. . . .. . .. . K X X Trichothecium sp .. . P X X Meristacrum sp . . . . K X Haposporium sp .. . . K x X X Dactylaria sp . . ... . P X X X Dactylella sp .. .... . P X Arthrobotrys sp . .. . p X X X X X IN = nonpredator, P = predator. Efforts to identify and incorporate resistance to disease tolerance, yield potential. root conformation, cocoyam root· rot disease in the susceptible local branching habit, and leaf retention. Highest fresh white cultivar are continuing. yields were from the improved clones: 42 t /ha at Seed yam production was initiated during the year Mujuka, 11 tJha at Meiganga, and 39 tfha at Bertoua. and a large yam germplasm collection was main· Local controls produced 21 tfha, 2 tfha, and 18 tfha, at tained. It should be transferred to our international those locations . body for maintenance and further uses. Advanced yield trials were established with 35 Improved sweet potato clones have been selected clones at Mujuka and 38 at Meiganga. All produced and tested in on·farm trials. Excellent yield improve· higher yields than the locals. Clone S081's 34 tiha at ments were recorded in several hundred comparisons Mujuka was highest , while clone 0844 was the best with the best local material. We also assessed im· yielder at Meiganga with 15.8 tfha. proved cropping techniques in farmers' fields and A reciprocal yield evaluation was performed with began multiplying elite clones. clones selected at Mujuka (SO m asl) and Babungo (1,130 m asl). Clones yielded best where they were Cassava initially selected. Clones 7839 produced 31.3 tfha of fresh storage r oots at Mujuka, while 79307·8 and 7i46 The Cameroon roots program has concentrated on each produced 27 .2 tfha at Babungo. Where storage major cassava diseases: mosaic disease, bacterial root yields were halved. storage root numbers reo blight, and anthracnose disease. We now have many mained the same with smaller r oot sizes. Mos aic and disease· resistant clones in yield tria ls. bacterial blight were severe on the Babungo·selected The cassava green mite, identified in 19SO near the clones at Mujuka.- J.A. Whyte, H.J. Pfeiffer, and S. Central Africa Republic border, is distributed over Lyonga the entire country. The cassava mealybug was ob· served in January 1985 within 15 km of Ekok on the Food Technology Cameroon. Nigeria border. We plan to increase em· phasis on pest resistance . To formulate research priontles for post· harvest technology, we investigated by Questionnaire post­ Genetic Improvement harvest handling, traditional methods of processing, and preserving and marketing root and tuber crops in Cassava genetic improvement proj ects are at three Cameroon. Seventy percent of those surveyed were locations that represent Cameroon's major agro­ women. An unusually high 89% (567) r eturned ecological zones. Mujuka is in the high rainfall, questionnaires. lowland zone on poor sandy soils; Nkolbisson, in Results showed that commercial cultivation was south continental land with medium rainfall on red not common. Farmers' holdings are small; 59~~ har· lateritic soils; Meiganga, in the highland savanna vested as needed. Physiological and pathological with moderate r ainfalL Secondary selection sites are losses were greater than pest losses of stored crops. used in the rainfall highland savanna at Babungo, Maximum average storage periods observed were 5 south continental savanna at Bertoua, and in the days for cassava, 3 months for cocoyam, 4 months for drier lowland savanna zone at Mbe. yams, and 5 months for sweet potatoes. Only fresh From the 15,000 seeds sown in three seedling cassava in any quantity was transformed into other nurseries, we selected 564 plants this year for clonal food products. Excess and rejected crops and their by­ evaluation. They were the best for disease and insect products were used as animal feed. Industrial use was resistance and agronomic characteristics. rare.-S. Lyonga, J.A. Whyte, and H.J. Pfei.ffer We evaluated 3,000 plants selected from the 1981-82 nurseries for disease and pest tolerance, storage root yield, and conformation and branching habits before advancing 385 clones into preliminary Cocoyams yield trials. Research work on cocoyam (Xanthosoma spp.) fo· From 450 clones tested in preliminary yield trials, cused mainly on control methods aimed at minimiz­ we selected 150 for the advanced yield trials based on ing pythi.um myriotylum root rot. 150 Root and Tuber Improvement Program Identifying resistant clones. Cameroon has quality. Many progenies are being multiplied for three local macabo types, white. red, and yellow yield, quality, and storage testing. fleshed. The white is susceptible to root rot; the red, Seed yam production. Although yams have high slightly t.olerant; the yellow resistant. The red and production potential in the Mbe plain (north white are commonly grown for tuber production. Cameroon, Sudan savanna, sandy soils, 500 m asl, We failed t.o transfer resistance from the low· 1,000 m annual rainfall), availability of good planting yielding, resistant local yellow type, primarily be­ cause of its low pollen fertility material has been an important limiting factor of yam (2- 5% ) and high ploidy (n = 2x = 52). We now are trying to incorporate production. The regular shortage of planting rna· terial could be overcome by systematic seed yam disease resistance identified in exotic germplasm production from minisetts. from Central America and their crosses with local cultivars. We have produced several thousand hybrid Invest.igations on seed yam production from mini­ setts were initiated in 1984 in two agro-ecological seeds using 750 to 1,000 ppm of gibberellin to induce flowering by the selected tolerant , high yielding zones with the two commonly grown D. rotundata, clones, which are being evaluated in yield trials. The white yams, Bakokae and Ngang, late and early tolerance levels observed in the selection series are maturing varieties. respectively. Minisetts of different sizes were ashed with or low. without a insecticide/fungicide mixture (Thioral) , Improving cultural practices. A "cropping pac­ and the cut surfaces allowed to dry before direct field kage" developed the last five years to reduce cocoyam planting on ridges 1 m apart with 50-cm spacing. root-rot disease was tested in farmers' typical Bakokae's germination rate was much lower "groundnut-maize-cocoyam-cassava" fields in the (35.6%) than ~gang's (69%) (Table 4.22)_ Bassa area (south Cameroon forest) w here root rot Increasing minisett size improved germination rate disease is high. but the insecticide/fungicide mixture did not raise it Fields with well drained soils fallowed at least four significantly. Average tuber yields were d irectly years were selected and cultivated by the farmers related to weights ofminisetts for each variety. according to their usual first season practice except Bakokae's seed yam tubers were oblong; Ngang's for the treated cocoyam planting material and the were longer and thinner. In the heavy soils, the many time they planted treated cassava. Average results " shoes" produced diminished the value of Ngang's from four farms (Table 4.21) show that cormels as marketable seed yams. planting material give significantly better aerial Mulching the seed yam setts with dry grass development and final cormel yield than corm heads "Dourou hut" directly after planting significantly or middle parts. Regrowth of cormels that remained improved emergence of Bakokae but not of Ngang. in the soil from former cocoyam crops was thinned to Plants with a "hut" had only one strong vine; those the same populations as the planted material. Their without a hut had several small stems originating mean vegetative development, cormel yield, and from the seed yam sett, which leads to faster em· commercial-yield percentages were medium. So cor­ ergence. significantly higher yields, and higher ave­ mels appear to be the preferred planting material rage weight per tuber.-S. Lyonga. A. Agueguia, and under diseased conditions. Planting cassava at onset J.A . Whyte of the first or second rainy season has no effect on cocoyam yield. At the locally used rate (+ or - 2,000 Vulgarizing (Popularizing) Technology cassava/hal with two plants per stand, the cassava did not compete with cocoyam even when planted at To improve the links between small farmers and the same time.- A.Agueguia, S. Lyanga, and J.A. researchers, on·farm testing is used. This approach to Whyte Table 4.21. Effect. ofindieated planting material on Yam Improvement vegetative development and yield of cocoyam tested on farmers' fields at A total of 93 accessions of eight species of yams Handole, Cameroon, 1983 (Dioscarea spp.) is being maintained in the field at No. of three locations. large Breeding for nonhardening trifoliate yam, D. leaves. ___Y ",i"."ld"",.:k",g"l,p",l"":;;n,,,t __ dumetarum. Post-harvest tuber hardening, due to Planting in 5 Cormels Cammer- material months Conns % cial lignification of cell walls, limits cultivation of D. dumetarum because it must be cooked so long. But D. Corm heads . .. . ... , 4.60 0.6\ 0.29 38 dumewrum's Corm middle parts . . . . . clustered tubers and low staking re­ 1.85 0.25 0.10 28 Germinated cormels. , .. 6.35 0.67 0.79 67 quirements bode well for its future. Nongerminat.ed corrneis 6.75 0.71 0.87 67 Two local cultivars. ex·Jakiri, high yielding with Natural regrowth. , . . .. 3.60 0.52 0.62 47 poor storage quality, and ex-Mujuka, low yielding with high storage quality, are being hybridized to Mean , .. , . . .. ... , .. ,. 4.63 0.55 0.53 49 obtain a high yielding clone with improve /..', I« 4 stab le aggregates, aggregate stability test, and Henin irtlirmrllOfl • 130- '" index. The instability index of De Leenheer-De Boodt ~ ""\, (CMWD) had the highest positive correlation. ",/ ". 10 20 Stepwise regression analyses gave the following K~' '" -"'" ~E:1:'hIAcf, :.... (R:Ot I ~" "--~ equations: K = 0.065 (CMD) - 0.018(E~) - 0.071 (Ow) , 0 • 1 0 : .... + 0.030 (SMWD) + 0_255_ The stepwIse regression • _i.. analyses explained 65% of K's variation with 0.01 0~~IO~~20---7.'0---4tO--~5O~~~ o f. ° 0 0 o 10 confidence. nme (min.) 20 '" The choice of suitable factors for the predictive Figure 5.1. We used simulated rainfall to illustrate its equations depends on the methodology and its adap­ collation with erosion with a sandy clay loam soiL tability to routine analysis. With that in mind. we Farming Systems Program 157 Table 5.1. Measured (K) and nomogram computed (Ke> erodibility according to textural class Textural class Minimum Maximum Average Minimum Maximum Average Sandy loam (24) ' ..... .... ..... 0.013 0.378 0.158 0.03 0.42 0.11 Sandy clay loam (16) . , . . . . . . . . · ... . ... 0.000 0.237 0.126 0.02 0.13 0.05 Loam. • iltJoam (4) ....... . . . .. · .. . .... 0.116 0.342 0.230 0.23 0.37 0.32 Clay loam (5) .. ...... . . .... . ..•... · ....... 0.123 0.355 0.283 0.08 0.29 0.22 Clay (7) . . .... . . .. ...... .. . ...•... . ... . .. .... 0.000 0.535 0.285 O.oJ 0.27 0.14 I Number of samples. selected three factors: CMWD (Instability index of root growth did not differ significantly in the O· to 5· De Leenheer-De Boodt), IRED (reduction of infil­ cm layers of the mechanized no· t ill treatments from tration rate) and Dw (wet density of aggregat es). root growth in the hand-planted control. But plowing Regression of K with those factors is as follows: K = significantly decreased root proliferation (Figs. 5.3. 0.063 (CM WD) + 0.002 (IRED) - 0.070 (Dw) + 0.071, and 5.4). which explains 63% of the variation with 0.01 con­ fide nce, against 23% explained by the nomogram Table 5.2. Maize dry matter grain yields (t/ha) as estimation.- A. Vanelslande, R. Lal, and D. Gabriels inftuenced by tillage treatments, 1982-84 Vegetative seasons Soil Compaction and Seedbed Preparation 2 3 4 The experiment. initiated in 1982, was continued the 1 May- Sept.- ApriL­ first season of 1984 to assess effects of mechanized Oct. 1982 Aug. Nov. July farm operations on soi I compaction and r oot growth. Treatment J.n.1983 1983 1983 1984 This report summaTizes maize grain yields and root H' . .. . .... . ....... . 5.6102 3.52 b 0.708 4.67 ab growth for four consecu tive seasons. Tillage treat­ MH ............... . 5.LOa 3.51 b 0.408 4.12 b ments studied were no-till with manual operations M . ....... .... .... . 5.44 a 3.35 b 0.37. 3.93 b (H); no-till , machine seeded, manually harvested P ................. . 5.48 a 3.97 0.58. 4.958 (MH) ; no·till , machine planted and harvested (M); lTreatments are averaged over mulch and no-mulch sub· and plowing with conventional machinery (P). Each treatments . H = no-till ..... ith manual operations: MH = no-till , t reatment was subdivided into with (+) and without machine seeded, manually harvested; M = no till. machine planted and harvested: p """ plowing with conventional (-) mulch. machinery. How tillage inAuenced grain yields four conse­ 2Values in the same column not followed by the 38.me letter differ cutive growing seasons is shown in Table 5.2. The first significantly 0.05. growing season (under irrigation) grain yields did not differ s ignificantly under different tillage methods. Dry metter oroin Yield (t/ho'!) The second season grain yield was greater in the I , Veqefotlve period 2. Veg;etofive period plowed than in any no· till treatment. During the third vegetation season a long dry period drast ically re­ 6 duced plant growth and grain yields. Dlfferences were slight among the low yields. In the fourth 4 season, as in t he second, plowed plots gave greater grain yie lds , but t he yield of the hand planted, no·till 2 treatment was similar to that of the plowed plot. Mulch significantly affected grain yields a ll four seasons , but averaged over tillage treatments the mulch effect was significant only one season (Flg. 5.2). 3 Vegetotlve perIOd 4. Yeoeklliw period There was no interaction between tillage and mulch 6 effects on grain yield . To observe roots produced by the various tillage 4 practices, we characterized maize-rooting patterns using a trench profile method and replicated root 2 observations three times in one profile of each treatment. oL..E~~B In a1\ tillage treatments roots wer e generally H MH M P H MH M P longest in the upper soil layers (0-15 em, Figs. 5.3 and Figure 5.2. Dry matter grain yield of maize as influenced by 5.4). Plowing reduced deep rooting to about 50 cm, tillage and mulch during four vegetative seasons (1982-84). compared with 80 cm in the no-till plots. Letters a and b indicate s ignificant differences (0.05) After two seasons of mechanica l impact by farm between mulch and no mu Ich within a tillage treatment, machinery (the third and fourth vegetation seasons), according to Duncan's Multiple Range Test. 158 Farming Systems Program In the third vegetation season, root growth was Rooti"9 density (cm/cm3 ) reduced much more in the 5- to 15-cm soi l layers of the o 201 2 0 I 2 2 o , completely mechanized. n o-till plot than in h and planted or plowed plots (Fig. 5.3). The reduced root ,, , ~ b ,b ,, 20 -~ growt h may have resulted from vehicular traffic H+ MH+ M+ p- during harvest or from the increased clay content in 40 , b , the 5- to 15-cm soil layers. Seven weeks but not nine \ MoAi.,..,." b 60 of 'oot d~" weeks after planting, rooting density 5- to 100cm deep 19.5 13.5 13.3 15.3 was significantly lower in the reduced mechanized, no-till plot than in the hand-planted plot (Fig. 5.3). In the fourth vegetation season (Fig. 5.4) differ­ ,b ,, ences among tillage treatments were clear in the 0- t o H+ MH+ M+ p- 15-cm soil layers six and nine weeks after planting. 40 Rooting density was significantly higher in the h and planted and the plowed plots than in the completely 60 { . .. '''''' mechanized plot. Root density was intermediate in 17.6 163 12.0 .,. .(-.c.m..,'.c.m ..z. t he reduced mechan ized, no-till plot. Due to the 00 9 weeks impact of farm machinery, d ifferences in root growth below 15 cm could not be determined among the Figure 5.3. Rooting density of maize within 5-cm soil layers mechanized, no-till, and the hand-planted (con t rol) and total root length in four tillage treatments seven and treatments. Root density was considerably higher in nine weeks after planting (vegetative season 3). Values ala given depth not followed by the same letter differ signifi­ the 15· to 25- and 15- to 20-cm soil layers of plowed cantly (0.05). plots than in other plots six and nine weeks after planting (Fig. 5.4). In the mulch no·till system, compaction by heavy farm machinery apparently mainly affects root growth negatively 10to 15cm deep (Figs. 5.3 and 5.4). ' :' Aootin9 density ( cm/cm3 ) 2 0 , , Z 0 2 00 0 During drought stress (vegetation period 3), root b ,b , 20 : "b growth was r educed in the 5· to 10-cm layers of the b , H+ MH+ M+ p- mechan ized plot (Fig. 5.3). But when rainfall was 4 0 MllllmUm r OOT dllpth distributed regularly (vegetation period 4), vehicular 19.9 17.5 I 204 traffic clearly influenced root growth (Fig. 5.4). -60 0 L..._ __J 0S.' b b ~ In t he O· to 5-cm soil layer during the third and 6 weeks fourth seasons, root growth was not influenced by ;; 0 ,----+- -;---, vehicular tr affic (Figs. 5.3 and 5.4). We assume that ! 20 the negative effect of compaction on root growth in b t he no-till system is r estricted to the upper 5- to 15-cm 4 0 ... M+ p- b soi l profile. Although root growth. was reduced by b • b Total ' 001 mechanical operations, grain yields did not differ 60 /(~:'2 significantly between the mechanized and the hand 31.0 329 28.2 so '---------' 2:5.2 ~e'r'· planted, no· t ill plots (Fig. 5.2), though the no-till plots 9 weeks had nonsignificantly higher yields each season. Our root and grain-yield data during four cropping sea­ Figure 5.4. Density of maize roots wi thin 5-cm soil layers and total root length, four tHlage treatments six and nine sons indicate that soil structure in no·till plots did not weeks after planting (vegetative season 4). Values at 8 critically lower plant growth or cause disintegration given depth not followed by the same letter di1fer signi­ of soil physical factors.- H. Frazen, R. Lal, and W. ficantly (0.05). Ehlers (University of Gottingen, West Germany) Soil Compaction and Plant Growth compaction. 2-, and 4-pass compaction treatments were 4.7,3.0, and 2.3 t/ha for no-tillage and 5.4, 1.6, and The effects of vehicular traffic on soil compaction , 1.4 t /ha for conventionally plowed plots, r espectively. crop growth, and yield were investigated in an Maize grain yields correlated significantly with bulk experiment initiated in 1982 (Annual Report, 1983, pp. density (negative) and with total porosity (positive) 143-144). The compaction treatments imposed in a (Tahle 5.3). Maize grain yield also correlated posi­ spli t-plot experiment were zero, two, and four passes tively with the equilibrium infiltration rate. Soil of a tractor-mounted, 2·ton roller as sub· plots super­ compaction reduced cowpea yields. In the second imposed on no· tillage and convent ional plowing as season 1983, yield of cowpeas was 1.3, 1.0, and 0.8 t lha main plots. The effects of compaction-induced for no· tillage and 1.3, 0.9, and 0.9 tfha for plowed plots changes on soil physical properties were assessed on for 0, 2-, and 4-pass compaction treatments. Cowpea yields of maize, cowpea , cassava, and soybeans. grain yields, like maize yields, significantly cor­ In the second season 1983, maize grain yields for no related with soil bulk density (Table 5.3). Soybean Farming Sysrems Program 159 Table 5.3. Soil physical properties and indicated crop alternately in the same row. First-year results were yields, 1983-84 IITA reported in the 1983 Annual Report. Soil Regression Correlation As in 1983, the major component of the soil-water property equation coefficient (r) balance in 1984 was evapotranspiration (Table 5.4), Maize which differed among treatments only during 4- 18 Bulk density. . .... . .. Y = 17.34-8.84X -0.80*** June and 16- 30 July. Water-use efficiency (WUE) of Total pDrosi ty . ...... Y = -6.33+23.7X 0.79*** intercropped maize and cowpea was significantly Infiltration rate . ..... Y = 1.46+0.55X 0.82*** higher than that of sole maize or cowpea under 1984's Penetrometer resistance Y = 7.80-1.68X -0.59' favorable rainfall (Table 5.5). Compared with seeding Soil Regression Equation sole maize or cowpea, seeding maize and cowpea property m b R alternatively in the same row increased \VUE 100(%, Cowpeas and in alternate rows, 71.4%. Planting cowpeas in the Bulk density . ..... . -1.43 3.1 -0.73*** same row with maize increased maize yield 22.3 '}~. Total porosity . . . +3.76 -0.69 0.72*** Planting maize and cowpea in alternate rows did not Infiltra ticn rate. . +0.02 0.57 0.67** affect maize yield; neither system affected cowpea Penetrometer yields. resistance . . .. . - 0.22 1.47 -0.47' The increase in WUE with intercropping resulted Soybeans from more effective use of soil water reserves. While Bulk density ...... . -2.03 4.22 -0.74*** the total season evapotranspiration was from 546 to Total porosity . .... . +5.37 -1.18 0.72*** 554 mm, and total runoff, 6.8 to 9.3 mm, the high WUE Infiltration rate. . .. . +0.04 0.47 0.88*** in intercropped treatments came from increased Penetrometer grain production. So One benefit of intercropping resistance . . . .... . -0.46 2.14 -0.58' maize and cowpea is higher WUE than with mono­ Cassava cropping, provided soil water is not limiting.- N. Bulk density -15.44 39.73 -0.62** Hulugalle and R. Lal Total porosity ... +4.67 -2.33 0.60** Infiltration rate . .... +0.30 11.75 0.52' Penetrometer Soil-water Balance and Water-use resistance . .. ..... -4.05 27.09 NS Efficiency by Cowpeas Seeding cowpeas after rice to use residual soil mois­ grain yields On second seasOn 1983 were drastically ture and increase land productivity is a r apid ly reduced by the compaction treatments. For example, spreading management system in Nigeria. the four-compaction treatment red uced soybean Early-maturing cowpea cultivars to follow rice and grain yields by 50%, 64%, and 75% for no-tillage and use the residual moisture are highly important. Yet by 47%, 47% and 65% for plowed plots for 0, 2-, and 4- little is known about water-use efficiencies (WUE) or pass compaction treatments, respectively. Soybean water-extraction patterns of cowpeas. This study was grain yields also correlated significantly with soil to monitor seasonal water balances to determine bulk density and infiltration rate. WUE and extraction patterns of three early-maturing Soil compaction affected fresh tuber weight of cowpea cultivars and compare them with patterns from a standard 9(H\ay cultivar. cassava and grain yields differently. Fresh cassava tubers harvested nine months after planting weighed IT82D-789, IT82D-8B9, IT82E-32, and VITA-5 were 18.2,16.1, and 9.5 tjha for no-tillage and 20.6, 24.2, and planted in a completely randomized block design wit h three replications. Runoff was measured with multi­ 12.5 t/ha for plowed treatments for 0, 2- and 4-pass divisor flumes. Soil-water potential and soil-water compaction treatments, respectively. The 2-pass com­ content were measured at O.Ol-m increments in the paction treatment had no consistent effect on tuber root zone. Drainage was inferred from soil water yield. The 4-pass compaction treatment, however, retention and hydraulic conductivity curves derived decreased fresh tuber yield 48% in no-tillage and 39% in situ. Precipitation and potential evaporation in plowed plots. Soil bulk density explained about w ere monitored at the experimental site. Evapo­ 38% of the variation in cassava tuber yields (Table 5.3). transpiration was determined by the soil-water So cassava can withstand somewhat more soil balance equation. compaction than grain crops.- B. Kayombo and R. The major component of the soil-water balance in Lal all treatments was evapotranspiration (Table 5.6); Soil-water Balance in Intercropped Root runoff and drainage were negligible. Seasonal evapo­ Zones transpiration was in the order of IT82D-789 = IT82D- 889 < IT82E-32 < VITA-5 (Table 5.7). Water-use In an experiment initiated in 1983 to study the root­ efficiencies of VITA-5 and IT82E-32, however, ex­ zone, soil-water balance of four maize/cowpea crop­ ceeded those of IT82D-789 and IT82D-889. Differences ping patterns, we used sole cowpea, sole maize, in water-use efficiency among varieties resulted intercropped. maize, and cowpea planted in alternate primarily from higher-yielding VITA-5 and IT82D-32 rows, and intercropped maize and cowpea planted under high rainfall.- N. Hulugalle and R. Lal 160 Farming Systems Program Table 5.4. Soil ,,'ater balance (mm) in the root zone oefour maize/cowpea cropping patterns, IITA, 1984 Change in Potential stored Evapotrans- Treatment Precipitation evaporation soil water Runoff Dra inage piration 21 May-4 June Sole cowpea. . . . . . . . . . . . . . . . . . . . . . . . . . . .. .• . . . . . t49.7 80.3 -1.4 ab 2.8a 0.1 a 148.4 a Sole maize . . .. ...... ... , .. . ........ ... . . •.. ... . .. . .... . .. . . . ....... 4.6a 2.6a O.O a 142.5 a Maize:/cowpeas in same row .. . . .. ... . . . .. . • . . .. ... ........ .. ...... 0.5 ab 2.2a -0.3 a 146.7 a Maize/cowpeas in alternate rows .... . ... . .. • . .. .......... . . .. , ... . . .. . -3.3 b 3.2a -0.4 a 149.4 a 5-18 June Sole cowpea. . . . . . . . . . . . . . . . . . . . . .• . . . . . . . • . . . . . 52.5 80.7 -2.9a 0.4 a O.O a 55.0 b Sole maize .... . . .... . .. .... . . .. .. , . . •... • . ..•....... .... .... .... . . . -12.0 ab 0.4 a -0.4 b 63.7 a Maize/cowpeas in same row .. . . .. . . ... ... . .. . . . ... . , ......... . .... .. . -3.gab 0.3a O.Oa 58.1 ab Maize/cowpe8S in alternate rows .... . ......• . . . •................. . ... . -7.7 ab 0.5 a -0.6b 59.1 ab 19 June-;) July Sole cowpeas .............. . . .... .. ... . ..••... . . 82.5 87.8 2.9a 1.3 a -0.5a 77.8a Sole maize . . . .. .....................•..•. . ..• . .......... . .... . .. . .. - 1.4 a 1.2 a -0.1 b 82.6 a Maize/cQwpeas in same row . . ......... •. . .• . ..•. .... 0.9 a LO a 0.2 a 80.8 a Maize/cowpeas in alternate rows . . .. ... •. .••. .. •..... 3.2 a 1.1a - 0.4 lib 77.88 4--16 July Sole cowpeas . . .. .. .... . . . . . .. .. .... . . _. . .. . . . . . 58.5 78.2 4.4 a 0.9 a 0.2a 53.5 a Sale maize .... . ... . . ........... . . . .. . • .• •. . . .. . .. . ............. 11 .3 a 0.6 a O.ta 46.8 a Maize/cowpea in same row . . . . . . . . . . . . • • . • • . . . • . . . . . .. . . .. . 6.0 a 0.6a 0.5a 51.6 a Maize/cowpea in alternate rows . .. ... . .• _ . . • . . .•. . . . . ......... . . . 3.7 a 0.8 a 0.38 54.4 a 17-30 July Sole cowpea. . . . . . . . . . . . . . . . . . .. . . . . .. • .. .. . .. . . 39.6 63.6 -13.1 ab 0.8 a - 0.1 a 51.8 ab Sole maize . . ... . ......... . ... . . .. ... . . .. .. _ . .. ................ . -24.0 c 0.7 a -5.7a 57.2a Maize/cowpeas in same row ....... , .. . ... • ............. . . . . .. . . . . .... -18.tbc 0.3a -2.0 a 55.9a Maize/cowpeas in alternate rows ............. . . .. . .. . ..... . ......... . . - 8.3a 0.8a - 0.1 a 47.0a 31 July-15 August Sole cowpea. . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . •. . . . 175.9 63.8 14.2 a 2.1 a O.Oa 159.6 a Sale maize .......... . . . . . ... . .. . . . .. . . .... .. .... . . ............... . . 12.5 a 2.0a - 0.2. 161.2 a Maize/cowpeas In same row ....... . ............... . . ...... . ... . . . ... . 11 .98 2.4 a O.Oa 161.6. Maize/cowpeas in alternate ro 8 . . .. . . . . ... . . . ......... . .............. . 12.1 a 2.9a -O.la 180.8 a 1 Means are grouped according to Duncan's Mul t iple Range Test. (0.05). 2 A positive value for drainage indicates upward flow from below the root zone. 3A positive value fo r cbange in soil-water storage indicates a gain in soil water. Table 5.5. Grain yield (kglha) and water·use The experiment consisted of seven plots each 10 m efficiencies of indicated maize/cowpea wide and 50 m long. The three cropping system treat· cropping patterns, 1984. Soil-water not ments studied were traditional farming of cassava + limiting maize grown on ridges made across the slope without Ware use fertilizer, maize-cowpea rotation under no-tillage Grain yield efficiency with fertilizer and herbicides, and the same rotation Cropping Maize Cowpea (kg ha-I mm-') and inputs under conventional plowing. Each treat· Sole cowpea ..... 1,162.6 a 2.1 c ment was replicated twice. The seventh runoff plot Sale maize . .. .... 1,175.3 b 2.t c was plowed and kept free of vegetation as a control to Maize/cowpeas in determine K and C factors of the Universal Soil Loss same row . ..... 1,437.2 a 894.1 a 4.2a Equation. Mai2e/cowpe8s in . alternate rows. 960.4 b Soil bulk density of the O· to 5-cm layer increased 1,004.4 a 3.6 b with t ime after initiating cultivation. Two years after Means are grouped according to Duncan's Multiple Range Test, the study was initiated bulk density was in the order (0.05). of bare fallow, plowed, no-till, and traditional t reat· ment. The surface soil in plowed and bare fallow t reatments was prone to crusting and had lower Runoff and Soil Erosion in the Savanna infiltration rates t han that in no·till and traditional plots . The saturated hydraulic conductivity mea· An on-going soil· erosion and runoff experiment es· sured on undisturbed cores in the laboratory was less tablished in 1983 (Annual Report, 1983, pp. 14&-149) than the equilibrium infiltration rate. The water near nore/ljaye village was continued during 1984. acceptance of the soil was in tbe order of natural Farming Systems Program 161 Table 5.6. Soil·water balance (mm)! of foul' cowpea varieties. 1984 Change in Precipita. Potential soil water Evapotrans- Period Variety tion evaporation storage2 Runoff Drainage! piration IT82D·789 4.10 1.30 O.Oa lOB.Ob 27 August--- .... . . .. . . .. .... . . . . IT82E-32 -12.2b 113.4 65.8 1.50 0.0. 124.10 IOSeptember .....•. . •. . • ...... IT82D·889 4.58 1.0. -0.1 a 107.8 b VITA·5 -3.2.b 2.0. 0.0. 114.6.b IT82O·789 4.6 c 0.5 a -0.1. 66.7 a 11- 24 ......... . . . • . .•. . • • .. ... IT82E-32 71.9 80.2 9.9b 0.9. -0.3. 8O.Bb September. .......• . .•. . . . . .. . . IT82D·889 2.7 c 0.9. -0.2. 68.1. VITA·5 17.9. 0.68 -0.2. 53.2 c IT82D·789 -1.9b 3.6. -0.2. 96.0. 25 September--- ....•..•..... . ... IT82E-32 October ...... . .•.......... . . IT82D-889 97.9 58.3 3.3. 2.2. -0.58 91.9b 8 -0.50b 1.2. -0.38 96.98 VITA-5 -2.5b 1.50 -0.20 98.7. 11'820-789 10.88 0.68 -0.38 85.70 9-22 . . . . _ ....•• . _• .. •.. . . . .... IT82E-32 • •..•.. . . ..... 97.4 IT82D-889 81.5 2.5a 0.98 -0.18 93.9 a October. _ . ... . •. 7.3a 0.8. -0.1. 99.2. VITA-5 2.9 a 1.0. -0.2. 93.30 IT82D-789 -34.50b 0.0. +0.2. 42.7 ab 23 October- .... . .. _ . . . .... . .. IT82E-32 vember . . .. ..•• . •... .. . .. IT82D-889 8.0 79.1 -36.2. O.Oa +0.1. 44.3 . 5 No -32.3.b O.Oa +0.1. 40.4 ab VITA-5 -30.9b 0.0. +0.10 39.0b IT82D-789 6-20 . . . . _ .... ....... ... IT82E-32 November .... .. . ...... . ITB2D-889 43.1 104.2 VITA-5 6.5 0.3 +0.4 36.7 1 Means are grouped according to Duncan's Multiple Range Test. (0.05). t A positive value for either soil water storage or drainage indicates a net gain of water into the root zone. Table 5.7. Yield, seasonal evapotranspiration, and 1984, however, the traditional system consisted of water use efficiency of four cowpea maize + cassava grown on contour ridges that varieties reduced water runoff and soil erosion. The least soil Days to Grain Water-use erosion was observed on no-till plots. Compared with mal.o. yield, Evapotrans- efficiency, the no-ti II system. water runoff and soil erosion were Variety urity kglh. -I piration, mm kg mm - 1 ha - 1 3.3 and 2.5 times more, respectively. from the tradi­ IT820-789 . . .. 60 589.8c 399.1 b 1.5 b tional system, 21.5 and 16.8 times more from the IT82D-889 . . .. 60 589.0c 402.4 b 1.5 b plowed treatment., and 45.4 and 43.6 times more from IT82E-32 . . . .. 65 781.9 b 415.0 ab 1.9. the bare fallow plot.-U. Sabel-Koschella , R . Lal. and VITA-5 . . _ . . . . 90 878.7. 435.5. 2.0 a U. Schwertmann Means are grouped according to Duncan's MultipleRange Test, (0.05). Effect of Mulch's Spatial Orientation We studied how mulch's spatial orient.tion affects soil properties and growth of root crops in an experi­ vegetation > traditional plots with mulch > tradi­ ment on a sandy Ultisol at Onne in southeastern tional plots without mulch > no-till > plowing > bare fallow. Differences in water transmission pro­ perties were also related to the differential activity of Table 5.8_ Effects of soil and crop man.gement on earthworm activity. Earthworm activity was highest runoff' and soil erosion at Alore, Nigeria, under na tural veget.tion followed by traditional no­ 1983-84 tillage. plowed , and bare fallow treatment. Soi1 erosion, Table 5.8 shows drastic d ifferences in runoff and Runoff, tlh. soil erosion among the treatments investigated. mm Treatment 1983 1984 1984 During 1983. the traditional treatment, managed entirely by the farmer, comprised maize + yams Traditional. . . . . . . . .• . . . . . . . . . 5.1 1.5 23.8 grown on mounds. Soil erosion was more severe from No-till. . . . . . . . . . . . . • . . .• . 2.3 0.6 7.2 this plot than from flat-seeded. maize-cowpea ro­ Plowed . . . .. .. 2.4 10.1 154.9 tation grown with no-till or the plowed system. In B.re fallow. . . . .• . . . . . . . . 33.9 261.6 327.1 162 Farming Systems Program Nigeria. The root crops were cassava or yam in 1983, Soil physical properties measured were bulk den­ cassava and cocoyam in 1984. Treatments , the same sity . soil temperature, soil·water potential. soil-water both years, were (1) no mulch, (2) mulch applied content, and water infi ltration. Soil chemical proper­ uniformly over the entire plots, (3) mulch incorpo­ ties measured were organic C, total N, Bray-l-P; rated 0.20 m deep, and (4) mulch concentrated in a NH.OAC extractable Ca, Mg, Na, K, and Mn; KCI 0.25-m strip in the row zone. Eupatorium odoratum extractable total acidity, and pH. Plant growth mulch (25% moisture content) was applied to all plots during the season was monitored as plant height, at 12 tlha in split applications. at planting and 150 ground cover. and root growth. days later. The soil was disc-harrowed 0.20 m deep Tillage reduced soil bulk density (Table 5.10). before planting. Infiltration increased only when tillage and mulch Soil physical properties measured were bulk den­ were combined. Soil temperature was lowest in un­ sity, soil temperature, soil-water potential. soil-water tilled. mulched plots , and soil chemical properties content, and water infiltration. Soil chemical proper­ improved most in those plots. Mulching increased ties measured were organic carbon , total 1', Bray P-1. cassava yield 32.3% in untilled plots (Ta ble 5.11). ammoniun acetate extractable Ca , Mg, Na, K, and Yield of tilled plots was not similarly affected. Yam Mn; KC1 extractable total acidity, and pH. Plant tuber yields were increased 27.5% by mulching tilled growth during the season was monitored by plant plots, and 13.3% on untilled plots. The height, ground cover, and root growth. plowed/mulched treatments produced the highest Concentrating mulch in the row reduced soil bulk yam tuber yields. Mulching increased yields of co­ density, increased infilt ration, organic carbon, total coyam more than either cassava or yam. Mulched CEC and exchangeable Ca. Mg, and K. Soil tempera­ cocoyam yielded 3.7 times more in untilled plots and tUre was decreased only the first three weeks after 4_8 times mare in tilled plots than the unmulched planting, Neither applying mulch on the surface nor controls.- N, Hulugalle and R. Lal incorporating it affected soil physical properties. Yam and cassava yields were not affected by the Effects of Post-clearing Land Management experimental t reatments, But cocoyam yielded most with mulch applied uniformly over the surface or An experiment was established in 1983 to in vestigate the effects of different land-clearing methods and concentrated in the row (Table 5.9), where yields were post-clearing, land-management systems on soil approximately double that from the unmulched control.- N, Hulugalle and R , Lal physical properties, crop growth, and yields (Annual Report, 1983). The experimental site was cleared Mulching and Tillage Effects during the dry season of 1982- 83. Land clearing methods were manual. with shear blade, with tree Effects of mulching and no-tillage on soil properties pusher, and with tree pusher/root rake. Each clearing and growth of root crops were studied in two experi­ treatment was cropped with a maize-cowpea se­ ments on a sandy Ultisol at Onne in southeastern quence or sown to a Mucuna cover in 1983. In 1984 all Nigeria , Root crops were cassava or yam in 1983. cassava and cocoyam in 1984, Treatments, the same both years, were (1) no-tillage/with mulch, (2) no­ Table 5.10. Effects oftillage and mulching on bulk tillage/no mulch, (3) conventiona l tillage/with density of 0- to 10-em soil at Onne. Nigeria mulch , and (4) conventional tillage/no mulch. Conventional tillage was disc-plowing 0.20 m deep. 1983 1984 Cotora n herbicide was applied before planting at 4 kg Tillage Mulch Yam Cassava Cocoyam Cassava a.i ./ha to all treatments, and Paraquat was applied on No-till ... . Yes 1.38. 1.40 a 1.33 a 1.37 a all no-till plots at 0.5 kg a.i./ha. All plots received No-till . ... No 1.42. 1.40 a 1.33 a 1.41 a Eupatorium odoratum L. mulch at 12 t iha (25% Plowed . . . Yes 1.29b 1.24 b l.24 b 1.26 b moisture content) in split applications at planting Plowed ... No l.31b 1.34 a 1.27 b 1.36 a and 150 days later. Means are grouped according to Duncan's Multiple Range Test, P - 0.05. Table 5.9_ Methods of applying mulch and yields of Table 5.11. Effect. of mulch and tillage methods OD root crops at Onoe, 1984 yieJds of root tubers at Onne Yield, t/b. Tuber yield, I/h. Cassava Y.m Cocoyam Yam, Cocoyam, Cassava, Treatment tubers tubers cormels Tillage Mulch 1983 1984 1983-4 No mulch .......... .. 27.9. 13.4 • 2.9 b No-till .... . . Yes 11.1 b l.lb 16.8a Complete surface mulch . . . . 30.3. 13.6a 5.5 a No-till .... . . No 9.8c 0.3b 12.7 b Plowed under. . . . .. . .. . ... . 28.2. 10.9. 2.7 b Plowed .... . . Yes 13.9 a 1.9. l3.lab Row mulch .. .. .. . . . . ..... . 24.3 a 12.8a 6.0a Plowed ... . . . No 10.9bc 0.4 c 14.5 ab Yields are grouped ac cording to Duncan's Multiple Range Test Means are grouped according to Duncan~Ei Multiple RBoge Test; (0.05). differences < 0.05. Farming By.rem. Program 163 were sown to a maize-cowpea sequence, with the no­ Soil Fertility till system and herbicides for weed control. Soil properties were measured during the dry Nitrogen Management in Various Cropping season ofl983-84 and the growing season of 1984. Soil Systems physical properties measured were bulk density, An experiment was started in 1982 at the Institute's penetrometer resistance, in situ water retention, and high rainfall substation at Onne, Nigeria, on hand· infiltration. cleared Ultisol (Typic Paleudult) soils. The experi· Soil properties measured before planting indicated ment has asplit·plot design with four replications and that compaction was highest after a maize-cowpea compares four cropping systems at five N rates (0, 45, sequence (Table 5.12). Mechanized farming, however, 90, 135, and 180 kg/ha, main treatments): (1) maize increased compaction so it was similar in all plots by followed by cowpeas, (2) maize followed by stylo· the end of 1984. Bulk density, measured two weeks santhes, (3) maize intercropped with cassava, and (4) after maize was planted, showed a mean increase of maize intercropped with cassava and pigeon peas_ 15~/o over preplanting density; two weeks after cow­ During 1984 cropping, maize density was decreased peas were planted, bulk density increased a further 50% in the maize/cassava/pigeon pea intercrop to 13.7% in manually cleared, continuously cropped allow pigeon pea to develop better. plots. Changes were not significant among other In the third year, main-season maize yields were treatments . Penetrometer resistance 'did not differ sustained at about the same level as in 1983 (Table among treatments; means were 240.7 kPa in the O· to 5.14). In both years, yields were lower than the first 20·mm depth and 270.6 kPa in the 50- to 70-mm depth. crop yield in 1982. Significant fertilizer effects were Soil compaction is governed more by post·clearing observed with the four cropping systems. Nitrogen land management than by the initial clearing response was observed from 135 kg/ha. Maize gr?-in methods . vield tends to be greater after macuna than after Maize yields in 1984 were reduced by voluntary cowpeas with low N applications. Yield of maize growth ofmucuna sown in 1983 (Table 5.13). Cowpea intercropped again was lower than sole-crop malZe. yields were not affected by the experimental treat· Reducing t he maize plant population 50% in the ments. The choice of a suitable crop is, therefore, intercropped system with cassava and pigeon pea had important in post-clearing land management sys· no effect on maize grain yield at low nitrogen rates. terns, and macuna needs to be managed differently.­ Only at high N rates did lower plant populations N. Hulugalle and R. Lal reduce grain yields. Table 5.12. Effects of post-clearing systems of land management on soil bulk density, ()"'100 mm depth Cropping treatment Bulk density (Mg m· ' ) Clearing method 1983 1984 Feb. 1984 May 1984 Oct. 1984 Mucuna Maize·cowpea 1.08 f 1.38 b 1.36 d Manual. . . . . . . . . . . . . . • . . . • . . • . . . • . . . .. . . . . .. Maize.cowpea Maize.cowpea 1.18 be 1.39 b 1.58. She.rblade .. . .. . .. . ........ ..... ... ... ....... Mucuna Maize.cowpea l.l5cd 1.41 b 1.47 c Maize-cowpea Maize-cowpea 1.22 ab l.38 b 1.46c Treepusher ........ . .... .. . .... . . ... . . . . Mucuna Maize·cowpea 1.09 ef l.34c l.38d Maize·cowpea Maize-cowpea 1.13 de 1.46 a 1.51 b Treepusher/root rake ........ . . . . . . . . . . - . . . . . Mucuna Maize-cowpea 1.24 a 1.47 a 1.46 c . Maize-cowpea .Maize-cowpea 1.26 a 1.47 a 1.45 c Note : Values in the same coLumn followed by the same letter do not differ s ignificantly (0.05). Table 5.13. Effects of post-clearing land management systems on oven-dry grain yield of maize and cowpea (t/ha·' ) Cropping treatment Clearing method 1983 1984 Maize Cowpea Manual ............• . ..• ... • . .....• .. .. . .. . _ . • • . • • . ..•. . .. . Mucuna 1\Ilaize-cowpea 1.7e 0.8. Maize-cowpea Maize-cowpea 5.2a 0.7 a Shearblade ... .... ....... ............ .. .. .. . .. ...... .. ..... . Mucuna Maize-cowpea 3.4 c 0.5. Maize-cowpea Maize-cowpea 4.5b 0.6a Treepusher .......................•...•..••......•... • ...... Mucuna Maize-cowpea 2.8d 0.6a Maize-cowpea Maize-cowpea 5.0a 0.4 a Treepllsher/root rake ....................................... _ . Mucuna Maize-cowpea L5e 0.4 a Maize-cowpea Maize-cowpea 4.5 b 0.6 a Note: Values in the same column followed by the same letter do not differ significantly (0.05). 164 Farming Systems Program Table 5.14. Effects of nitrogen rate and cropping Table 5.16. Effects oftillage and applied N on grain system on maize yield (kg/ha), Onne, yield (kg/ha) of main season maize Nigeria, 1984 (variety TZPB) grown on degraded Alfisol Mix cropped (Oxic Paleustalf) fol1owing one year of Mucuna utilis cover crop :-1- Mono cropped With treatment After After With cassava and N-rate, kgjha cowpea mucuna cassava pigeon pea Mean kg/ha Tillage Paraplow No-till Mean o . " ... 646 691 552 776 666 o 4,722 3.460 2.509 3.564 45 1,659 1,991 1,554 1,428 1,658 30 ..... ... . 5,472 4,498 4.077 4,682 90 ..... 2,331 2,59i 2,224 1,612 2,191 60 .... .. ....... . 6,479 5,204 5,095 5,592 135 ... . . . 3,165 2,848 2,924 2,014 2.738 90 ......... .. . .. 6.451 5,508 5,513 5,523 180 ...... 3,532 3,363 2,975 2.052 2.980 120 ............. . 6,421 5.616 5,956 5,997 Mean . ... 1,996 2,030 1.797 1,443 Mean .. . .... _. 5,909 4,857 4,630 LSD, 5% LSD, 5% N treatment means . . .... . .. ........ , . . .. 217 Tillage .... . . _ .. . ... . ..... . . . . .... . ...... . .. . 1,299 Cropping system means ........ . ....... . ........ 139 Nitrogen rate meBns . .. .. ...... . _ .. .. .....• . ... 707 N treatment means for same cropping system . . ..... 294 Nitrogen fates for same tillage treatment ....... . 1.225 N treatment means for different cropping systems . , . 340 Maize density 20,000 instead of 40,000 plants/ha yield differences were observed between tillage treat­ As in 1983, cassava intercropped with maize at ments with no applied nitrogen; conventional tillage 10,000 plants/ha showed no significant response to gave the highest yield; no-tillage, the lowest. The applied N (Table 5_15). At high N rates, tuber yields higher maize yield with conventional tillage may tend to be lower . No significant effect of pigeon pea stem partially from plowing under the mucuna cover intercrop was observed as it was severely suppressed crop, which should supply nitrogen. Wit h applied N. by the cassava and maize crops.- A. van der Kruijs, yield differences between paraplow and no-till treat­ B. T. Kang and J . van der Heide ments were minimal. Although conventional tillage was the best with high rates of applied N, maize yields Tillage and Nitrogen differed little among conventional tillage, paraplow­ Earlier studies have shown that on compacted ing. and no-tillage. Alfisols, good seedbeds can be prepared with pal'a­ The residual tillage effect on maize yield is shown plowing and conventional tillage, although t illage in Table 5. 17. The effects of conventional tillage land paraplowing apparently lasted more than one maize may inc rease erosion. To further test the effect of fa llowing the leguminous cover crop on compacted crop. But the mucuna cover crop in the no-till treat­ Alfisol, we carried out a trial on an Egbeda soil (Oxic ment appears to benefit only one maize crop.- B.T. Pale usta lf) at UTA . The experimental area was Kang and C. Garman planted for one year with Mucuna utilis before the experiment. The trial has a split-pl.ot design with Effects of Mulch and Nitrogen on Maize three replications. The first. trial was during the main The effects of mulch and applied nitrogen on maize season after mUCllna fallow. A sequential maize crop were studied on a loamy sandy Alfisol (Ox.ic was planted during the minor season. The only tillage was before planting main-season maize. Maize still r espon ded significantly to applied nitro­ Table 5.17_ Tillage residual effect and effect or applied gen and tillage on this compacted soil even after a N on grain yield (kg/ha) of minor season one-year mucuna fallow (Table 5.16). Large maize maize (variety TZPB) grown on degraded A1fisol (O,,;c Paleusta1f) plot previously cropped with maize after one year of Table 5.l5. Effect of nitrogen rate and intercropping Mucuna utilis cover crop~ 1984 on dry cassava tuber yield (t/ha) at Onne, N-rate. 1983-84 kgjha Tillage' Paraplow l No-till Mean N- Intercrop with o 3.203 2.714 2,366 2,761 treatment, Intercrop maize and 30 . .. ..... ..... . 3.463 3,329 2,688 3,225 kg/ha with maize pigeon pea Mean 60 ............ .. 4,153 3,529 2,864 3,515 o . . . . . . . . 6.66 7.21 6.94 90 3,459 3,930 3.449 3.609 45 .. . . . . .. .. . . . .. . . . . . . 8.76 7.05 7.91 120 4,958 4.307 3,400 4,222 90 ...... _ . . . . • . . . 7.32 7.24 7.28 Mean ........... . 3,847 3,558 2.997 135 .. .. . .. .. . .. 6.71 4.96 5.84 LSD.5~o 180 . . . . . . . . 7.46 7.54 7.05 Tillage means. . . . . . . . . . . . . . ..... _. ...... .. .. . 761 Nitrogen rate means. . . . . . . . ................ . . 572 Mean ..... . 7.26 6.87 Nitrogen rates for same tillage treatments ...... . 990 LSD. 5% . .. . . ... . ....... . ns ns lTillage and paraplowi ng done before main season crop. Farming Systems Program 165 Paleustal!) at Mokwa in the southern Guinea sav· significant effect on maize yield. From 1982 to 1984 anna region of central Nigeria. The experiment was maize yielded more grain under conventional and carried out with a split·plot design and four reo manual tillage than under strip or no tillage. plications. Residue mulch rates were main plot treat· Effects of tillage on sorghum grain yields were ments with nitrogen rates as subplot treatments. inconsistent.. In 1981 sorghum under conventional Mulch rates were 0,5, and 10 t/ha and N rates were 0, tillage outy ielded that under other treatments, but in 30,60,90, and 120kg!ha as calcium ammonium nitrate 1982 strip tillage, and in 1983 manual tillage, out· (CAN). Phosphate and K were applied uniformly as yielded the other treatments. In 1984, that on strip single superphosphate and muriate of potash, respec· and no-tillage yi elded significantly more than that tively, at 60 kg P,Os/ha and 60 kg K 20 /ha. The trial under conventional a nd manua l tillage. The higher was under zero tillage. infestation of Striga in the tilled plots probably con· Maize responded significantly to applied nitrogen tributed to the lower 1984 yields. (Table 5.18). Yield difference due to mulching was The four·year results of residue management treat· significant. The mulched crop was taller and more ments with sorghum (Table 5.20) showed that burning vigorous than the unmulched. The mulch provides residue gave t he best yield followed by removing crop better soi) moisture and temperature regimes and residue ; mulching gave the lowest yield, which was reduces weed competition. Similar results were ob­ attributed to the higher infestation of cercospora leaf tained in 1983. Adequate nitrogen and, particularly, disease in the mulched plots. That disease was trans· mulching are important inputs for maize production mitted with the sorghum stalks used as mulch. in this ecological zone , as evidenced by this trial.­ Burning crop residue also gave higher maize grain E.N.O. lwuafor, B. T. Kang, and L. Nnadi yields in 1981 and 1982 but mulching gave higher maize grain yields in 1983 and 1984. Burning crop Tillage and Crop Residue Management residues in the field may be beneficial for crops like In vesti gations of the effects of tillage and crop resi· sorghum as burning prevents transmission ofd isease. due management on maize and sorghum on an Alfisol But mulching likely will continue to give higher (Oxic Pa leusta l!), conducted at Mokwa since 1981, maize yields.-E.N. O. lwuafor, B . T. Kang, and L. were continued in 1984. The experiment was carried Nnadi out with a split· plot design with four replications. The four·year results are shown in Tables 5.19 and Evaluation of Phosphorus Fertilizer 5.20. During the first year, tillage treatments had no Sources We expand ed the phosphorus fertilizer·source trials to Ultisols at !\teje and Owerri in southeastern l\igeria and to an Alfisol at Samaru, Zaria. The expanded tr ials compared Togo phosphate rock (PR). partially acidulated materials PAPR·25 (26% acid· Table 5.18. Effects of mulch and applied nitrogen on ulation), PAPR·50 (50% acidulation) , a nd single maize yield at Mokwa, 1984 superphosphate (SSP). At all sites P 20 5 was applied Grain yield, kg/ha at one of 0, 30, 50, a nd 90 kg/ha in trials that used a Nitrogen rat.e, Mulch rates, t /ha randomized complete block design with Ii ve repli· kg/h. o 5 10 Mean cations. At Owerri, maize and cowpeas were sequen­ tially cropped ; at Samaru, they were relay cropped. 0 . . .. ....... . 331 827 631 590 30 ....... . .. ...... .. 569 1,492 1,533 1,198 Maize yields were low in the highly acid soils at 60 ... .. .. . . . ....... . 2,054 1,996 2,293 2,1l4 Nteje a nd Owerri. Phosphate rock marginally in­ 90 . ... . .. . .. . ... .. .. 3,179 3,485 3,534 3.399 creased yields at both locations. As the solubility of 120 ....... . ......... . 3,338 3.720 3,695 3.585 phosphorus materials increased, yields increased. Mean ..... . .....•.... 1,890 2.304 2,337 PAPR·50 and single superphosphate (SSP) signi· LSD, 5% ficantly increased maize yields (Figs. 5.5 and 5.6). At Mulch rate . . .. . ....... . .......... . ... . . . . ... . 399 Nteje, maximum yields were not r eached with 90 kg Nitroge n ferti"lizer rate .. . . .. .• ... • .... . .• . .. • . . . 380 P zOo/ha, but at Owerri , 60 kg P , O. !ha from all Table 5.19. EfFects of tillage on maize and sorghum yields, Mokwa. Nigeria, 1984 Maize yield, kg/ha Sorghum yield, kg/ha Tillage method 1981 1982 1983 1984 1981 1982 1983 1984 Conventiona l tillage ......•. .•... . ... 2,745 4,137 4,916 4,280 2,514 2,331 1,842 1,037 Manual tillage ........ . ..•..•...... . 2,493 3,843 4,285 4,210 2,404 2,413 1,900 1,383 Strip tillage ...... .... ...... . ........ 2,795 3,381 3,716 2,992 2,181 2,461 1,714 1,573 No-tillage . . .... .. .. . . .. . .. .. . .. .. . . 2,525 3.299 3.419 2,937 2,056 2,039 I,B22 1,431 LSD,5% . . .. . ....... . .. .. .... . .... . n .s. 336 453 458 n .s . 267 n .S. 307 166 Farming Systems Program Table 5a20. Etrecta of crop residue management on maize and sorghum yields, Mokwa, Nigeria, 1984 Crop residue Maize yield, kg/ha Sorghum yield, kg/h. management method . 1981 1982 1983 1984 1981 1982 1983 1984 Mulch........ . . . ....... . .. . .. . . ... 2.525 3,615 4,472 4,166 2,124 1,923 1.633 1,225 Residue removed. . . .,.. .. . . . ... 2.615 3,540 3,664 3,091 2,229 2,451 1,785 1,351 Residue burned.. . .. . .. . . . . . .. • . .. . .. 2,705 3,840 4,116 3,559 2,514 2,555 2,022 1,492 LSD, 5% .. .. .. .. . .. .... .. .. .... .. .. In 221 n.s. 319 n.8. 515 254 290 Y;eld ( kQ tho ) Zaria; the highest P rate at Owerri depressed yields, perhaps because ofP·induced zinc deficiency. SSP That sulfur is needed for maize production in the 1200 I LSD (5'10) sa vanna is shown in Table 5.21. From the derived sa vanna at Ikole and lkare in southern Nigeria to the southern Guinea savanna at I10rin and Bida Junction in northern ~igeria, maize responded more to sulfur­ 1000 containing phosphorus sources (SSP and PAPR-5O) than to nonsulfur·containing sources (15·15-15 and diammonium phosphate). PAPR-5O was just as effec­ tive as SSP at Bida Junction, but was more effective ---- PR" -------'*--­ at Ikole and Ikare where its relative agronomic 800 ....... ... _.. .. effectiveness (RAE) was 120%. On the other hand, PAPR-50 was 70% as effective as SSP at Ilorin. " Among phosphorus sources tested , 15-15·15 was the least effective.-S.K. Mughogho, B. T. Kong, and U. eoDL-__ Mokwunye ~ __ ~ __ ~~~~~~ __ ~ __ ~~-J o m ~ w eo Phosphoru, oppl"d ( kO PAI/ha) Evaluation of Nitrogen Sources Figure 5.5. Effects of rates and sources of phosphorus on Three trials were started on Alfisols at Mokwa and grain yields of maize grown on an Alfisol at Nteje. Zaria and on Ultisol at Owerri to test calcium am­ monium nitrate (CAN), prilled urea (urea), and urea Yield (kg/ha) supergranules (USG) as nitrogen sources. Nitrogen 4000rI------------------------------- rates were 0, 40, 80, 120, and 160 kg/ha. The trials were set up witb a randomized complete block design and L , SD(5%) five replications with maize TZSR·W as the test crop. Response was highly significant at Owerri (Fig. 5.8), with no significant differences among nitrogen 3000 2000 1000 o 20 40 60 80 Phospnorus applied ( kg P20S / 00) Figure 1>.6. Effects of rates and sources of phosphorus on maize grain yields in the subhumid area. phosphorus sources gave maximum yields. Results were similar on the Alfisol at Zaria. The less· soluble 1200 '---'----'---'---'---'---'---'--"""_-' phosphorus (PAR·25) did not effectively increase o 20 40 w 80 yields. But the more soluble PAR-50 and SSP in­ Phosphorus applied ( kg P20s tha) creased yields significantly (Fig. 5.7). As at Nteje, 90 Figure 5.7. Eff'ects of rates and sources of phosphorus on kg of P ,O.!ha did not give maximum maize yields at maize yieJds in the humid area. Farming Systems Program 167 Table 5.21. Relative agronomic eft"ectlveness (RAE)' MoiZlqfOin yitld ( ~liI / ho) of indicated phosphorus sources for maize Location PAPR-50 OAP 15-15-15 !lorin . .. .. . . . ...... . . . . . . . . . . 70 55 45 ILS D ( ,...) Ikole .. .. _ . . . . . . . . . . . . . . . . . . .. 120 75 71 Ikare . .. . _. . . . . . . . . . . . . . . . . . . . 120 76 71 2.000 Bid .. Junction . _. . .. . . . . . . . . . . . 99 17 10 (Yield with P fertilizerH Yield with no P applied) 00 'RAE (Yield with SSPHYield with no P applied) x 1 1.000 sources. Even at 160 kg Nlha from CAN or USG, maximum yield was not reached. but the highest yield with urea came with 120 kg N/ha. Cowpeas grown after maize showed no response to residual N. Higher urea rates seemed to suppress cowpea yields. 200 Figure 5.9 shows maize response to applied nit­ rogen at Mokwa where CAN and urea were signi­ °O~------~40~------~80~------~'20~------~'OO' ficantly superior to USG (0.01). Timing of appli­ Ni"~ QP~tllQ/ho1 cations rather than N sources likely caused the Figure 5.9. Effects of rates and sources of nitrogen on yields difference. The first N was applied 4 WAP and side­ of maize on an Alfisol at Mokwa. dressing was6 WAP, which may affect release of urea from USG. Maize responses from urea and CAN reach Nab.. t,o.n yield ( kQ/ho) maximum at 120 kg Nlha, while response to USG was 3POO almost linear. Applied nitrogen significantly increased maize I LSD (5".1 yields at Zaria (Fig. 5.10). The responses were linear with USG, hut with CAN and urea, responses were like those at Mokwa. There were no significant differences among the nitrogen sources.- S.K. 2.000 Mughogho, B. T. Kang, and PLO. Vlek Nutrient and Water-use Efficiency of Maize and Cassava The low available nutrient and water-holding capa­ cities of highly weathered soils limit normal crop 1,000 O'-- -----'4Q'--------'OO'---- ---'I2Q'--------,.oJo Nilroqen oppli~ (kQ/hol Jr,Coizl qrain ~Ie ICl ( kq/ho 1 Figure 5.10. Effects of rates and sources of nitrogen on CAN yields of maize on an Alfisol at Zaria. ',200 ILS D (5%) growth. During the second year of the IIT Aj BuntehofjGTZ collaborative research project, we 1,COO I- established a field experiment on an AI6soi (Oxic Paleustalf, Egbeda Series) at Ibadan to study nut­ rient and water·use efficiencies of cassava and maize. eoo Soil-solution probes and tensiometers were installed in soils at various depths up to 150 cm. Plant samples 600 were taken at several growing stages to determine dry-matter yield and nutrient contents. Soil-solution samples at various depths were taken periodically for chemical analyses. Soil-moisture regimes were moni­ tored by recording daily tensiometer readings. Both 200 cassava and maize received three applications of N, P, and K for a total of 120, 60, and 100 kglha , °O~-----~40~----~80~------~'±20~------~60~ respecti vel y. Nitroqen gpplitd (k9/1'o) Nutrient uptake data of maize differed widely Figure 5.S. Effects of nitrogen rates and sources on grain between fertilized and nonfertilized plots during yields of maize on an AI6.sol at Owerr •. early growing stages. Shortly after the third fertilizer 168 Farming Systems Program application on maize (six weeks after sowing), N soil (0-15 cm) was Mokwa loamy sand (pH 5.7 , field uptake was 50% more from the fertilized than the moisture capacity 8%)_ non fertilized plots and K uptake was 100% more. Dry-matter yields as influenced by lack of soil These wide differences indicate maize's large de­ moisture at different stages of growt h are given in mands for K and N during its early growth stage (four Table 5_22. Soil moisture deficits during the early to eight weeks after sowing)_ stage significantly affected cowpea growth, but Moisture-depletion zones under cassava and maize drought during flowering and post-flowering stages occurred at different depths (Fig. 5_11). Cassava had little orno effect. The uptakes ofN, p, K, Ca, Mg, extracts soil water mainly from the 50- to 1oo-cmzone; Zn, and Mn by cowpeas during early growth stage maize , from the 30- to 6O-cm zone. During dry spells, were significantly lower under drought stress_ But neither crop was able to utilize the soil water stored the differences in nutrien t uptake by cowpeas be­ in both upper and deeper layers of the soil profile (Fig. tween the stressed and nons tressed treatments dur­ 5.11). ing later stages of growth were not significant (data That poor water conductivity within the soil profile notshown)_ was described in the 1983 Annual Report.-M. Weber In the case of soybeans, drought stress at both early and A.S.R. Juo growth and post-flowering stages significantly re­ duced growth and yield_ But drought during the Effects of Soil-moisture Deficits flowering stage did not reduce yields significantly. The uptakes of N, P , K, Ca, Mg, Zn , and Mn by Drought stress from frequent dry spells is a major soybeans were significantly affected by lack of soil constraint for cowpeas and soybeans in sandy moisture at the early and post-flowering stages of Inceptisols and Alfisols in the subhumid regions of growth_ Drought stress during the flowering stage West Africa. While various field practices, such as lowered N uptake by soybeans, but did not affect tillage and mulching are being studied by IITA uptake of other nutrient elements significantly (data scientists, we conducted a supplemental growth not shown). chamber experiment to identify the specific stage of The results show that conserving soil water in plant growth where soil-moisture deficiency may sandy soils in the subhumid regions for cowpea critically affect cowpea and soybean growth and during the early-growth stage is critically important. nutrient uptake. Soybeans in the suhhumid regions face high risk Two soil-moisture treatments, field capacity (n9 without improving soil management to remedy fre­ drought stress) and two-thirds field capacity (drought quentdeficiencies in soil moisture_- M.B. Niyungeko stress), were used at three stages of plant growth: and A.S_R. Juo early, flowering, and post-flowering to maturity. An early maturing (63 days) cowpea (TVx 3236) and a promiscuously nodulating soybean (TGm 579) were Soil Microbiology used_ The soil (10 kg per pot) received adequate Decomposition and N Contribution of nutrients(N, P, K, Ca, Mg,S, Zn, Cu, Mo, and B). The Prunings Decomposition studies were conducted in a field under alley cropping with Gliricidia sepium, Scil W01 ... suction 200 ' 00 ~ ooo0r-__~ ~~ ___' ~oo~-, Flemingia congesta, and Cassia siamea as hedgerow .. < • plants_ Fresh leaves and small twigs from legume cutbacks and prunings were sealed in mesh bags, weighed, and distributed in the field cropped to maize_ The cutbacks were made of prunings after shrub established for two years_ For each shrub, three bags 50 each ofprunings were collected on days 7, 14,28, 60, and 120 after cutback. Contents of the bags were oven dried, weighed, ground, and analyzed_ Dry matter loss of the cutback was 96%, 50%, and 45% for Gliricidia, Flemingia , and Cassia, respec­ tively, during 120 days (Fig. 5.12). The fast decom­ position of prunings, particularly of Gliricidia, re­ ...c .... ,. ..... "1" sulted from high N contents and their succulent 8~' -4~ C2o7y .,"k"f"tf l-8.My nature_ In 120 days, Gliricidia prunings decayed completely, Cassia lost 85% of its matter, and Flem­ ingia 73%. Their average rates of dry matter loss per week ranged from 2_7% to 5_8%_ Cutbacks and prun­ '00 1\IfJ!---------------------' ings of Gliricidia, Flemingia, and Cassia released 252, Figure 5.11. Soil moisture profile under cassava and maize 75, and 120 kg Nlha, respectively, during decom­ monitored during dry spells in 1984; a, b, c, and d indicate position (Fig_ 5.12)_ They increased maize grain yield monitoring sequence at two.day intervals. 15% for Gliricidia, 22% for Flemingia, and 50% for Farming Systems Program 169 Cassia. The mulch effect of the persistent Cassia biomass as an indicator of soil life and soil fertility . leaves, perhaps through moisture conservation and The increased biomass C in the alley·cropped plots weed cont rol, was more heneficial than Gliricidia's can be attributed to the organic materials supplied high N content. through litter fall, prunings, a nd crop residues. Microbial biomass was evaluated by the fumi· Cassia maintained a relatively high soi l biomass C, gation method (Table 5.23). Scientists use microbia l likely hecause it decomposes slowly. N applied at 90 kg/ha promoted soil biomass C, essentially by in· creasing 1eguminous dry matter production. Biomass Table 5.22. Dry matter yield (t/ha) of cowpea. aod C correlated with soi l N (r = 0.77) and organic carbon soybeaos (g/plaot) as affected by drought (r = O.70).- C. Yamoah alld K. Mulongoy stress (! field capacity) Growth stage I Nitrogen Fixation by Leucaena leucocephala Moisture m We previously reported that inoculated and well status I "Grain Stover nodulated Leucaena fixed 224 to 274 kg N/ha in six Cowpea (TVx 3236)-63 days months at UTA (Annual Report, 1983). We again Field capacity. 0.56 6.79 6.06 7.86 evaluated N fixation using the 16N technique. N fixed , field capacity .. . 1.24 7.41 6.37 7.70 in six months ranged from 112 to 146 kg/ha, a nd S.E .. . .. ... .. . 0.34 NS NS NS represented 43, 45, and 40% of the nitrogen in lea ves, Soybean (TGm 579)- 95 days stems, and roots, respectively. We underestimated l'\ Field capacity. 0.64 4.43 8.54 17.28 fixation because uninoculated plants were nodulated ~ field capacity . . . 1.30 5.26 14.41 22.11 by indigenous rhizobia (35% of the nodules) and from S.E .. . . . .. .. . . . . . 0.31 NS 2.91 NS inoculated plots (65% of the nodules). II = Early groy.th stage; II = flowering stage; III = Post The benefit of Leucaena·rhizobium symbiosis on a flowering stage to maturity . subsequent maize crop was assessed at UTA. The Dry we~ht lost (%l N released (~g /hol 100 I-----------=:::::~~ 160..----------------, Gliricidia pruning 75 50 80 Cassia cutback 60 Flerringia 25 40 pruning 60 120 Doys Days Figure 5.12. Dry matter loss and K release during field decomposition of Gliricidia sepium, Flemingia congesta, and Cassia siamea at []TA. 170 Farming Systems Program Table $.23. Soil biomass C as affected by nitrogen and 61 % of inorganic N equivalent in plots with the prunings of selected legumes in a)ley prunings added (Table 5.24). Nodules of the inocu· cropping systems lated plants seemed to be important sources of N for Biomass C ' ~glg maize. T he highest inorganic N equiva lents were in Gliri· Flem· plots that received prunings from previou s ly in­ Treatments Control cidia ingia Cassia ocu lated or N fertilized Leucaena plots but they Prunings removed ... . . . . 81 92 96 89 represented only 32% of the N in the prunings, Prunings retained .. . .. . NA 121 101 133 indicating t hat maize did not use Leucaena N ef­ Prunings retained + fic iently. N-recovery studies are needed to assess t he 90kg NJh • .. . . . .. .. . . 112 136 127 155 fate of pruning N and appropriate management Means of biomass C in t.wo cropping seasons. practices. including time of pruning and placementof NA = Not applicable. cutbacks to maximize benefits from the tree legume in alley cropping systems.- K. Mulongoy and N. experiment was conducted in plots where effects of Sanginga inoculation with rhizobia and applied n itrogen on Leucaena were sturued the previous year . At 24 weeks after planting (WAP). Leucaena Residual Effect of Legume Inoculation was cu t at the so il line. Glyphosate was sprayed onto t he stumps a t 3 kg Vigna unguiculata, Arachis hypogaea, Cajanus ca· a.i ./ha to retard Leucaena regrowth. The mai n plots of jan, Centrosema pubeseens, Leucaena leueoeephala. t he previous experiment were then divided into t w O and Psophocarpus polustris were inoculated in 1982 subplots with Leucaena cutbacks either r emoved or at UTA , Ibadan , and at Fashola (70 km north of [ITA) di stributed even ly on the surface. Ma ize was sown with strains of rhizobia from various origins . At both two week s later in the plots and in an adjacent fie ld locations , inoculat ion with some strains increased where urea was applied at 0, 40, and 80 k g l\ /ha for total N, grain , and/or dry matter production of the comparisons . legumes (Annua l Report, 1982). Maize contained signi fica n t ly more l\ in plots Popu lations of introduced rhizobia were assessed where Leucaena prunings were a pplied . The highest at planting time and 10 mon ths after inoculation . At maize N contents were in plot s with prun ings col­ both loca tions, cowpea and Leueaena rhizobia num· lected fr om Leucaena previously inoculated with ber s in uninoculated soils were 2.5 x 10" and < 10 rhizobium. Pruningsfrom inoculated plants added 2.5 rhizobia per gram of soil, r espectively. They were also times more :-.I t ha n those from uninoculated plants the same at planting time and 10 months later. The and maize·yield increases were 1.4 to 2.4 t j ha (Table populations in inoculated plots recorded 10 months 5.24). Maize N and the amount of:-.l added by pruning after planting increased to an average of 2.5 x 10' did not cOl·relate. In plots where prunings were cowpea rhizobia and 2.5 X 103 Leucaena rhizobia per removed, N contribution of Leucaena roots, nodul es . gram of soil. To evaluate the infecti ve ability of and litter to ma ize grain yield was equiva lent to an inoculant st rains, 10 months after inoculati on, ,\'e average of 36 kg urea Nlha, and represented about collected surface soil 0 t o 20 cm deep in both fields Table 5.24. Effect of Leucaena cutbacks on maize N and grain yield as influenced by previous N and inoculation with Rhizo bium, UTA, 1984 Maize Pruning grain N Inorganic N MaizeN, yield, added , Efficiency. I equivalent I Treatment kgjha t/ha kg/h. % kglha Cutbacks removed U ninoculated ....... . . . . . . . . . . . . . . . . . . . . ... ..... 40 1. 5 a 37 25 150 kg N/ha. _. .. . ... . . . . . . . . . . . . . · · · · .0 · . · • · . • 0· . 44 1.9 a 40 32 Rhi.o bium IRe 1045 ..... . . . . .. . .... 36 2.5 a 53 42 Rhi.o bium IRe 1050 . . . .. . .. .. .. . . ... . .. . . . . . ..... 52 2.5 a 54 43 Cutbacks retained Uninocula ted ... . . . . ... ... . .• . . . . .. .. , . , .. ... . " . , . , .. " 36 2.2 174 47 38 105 kg Nih •.. . . . . . . . . . . , . . . . . . . . . . . .. 0. · ·· · .· . • . ·· 64 3.3 448 71 56 Rhizo bium IRe 1045 92 3.6 398 76 61 Rhi.o b i um IRe 1050 . . . . . . . .. ... . . .. . .. .. . . . . . . . ... . . 72 4.6 445 98 78 LSD,5% Cutbacks removed .... . .. . . . . . . . . . . . . . . . . . . .. . .. . . 16 0.5 ND" ND ND Cutbacks retai ned .. .. . .. . . . .. . .. , .. .. , .. . 28 0.3 82 ND ND ' % effic iency was calcula ted wi th maize yield (4 .7 t lha) in plots fe rtilized with 80 kg N/ha as reference. Inorganic N equ iva lent is % efficiency x 80 k g N/ha. llNo t determined. Farming Systems Program 171 from the plots where legumes had been inoculated N Uptake in a Live Mulch with rhizobium strains mc 1050 and Vi F2. mc 1050 was isolated in 1981 from Leucaena at UTA, and Vi F2 We previously compared the N contributions of three from cowpea VITA 6 grown at Fashola. live mulches in newly established and long.term The pots of soil represented 12 legume· rhizobium maize plots (Annual Report, 1982). Psophocarpus combinations and were set in a randomized complete palustris was the best of three in the trial. Live block design with four replications. Surface· mulches and maize competed for N in the newly sterilized seeds of Leucaena and cowpea VITA 6 were established field , and about 30 kg N/ha was contri­ sown. Nodulation was generally improved by in· buted on long-term plots. We needed to monitor the creased rhizobial populations in the sOtI after in· three-year evolution of N uptake by maize in a oculation and legume cultivation. Each plant had at psophocarpus live·mulch lot to determine when live­ least 15 nodules in soils where the rhizobia had been mulch N inputs become beneficial to the food crop. introduced. Table 5.25 shows proportions of nodules The experimental design was a randomized com­ formed by inoculant strains at 1.5 and 10 months after plete block with four replications and nitrogen (0 and sowing. Nodulation due to mc 1050 was improved, 120 kg N /ha), live mulch (with and without psopho­ particularly in Fashola soils where it averaged 89% carpus), and food crop (with and without maize TZE 10 months after introduction. Rhizobium mc 1050, 4). All plots received a basal application ofP as single isolated from lITA, established well at Fashola and superphosphate and K as muriate of potash at 60 was more competitive than indigenous rhizobia kg/ha in 1982. The first year, psophocarpus was sown there. The strain is being used successfully as in· between rows of maize at spacings of 50 em within oculant for Leucaena. rows ; the following cropping seasons, maize was A field trial was conducted to evaluate residual established in the live mulch after strips had been effects of inoculation and legume cultivation by opened with paraquat. Psophocarpus's climbing ten· planting a cereal in plots inoculated 10 months dency was controlled by spraying the cover crop with earlier with six rhizobial strains. The field was mowed a growth retardant three weeks after sowing maize. and branches of Leucaena and pigeon pea were Plots were 3 x 10m and maize was planted at spacings removed. Maize TZB was sown by rolling injection of25 X 75cm. planter. Adjacent plots were planted and fertilized Nodulation and nitrogen fixation of the legume with 0, 25, 50, or 75 kg N/ha for reference. The were assessed in October 1983. Where no N was experimental field was irrigated as needed. At 7 W AP, added, roots of the legume produced only 20 nodules maize-shoot weights in the various treatments were from 0 to 15 em depths in 1-m' quadrat. Nodules were compared, and the residual contribution of the Ie· fewer in plots fertilized with N. Nitrogenase activity gumes was assessed in terms of N equivalent (Table was also low, about 20 micromoles C, H, /hm '. At 5.26). In legume plots where urea WaS not applied. N maize harvest. psophocarpus l\ ranged between 62 equivalents ranged from 42 to 64 kg/ha and were not and 102 kg/ha. It was not affected by N fertilization significantly influenced by inoculation. In plots but tended to be less in maize plots. N contribution where the legumes had been fertilized with 150 kg was calculated as the difference between N contents N/ha, N equivalent averaged 68 kg/ha, so legumes of maize in live·mulch plots and in plots with no live improved soil fertility , and introduced rhizobia per· mulch. N contribution of psophocarpus was essen· sisted, multiplied, and ensured proper nodulation of tially negative in 1982 and 1983, indicating that the subsequent legume crops.- K. Mulongoy and M.M. newly established cover crop competed with maize for Mafuka N (Table 5.27). Table 5.%5. Proportions ('Yo) of nodules formed by Rhizobium IRe 1050 and ViF2 at 1.5 and 10 months after legume inoculation at two locations in West Africa Time after Rhizobium ViF2 Rhizobium me 1050 inoculation, months Legume host Fashola IITA Fashola lITA 1.5 1 ••••• ••.••••••• •• ••••• • • • •••• • • ••• ••• Arachis hypogaea 36 40 0 0 Cajanus cajon 75 85 KD ND Centrosema pubescens 0 ND 0 ND Leucaena leucocephala 0 0 100 100 Psophocarpus palustris ND ND ND NO Vigna unguiculata 50 68 8 3 10' ........•..........•................. A . hypagaea 36 45 91 9 C. cajon 30 33 87 3 C. pubescens 3 44 81 3 L. leucocephala 7 28 97 58 P. paiustris 11 14 90 28 V. unguiculata 43 20 87 16 ND :oa Dot determind because there were fewer than 10 nodules per plant. I = (n the field. 4! = In potted soils from the field trial. 172 Farming Systems Program Table 5.26. Effect of six legumes on fresh shoot weight of a subsequent maize crop seven weeks after planting, and expressed as N equivalent. IITA Centrosema , Previous treatment Cowpea Arachis Cajanus kg Nlha Leucaena Psophocarpus Uninoculated. . . . .... . 47 64 47 42 52 48 150 kg :-l/ha ........... . 56 73 64 64' 82' 77 Inoculated 2 .••. 55 54 52 42 44 64 lSignincantly different for the same legume (0.05). 2Means for six rhizobial isolates. Table 5.27. Kg/ha ofN uptake by maize and N contributed (values in parentheses) in a psophocarpus live mulch established in 1982 and cropped every year (field I) or left fallow until 1984 (field 2), UTA Live Field 1 Field 2 Urea, kg Nlha mulch 1982 (F) ' 1982 (8)' 1983 (F) 1983 (8) 1984 (F) 1984 (8) 0.... . . . .... .. .. .. .. . no 42 48 46 42 58 58 0 . ............... . ... . ...... .. yes 29 (-13) 44 (-4) 45 (-1) 45 (+3) 72 (+14) 102 (+44) IW ........................ . ... . no 54 72 68 64 70 ND IW .. ... . ............... . .... . yes 43(-11) 65 (-7) 62 (-6) 55(-9) 86(+16) ND LSD, 5% for same urea treatment .. . 13 21 20 15 14 For different UTea treatments ...... . 10 19 16 18 II l(F) = first season; (8) = second season. In 1984, psophocarpus transferred about 15 kg N/ha found in short· term fallows in the humid and sub· to maize, but that was less than 20% of the cover crop humid tropics of West Africa. Two experiments were N at maize harvest. The positive N contribution in conducted this year at the same sites as in 1983 and in 1984 probably was from organic matter (plant litter the same way as reported in the lITA Annual Report and earthworm casts) accumulated under the mulch. for 1983, except that a new herbicide, sethoxydim, In plots of psophocarpus established near the experi· was added. mental field in 1982 and left fallow until the second All treatments except sethoxydim gave good to season in 1984, live mulch contributed 44 kglha. excellent control of C. odorata. Paraquat applied at Amount oflitter under mulch was difficult to evaluate 1.0 kg or 5 llha gave excellent control of C. odorala because much of it was trapped in the top soil. Fallow and P. maximum throughout the cropping season, plots had nearly twice the earthworm casts as plots resulting in maize yields comparable to those in plots cropped to maize every year (528 to 285 g/m'). The treated with glyphosate at 3.6 kg or 10 IJha. I t is earthworm casts contained 0.42% N or reserves of suspected that the C. odorata and P. maximum stands 12.0 and 22.2 kg N Iha. Maize grain yield was generally controlled by paraquat must have originated from low, particularly with no added N, averaging only seeds from the last seasons's weeds rather than 1,000 kg/ha, with or without live mulch. from established tussocks of the perennial weeds. These data confirm the earlier observation that N Glyphosate at 0.7 kg or 2 llha plus 2,4·D was more contributed by live mulch is not its major attribute effective on C. odorata but less so on P. maximum than the first cropping seasons. It may be important in glyphosate at the same rate plus MCPA. Glyphosate weed control, soil protection, and stimulating at 1.08 kg/ha followed sequentially by 2,4.D at 1.5 biological activity. In this study, psophocarpus fixed kgJha was the only treatment that gave excellent approximately 15 kg NfhaJyear. Its ability to fix control of C. odorata and good control of P. maximum. nitrogen needs to be improved so it does not compete Sethoxydim alone or plus 2,4·D was ineffective on P. with maize for soil N. - K. Mulongoy maximum. resulting in a significant redu.ction of maize yield. The yield of plots treated with seth· Cropping Systems and oxydim was no different from that of plots receiving other treatments. in spite of this chemical's poor Management control of C. odorata. Sethoxydim plus 2,4·D con· trolled C. odorata effectively.- J.A. Poku and J.O. Weed Management Akobundu Chemical Control of Fallow Vegetation Weed Control in Soybeans A project begun in 1982 was continued in 1984 to Herbicides were evaluated for weed control in soy­ identify suitable herbicides and herbicide mixtures to beans at Mokwa in the southern Guinea savanna and control such perennial weeds as Chromolaena odo­ at Ikenne in the forest transition zone. The experi· rata and Panicum maximum, which are commonly ments had a randomized complete block design with Farming Systems Program 173 four replications ; soybean cultivar TGx 536-02D was Weed Control in Cowpeas planted at both sites. Herbi cide injury to soybeans ranged from 5"/0 for a Kat many herbicides can control Euphorbia heter­ formulated mixture of pendimethalin plus linuron at ophylla, which poses a serious problem in cowpeas. To 1.25 kg/ha to 38% for pendimethalin at 3.0 kg/ha determine the best herbicide management program under co nventional tillage at Mokwa (Table 5.28), for this weed , we conducted experiments in the first where the growing season was rela t ively wet . The and second seasons with cowpea culti var IT 82E-9. results of these and previous studies indicate that The experimen ts had a randomized complete block pendimethalin at 3.0 kg/ha is toxic to soybeans. design with four replica tions. Scepter, a new herbicide, gave excellent control of Scepter, a pplied pre-emergence atO.l to 0.25 kglha, Euphorbia heterophy/la. The grai n yields of plots was t he only herbicide that controlled E. heterophylla treated with it were comparable to yields of hoe­ without crop toxicity (Table 5.29). In plots receiving weeded plots and those receiving the standard treat­ t his t reat ment, gra in yields were as good as t hose of ment chloramben plus alachlor (tank mix) at 2.0 + 2.0 t he weed-free control plots and higher than in t he kg/ha. In previous studies E. heterophylla had proven unweeded control. Th" results were just as good a t to be resistan t to most herbicides recommended for 0.15 kg/ha as at the higher and more expens ive rate of weed control in soybeans. Pre-emergence application 0.25 kglha. A bioassay of soil samples taken at cowpea of scepter appeared to be more effecti ve than post­ harvest from plots treated with scepter sho wed no emergence applications three to four weeks after toxicity to maize or rice~ an important consideration planting_ Scepter also satisfactorily controlled in cropping systems where these crops follow cow­ Brachiaria lata initially, but plots wer e reinfested peas treated with scepter. A tank mix of scepter and with the weed within six weeks . metolachlor or pendimethaJin should greatly im­ In plots treated wi th a form ulated mixture of prove chemical weed control in cowpeas and minim­ pendimethalin pl us linuron at 1.25 kglha, weed con­ ize yield losses caused by weeds. trol and grain yields were good at Mokwa but poor at Ikenne. Lower yields at Iken ne resulted from lower Tillage and Weed Control soybean stands and higher weed infestation. The Cowpeas and soybeans. Two experiments were predominant weeds at the Mokwa si te was B. /ata a nd begun, one with cowpeas at Ibadan and a nother with at Ikenne B . lata, E . heterophylla, and Commelina soybeans at Ikenne, to determine t he efficacy of pre­ bengha lensis. - J.A. Poku, D. Shannon , and [.0. emergence herbicides in these crops under con­ Akobundu ventional t illage, strip tillage, and no-tillage, a nd to Table 5.%8. Evaluation ofberbicides for weed control in soybeans under conventional tillage, Mokwa, Nigeria. 1984 Control of Weed dry Soybean Rate Crop injury broadleaf weeds weight at Height Yield at 12% applied. at 4 WATJ., 4WAT, 8 WAT, Stand/m2 at 8 moisture, Herbicide kg/ha 1 % % kg/ha at harvest WAT, cm kg/ha Pendimethalin + Linuron ..... ... . . .. . 1.25 5 79 362 17 77 2. 117 Metolachlor ..... . . . .... .. .. 2.00+ Chloramben . ... . •. .• . ... . ... 2.00 24 80 362 15 76 2,076 Scepter .. . . .... . ........ • .. . 0.15 23 88 350 14 71 1,972 Alachlor ...... . . . . ... 2.00 10 73 735 14 71 1,955 Pendimethalin + . . ... 2.00+ Chloram ben .... 2.00 36 89 539 11 64 1,949 Alachlor .... .. ....... 2.00 + Chloramben . .. .... __ ... .. ... 2.00 11 91 332 14 79 1,876 Metolach lor - . . . . . . . . - . . . . . . 2.00 14 58 715 14 76 1,831 Pendimethalin . ... , ... ...... 2.00 + Metobromuron . . _ .. -' . -,., 1.50 28 89 175 14 57 1,778 Metolach lor + Metobromuron . , .... . .... .. 2.50 29 70 894 12 67 1,742 Pendimethalin .. . . . . . . . . . . . . 1.50 15 79 401 12 66 1,703 Pendimethalin .. . . , . .. ...... 2.00 25 73 870 10 63 1,575 Pendimethalin .. . ... ....... . 3.00 38 80 415 12 53 1.384 Two hoe weeding. (3 & 6 WAP3) . .. . ....... 0 75 210 15 72 2,435 Untreated control .. _ ...... . .. . . _ ..... . ..... . . 0 0 957 14 72 1,34 1 LSD,5,,/O ..... .. ...... . .. ...... .. ......... .. ..... . . . . . . . . . . . . 480 os 10 ns l All herbie-i.des were applied before crop or weeds emerged. 2WAT = Weeks after treatment. 3WAP = Weeks afte r planting. 174 Farming Systems Program Table 5.29. Effect of herbicides on grain yield and Euphorbia heterophylla control in cowpea, lIT A, second sesson, 1984 Rate CTOP injuTy E. he!erophylla Grain yield applied, Time rating, % dry weight, kg/m 2 at 12~o moisture, Herbicide kg/ha applied' 2WAT' 6WAT% kg!ha Scepter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.25 PE 0 4 658 Scepter .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.15 PE 0 7 545 ScepteT .. . .. .. .. . . .. .. .. . . . . . . . .. . . • . . . .. . . . .. 0.10 PE 0 32 450 Scepter . . . .. . . . . . . .. . . . . . . . . . . . .. . . . . . . .. . . . . . 0.30 PP 4 10 443 Oxadiawn . . . . . . . . . . . . • . .. . .. . . . .. . . . . . . .. . . . . . 1.00 PP 4 152 235 Oxadiawn........ . . ..... .............. .... .... 0.50 PE 6 164 211 AciftuoTfen.. .. .. .. . .. . . .. . .. .. .. • .. . .. .. • .. . 1.12 PE 9 94 142 Oxyftuorfen .. .. . . . ..•... . ........• . ......•... 0.25 PE 73 114 134 Metribuzin . . . . . . . • . . . . . . . . . . . . • . . . . .. . . . . . 0.50 PP 0 105 104 Metribuzin . . . . . . . . . . . . . . . . . . . . . . . • . . . • . . . . 0.25 PE 13 144 89 AciftuoTfen. . . . . . . . . . . . . . . . . • . . . . . . . . 1.56 PP 0 105 104 Oxyfluorfen . . . . . • . . . . . . . . . . . . . . . • . . . . . . . . 0.50 PP 28 125 81 Weed-free control . .... . . . . . ................. . . . .. . '. . . . 0 0 586 Un weeded contro] . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 0 147 146 LSD, 50;., . .. . . ... . ...... . . ....... . . . ..... . .. . .. . .. .. .• ... . . • • • . .. . •.... . ....... .•. . •. . . . 56 185 1 PE = Pre-emergence. PP = 7 days before planting. 2WAT = Weeks after treatment. assess crop performance in these systems. Tbe cow· the lowest rate, 1 kg/ha of each herbicide. A tank mix pea cultivar planted was TVx 3236 and the soybeans of atrazine plus pendimethalin was more elIective in were TGm 53EH12D. Both experiments had a split·plot controlling Rattboellia erallata, the predominant design with four replications. weed in this experiment, than was a tank mix of Though ithad no effect on weed biomass, tillage did atrazine plus metolach lor. influence cowpea yields, which were hi gher in can· Maize yield was lowest under no·tillage but not ventionally tilled plots than in strip and no·tillage significantly different from that of maize under the plots . Cowpea·shoot dry weight per plant was also other tillage systems. Low yields under no· tillage can highest under conventional tillage. Because of stand be attributed in part to weed competition (as sugges· reduction in plots treated with a formulated mixture ted by yields in hoe·weeded plots being about the of metobromuron plus metolachlor (Galex) at 4.0 same for all tillage systems) and to reduced effective­ kg/ha, yields were lower under that treatment. ness ofpre-emergence herbicides, caused possibly by Uncontrolled weed growth significantly reduced mulch in the no·tillage system intercepting some of yields. Cowpeas in hoe·weeded plots yielded more the herbicide. than in plots treated with chemicals. Soybean yields and weed biomass did not vary with tillage system. But yield and weed control were much Control of Imperato cylindrico improved by herbicide treatment. Soybeans tolerated A study begun in 1983 was continued this year to a high rate (4.0 kg/hal of formulated mixture of determine the long·term effectiveness of chemical, rnetobromuron plus metolachlor. As observed in this mechanical, biological , a nd integrated methods of and previous studies, no herbicide treatment was effective on E. heterophylla and Commelina controlling T. cylindrica (spear grass). The results, spp. given in Table5.30, indicate that spear grass was most Maize. An experiment was begun at Ikenne to thoroughly suppressed by Mucuna utilis at 67 weeks study the effects of till age and weed control on the after treatment. The cover crops Psophocarpus pal· performance of maize cultivar TZSRW. The experi· ustris and Pueraria phaseolaides, because they ment had a split plot design, with tillage practices showed better establishment, reduced spear grass (conventional tillage, strip tillage, no-tillage. and stands from 20 to 40% and 2 to 46%, respectively. The disk harrow) as main treatments and various weed population of spear grass increased, however. under control practices as subtreatments. l'o·tillage plots tillage (ridging). had more weed biomass than disk harrow and con· Glyphosate at 1.B kg/ha , followed by tillage one ventional tillage plots at six weeks after planting and week after treatment, was more effective than delay· at maize harvest. This result underlines the impor· ing tillage by four weeks and was just as good as tance of cultivation as a weed control practice. glyphosate at 3.6 kg/ha. Mucuna cover was the only Atrazine at 2 kg/ha plus pendimethaJin at 2 kgJha treatment that significantly reduced the fresh weight controlled weeds just as effectively as hoe weeding, of spear grass rhizomes at 67 weeks after treatment. resulting in campa Table yields. Yields with this But this percentage of dead rhizomes in mucuna plots herbicide treatment were the same regardl ess of the was no greater than in the slashed control plots.­ rate, so farmers can save considerably by applying at J.A. Poku and LO. Akobundu Farming Systems Program 175 Table 5_30. Response of Imperata cylindrica to kg/ha. Maize did not lodge in plots t reated with a tank indicated control practices, IITA. 1984, aJl mix of atrazine plus metolachlor, and the dry weight herbicides applied before planting. of volunteer mucuna at maize harvest in those plots Imperata , was lowest. But weed control was best in plots treated standfm2 with a formulated mixture of atrazine plus metolach­ 12 days Stand lor. Paraquat as a directed spray did not prevent before fres h mucuna from having a smothering effect on maize. Herbicide treat· reduc- wt . of Dicamba controlled volunteer mucuna as effectively rate, ment in 67 tion, rhizomes. as a tank mix of atrazine plus metolachlor. One Control kg/ha 1983 W AT ~~ 8/10 em handweeding fiv e weeks after planting did not Glyphosate fb i no prevent volunteer mucuna from affecting maize tillage . . .. ...... 1.8 101 42 59 6.2 adversely.- J .A. Poku and 1.0. Akobundu G1yphosate !b tillage 1 W AT' . 1.8 133 27 80 5.1 Glyphosate !b tillage 4 WAT .. 1.8 i8 40 60 6.8 Cropping Systems Glyphosate During 1984, emphasis in multiple cropping r esearch !b no· t illage CDA . 1.8 101 55 33 7.8 was devoted to performance of some UTA major Glyphosate !b no tillage ... . .. 3.6 86 21 76 mandate crops in two· crop and complex·crop mix­ 4.6 Psopho cO"'er ..... 101 58 42 5.7 tures on farmers' fields and on station. Specific on­ Peuraria ..... .. .. 114 41 46 5.3 station research to delineate some of the plant and Mucuna cover . ....... . .. 101 1 96 2.1 management characteristics required for intercrop­ Ridging . . . . . . . , ...... ... 78 78 -16 6.9 ping cassava , maize, and cowpea was conducted to Slashing . ... . ... .. . ..... 85 79 2 6.5 furnish background information for on· farm reo LSD,5% ..................... 18 2.4 search. Emphases in 1985 will be devoted to bench­ Ifb = Fol lowed by. mark cropping-system research, crop-rotation 'l WA1' = Weeks after treatment. sequences, and to responding to technological needs JCDA = Controlled droplet application. of the On-Farm Research Team. Response of Maize to N Fertilizer and Cassa va/Cowpea Intercropping Ground Cover Management The great difference in maturities between cassava Data on the performance of maize on an Alfisol that (more than 12 months) and cowpea (60 to 90 days) has been continuously cropped with a live mulch for should permit compatible intercropping of the crops. four years show that maize yields were reduced under The development of each crop in a mixture may be conventional and no·tillage when no nitrogen ferti· modified by planting pattern and level of nutrition. lizer was applied . Under both those tillage systems, We established three experiments in a 4 X 2 X 1 inputs such as pre·emergence herbicide were applied; fa ctorial arrangement in a randomized block design in four replications with planting patterns and va­ plots under live mulch received no inputs. These and previously obtained results indicate that live mulch rieties as described in Table 5.32. Cowpea varieties offers a ll the advantages of no-tillage and reduces the were TVx 3236 (semi-erect) and IT 1948·OlF (spread· inorganic nitrogen requirement of maize more than ing). TMx 30572 was the cassava variety. We used any of the other tillage systems. three growing seasons with crops established in May for the major season, in September for the minor Mulches for Weed Control season, and in December (with supplementary irri­ gation) for the off season. Studies conducted at !ITA have shown that Mucuna Lodging, particularly in the double- and alternate­ utilis produces good mulch, conserves the soil, and row plantings in the dry season, caused unreliable smothers weeds. A dead mulch system using this cassava root yields. During the other two seasons species, however, creates the problem of controlling cassava was, in general, unaffected by cow peas volunteer mucuna in maize. An experiment was (Table 5.32). Spreading IT 1948·01F significantly conducted in 1984 to determine the ability of various out yielded TVx 3236 in intercrop. Weak inter·specific herbicides to overcome this difficulty. In a plot that interaction between intercropped cassava and cow­ had been in mucuna fallow during 1983, the mucuna pea was indicated by low residual correlation values was killed with a broadcast application of paraquat from bivariate analyses (+0.29, +0.22, and -0.19 for at 0.5 kgJha. The experiment had a randomized com· major, minor, and dry season, respectively). Those plete block design with four replications. values compare favorably with the value for cassavaJ As shown in Table 5.31, a tank mix of atrazine at 2.0 maize, another highly compatible association. kg/ha plus metolachlor at 1.5 kg/ha applied five Economic analyses indicate that two successive weeks after planting was more effective in controll· crops of cowpeas during a given year may be as ing volunteer mucuna than a formulated mixture of beneficial as one crop of cassava intercropped with the same herbicides applied pre-emergence at 2.5 cowpea in the bimodal rainfall zone (Table 5.33). 176 Farming Systems Program Table 5.31. Effect of Mucuna uti/is dead mulch on weed control and maize yield Herbicide Herbicide Weed fresh weight. Weed dry weight, Mucuna dry Maize and time rate, gjm2 at 4 W Ap t g/m 2 at harvest weight. lodging, Yield, applied kgJha Broadleaves Grasses Broadleaves Grasses kg/h. % kg/ha Primextra 2 pre-emergence ...... .. 2.50 28 5 45 21 1.390 48 3,036 Atrazine+ ....... . .. . . 2.00+ metolachlor 5 W AP .... 1.50 237 75 15 \61 227 0 3,252 Paraquat 5 W AP . . ... .. 0.50 266 115 68 109 1,556 49 2,620 Dicamba 5 W AP ....... 0.28 194 58 84 181 359 6 2,563 2,4-0+ . ............ .. 0.75+ metolachlor 5 WAP .. . . 1.50 204 47 36 84 580 9 2.865 Weeded once 5 WAP .. .. .. ... .. .. . 170 93 157 58 895 29 3,033 Unweeded control . ... . ... ... ,., ..... 186 68 110 136 1.174 49 1,875 LSD,5% ............ . ... .......... 115 53 ns ns 744 26 ns I WAP = Weeks after plant.irig. 'lPrimextra is the tradename of a Connulated mixture of atrazine and metolachlor. Table 5.32. Effects of planting pattern on cassava and cowpea yields (t/ha in three seasons) Dry Major Minor Planting Wet season season pattern Cassava Cowpea Cassava Cowpea Cassava 32.0 1.84 27.1 2.21 • Cowpea Sole .... ............ . ....... . .. . . .... . . . 25.8 1.12 Cassava & cowpea on same row (1m) . ..... . . 23.7 1.24 22.1 2.01 17.8 1.06 Cassava & cowpea on alternate rows (lm) ...... . 32.2 1.11 20.6 1.83 20.2 0.91 Cowpea on double rows (40 em apart) between two cassava rows (2m apart) . . .. . . . . . 23.9 0.86 28.3 1.23 2f2 0.92 LSD,5~~ ..................... . ......... . . . . . 7.1 0.32 ILl 0.27 8.9 0.18 Table 5.33. Economic analyses of cassava/cowpea production in different planting patterns (established in September minor season) Cowpea Ratio Cassava value as crop Cowpea Cassava, Income/ha benefit/ % ofcassava+ Planting First Second total t/ha Gross Net cost cowpea Sole cowpea . .. . . . . . . . . . . . . . . . . . . . 1,118 2.161 3,279 4,9192,964 1.5 Cassava/cowpea on same row . . ... .. 1.060 0 1,080 17.8 4,280 2,349 1.3 63 Cassava/cowpea on alternate rows . . 912 0 912 20.2 4,398 2,.187 1.3 69 2 rows cowpea, 40cm apart between 2 cassava rows 2m apart . . 923 0 923 27.2 5,465 3, 554 1.9 75 Sale cassava . . . .. . .... .. . . .. . . . .. 0 0 25.8 3,8702,603 2.1 100 Note : 1.50 naira/kg cowpea and 0.15 nairafkg cassava Cost of production Na ira /ha Sole cassava = 1,267 Sole cowpea = 1,061 First season and 894 second season Cass8va+cowpea = 1,911 Second season =2,078 First season Unlike cassava/maize where the net benefit from Cowpea Grain Yield in Cassava-maize maize was only about 30% or less, cowpeas net was 30 System to 50% of the cassava/cowpea intercrop, because Attempting to produce two crops of cowpea. through cowpea grain has a high cash value (1.5- 2.00 unpruned cassava, we used cv . 30572 cassava and naira/kg). Still farmers may resist cassava/cowpea cowpeas in three planting patterns: in alternate systems that reduce cassava yields because cowpea rows, in the same row, and in strips (2 rows of yields are reliable with assured pest control cassava- 2 rows of cowpea and 5 rows of cassava- 5 measures.- A. Mba and H.C. Ezumah rows of cowpea). Crop spacings, popula tions, and Farming Systems Program 177 yields are given in Table 5.34. Cassava population Table 5.34. Cowpea grain yields as affected by was maintained at 1O.OOO!ha except for 5,000/ha in t.he cowpea/cassava planting patterns and strip plots where the area of each crop was reduced by cowpea populations (TVx 3236) 50%. The eight treatments (Table 5.34) were re­ Cowpea plicated four times; plot area was 10 m x 12 m. Spacing, m popula- Grain. Highest cowpea grain yields were from cassava and Planting Cassava Cowpea tion, '()(x) kgiha cowpea on the same row. Low yields from strip Alternate rows. 2 xO.5 2xO.2 50.0 850 arrangement. likely stemmed from the 50% reduction 2 xO.5 2xO.3 33.5 947 in area. Except for a paltry 238 kg/ha from the 5·m 2xO.5 2xO.4 16.5 910 strips, second season cowpeas produced no grain, Same row .......... 1 X 1.0 1 xO.2 50.0 1,465 even 3-m strips were not wide enough to prevent 1 x 1.0 1 xO.3 33.5 1,154 shading. We need to determine the minimum strip 1 x 1.0 lxO.4 16.5 1,133 width and length for different crops because strip Strips cropping is an alternative to mixed cropping.- H. C. 5 rows ........• . . 1 x 1.0 lxO.2 25.0 816 2 rows .......... . 1 X 1.0 1 xO.2 25.0 898 Ezumah S.E ....... . 57 Factors Affecting Extra-early Cowpeas I1:)D, 5~~ ........................ . ...•.. 119 CV~~ ....................... . .. . . _. . .. .. .... . 11.2 A trial on cowpea-population density and row spac­ ing was established at IITA (humid/subhumid tran­ sitional zone), Samaru (northern Guinea savanna) Tab1e 5.35. Effect of between-row spacings on yield and ~allam Maduri (Sahel savanna) wit.h particular and yield components of IT 82E-60 reference to newly developed, early-maturing (about cowpeas at IITA, 1984 60 days) cowpea genotypes, including the white· Between·row Grain yield, kg/h. seeded IT 82E·60. Its canopy is sparse and plant spacing, em Pods/plant g!planl g/pod architecture erect or semi·erect so it should support Season 1 more than the recommended 66,666 population/ha. In the study, the row spacings were 30, 50, and 75 cm; 30 .. ........ . 1,143 7.87 50.4 6.40 50 .. . .. . .... 1,020 plant populations, 67,000, 133,000, 267,000, and 8.00 52.0 6.50 75 .. .. . . .... . 1,012 8.68 66.8 ;.70 400,000 plants/ha. The treatments were factorial LSD, 5% .... . NS NS 8.58 1.10 combinations of row spacing and density in three C,V. , o/~ . .. . . . . . , .. . 14.9 19.3 1.80 replications. The design was a randomized complete Season 2 block. At the IITA site, the experiment was repeated 30 .. . .. . . . . . . . . . . . , during the second season. 510 5.21 3.2 0.62 50 .. . .. . .... . .. . .. . 6i7 6.28 4.6 0.73 Results from the northern Guinea and Sahel were 75 ..... . . . . . . . . . . . , 721 6.63 4.8 0.73 inconclusive because water logging (moisture stress) LSD.5% .... 165 NS 1.13 0.0; prevented establishing appropriate stands in C.V .. % . .... . .. . . . . 17.3 30.6 11.9 Samaru, and lack of timely insect control in Mallam Maduri resulted in low yields. At ~allam Maduri, row spacings did not affect yields significantly. Table 5.36, Effects of plant density on grain yield of Though low, grain yields declined with increasing IT 82E·60 cowpea during two seasons popUlations from 300 kg!ha at 67,000 plants to 130 Kg/h., grain kgjha at 400,000 plants/ha. Population. At !ITA, less than 70% of the germinated cowpeas '000 Season 1 Season 2 survived in populations exceeding 67,000 despite 67 .... . 1.095 467 good management. Stand counts were taken 10 days 133 ..... . 1,031 425 after planting. During the 1983 major season, rainfall 267 .. . .. . ...... .. ..... .. ... .. ... . 1,093 522 was normal, moisture adequate, and plants grew and 400 .. . .......... . .. . . . .......... . 1,017 821 branched normally. Neither plant population nor LSD, 5% . . .... . .• . ...• . .... .. • , .. . NS 177 row spacing (Tables 5.35 and 5.36) affected cowpea C.Y., %. ...... . .... . .... ., . . . ., .. . 14.9 17.3 yields. Competition from closer within· row spacing lowered survival rates and was stronger than from closer between· row spacing. Increasing row spacing to 75 cm increased pods/plant and grain weight per grain yields per pod and per plant were drastically plant and per pod (Table 5.35). Those compensatory reduced (Table 5.35). relationships resulted in small differences in total Populations higher than 67,000 plantsJha are yields/unit area during the first season. needed for extra-early IT 82E-60 type of cowpea when The second 1983 cropping season was particularly moisture is limiting. Under drought stress, grain dry. Plants grew poorly and grain yields were low. yield increased with increasing cowpea population in Higher plant populations gave significantly higher the bimodal rainfall area of humid/subhumid tran­ yields (0.01). Number of pods per plant was much sition zone represented by Ibadan. These results need lower than during the major planting season and to be confirmed for the Sahelian wne. 178 Farming Systems Program Cassava/Maize in Two- and Four-crop Table 5.38. Effects of maize varieties and populations Mixtures on cassava lodging, % Cassava/maize intercrop. In earlier (1981 through Maize population, '000 1983) Annual Reports, maize was dominant over Ma.ize 20 40 80 Mean cassava with dominance moderated by maize's archi­ Cassava lodging, % tecture, population, and planting pattern. Cassava's TZSRW. 17.6 23.5 30_0 23_7 fixed population was 1O,OOOlha (1 m X 1 m). TZE .... 17.4 16.1 18.6 17.5 1ot1ore detailed information from experiments in Mean .. 17.4 19.8 24_3 20_6 which both cassava and maize populations and archi­ LSD for comparing means. . . . . . 0.50 0.01 tectures were varied are reported. Two cassava (TMS Maize variety . . . . . . . . . . . . . . . . 2.00 2.65 30572 and TMS 30001) and two maize varieties (TSZR Maize population. . . . . . . . . . . . . 2.48 3.31 and TZE) with contrasting growth habits were inter­ Maize variety X maize population. .. 3.63 4.91 cropped with cassava at three populations, 10-, 20-, C.V., %_ . _. . _. . _. ......... _. . _. . . . . .. 3.56 and 40- x 103/ha. The 2 X 2 X 3 X 3 treatment combinations were arranged in factorial combi­ nations and replicated three times. Increasing cas­ Data from these studies suggest that sparsely sava populations had no effect on cassava root yield vegetative, short maize types should be improved for but h igh populations increased roots/plants. so t he higher grain yields when used in intercrops. Maize increase in roots was at the expense of root size (Table that reduces cassava yields the least should be 5.37)_ preferred as returns from maize are only 20 to 30% of Neither cassava variety nor architecture had any total income from the mixture. effect on maize. Increasing maize population to 40,000 Cassava intercropped with TZSRW (tall , highly reduced cassava root and increased maize grain vegetative, late maize) lodged more, especially with yields (Table 5.37). Grain yield of the spreading, high maize population than with Pool 16, a shorter, highly vegetative maize type declined beyond sparse, early maize (Table 5.38). Lodging expressed as 40,OOO/ha. Cassava lodged more in high maize popu­ main-stem angle from vertical correlated highly with lations and when mixed with the highly vegetative root yield (Fig. 5.13). Cassava lodging caused by TZSR maize (Table 5.38). maize shading might be a major factor in reducing In a follow up study, we screened maize varieties for cassava yields when cassava is intercropped with growth habits compatible with intercropping with highly vegetative, tall maize, especially at high cassava. The yield data indicated that short maize populations. Such high lodging rates are not usually plants that mature early are most compatible for observed on farmers' fields because of low maize intercropping with cassava. Even though cassava populations. The total leaf area of cassava intercrop' takes more than a year to mature and maize few er ped with TZSRW was displayed over a higher vertical than 100 days. cassava root yield was 22% less in length than when intercropped with the shorter Pool cassava/maize than when cropped alone (26.2 to 31.9 16 maize (Fig. 5.14). Though the taller (etiolated) t/ha). The interaction between maize variety and cassava associated with TZSRW is more likely to maize population for maize yield from cassava/maize lodge, it may intercept more light through the tall was significant because Pool 16 maize yielded more at maize.-H. C. Ezumah, T.L. Lawson. and J. Mareck 6O,000/ha than at 3O,000/ha, while the reverse was true for TZSRW. Higher cassava root yield (0.01) was Subsidiary Crops to Increase Productivity obtained when intercropped with Pool 16 maize than of Cassava/Maize with TZSRW. Though cassavalmaize intercrop is popular in Africa's humid!subhumid zone, one is seldom grown alone, but rather in mixtures with fruity and leafy Table 5.37. Effects of maize and cassava populations on cassava and maize yields and vegetables. The Root and Tuber Improvement agronomic characters Program at IITA has produced several genotypes of cassava that need to be evaluated in intercropping. Cassava. lVIaize. Among them are both sparse- and dense-canopy types. Tubers. Tops, Tubers/ grain Subsidiary crops commonly grown with cassava Pop./ha cassava '000 t/h. tiha plant t/ha and maize in complex mixtures are low-canopy, 10 ... _ ... 11.88 16.04 4.67 1. 75 spreading melon (Colocynthis vulgaris L.), okra 20 ....... ............ 12.21 18.42 5.20 1.82 (Abelmoschus esculentus), and many other vegetables 40 ........ 12.46 17.83 5.83 1.65 LSD,5% .. ........ . such as celosia and amaranthus for leaves and pump­ NS NS 0.61 NS kin for fruits, seed, and leafy parts. POE./ha maize '000 This study was set up to determine yield advantages 20 ........ ........ 13.96 18.29 5.69 1.65 of complex mixtures over two-crop cassava/maize, 40 _. .... _. ........ 11.75 17.17 5_24 2_18 cassava's architecture (dense or spreading canopy), 80 ........ .. . . . .. . ... 10.88 16.83 4.78 2_29 LSD,5% ... . . .. . . . . . . . and changes in the microenvironment (soil moisture, 1.54 NS NS 0. 38 nutrients, temperature, and light) from adding okra Farming Systems Program 179 Root per plant ( 110 ) of the component species, and the calorie yields. We also monitored the productivity of each crop mixture 30 • • in gross yield, income, and labor for each system. Y=2..!'J7-0.248X • • """0.8, R.:::O.64 Soil temperature, Egusi melon improved the soil micro-environment in cassava/melon. cassava/ 20 maize/melon. cassava/okra/melon. and in the four­ crop mixture. Soil temperatures in the mixtures were • 7 to 9" lower than in nonegusi plots. Here are the 14 1.0 • • crop combinations we evaluated; cassava/maize,f okra/melon, cassava/maize/okra. cassava/maize/ melon, cassava/maize, cassavaiokra, cassava/melon, °O~I~-----2~----73----~4----~5----~6~--~7~ cassava/okra/melon, cassava (sale), maize (sole), Estimated cass.ova stem anqle to yertical (1·7) okra (sole), melon (sole), maize/okra, maize/melon, Figure 5.13. Effect of cassa va stern lodging at 5 months on and maize/okra/melon. We used two morphologica lly root yield at 12 months. different cassava cultivars (sparse-canopy and dense­ canopy types). Height (em 1 The mulching effect of melon may enhance soil moisture·organic matter relationships and maize seedling growth. Lal (1974) demonstrated that maize is affected by high soil temperatures at seedling stage. 120 - Soil moisture, Competition for soil moisture appeared to be greatest in the top 10 em between 30 and 60 days after planting. The melon component in such mixtures conserved soil moisture more than 60 100 DAP when melon had attained maximum LAI, prob· ably because of its profuse leaf growth. Okra grew too slowly to shade the soil enough to prevent soil· moisture loss, which leaf·water potential values 80 confirmed. Light interception. Only the maize component TZSR-W POOL 16 reduced photosynthetically active radiation (PAR) reaching cassava, okra, or melon. Maize reduced the 60 PAR reaching cassava 22, 29, 47, and 16 at 4, 6,8, and 10 weeks after planting. Only 50% or less of okra and melon's PARs reached their sale· cropped counter· parts. Linear correlation between PAR and cassava biomass {r = 0.611, cassava root yield (R = 0.66), and 40 t- ne t assimilation rate (r = 0.70) were highly signi· ficant. A continuous crop cover for up to nine months with melon produced a LA! of 3 within six weeks, while cassava needed ab"mtsix months for maximum 20 LAI (Fig. 5.15). Soil nutrients, Melon and okra increased only the quantity of easily decomposable crop residues without increasing organic matter content at the end o , of the growing season. Sharp decreases in P and K, o 2 3 4 o I 2 3 4 especially at 8 and 24 weeks after planting (W AP), Leaf oreo ( X 103 1 indicate that complex mixtures would requre P and K supplements at those growth stages. Figure 5.14. Vertical distribution of cassava leaf area intercropped with maize at 60,000 population per hectare. Economic yields. Maize reduced cassava's fresh storage root yield 28%, while okra, melon, or both, mixed with cassava, reduced cassava's storage root and melon to cassava/maize. yield only 3, 6, and 9% respectively. Maize in the We monitored changes in soil temperature, soil cassava/maize mixture yielded 19% more than sole moisture (and plant·water status), soil organic car· maize at equivalent populations. Okra's fresh fruit bon, :-.1, P, and K. and light interception patterns at yield was reduced 72% mixed with cassava; 89% with selected growth period. Other factors we considered cassava/maize; and 56% with cassava/maize/melon. were the distribution of leaf area index (LAI) in the Melon seed yields decreased 56% in mixtures with various crop mixtures, relative competitive abilities cassava and 76% with cassava/maize. Maize yields in 180 Farming Systems Program leef area indell planting. The second factor , cowpea varieties in­ 3,----------------------------------, cluded a semi-erect, determinate one (TVx 3236) and a spreading day·length sensitive one (VITA 5). The third factor was four N levels (0, 40, 80. and 120 kg/hal. The maize and cowpea populations were 53,333 (75 x 2 25 em spacing) and 66,667 (75 x 20 em) per ha, respective Iy. Results of univariate analyses for two years showed a highly significant negative residual correlation (based on bivariate analysis) of -0.276 (0.006) in­ dicating strong in terspecific competition. Maize variety and N rate were highly significant (0.01). Two o 2 4 5 6 7 8 9 interactions were significant, cowpea variety x N Weeks after plantinq rate (0.02) and maize variety X cowpea variety x N Figure 5.15. Leaf area indices of cassava and indicated rate (0.03). subsidiary crops. The variety interactions give managers a choice of varieties to reduce the interspecific competition between maize/cowpeas. The variety differences are mixtures with the different cassava cultivars did not related to N fertilization. Unlike maize, cowpea's differ except between four and nine weeks after yield to increasing N varied with maize and cowpea. planting. Maize's competitive ability was higher than The grain yield of Pop 49 associated wi th VlTA 5 okra's or melon's. Melon was slightly more com­ decreased with increasing N but tended to increase petitive than okra. when associated with TVx 3236. While yield of TVx Among the crop mixtures evaluated, cassava/maize 3236 associated with TZPB was not affected by N yielded the most calories/ha/day (18.6 X 10' Kcals). fertilizer. yield of VITA 5 declined (Fig. 5.16). With Total net return per unit area of land was highest TZSR-W the trend was not consistent. when the cassava/maize mixture included melon. The cost of producing maize/cowpea was estimated Okra in cassava/maize mixtures reduced total net at 1,360 naira/ha. Intercropping maize and cowpea income slightly but farmers may choose that mixture was, in general, beneficial with higher benefits from for dietary needs. - J.C.J. Ikeorgu (National Root combinations t hat gave high maize yields. Average Crops Research Institute, Umudike, Nigeria), H.C. Ezumah, and T.A.T. Wahua (University of [badan, Nigeria) Pruning Cassava to Double Crop with Maize o • TVX 3236 Cut at certain heights. cassava will recover and o VITA-5 produce root yields equivalent to those from Ull­ - POP-49 pruned plants (Annual Report, 1981). That finding led TZSR - W us to experimenting with two maize crops through --- TZPS one cassava crop. We cut cassava stems back to 30 em height soon after the first maize crop was harvested. Briefly, the second maize crop produced hardly any grain due to poor establishment attributed to bird. • rat, and millipede damage. Cassava yields were i reduced about 50% the first year and 30% the second. The system may become viable if scarce labor forces mechanica l harvesting on large scale farms, in which case cassava might be topped at maize-ear height.­ H.C. Ezumah and T .L. Lawson Effects of Maize/Cowpeas and Nitrogen The 1983 Annual Report indicated that maize domi­ nates cowpeas, especially with high N levels. TVx 3236 cowpeas outy ielded VITA 5 with high N. 2000~~O~--~~~----~~~----~600~----~roo~----~800 The 3 x 2 x 4 factorial experiment in three Cowpea I '9 1 ha) randomized complete blocks of 1983 was repeated in Figure 5.16. Bivariate graph showing maize X cowpea x 1984. The firstfactor, maize varieties, includeda short nitrogen interaction on maize and cowpea grain yields. The population and two tall, streak·resistant varieties, circle represents the standard. error of a mean of each one with more lower leaf area eight weeks after treatment combination. Farming Systems Program 181 benefit was 1,689 nairajha at prices of 500 and 2,000 especially when maize is managed at recommended naira per ton, respectively. . levels. Then returns from cowpeas are a bonus. The The LER values for maize x cowpea X Ntreatment mean LER of 1.38 is attractive when not obtained at combinations in this experiment exceeded 1.0 with a the expense of maize yields.- H.C. Ezumah and mean of 1.38 (Table 5.39). Treatment combinations Nguyen Ky Nam with high maize yields gave the highest LER. It appears that cowpea intercropped with maize had no Mixtures of Maize, Cowpea, Okra, and effect on maize because for the two years maize Cassava averaged 4,114 kgjha, 3% above its monocrop 4,011 kgjha from unreplicated plots. But inter cropped Because traditional farmers in the humid tropics of cowpeas averaged only 496 kg, or 35% of t heir Africa generally grow crops as complex mixtures, we mono crop 1,413 kgjha average from unreplicated need to understand the basis for complementarity in plots. the mixtures, so we designed experiments to provide As expected, cowpeas had little effect on calorie data for statistical models of complementarity. yield, which depended on maize. Mean calorific value We studied how increasing subsidiary crop popu­ is 14.85 x 106 jha. lations would affect cassava and maize yields. We also Decline in total protein with increasing cowpea grew subsidiary okra and cowpea at 25, 50, 100, and yield resulted from high cowpea yields being with the 150% of their combined optimum monocrop popu­ lowest maize yields. The cowpea-maize protein ratio lations, each contributing 50%. We fixed monocrop is 2.4, much less than the maize/cowpea yield ratio of populations of cowpea and okra at 40 and 50 x lO3fha, more than 8.0. respectively, so their combined per-ha populations When N rate is reduced to benefit cowpea in were 5,000 okra + 6,750 cowpea at25%; 10,000 okra + maize/cowpea intercrop, maize yield is reduced signi­ 12,500 cowpea at 50%, etc. The cassava and maize fican tly and cowpea yield is increased significantly were managed at their optimum populations in mix­ (Tables 5.40, 5.41,5.42). Cowpea yield was still too low ture, i.e., 10,000 cassava and 40,000 maize. Two (only 35% of its monocrop yield) so its increase did fertilizer levels, 300 kg/ha of 15-15-15 broadcast at not compensate for 61% reduction in maize from planting or the same supplemented with side-dressing reduced N. Although cowpeas sell for about four four weeks after planting were superimposed, for a 4 times as much as maize, as an intercrop cowpeas' x 2 x 2 factorial experiment with three randomized value comes from their not reducing maize yields, complete blocks. Table 5.39. Maize and cowpea yields, LER, naira, calorie, and protein values at indicated treatment combinations Maize Cowpea Maize Cowpea N, yield, yield, Naira 1 Calorie Protein, variety variety kg/h. kg/ha kg/h. LER ha x 10- ' kg/ha POP-49 ... ....... . ., .. . ........ ... .. ... TVX3236 0 2,596 547 1.03 1,032 9.45 367 POP-49 ... .. , .. . .. . . .. . .. .... .... TVX3236 40 3,434 614 1.29 1,585 12.47 461 POP-49 ... .. , ..... . . .. . .. ........ TVX 3236 80 4,043 480 1.35 1,622 14.60 488 POP-49 .. ... - . . ..... .... TVX3236 120 4,382 591 1.51 2,013 15.84 545 POP-49 .. . .. ...•.. VITA-5 0 2,472 717 1.12 1.310 9.07 394 POP-49. .... . .... VITA-5 40 3,071 572 1.17 1,320 11.16 417 POP-49 .. . .. . .. _• . VITA-5 80 3,855 616 1.40 1.800 13.97 501 POP-49 ..... , . . .• . .• . ... VITA-5 120 4,065 477 1.35 1,627 14.67 489 TZSRW ......... . .. . ... TVX3236 0 3,454 502 1.22 1,371 12.50 488 TZSRW ....... . .. , . . .... . _• . .• . . _ ... TVX3236 40 4,206 399 1.33 1.541 15.15 485 TZSRW ...... .. .... _ ..•. . , • .. •..... TVX3236 80 4,724 416 1.47 1,834 17.01 538 TZSR-W .... , . . .• . .• .. . • . .. . .. . . .... TVX3236 120 4,595 443 1.46 1,824 16.55 532 TZSR-W ...... . . , . . , ... • . .. • . .• ...• .. VITA-5 0 3,150 543 1.17 1,301 11.56 531 TZSR-W .. ..... . ...... ..... . .... ... VITA-5 40 4,363 482 1.43 1,786 15.74 519 TZSR-W .... , .. ..... ... ... .... . VITA-5 80 4,596 440 1.46 1,818 16.56 531 TZSR-W .... , .. ..... ..... . ..... VITA-5 120 5,554 360 1.64 2,137 19.95 603 TZPB ...... , .. ........... TVX3236 0 3,101 590 1.19 1,371 11.27 424 TZPB ........ ..... .... .. TVX3236 40 4,158 448 1.35 1,615 15.00 492 TZPB ...... TVX3236 80 4,977 420 1.54 1,969 17.91 562 TZPB ...... .... .. . . . .. TVX3236 120 5,588 375 1.66 2,194 20.08 610 TZPB ........ . . ... . . . . . VITA-5 0 3,421 657 1.32 1,685 12.44 469 TZPB ......... . . ... . .. . . VITA-5 40 4,235 444 1.37 1,646 15.27 498 TZPB ..... ... . ........ . .. VITA-5 80 5,428 435 1.66 2,224 19.52 608 TZPB ......... . . . . . . . . . . . VITA-5 120 5,269 341 1.55 1,957 18.93 572 Mean ................ . . . . . . . . . . . . . . . . . . . . . . . . , . . 4,114 496 1.38 1,689 14.85 498 11 naira = around $1.30. 182 Farming Systems Program Table 5.40. Effect of maize variety on maize and cowpea yields under intercropping, UTA, Plantain and Cooking 1983 and 1984 Bananas Maize, I Cowpea,! Variety kg/h. kg/ha Plantain production activities increased markedly in 1984. Requests for assistance in setting out new POP·49. . .. . . ...•..... . .. 3,490 577 TZSR·W ... . ....... .. .... . 4,330 448 plantations came from Congo, Gabon, and Nigeria. TZPB ........... . .............. . 4,522 464 Many Nigerian farmers established new fields. The SEm .... . . . ... . .. . . ........ . . . . . . 80 14 demand for suckers exceeded supplies so many state ministries of agriculture established multiplication IThe effect of maize variety is highly significant on both maize and cowpea yields. plots. Technicians manning the plots were trained in specially arranged courses at Onne. The ISADAP Table 5.41. Effect orN rate on maize and cowpea project at Owerri, Imo state, requested assistance in yields under intercropping, IITA, 1983 and establishing a rapid multiplication center. Growing 1984 awareness of plantain as a commercial crop has Maize. 1 Cowpea, I stimulated demand for more advice to farmers, which N rate kg/h. kg/h. kg/h. provided W ARCORP (West African regional 0 ... . . ... . ....... . . ... .... . . 3.032 593 Cooperative for Research on Plantain) participants 40 .... . . .... . .. .......... ... .. . 3,911 493 closer contact with farmer s and led to some on·farm SO ... . . . ... .. . .. .. .. .... . .... .. 4.604 468 research. lW . .. ......................... . 4,909 431 The annual meeting of WARCORP was held in SEm .... ...... .... . ....... . ..... . 93 16 Libreville, Gabon. Gabon's Secretary for Agriculture IThe effect of N rate is highly significant on both maize and and lITA's Director General spoke. Thirty·seven cowpea yields. members and observers attended the four·day meeting.-G.F. Wilson Table 5.42. Effect of cowpea variety x N interaction I Proximity ofMulcb Source on maize and cowpea yields under intercropping,llTA, 1983 and 1984 This experiment, started in 1982, evalua tes two Nr.te, Maize, Cowpea. methods of growing mulch·source plants close to Variety kg/h. kg/h. kg/ha plantain. The aim is to r educe labor requirement for TVx3236. ..... . . .... . . . .. 0 3,051 546 mulching while retaining mulching benefits with 40 3,933 487 organic mulch . SO 4,581 439 In one method we divided the level with 8·m strips 120 4,855 470 of plantain, separated by 8·m strips of mulch·source VITA·5 . .. . ... .. . .. . ........ 0 3,014 639 plants. Thus half the land was in mulch source. In the 40 3,890 500 other method, we used double rows (2 m apart) of 80 4,626 497 plantains separated by 4·m intervals with the center 120 4,963 393 2 m planted to mulch sources (Eupatorium adoratum, SEm ........ . ...... . ..... . . . . . . . .. . 131 23 Flemingia congesta, and Pennisetum purpureum) . I Cowpea variety x N rate interaction significant . Control plantains had no mulch source and were not mulched. Higher yields from double rows of plantains (plant Multivariate analysis of yield data showed that crop + first ratoon + second ratoon) resulted solely effects of maize variety and subsidiary crop density from the higher plant population (1,667 plantsfha were both highly significant. TZSRW yielded more against 1,250 plants/hal. Yields per plant did not than twice as much as Pool 16. Surprisingly, maize did differ significantly. Mulching was not superior to the not affect cassava or okra yields. Cowpea and okra unmulched control, and there was no response to type yields increased with densities and did not affect of mulch. Results from this experiment do not support cassava nor maize yields, nor were they affected by those of previous trials where plantain responded to fertilizer treatment (Table 5.43). That observation mulch. Apparently 50% and 40,% of the land did not was supported by results of multivariate analyses, supply enough mulch to stimulate higher yields.­ showing no significant residual correlation between G.F. Wilson and R. Swennen cowpea or okra and maize or cassava (TabI5.44). One year's data indicate that including such sub· sidiary dietary protein·rich and vegetable crops as Flemingia congesta as Plantain Cover Crop cowpea and okra with maize and cassava does not We know that mulching enhances plantain yield. The reduce yields nor provide adverse intercrop com­ mulch also prevents soil erosion, the most serious petition. Instead, adding such crops increased total problem in clean· weeded plantain. Though mulching gross revenue (Table 5.43).- H.C. Ezumah and is desirable, it is costly and the organic matter Nguyen Ky Nam traditionally used in the tropics is often scarce. A live Farming Systems Program 183 Table 5.43. Effects of maize varieties, fertilizer, and subsidiary crop density on yield (t/ha) and revenue Crop Fertilizer. density. Naira/ha. Maize variety 15·15-15 % Cowpea Maize Okra Cassava gross TZSRW . ......... . . . .. . .............. ... . 300 25 0_38 4.25 0_25 10.2 4.492 TZSRW .. . . . .. . . .. . . ... . . .. . ............. 300 50 0.37 3.74 0.74 15.0 5,072 TZSRW . .....•..•.. ... .. • ... • . . . . . 300 100 0.54 3_77 1.10 16.7 5,772 TZSRW .. . ... . .. . . . •. ... . . . .. . . . . . . . . . ... 300 150 0_51 2.52 1.01 17.1 5,145 TZSRW ... .. . ... . ..... . .. ...... . .... .. ... 300+45 urea 25 0.35 3.26 0.83 21.0 5,863 TZSRW ... .. .. . . . .. . . .. . . .... ... ......... 300+45 urea 50 0.39 3_36 1.28 19.2 5,692 TZSRW ... . ............ . ... .. .. . .. .... . .. 300+45 urea 100 0.44 3.41 0.85 19.0 5,865 TZSRW ..... . ............... .... ......... 300+45 urea 150 0.47 4.54 1.58 14.8 5.862 Pool-l6.. ....... . .. . . ..... .. ........ . .... 300 25 0.59 1.38 0.57 15.2 4,315 Pool-16 . .. .. ... • . . • . . . . ... • ..•. . .... . . _ . 300 50 0.54 1.37 1.35 14.1 4,245 Pool-16 . . . .. . _ . .. . • _ .. •. __ •..• _. .• _. . 300 100 0.77 1.51 1.01 15.5 4,904 Pool-16 . . . . .. . .•. . , .. . , . .. . .. , . .. , .. . _. . . 300 150 0.50 1.20 1.53 19.0 5,089 Pool-l6 . . .. , . . .• . . , ... . . _ . ... • . . _ . .. . .... 300+45 urea 25 0_64 1.41 0.69 26_8 6,191 Pool-16 ... _, . _ . . _. , .......... __. . _3 00+45 urea 50 0.59 1.37 1.12 17.2 4.761 Pool-16 _ . _. • _. . • .. •. . . . .. _ . . . • , ..• . . . _. . . 300+45 urea 100 0.71 2.22 0_72 21.5 5,967 Pool-16 . . . _. . _., . . , .... _. . , ... . .. .... _ ... 300+45 urea 150 0 .70 2.32 1.85 23.5 6,608 Mean . .. ...... " . , , . ... . , .... ,' •.. , ... , .. , .. " . . . . ..... , . . .. . _, ,. 0 .54 2.60 1.03 17.9 5.353 SEm ........ _. .. _. . ... . .. . _. ........ . .... . .. ...... . .... .. _. ... . .. 0.60 0.59 0.34 3.7 Table 5,44. The residual correlation matrix of the Burning destroys the fallow residue leaving bare soil. four-crop complex mixture computed Mulching increases plantain yields, so it has been from the variance-covariance matrix hypothesized that retaining bush-fallow residue Cowpea. , ... . 1.000 would improve crop growth and yield. Because re­ Maize . . "., . -0.055"" 1.000 sidue from the initial clearing does not persist long Okra . . _. _. . _ -0.318"" -0.010n• 1.000 enough as mulch for the ratoons, a live source for Cassava .. , ._ 0_327"" -0.192n. 0_128n• 1.000 recurrent supplies is n eeded. To that end, the prin­ Cowpea Maize Okra Cassava ciples of alley cropping are being adapted to bush­ fallow clearing and management for plantain. A 12- year-old fallow was slashed manually and its residue cover crop is the most obvious alternative to mulch managed six ways: burned, with and without ferti ­ but it should be pruned periodically to provide lizer added; retained, with and without fertilizer enough ground cover to prevent erosion without added; and residues from fallow and bush fallow with competing with the plantain. and without fertilizer added. Flemingia congesta, a deep· rooted shrub with good The last two treatments were managed as alley biomass production, is being evaluated as a potential cropping with the land on the sides of the plot Ii ve cover and compared with mulch from Pennisetum providing plants later cut and applied to the plot as purpureum and with a clean-weeded, unmulched mulch. Trees were allowed to regrow but were pruned controL The plantains were spaced 2 x 2 m with F. regularly. Weeds were sprayed twice during the first congesta planted in the interrow space 0.5 m from the year of growth. plantain . Fertilizer was applied every sixth week for There were five replications with 20 plants per 10 months. Total fertili2er applied was 450 kg Nlha treatment in a randomized complete block design. To and 825 kg K,O/ha. Ground magnesium limestone (300 g!rnat) was applied at. planting. Furadan 10 G (nematicide/insecticide) was applied every four Table 5.45. Yields and percentages of plants months at 3 g (aj.) per mat. harvested under indicated mulching Results to date (Table 5.45) indicate that P . pur­ systems pureurn is superior to live F. congesta as mulching. Plants Plants Results from F. rongesta and the unmulched control Yield. harvested, lost. did not differ although live F. congesta cover did not Treatment t/ha O/' . 0 % stimulate yield, it may prevent soil erosion and No mulch (Control) 15.5 b ' 61.73 19_2 a degradation. Definite conclusions will come after the Live cover crop experiment ends.-R. Swennen and G.F. Wilson (F. congesta) . .. __ . _. ___ 14.5 b 57.58 9_2b Mulch Bush Fallow for Alley Cropping (P. purpureum) ....... . 19.9. 74.2 a 9.28 'Values in the same column followed by the same letter do not Traditionally the slash and hurn method is used to differ significantly (0.05). clear land for plantain in bush-fallow agriculture. t--.'ote : Data are incomplete; some plants stilI to be harvested. 184 Farming Systems Program date only height measurements of five-month-old constraints be removed as early as possible so Africa plants are available (Fig. 5.17). could participate in the international exchange of Preliminary results indicate that burning is not germplasm without the risk of introducing new advantageous for growth. Fertilization generally diseases. significantly increased plant height. The best growth The technique also cou ld facilitate rapid multipli­ performance was by plantains tha t received ad­ cation of new cultivars. especially disease-resistant ditional mulch . types_ The resolution led to !ITA being asked to Heights of plants five months after planting cor­ develop the facilities and provide training for relate positively with future yields, so the significant Africa's scientists and technicians. In 1.983 six young differences in growth may indicate significant yield scientists were trained in plantain/banana meristem differences.- G.F. Wilson and R. Swennen cu ltu·re at the tissue-cul ture facilities in Tbadan. Those facilities were not large enough for plantain Plantain/Banana Tissue-culture research, so a tissue· cuLture laboratory for plantain/ Laboratory hanana was built a t the high-rainfall station at At the 1981 meeting of the International Association Onne. Itis now functional and can cope with two train­ for Research on Plantain and Other Cooking ees at a time in four- to six-week courses. The labora­ Bananas (IARPCB), it was noted that Africa lacked tory aims to develop specia l techniques applica ble in the skills and facilities needed formeristem culture, a developing tropical countries.-D_ Vuylsteke and technique necessary for producing disease-free plant­ G.F_ Wilson ing material and for rapid multipli cation_ A re­ solution passed at the meeting requested that these Response ofCultivars to Meristem Culture At the rate plantain production is expanding in West Length of pseudo stem ( em) Africa, the convent ional method of propagation will not supply the suckers needed. A possible solution is propagation by meristem culture, that is, inducing • With fertilizer meristems t o proliferate buds that are cultured into W Without fertilizer plantlets_ Preliminary observations on various cul­ tivars' response to meristem culture (Table 5.46) indicate potentials of 24 million to 50 million plantlets/per annum of Agbagba and French Sombre cultivars, the most popular ones in West Africa. In addition to meeting the sucker requirement, the technique could be used to produce planting material free of the pests and pathogens often transmitted by contaminated suckers.- D. Vuylsteke and G.F. Wilson A Field Technique for Rapid Multiplication ILSD In addition to multiplication by in vitro meristem .05 80 cu lture, we are attempting to improve the existing field technique for rapid multiplication. Strong api­ cal dominance in pla ntain delays sucker development until after Aoral initiation, so suckers are usually not available for about a year, when the mother plant Table 5.46. Proliferation rates of some plantain cvs. tested for irH,litro multiplication at I1TA. The number of neo~formed meristem tips per explant was counted after each A B subculture period ors to 10 weeks Treatments First Second Second Figure 5.17. Effects of fer t ilizer, mulching, and burning inoculation subculture subculture (alone or in combination) on length of pseudostems, of a Cultivar tips/explant tips/explant tips/explant plant crop five months after planting. Agbagba ... _. HHO 10-35 15-·35 A. Burning after land clearing; no fertilizer French Sombre ..... B. As A, with fertilizer 20-40 15-35 15-35 N'Jock Kon _. . _. . _. 20-35 20-45 C. No burning after land clearing ; no ferti lizer 5-35 Batard .... _. . _. . .. 10-30 15-35 15-35 D. As C. with fertilizer Orishele .... . . _. . _. 10-25 10-40 10-35 E. No burning after land clearing, double amount of Mimi Abue __ . __ . __. 15- 30 5-20 15-25 residue; no fertilizer Bluggoe (banana) _. . F. As E, with 25-60 5- 45 5-45 fertilizer. Farming Systems Program 185 matures . The present field technique consists of and loosens it, and drops it back in the same position. decapitating the mother plant when it is ahout six It loosens all the soil without disturbing surface months old and destroying the main apical meristem. organic matter (old crop and weed residues). The That destroys its photosynthetic mechanism, so more second season the no-till system of treatment was suckers might he produced, and earlier, if apical used. (4) Paraplow every other year; tillage 25 em dominance could be suppressed and the foliage re, deep with a paraplow in alternate years just before tained. Also mounding up should provide a better planting the first season crop, followed by no-till as in environment for sucker development. treatment number 1 the next three seasons. (5) Chisel The treatments, four and six months after planting, plant; chisel 20 cm deep under planting position were (1) decapitating and destroying apical meri­ keeping soil in chisel track and planting on top of stems; (2) decapitating and destroying apical meri­ chisel position. (6) Strip till; plant in strips tilled 10 stems by mounding up; (3) destroying meristems but cm wide, 13 cm deep, and 75 em apart with a rotavator. retaining foliage; (4) destroying meristems but re­ A1 1 operations except the tillage are done exactly taining foliage and mounding up. Results (Table 5.47) the same on all plots except for the no-till, hand showed that destroying meristems and retaining treatment. Treatment 1 (no-till tractor) lists the foliage produced twice as many suckers as decapitat­ equipment used on all treatments except Treatment 2, ing and destroying meristems. Retaining foliage which lists its equipment. Harvesting was simulated allowed photosynthesis to continue and to produce by driving a 8O-hp tractor over the rows in the same energy for sucker development. The technique will be pattern as a picker-husker with a bin trailer attached. introduced to farmers specializing in sucker Yield followed the same pattern in 1984 as 1983 production.-G.F. Wilson, D. Vuylsteke, R. Swennen except that the 1983 yield with the paraplow every second year was lower in 1983 when it was not tilled. The 'reduced yield on the no-till plots resulted from poorer plant establishment, which increased weed Farm Mechanization pressure (Table 5.48). Cowpea yields followed the same pattern as maize but did not yi eld in relation to Minimum Tillage Trial plant population (Table 5.48). Plant population of the no-till tractor treatment almost equaled the highest­ This minimum tillage trial has completed three years yielding treatment, but it was significantly lower so (six cropping seasons) with maize the first season and deeper tillage positively affected cowpea yield. cowpeas the second each year. The 1983 and 1984 In 1983 planting depth among treatments was not results followed the same general trend. equal and yield exactly paralleled depth, i.e., highest The experimental design is randomized complete block with six treatments and eight replications. The treatments were : (1) no-till tractor; a continuous no­ till using a four-row, no-till rolling injection planter, Table 5.48_ Third year results with maize (top) and cowpea (bottom) under indicated a 10-meter boom sprayer, a broadcast fertilizer appli­ minimum tillage systems. 1984 cator. aDd an applicator to sidedress fertilizers. (2) No-till hand ; a continuous no-till system with hand­ First season operated equipment, a rolling injection planter, and maize manual fertilizer applicator. The plot is sprayed with Yield, Plants Infiltration a boom sprayer that extends above half the plot from Treatment kg/ha per ha rate, mm/hr each side. (3) Para plow once a year; tillage 25 cm deep Strip till ........ 4,937 a' 46,084 a 233 b with a paraplow just before planting the first season. Paraplowalternate The para plow sliding under the soil at an angle raises years ............ 4,925 a 46.067 a 239b Para plow once a year ........... 4,780 a 46,717 a 297 b Chisel plant .... .... 4,737 a 45,845 a 450 a No-till hand ........ 4,599 a 41,783 346 ab Tllble 5.47_ Number of suckers at least 50 em tall No-till tractor ..... . 4,118 38,548 225 b produced by plants treated four months after planting Second season cowpeas Treatment Suckers/stool' Strip till ..... .... . . 1,108 b 118,666 e Decap~tat~ng + mounding-up ___ ............. . 4.0a2 Paraplowalternate Decapltatlng ........ . .. . .............. . .... . 4.6. years . .... . . . _. . . 1,234 a 129,244 ab Destroying meristem + foJiage Para plow once retaining .. . .. , , , . , . . , . ..... . .. .. ... .• . . . 9.3b ayear ... . .... .. . 1,238. 130,222ab Destroying meristem + fo liage Chisel plant ... . ... . l,I46ab 125,022 be retaining + mounding-up .. . .. .. ... .. _ . .. . . . 11.5 b No-till hand .... ... . 994c 118,888e IAverage. No-till tractor . . .. . . 1,093 be 136,578 a 2Val uea followed by the same letter do not differ at 1% level of I Means with the same letter after them are equal by Duncan's confiden ce. Multiple Range Teat, 0.05. 186 Farming Systems Program yields were from the deepest planted plots. In 1984 we Table 5.49. Indicated. results from experimental attempted to plant all treatment the same depth so mechanical weed control in no-till depth was not a factor . cowpea, second season, 1984 Cone penetrometer readings divide the six treat­ Weeding t ime. Yield, ments into two distinct groups, one with higher Weeding method man hr/ha kg/ha readings (no-till tractor, no-till hand, and strip til­ Hoeing2 x . ....... . 362.54 d 1,253.96 • loge); and one with lower readings (para plow once a Brush cutter 1 X . .. . 82.62 c 950.32 ab year, para plow alternate years, and chisel pla nt). The RotavatoT 1 x ....... •. .• .. 54.84 be 894.14 abc deeper tillage did not seem to influence maize yields. Cutter bar I X ....... .. .. . . 17.80 • 877.87 abc Shallow, narrow tillage that prepares good seed-soil Mowerlx . ....... .. ..... . 13.54. 750.15 bc contact seems to be the important factor. No weeding ........ • ...... 0.00. 493.18 c Cone penetrometer readings made where the t rac­ C.V., (7~ . .. . .. . .• .. . ...•. . . 26.43 27.65 tor wheel had not passed and where it had during planting showed no significant difference. So the 45- hp tractor used for planting did not cause soil slasher pruned 1 hectare of 4-meter row alley crop in compaction. less than 1 hour. A brush cutter that required a Effects of different tillage systems on water in­ continuous trickle of water on the blade was practi­ filtration are shown in Table 5.48. The chisel treat­ cal. Among all cutlasses , the jungle bolo or the ment produced a significantly higher infil t ration rate standard cutlass is best, representing all other cut­ than any other treatment except the no-till hand ting tools for manual cutting. The thick-bladed treatment. High infiltration rates should permit more woodman 's mate resulted in many splits in cutting water storage during the first heavy rains and thus Leucaena. The tractor-mounted slasher is faster but it allow earlier planting. shred the r emaining stems to pieces causing ex­ There were no significant differences in weekly soil tremely long "die back." Bigger angles of cut (from moisture from 0 to 75 mm or 75 to 150 mm for any vertical) cause shorter die back , and cut stems with treatment either cropping season. Bulk density fewer splits also have less "die back." Other factors measurements showed no significant differences, measured like the number of sprouts, average dia­ although the no-till hand treatment tended toward meter, and height of cut seem to have no cutting lowest. effects. Narrow strip tillage in the seed row is expected to The same tools were used to determine total provide adequate tillage for sustained high maize amount of pruning work per hectare per season. The production on kaoliniti c soils. The parap low and first pruning time included removing branches in to deeper tillage may increase yields when strip tillage one 4·meter-wide alley. The first pruning was done is used in conjunction with the plow. We sh a ll befo re maize was planted; the second, before attempt to test that theory in 1985.- C. Garman, A. Leucaena shaded the growing maize. The second was Ju o, N. Hulugalle not laborious because there were no big stems to handle. The tractor· mounted rotor slasher was re­ Machinery for Weed Control placed by a brush cutter for the second pruning as it Because weeding is slow, hard work, we experimen­ can enter a field without disturbing the crop. The ted with the following engine· powered equipment for rotor slasher with brush cutter on second pruning weeding in 1984 : (1) a 2.5·hp brush cutter for bush took 22.78 man hours per hectare; the brush cutter clearing, slashing, inter-row weeding, alley crop alone, 93.29 man hours. The standard cutlass, still pruning, and harvesting; (2) a 7-hp rotavator for the best cutter tool, required !l8.57 man hours per tillage and inter· row weeding ; a 5-hp cutter bar hectare. mower for slashing. inter-row weeding, and harvest· In another experiment, we recorded the weight of ing; and (3) a 5-hp rotary mower for slashing, inter­ Leucaena stems and leaves at different cutting row weeding, and harvesting. heights. Cutting rates were recorded also. Table 5.50 All the power equipment drastically reduced weed· shows the results. Again, the second pruning after the ing time but also reduced yields compared with hand rotor slasher was with the brush cutter. Weights of weeding (Table 5.49). During the test the cutter·bar stems and leaves are lower after the rotor slasher mower broke down often. The rotavator excessively because it shreds them and throws them all over the damaged the crop. The brush cutter required the most fi eld. Caution is needed with the tractor·mounted hours of the powered weeders. In future tests we will rotor slasher because it throws Rying bits and pieces compare a suitable power weeder with chemical weed of Leucaena stems at high velocity. The operator controJ.- N.c. Navasero needs a shield for protection. Modifications are needed so the tractor-mounted Tools for slasher could be used for the second pruning. The Alley-crop Pruning alley crop must be cut regularly or it will grow over We tested severa) cutting tools for pruning Leucaena the alley. Neglected, it will heavily increase pruning as an alley crop, the experiment was on Leucaena that labor and may even make the rotor slasher unsuitable was more than one year old. A tractor· mounted for the big Leucaena stems.- N. C. Na vasero Farming Systems Program 187 Table 5.50. I ndicated d ata from prunin g a n a lley c r o p one season with t h e tools na m ed , 1984 Weightof stem. We ightof stem. Weight of leaves. Weight of leaves. Rate. hr/ha Rate. hr/ha kg/ha a t kg/ha at kg/ha at kg/ha at Tools at25-cmcut l at 50·em c ut :? 25·cm CUl3 50·em cut 3 25-cm cut 50·em cut Rotor s lasher :! . 6.7 b 6.7b 669.8 b 413.4 b 991.0 b 625.0 b Brush cutter (2.5 hpj .. 7.4 b 8. 1 b 1677.0 b 770.5 a b 1569.5 a b 989.9ab Heavy duty shear .... 39.6 a 87.0 1677.7 ab 770.8 ab 2191.0 a 1381.8 a Woodma n's ma te ..... 44.5 a 50.0 a 1662. 1 a 855.6 a 2026.0 a 1381.8. Thick·bla ded cutlass . 37.0a 37.4 a 2185.2 a 1007.3 a Ji51.7 a 1104.7. Standard thin- bladed cutlass ..... 37.4 a 36.3 a 1966.0. 904.2 a 1475.0 ab 930.3 ab C.V. . % ...... 35.57 32.9 28.2 28.0 21.6 21.6 IS preading time not included. 2Second pruning was by brush cutter. 3Roto r-s lasher s tems and leaves were chopped and spread widely (not easy to co ll ect). Cowpea Harvesting E xperimen t the number of persons needed to carry out accura te. re liable, convenient, and repeatable infiltration We prev ious ly reported (Annua l Report. 1982) that tests. manually picking cow peas that yield from 1.0 to 1.2 The standard ASTM double·ring infiltrometer was t/ha requires 451 man hour/ha. When a cutter bar is the bas is of the infiltrometer. We used two Aoat va lves used , the higher its clearance. the greater the cowpea to contro l the wate r level. The fl oat va lves had a 9.5· losses. Lo wered . it e ither c logs or s tones break it. mm sta inless ball to c lose a 5-mm ho le. The Aoats \\'e re Since we found that peeling and windrowing re· adjusted to maintain 90·mm water depth a bove the duce losses to minimum and require less labor than soil s u rface (Fig. 5.18). The ca pacity of t he Aoat va lve pod pi cking, we have tried harvesting cowpeas with is 2.4 li ters per minute or an infiltration rate of2.000 t raditional too ls that were improved for thres h ing mlll/hr with a head ofl. 75 m of water. Higher rates can other crops. Here are four ha r vest ing methods we be achieved by making th e orifice in the float va lve tried t h is yea r : (1 ) mower windrow, "hampasan" la rger or by increasing the height of the feed ta nk s. thresh. s ieve cleaning: (2) manua l windrow, mach ine The infi ltrometer was equipped to handl e four sets thresh and clean: (3) hand pu ll "hampasan' · combine, of infiltration rings a t once, each with a sepa rate sieve c lean ; and (4) hand pu ll , "hampasan" combine. supply tank feed ing the center rin g (Fig. 5. 19). Wate r machine c lean. The hampasa n is an improved rice depth in the supply tank is monitored by a strip·chart. thresher. water· leve l recorder made by A. Ott Kemptem. type R We modified t he mower windrower 's fl ymo to blow 20 (20.302) with a spec ia l paper feed rate of 30·60·120 the cut cowpeas to the s ide in a row. The cl eaning mm/hour and a spec ial float gearing of 1 to 1. It s ieve tray we used has a J -inch, wire-mesh bottom. The 8·hp threshing machine has adjustable rpm and a straw walker and blower for cleaning. Th e mower wincirower was more than e ight times faster than manual windrowing, but making neat bundles of cow peas for the hampasan thresher took longer than a ligning those pu lled by hand . Poor performa nce of th e threshing machine increased labor t ime with it. Further work is needed to eva luate other thresh· ing mach ines for fas te r , more efficient cow pea threshing.- N.C. Na.uasero Automa tic Recording Soil Wate r Infiltrome te r ITTA conducted approximately 800 infi ltration tests in 1983 as one of the met hods of assessing effects of cropping systems a nd ti l1 age practices on soil con­ ditions. Testing in 1984 was to develop a method of keeping constant water levels in the standard ASTM . Fi gure 5.18. The two Hoat \, ~ll ves ins talled on the double­ manua l doubl e·ri ng in fi Itrometer and simu ltaneously ring inJiltrometer with hoses attached. The oute r Hoat is keeping an accurate record of water infi ltrated with shaped to fit between the two rings. The Hoat push rod is time. The automatic recording infiltrometer deve­ placed at the center of gravilY. The Aoats are adj usted to loped and tes ted redu ced human errors and decreased mainta in 9O-mm water depth above the soil surface. 188 Farming System,s Program. The practica l workshops emphasize field case studies. To s how t hat we co ns id e r t rainers an d trainees equally qualified, we do most of the t rainin g in workin g groups. Formal lectures mainly consist of guidelines for the trainees' activities. The groups present their resul ts in plenary sessions and di scuss each other's results. The work groups copy OFR teams, and are expected to come up with a fini shed case study. Th e first 1984 workshop in July was on ex ploratory surveyin g a nd des igning on-farm trials. It was jointly organized with GTZ at Nyankpala in the southern Guinea savanna of Ghana. Part icipan ts were na­ tiona l and ex pa t riate staff from seve ral African countries. whose GTZ-funded development projects emphasize on-farm research. It conc luded with parti­ Figure 5. 19. Th e trai le r with its main ta nk , the four c ipants writing survey reports based on a t luee-day measurin g tanks with thei r recorders. the pump to transfer water from a nurse tank to the main ta nk . The pump a lso exploratory survey and designing on-farm trials. transfers water fro m the main tank to t he measuring tanks. Major problems for which the groups sought so­ Some of the hoses that go to the infiltration rings a re shown luti ons were erra t ic ra infall , s hallow soils . low so il connected to t he measuring tanks. fertility, a nd unre li able suppli es. Among the severa l innovations they proposed were temporari ly tied ridges to improve water manage­ ment a nd inte rpla ntin g the last pre-fallow crop with required 10 minutes to set up each ring. Th e first time pi geon pea to improve fallow. water is placed in the infiltrometer rings, it takes 30 A bilingual training workshop was he ld in Boua ke. seconds for the recorder to s ta rt a n accurate record. Ivo ry Coast, in September for nationa l OFR teams From then until the test is finished. everythin g is from that country a nd Nigeria. Most of t hem had a utomatica lly o perated . partic ipated in a n ea rlier workshop in [bada n in 1983. We tested whether water must be of equal he ights and all were active ly engaged in on-farm resea rch in in the cente r and buffe r rings and found t hat 10-mm their home institut ions. The University o f Florida­ difference in wate r leve l ca used up to 6% variation in based Farming Systems Support Project provided two infiltration ra te. Fifty percent of a ll tests showed resource persons. more t han 20/0 variation in the rate with a IO-mm The workshop subject, On· Farm Experimentation. difference in water he ight . so equa l leve ls are con· built o n the 1983 [bada n exploratory-survey ing wo rk· s ide red importa nt. shop. Parti c ipants first studi ed the report of a n The infiltrometer mate ria l cost $7,400; it saves exploratory survey by one of the Ivory Coast teams in $3.850 a year in labor cost but t he major return on the the forest-sava nna transition zone, t hen verified th e in vestment is accurate information from infiltration results in the fi e ld before designing on-farm tria ls. tests.- C. Garman, Y. Sa.bel·Koschella, and M. Most of the tr ia ls they des igned were to control t he Ogida.n weed problem identified in t he survey. Besides training the pa rtic ipa nts, off·IITA campus work shops in other countries a re pa rti cul a rly useful for the host teams. Tea m members benefit greatly from insights of a group of criti ca l but sympathetic On-Farm Research and outsiders who have their own experience as back­ ground for the local farming system. Socioeconomics Participation in field research and OFR net­ Cooperation with National OFR Teams works. Tn addition to organizing training work­ shops, JTTA staff took part in a two-week ex plora­ Cooperative on-farm resea rch (OFR) activities con­ tory survey by an Ivory Coast team. It form ed th e tinued and expanded during 1984 in three areas: (1) basis for the Bouake workshop as well as the design of Training sc ient ist s from National Research Insti­ the team's 1985 on-farm trials. tutes and from development proj ects in OFR ; (2) parti· In Nigeria the fourt h Nationa l OF/FSR workshop. cipating in fi e ld research and networks of national which form ally la un ched the Nigerian FSH Network , OFR teams: a nd (3) developing OFR training was sponsored by the Ford Foundation. material s. Development of OFR methods and training Training of scientists in OFR methods. Senior materials . During 1985 we hope to complete a scientis t s a nd developers from several countri es manual to serve as the major source material for ow' attended two major training workshops held in 1984. future OFR training activities. It will be based on Farming Systems Program 189 three major workshops held to date, IITA staff ex­ Table 5.51. Seed yield of cowpeas under farmers' periences as participants in exploratory surveys, the management, on-farm trial, second trials that National OFR teams designed, and on-farm season, 1984, Ijaiye research carried out directly by IITA staff.-H.J. W. Plant %of Mutsaers, M.e. Palada, and W. Vogel Cropping density Yield, best pattern X lOOO/ha kgjha yield Preliminary On-farm Testing, Second Maize+cowpea .. .......... 42 388 49 Season, 1984, Ijaiye Maize+cassava+cowpea ... 44 219 27 Maize+ cowpea + vegetables 27 171 21 Preliminary on-farm trials were established the Cassava+cowpea ........ 72 194 24 second 1984 season to develop the research team's Sole cowpea ............ ... 158 1,227 56 ability to manage on-farm trials and to obtain infor­ mation for succeeding trials, farmers ' management level, and willingness to cooperate. Group interviews Table 5.52. Seed yield and maturity of cowpeas under before the informal exploratory survey gave the researcher-managed, on-farm trial, research t eam important information on production second season, 1984, Ijaiye practices and major constraints on specific crops to be addressed. After groups of 5 to 10 farmers were Seed yield, Maturity, interviewed, nine joined the cowpea trial and nine, Cowpea kg/ha days the maize trial. ITA82-E60 ...... _. . ....... . 769 58 Cowpea trial. Cowpea had not been grown in the TVx3236 ..... . ......... _. . . _. . . 877 71 area for 10 years or more. Farmers interviewed said IfeBrown ............. .. 1,162 65 cowpeas produce vigorous growth, but no pods. They Local control .......•.. . • ....... 484 80 were not aware that insects attack cow peas and damage flowers before pods form, but they were willing to grow cowpeas if insect damage could be minimized. Still yield data were collected from only dollar. Should the rate of 1 naira = $1.30 be adjusted five cowpea farmers. Farmers used IITA cowpea to 1 naira = about $0.33, the share of imported inputs cultivars TVx 3236 and ITA 82-E60 with three spray­ would rise from 26 to 58% of variable costs and the ings of Decis and Rogor to control thrips, aphids, and break-even yield would go from 300 to more than 500 pod borers. Most farmers intercropped cowpeas with kg/ha (Table 5.53); credit would be a stronger con­ maize, cassava, or vegetables, so plant populations straint, and good management more essential. were suboptimal. Yield reduction was 50 to 80% (Table 5.51). Cowpea yields ranged from 171 to 388 Late season maize, Farmers interviewed seldom kg/ha with intercropping. The farmer who planted grow late maize because of erratic rainfall and high cowpeas as a sale crop at recommended plant density streak disease the second season. Most farmers were obtained yields of719 to 1.227 kgfha, which are lower interested in growing early maturing, streak­ than from the researcher-managed cowpea variety resistant maize, but when interviewed most of them trial (Table 5.52). More information is needed on already had planted their own variety. Only three insect damage on farmers' fields. Damage seemed to planted streak-resistant TZESR-W maize. Yields be substantial on resistant varieties thought to be under farmer management ranged from 1.5 to 2.4 tfha. protected with insecticides. The potential yield is 3 to 4 t fha under top manage­ Farmers' reactions support the agronomic and ment. Farmers'low yields resulted primarily from too economic assessment that cowpeas can be grown few plantsfha (18,000 to 25,000). There was no in­ successfully and profitably in traditional intercrop­ cidence of streak so early maturing, streak-resistant ping or as a sole crop. All farmers are keen to participate in t he program in 1985. Cowpeas are a highly valued source of protein, which otherwise Table 5.53. Crop budget (naira) cowpeas, zero till, must be purchased. Cowpeas are grown when demand season, 1984, Ijaiye. Ibadan for labor is beginning to taper off and the risk of Present Alternative drought is low. exchange rate. exchange rate , Although only one farmer grew cowpeas as a sole 1 naira = $1.30 1 naira = $0.33 crop, his crop budget shows constraints to rapid No chemical ................... . 421 421 adoption. Timely chemical applications require high Variable costs management or a well functioning extension service. Chemicals (herb. +insectic.). . . .. 147 588 Late spraying can result in total crop loss. Both Subtotal. . . . . . . . . . . . . . . . . . . . . . .. 568 1,009 chemicals and seed require relatively large cash Share of chemicals, % . . . . . . . . . . 26 58 outlays. Management, extension service, and credit Allocated costs. . . . . . . . . . . . . . . . 32 32 are severe constraints. Another problem could be Total costs. . . . . . . . . . . . . . . . . . . . 600 1,041 elimination of subsidies. All imported inputs now are Break-even yield, kg/ha cowpea subsidized through a strong naira rate versus the at 2 naira/kg. . . . . . . . . . . . . . . . 300 521 190 Farming Systems Program varieties should insure maize production the second mines, experience with Peace Corps workers, de­ season. velopment programs (ADP), research programs Cassava. Interviewed farmers used several cas· (ACRE), household composition; tobacco growing. sava varieties. Some are late maturing and suscep­ and road connections. Farmers prefer uplands to swamps, citing lower tible to insects and diseases. Cassava green spider mites, mealybugs, grasshoppers, and mosaic virus are labor requirements, fewer hazards, intercropping for food variety, and more food and firewood , a major the major pests in the area, but farmers do not recognize that they reduce cassava yields. They source of cash income. But several farmers had only a recognize weeds as a major agronomic constraint on swamp plot with no access to a nearby upland plot. cassava product ion. They were interested in growing All farmers interviewed understood the potential early maturing varieties with dense canopy to reduce value of the swamp but lack of water control and weed competition, but it was not possible to establish labor for tillage left most swampland unused. Needed cassava trials in 1984.- M.C. Pa/ada, W. Vogel, and was a system to control water with farmers trained to P. Ay use and maintain the system. The initial phase of the project requires a land·water specialist. An agro· economist should also be involved to evaluate the Proposed On-Farm Trials and Studies water-control system in relation to individual house­ After the constraints were identified, the research holds. Potential on·farm research was proposed to team proposed possible technical interventions for find ways to reduce the labor constraint and to on·farm testing in 1985 and beyond. The interven· increase food production and net income. Research to tions were matched against the constraints iden· be done should be determined in close cooperation tified. Socioeconomic studies will be done with the with farmers, whose ideas would predominate. agronomic trials to determine both the social and Population pressure has reduced the fallow period economic feasibilities of the new technology and how on uplands. That, in turn, reduced soil fertility. farmers' and families assess the improved practices. Yields on uplands could be increased with fertilizers . Consideration will be given to gender variances in deeper tillage, herbicides, and improved varieties but accepting and managing the new technology. We supplies of fertilizers and herbicides could be un· hope to have favorable results for the 1985 Annual certain. So such other soil management techniques as Report. --M.C. Palada, IV. Vogel , H.J. IV. Mutsaers, use of compost, farmyard manure, and crop rotation and N.D. Hahn also should be investigated. All parties involved agreed that to be able to obtain Agro-socioeconomic Survey of Sierra improved varieties, credit, and other inputs (trained Leone Inland Valleys animals, equipment, tools , fertilizers, insecticides, herbicides) and to transfer knowhow on improve· After the early 1983 inventory study of the Wetland ments, close cooperation with the MANR, seed farms, Utilization Research Project for West Africa, bench· ADP, and ACRE is essential.- T. Gebremeske/, C.A. mark regions were indicated in Sierra Leone. de Vries, and Mohamed JaUoh Nigeria. and Benin Republic based on these broad criteria: physical factors (climate, hydrology, topo· graphy. soils), agronomic factors (traditional rice A Small Farm Model growing), and socioeconomic factors (population !ITA has evolved to the stage where it is ready to density). transfer to farmers technologies it has developed the In June/July 1983. an agro·socioeconomic baseline last several years. A prerequisite to transferring survey was conducted in two Sierra Leone valleys, technology is determining whether the technology is the Matunjara Mannah valley (9 km northeast of acceptable to the majority of farmers and whether it :\1agburaka) and the Romankeneh valley (10 km will improve their socioeconomic conditions. To south ofMakeni). Each was selected by hydrologists answer such questions, the Farming Systems and soil scientists. Program (FSP) uses on·farm adaptive research The survey was a collaborative effort by UTA, (OFA R) and whole· farm modeling. MANR in Sierra Leone, and the Royal Tropical The concept of whole· farm modeling is based on the Institute of The Netherlands. Agro·socioeconomic premise that a farm, especially a subsistence farm, is data on Sierra Leone and particularly on the Makeni composed of highly integrated activities and com· area were from previous surveys, followed by group ponents, so small changes in one activity or com­ interviews in both valleys to identify the swamp plots ponent cause major rarmfications throughout the and their owners for individual interviews with them. farming system. When a whole·farm model is used to Both valleys were considered suitable for the study the farm problem, it should include all the project (relatively high population density with interrelationships among production and consump­ farming systems a nd farming constraints generally tion processes, how they depend on available reo r epresentative of a larger area, and at least some sources and, most important, the farmer 's goals, farmers interested in cooperating with the project). skills, attitudes, and preferences. A suitable whole· The valleys differ in several respects: distance to gold farm model must include all direct and indirect effects Farming Systems Program 191 the technology being evaluated have on the whole likely be improved with small-scale machines. farming system. Otherwise agrobiological scientists The high marginal value product of additional cannot determine how to make the technology labor indicates that job opportunities for landless acceptable. people who migrate into the area for seasonal hired The approach is not a substitute for testing tech­ labor are excellent. nology at experiment stations or on farms. The The high marginal value product of additional approaches are complementary, For example, on­ capital indicates that the potential return on loans to farm testing can provide necessary data for modeling, farmers is high even at very high capital costs. and some mathematical models can provide infor­ It is seasonal demand for labor, not total demand, mation about the value of some resources that are not that constrains farm expansion. Thirty to 40~~ of the exchanged on markets. total household labor available is not used. So oppor­ Linear programming (LP) was used in 1984 to tunities exist to increase farm income by creating develop a computer model of a small subsistence farm. productive activities that use the underemployed. We hope the model will be duplicated in other farming Weeding is the major agronomic problem that regions in IITA's mandate. limits growth of small farms. Technologies that sol ve A basic LP model is usually used to determine how the weeding problem would create opportunities to scarce resources should be allocated to optimize a expand farm size and improve management.. predetermined objective. It is assumed that farmers When farm operations are constrained by lack of have three objectives: (a) to produce enough food for available hired labor, crop production is dominated their households; (b) to produce a combination of by such crops as sorghum with relatively low labor crops that minimize risks of crop failure; and (c) to requirements. When the only constraining factor is maximize income from farm surplus. The model land, crop production will be dominated by labor­ maximizes net revenue after insuring for food pro­ intensive crops. Farmers with labor, capital, and land duction and against risk. Mean absolute deviations increase production both intensively and in crop yields and prices are used as a proxy for risk. extensively. Data for crop combinations, crop yields, and mar­ All farm families purchase food to satisfy house­ ket prices are from the Dorin Agricultural hold nutritional requirements. Net revenue , farm Development Project (IADP). Labor input and costs size, and risk are correlated. Higher income is always of hired labor, planting materials, and fertilizer were accompanied by larger farms and greater risk. And obtained from the socioeconomic unit's survev data. risk increases more at higher than lower levels of Kilocalories of energy and grams of protein r';quired income.-T. Gebremeskel are from published sources. The quality and nature of some of the data used Results ofImproved Technology were not satisfactory, particularly for distribution of In the traditional forest zone of southweste m Nigeria labor inputs, crop yields, and household consump­ three farmers were intensively observed to analyze tion. For that reason, emphasis was sillfted to their adoption behavior and the consequences of making sure that the model is structurally sound and adopting improved farm technology. The study was to produces logical, consistent results. The results provide insights into priorities farmers use in select­ obtained, except for crop combinations, were con­ ing various elements of improved technology and sistent with common knowledge and confirm that the how they incorporate the technology into their model can be used for its intended objective when farming systems. necessary data a re available. The three case-study, small-scale farmers were a The analysis focused on monitoring land, labor, subsistence farmer who used traditional farming and operating capital (credit) and their effects on net practices, a part-time, semi-commercial farmer , and a revenue, risk, and farm organization. The analysis full-time. semi-commercial farmer who combined assumes that at this stage only indigenous tech­ t raditional farming with modern inputs. The three nology is available to the farmer. A summary of the farmers had similar family labor supplies, were results follows and Tables 5.54 and 5.55 provide exposed to similar production opportunities, market quantitative details for the solutions obtained. possibilities, and social environments, and they had The numbers 1 through 8 on the extreme left ofthe similar access to extension information and nec­ tables identify alternative plans. Plan 1 was derived essary supplies. They represent the three most typical to determine the effect of la bor scarcity on the farm, farmer strata in Nigeria. Table 5.56 describes their therefore the farm is not allowed to hire labor. Plan 2 adoption levels. was derived by relaxing that assumption. Plan 3 was All three used improved seed/planting material of derived to determine the effect of land scarcity, maize and cassava crops. After accepting improved therefore available land was limited to 1.5 hectares. seed/planting material, they took up other tech· Plans 4 to 8 are alternative plans with different nology in this order: fertilizer use, tractor plowing combinations of income and risk. (custom hired), then use of herbicides. Next in· The high added value (marginal value product) of novations adopted were building a storage crib for additional labor and capital suggest substitution maize and concrete pits and platforms to process options. The economic viability of small farms would cassava tubers. 192 Farming Systems Program Table 5.54. Physical inCormation of alternative farm plans Peak labor Household demand Labor used, Farm Land used for production, ha Total period' (MD) Farm size, 1st year 2nd year land, 1st 1st 2nd Cropping plan ha Yam Maize Sorghum Cassava Yam Maize ha year year year plan2 ] 3 . •. . .•.•••. .••••. 1.04 0.06 0.32 1.01 0.03 0.95 0.95 10.0 4,5 155 158 MSYM, SYM, MSC,MSYS 2 .. . . , ..... ... .. . . 1.82 0.74 0.54 1.25 0.03 1.05 1.95 10.0 3,4 198 178 SYM , MSC, MSYS.MYM 3 . .. . ... . .... 1.50 0.26 2.11 0.61 0.04 0.61 0.61 1.5 3,4,8 190 95 MSYM, MM MYM,MCM 4 . . . .. . . . . 2.67 1.50 2.55 0.12 1.05 1.05 3.0 2,3,4.8 217 178 SYM,MSC, MSYS 5 . . . . . . . . . . . . . . 2.49 1.31 0.27 2.47 0.03 LI6 1. 16 3.0 2,3,4,8 219 188 MSYM , SYM, MSC, MSYS 6 .. . . ............. 2.37 Ll7 ] .17 2.34 0.03 1.17 1. 17 3.0 2,3,4,8 235 169 MSYM, MSC MSYS 7 .. . . .... .. . 2.32 1.65 0.63 2.18 0.03 0.63 0.63 3.0 2,3,4 232 100 MSYM,MSC, MSYS,MYM 8 ............ 2. 39 1.91 0.05 2.32 0.03 0.46 0.46 3.0 2,3,4 223 59 SYM,MSC, MSYS,MYM 11 = January- March; 2 = April; 3 = May; 4 = June; 5 = July; 6 = August; 7 = September and 8 = October-December. [n year 2, plans 2,4,5 and 6 experienced labor constraint during periods 3 and 4. 2For the cropping plans M represents maize; Y = yam; S = sorghum: C = cassava ; P = cowpea. MSYM stands for a cropping seque nce of ma ize. sorghum, yam, and second-season ma ize. 3Labor hiring permitted for all plans except plan 1. Table 5.55. Financial information of alternative (arm Table 5.56. Technology adopted and consequences, plans (credit available annually is 200 on-farm tests, Ijaiye. Nigeria naira for (arm plans 1 and 2; 500 naira for Tradi· Semi-commercial others). tional fa rme,r Financial Credit Value of Item farmer Part time Full time characteristics, used, additional naira nana credi t, naira Adoption Farm E' A' L:.A' 1st 2nd 1st Area planted to improved plan L:.E year year year:J seed I % . .. . ..... . .. . .. 50 75 75 1. . . ...... 796 118 169 Area fertilized, % ....... . a 60 75 2. . .... ... 1,851 265 200 12 8.00 Area tractor plowed, ~/o .. . a 40 75 3' ........ sao 90 187 Area sprayed with 4. ........ 2,776 529 2.20 500 12 4.98 herbicides 0;0' . ... . .... a 60 75 5 ......... 2,705 375 0.97 500 38 2.42 Maize storage capacity. 6 ......... 2,576 250 0.50 500 40 2.19 tons . ..... . .. .......... 0 4 8 7. .. .. . ... 1,8a6 25 0.21 500 5.06 Cassava processing pit & 8 ......... 1,717 0 500 4.07 dry Roor . ...... . . . . . a 0 Consequences 1 Net revenue. 2Mean absolute deviation or net revenue. Area cropped, ha . .. . . ..... 1.7 2.8 4.3 3Credit is not a constraint for any of the plans during model year 2. Cash costs, $/yr . . .. . ... . .. 150 780 910 " t:,A _ eh&n,* in A Labor input, man-daysJha .. 175 91 80 t:, E chang" in E Total crop value, S . . .. . .. . . 2,130 2,850 5,580 6Each additiona l h ecta re will increase plan 3'8 DeL revenue by 113 Value of tree crops, $ .. . .... 230 185 naira per year. The value ofl-he marginal physical product ofJand Contribution oflivestock: $. 40 125 45 iii rero for the other plans. Income from trading employment, $ .. . .. .. ... 2,150 4,090 2,210 Crop production sold <}-~ .. . . 15 50 70 Progressive fanners either borrowed cash or Measures of efficiency sought part·time jobs to finance the fann inputs. Return to land. $/ha .. ..... - 15 102 522 Seeking a nonfann, part·time job is a step toward Return to capital, S .. . ..... 0.7 1.3 3.2 commercial farm operations. Return to labor & These farmers at first only partially adopted mo· management. S/m.d. . . . .. 6.6 7.8 13.1 dern farm inputs. They used improved technologies "Calculated on the basis of arable crops only. Farming Systems Program 193 for crops to be marketed but traditional hand-hoe and the severe drought of 1983 second-crop, so the tech· short bush-fallow systems on crops produced for nology farmers chose was financially sound. Modern household consumption. When they use modern technology permits the part-time farmer to contri· inputs, farmers' crop management practices change bute 50% of his total farm production to the market, significantly. They plant sole or two-crop mixtures and the full·time farmer, 70% which compare with more densely and at opt.imum time, which is a signi­ only 15% by the full·time traditional farmer. ficant finding that indicates how producer markets Table 5.56 shows the low productivity of a tradi­ influence the adoption of improved technology. tional farm system with negative returns to land and An important result from tractor plowing and returns of no more than the current rural wage rate to using herbicides is releasing labor, particularly for family labor and management. That is a reason for the planting periods, so the semi-commercial farmers free access to community land with no rental charge. could increase cropped area up to 150% . The mid-year The low returns tempt the traditional farmer and his labor peak (Fig. 5.20) from harvesting first·season family to seek off-farm jobs in neighboring urban crops and planting second-season crops coincided centers. With farmer access to modern technology with long school vacations, so children supplied the improving, however, returns to farm labor double the additional labor that the semi-commercial farmers rural wage rate and returns to capital increase more needed. than four-fold, so the improved technology should The cash (costs) needs of a farmer moving from stabilize the farming sector and reduce rural·urban traditional to semi-commercial management in­ migration.-M. Ashraf crease five- to six·fold, an amount equal to his wife's annual cash earnings from her trading business. Agro-economic Evaluation of Cereal-based Although marginal returns to the increased cash Systems invest ment in ar able crops could be up to 350% for the Work begun in 1982 and 1983 to identify possible full·time farmer, they can be as low as 14% during the improvements in cropping·systems in the Bida transitional stage, so making the initial jump is risky. Agricultural Development Project (ADP) was con­ Also the overall return to capital was low for the part· tinued with on-farm experiments and socioeconomic time farmer though high for the full-time, semi· surveys. The 1983 Annual Report described how commercial farmer. But these returns came during production was increased by modifying the rice-based cropping system and using short-season cowpeas after paddy during the dry season. Productivity improvement described here is from Man- days the upland cereal-based system with short·season 80r-------------------------, cowpeas included during the main season. About 40% ofthe farmers in Niger state have poor, upland soils on whlch they grow low-yielding sor· 70 ghum, millets, and egusi melon. Due to low returns from small acreages (1.3 hal, they seek hired-labor, off·farm jobs to supplement their food supplies, es· 60 pecially to buy cowpeas. On·farm tests of extra·early maturing cowpeas showed that two cowpea crops can be grown by spraying a minimum of three times. 50 Average yields from the on·farm tests were 1 t/ha of beans for the crop planted in May and 800 kg/ha for the crop planted in August. The rainy season starts in 40 April/May and ceases in October. Most ofthe cereal farmers who grew a single crop of cowpeas got substantially higher yields and a few 30 who grew two crops in succession got still higher production. Gross r eturns increased more than 70o/~ with one cowpea crop and more than doubled with 20 two crops per season. Returns to capital increased more than 800%. Labor productivity increasing 50% made it competitive with labor used with cassava and 10 yam-based systems. Since labor on cereal-based small farms was avail­ able to clear and plant an additional half hectare, and OL-L-L-L-L-L-L-L-L-~~~~ JFMAMJJASOND the project distributed spraying equipment and chemicals, farmer adoption of cowpeas has been high. Months A survey in late 1984 indicated that nearly 25% of the Figure 5.20. Labor input in man-days for farm activities by project's 65.000 farmers had grown new cowpeas representatives of the types offarms in Ijaiye area, Kigeria. during the 1984 main season compared with 7% who 194 Farming Systems Program grew local cowpeas before new va.rieties were Table 5.57. Cbanges between 1975 and 1984 in introduced.-M. Ashraf percentages of indicated crops marketed in six Nigerian states. (Based on New Cash Crops in West Africa interviews ot 350 farmers) Trend analysis of market production. Percen- Information needed to analyze trends in agricultural tage of production in West Africa is scarce. Continuous data produ- Percentage usually are not collected so existing statistics would cers in Primary crop not yield reliable evaluations of development trends. Crop 1975 1983-84 sample areas Working with the University of Nigeria, we, there· With high market increase fore, developed an alternative method to assess Maize ....... 21 40 96 Six states changes in production for the market for the Sorghum . .. . . 18 41 42 Borno. Kaduna, 1975- 1984 decade. We collected data on 33 crops from Kwara, 350 farmers in six states in Nigeria: Anambra, Benue, Tomato . 39 49 15 Benue. Borno, Borno, Kaduna , Kwara, and Ondo. The average Ondo marketed proportion of all crops increased from Okra .. . 33 43 54 Six states about one·third in 1975 to half in 1984. Of special Stable market foodcrops interest are differences among single crops. Table Rice ... . . . .. . 55 61 33 Anambra. Dndo 5.57 shows trends for IITA·mandated crops and other Cowpea .... . . 51 59 19 Borno, K wara. crops that farming·systems researchers commonly Ondo find in farmers' fields. Cassava ... . . 43 44 68 K wara, Benue The method we used to collect data was to ask (incr.) Anambra, Ondo, (stable) about specific crops in interviews in villages and with yam . ...... . . 28 29 67 Kwara, Benue those in farm· research programs. It was possible to (iner.) Anambra, collect complex infonnation by farmer recall, pro· Ondo (stable) vided specific conditions were observed: Farmers Groundnut .. . 46 53 45 Borno. (increase), were a ble to answer if questions on changes in Benue (stable) production and marketing were raised for each single Bambara nut . 29 33 7 Benue crop separately. The study, therefore, started with a Pigeon pea ... 31 32 20 Anambra crop inventory and screening the number and kind of Crops with market decrease products on each respondent's farm. The respondent Cocoyam ..... 38 33 43 Anambra. Ondo was then asked the proportion of each crop consumed Millet 30 21 13 Benue. Kaduna and marketed, starting with the last season before the interview. The interviews were done in 1983 so infonnation was obtained for 1982--1984. The 1983 and 1984 data on trends were derived from a production· Senior team members periodically supervised the and-market study in 40 Nigerian villages with agri­ field work. cultural economics students. The data collection We evaluated information from 300 of the 350 included questions on short-term changes in pro­ farmers, using computers at !ITA and the University duction plans. In addition feedback from on-farm of Freiburg in Germany. We used data on 10 major research programs in the lIorin Agricultural crops: maize, yam, cassava. sorghum, groundnut. Development Project in Kwara state and in villages melon seed, okra, cocoyam, pepper (capsicum), and near Nsukka in Anambra state were used. The com­ cowpeas. Study costs included data sheets, travel bined evaluation gave us data for 10 years with 8 allowances for enumerators and supervisors. mi· instead of 10 years ' recall by farmers. nimum salaries for the students. University input for Test interviews had shown that collecting data by three months, IITA junior staff time for data input farmers' recall demanded concentration by the re­ and processing. and computer time at IITA a nd in spondent and the interviewer. Chances for success Freiburg. Using the home areas of the enumerators as are good only if a good relationship between re­ study sites lets us bypass the usually long preparation spondents and interviewers is established. We, there­ time and investment in contacts. fore , chose the best seven final-year students in There are no reliable sample frames for studies of agricultural economics and sent them to their own this type in Nigeria. [nstead of spending a large villages to collect data. We encouraged them to proportion of time and funds to determine sampling include friends and relatives among respondents, to procedures, we assumed that a large number of high capitalize on their social background in the villages. quality interviews would compensate for possible The students were trained to use precoded question­ biases. But each interviewer had these sampling naires that were developed by an agricultural econo­ rules: to include young and old farmers, to avoid bias mist, an agronomist, a rural sociologist, and the toward your age group, and include the smallest and students themselves. Students were told that the the biggest farms and male and female farmers. quality of data returned to the research team would The proportion of all 33 crops marketed increased be related to the grade of their examinat ion degree. from about one-third to more than half during the Farming Sysrems Program 195 decade. That corresponds to the increased proportion cultivars may be preferred.- H.e. Ezumah, F.I. of urban population. Relatively fewer peasant far­ Nweke (University of Nigeria, Nsukka), and P. Ay mers produce food for the increasing total population. Transferring New Crops to Farmers Differences in changes of percentages marketed A general crop survey of 194 farmers in 19 villages of varied widely among crops. The proportions of early !lorin ADP in Kwara state of Nigeria found not a planted maize and sorghum marketed has doubled single one who had ever grown soybeans. Still a since 1975. Likewise oranges and mangoes increased successful test of soybean production under farmer from about one-third to two-thirds. Significant in­ conditions could be organized with prospects for creases were also found for peppers (capsicum), loca1 market outlets and increased farmer incomes eggplants, tomatoes, okra, and sugarcane. The high from the new crop. Information from three research increase of sorghum results somewhat from new programs were combined and used to initiate and demands by brewers who substitute sorghum for expand soybean production in the project area. imported malt raw materials. Government incentives The three approaches included selecting suitable put maize prices higher than world market prices for varieties for the region on the project's research farm the second half of the decade. Still, a large proportion in cooperation with IITA soybean breeders, assessing of maize used for animal feed, especially poultry, is potential local market demand for soybeans in the imported. Sorghum marketing has increased more project area, and establishing production trials on than maize marketing. farmers' fields. Proportions marketed did not change much for a To introduce soybeans into the !lorin Agricultural number of food crops. Increases were slight for rice, Development Project Area, we used economic con· cassava, groundnuts, and cowpeas. siderations: high imports of soybean·based animal Our survey data for K wara state from a cooperative feeds for poultry farms and soybeans' potential to on·farm research program with the !lorin supply protein for Nigeria's increasing population. Agricultural Development Project indicated that the UTA's grain legume improvement program in· number of cowpea producers as well as total pro· cluded the research farm in the multilocational trial duction decreased during the decade. The small program and made international comparisons of crop sample for cowpeas (unlike for maize) gave us weak performance available. From the international trials, data. Less than 2()% of the sample farmers grew we selected TGx536·02D and TGx 713-()9 from UTA cowpeas. In addition to marketing information, we and Samsoy I from the Institute of Agricultural estimated farm sizes (each single plot was estimated) Research in Samaru to be tested on farmers' fie Ids. and gathered general socioeconomic information, TGx 536·02D performed best. A previous UTA which is being studied.-P. Ay, F.l. Nweke, H. socioeconomic survey had shown that soybeans are Ezumah, and N.B. Mijindadi used in the northern part of Nigeria. In West African Improved maize on farmers' fields. In cooper· savanna, locust beans from a leguminous tree are ative studies with the University of Nigeria, Nsukka, used to add flavor to local soups in daily dishes. lITA improved TZSRW maize in mixtures with cas· Interviews with the 194 farmers had shown that sava, okra, tomatoes, pepper, assorted leafy vege· locust bean trees were an important cash crop in the tables , and local cowpea (Vigna sp.) was evaluated area but expansion was limited because trees do not on farmers' fields. Ten farmers were supplied with establish quickly. Between 1982 and 1984 the price for enough improved seeds for a 10- x lO·m area and locust beans doubled with highs of more than 600 enough fertilizer to apply 400 kgjha of 15·15·15. The naira/ton in local markets. University of Nigeria staff assisted farmers. Soil in To assess substituting soybeans for the highly the area is highly weathered , acidic Nkpologu series demanded locust beans, we sought cooperation with (Paleustalt, pH = 4.7). the home economics unit of the Kwara State Ministry Many of the farmers used less than the l() x 10 m of Agriculture. They provided a special recipe for required at populations ranging from 5,()()() to Hdawadawa" with soybeans. Then with the IADP on· 2(),()()()!ha maize. Harvest was, however, taken from 3- farm research team the recipe was explained and x 5-m a reas and yield converted to kg!ha. Most discussed in detail with eight women ·processing farmers preferred maize at the low populations, "dawadawa" or Hiru" in four villages of the project apparently so they could intercrop with many other area. Comparisons of the end products from locust food crops in complex mixtures. Those who planted at beans and soybeans showed that soybeans not only higher populations got maize yields exceeding 1 t!ha could fully substitute for locust beans but had some (Table 5.58). The fertilized improved maize yielded advantages. The lighter color of soybean Hiru" might four times as much as when not fertilized. Contrarily. have a price advantage in the market. Processors the local fanners' variety yielded only 21 % more than might profit more because soybean processing re­ without added fertilizer. Variations in weeding were quired less time and energy. And soybeans' shorter wide. Farmers tended to manage the fertilized plots cooking time required less firewood. Immediate reo better where they expected higher yields. This simple suit of the test was that the IADP staff members trial indicated that improved varieties should be part involved started to use soybeans for Hiru)' in their of a complete production package; otherwise local households. All eight women who precessed soybeans 196 Farming Sysl£ms Program Table 5.58. Yields (kg/ha) of improved and local maize in farmers' fields in an acid soil (Nkpologu series, Paleustult) Farmers Treatment 1 2 3 4 5 6 1 Mean TZSRW + fertilizer ..... . . . . . . . . . .. . .. ... .... .. ... . 1,090 1.510 2,089 139 3,052 440 1,110 1,356 Local + fertilizer ..... . . .. .. . .. .... .. .... ... . .. . . . . 165 1.098 691 83 1.226 325 1,233 774 TZSRW. no fert.iJizer . . . . . . . .. . . . . . . . . ... . . .. .. . .. . . 105 724 361 10 511 120 319 317 Local. no fertilizer . . . .... .. . . .. . . . . .. .. . ... . , .. 320 1.194 1.039 21 1,297 371 229 639 Mean ....... , . . . .. , . . . . ... . .... , .. ..... , . . , . , ... . 570 1.132 1,047 63 1,537 314 738 in four villages said they would like to use soybeans ing innovations. Although new crop varieties regu· raw material when it is available. larly find their way to farms, massive adoption is rare The results indicate good potential for soybeans and adoption of new farming practices lags far production on small farms, as local processing mar· behind. keting channels already exist. Inefficient extension services are only a partial explanation because Africa in the more distant past Production Trials on Farmers' Fields has had mass adoption of new crop species wi th no As soybeans were completely unknown to farmers of extension-service help. Examples are cassava, maize. the research area, how to introduce the new crop various fruits, and cocoa. Recently savanna-area remained a problem. Whi ch system would best in· farmers rapidly adopted fertilizer, particularly for tegrate soybeans? Other legumes grown in the area their maize crops . are cowpeas and groundnuts, with dill'erent crop On· Farm Research (OFR) is now promoted as a mixtures. Instead of trying to substitute soybeans for systematic nsearch method" for innovations with the groundnuts or cowpeas, we recommended the most searching done in the real farm environment in close successful method from the research farm to 16 cooperation with farmers . carefully selected farmers, based on experience the On-farm search methods include continuous study two preceding seasons. They were told to plant the offarming systems and fa rmers' conditions to identify soybeans in rows 75 cm apart as sole crop. Only nine key areas for progress and testing innovations that of 16 farmers followed the instructions completely; promise success under farmers' conditions with seven intercropped soybeans in existing production farmer cooperation. systems. In some cases, intercropping was even better Because OFR must be intensive, it is carried out in than sole cropping. The maize/soybean intercrop small pilot r esearch areas that are representative of gave nearly complete weed control. But soybeans did wider "target areas." not compete with pepper and became completely Proof that an innovation is transferable is its overgrown. Cassava yield data will be available next adoption by farmers in pilot areas. In that regard, year. Some maize was harvested green before OFR overlaps with extension, so OFR staff and measurements were taken. But soybeans in mixture extension staff must act as partners. Formal cooper­ compared well with sole crop yields . Fa rmers sa id the ative agreements between a research institution and maize yields ranked like yields from other crop the extension service would sanction and strengthen mixtures. The big advantage seen was no weeding their cooperation , Extension can transfer the suc­ costs because sO}'beans covered the soil completely. cessful innovations outside pilot areas by their usual methods. A more detailed assessment of the potentials for weed suppression by soybeans in mixed cropping with UTA's FSP is now setting up model pilot areas in Nigerian ecological zones to test mainly !ITA ma­ maize is planned for 1985, as well as monitoring terials and practices under farmers' conditions and to competitions. A socioeconomic assessment of the production possibilities is scheduled for further on­ increase our experience with OFR methods. Such "backyard testing" will enhance our ability to train farm research. Meantime the commercial department of IADP, and cooperate with OFR teams in national programs. The pilot areas. all within reasonable distance from extension officers, and researchers had in creasing the Institute. are in the savanna, transition, and demands from villagers and large-scale producers for forest zones. soybean seeds. Estimated demand for seed in IADP In each area we use the following approach : (I) a area was more than 5 tons for 1985. A large proportion brief informal survey by a team of scientists, field of it can be directly related to the on-farm testing.- P. Ay. assistants. and extension staff, resulting in a list of Y. Isola, p, Oyehan, and M. Weber priority research objectives ; (2) consideration of the total farm unit including household factors and On-Farm Research women's role in decision making and work in pro­ Establishing OFR Pilot Areas duction and utilization; (3) choice ofinnovations for testing and designing on·farm trials in close coopera· African agricul ture does not yet rank high in adopt· tion with institute scientists; and (4) testing and Farming Systems Program 197 eventually pilot~scale extension of proven second. Rainfall during the second season is usually innovations. short and erratic, which limits late season maize and We recognize that additional studies, such as an cassava and sometimes results in crop failures. inventory of soil resources and labor profiles, will be November through February is the dry season. needed. Soils in the research area are light textured and A major contribution of informal surveys is that a derived from the southern basement complex. Major team observes a farming system and shares assess­ soils belong to Apomu, Egbeda, and Iwo series in the ments, which reduces errors. Several surveys by orders of alifsols, entisol, and inceptisols. Soils national teams have demonstrated that such team generally are oflow inherent fertility. work has merit. The maj or arable crops in the area are maize, The team work will strengthen what will continue cassava, and yams with minor crops of tomatoes, to be our greatest contribution and the bulk of our peppers, okra, eggplant, and amaranth, which contri­ work, on-farm testing and evaluation of new bute significantly to cash income. Intercropping is technologies. the basic cropping system with fields cropped three to Ijaiye area, transitional zone. The area com­ four years and then left fallow two to four years. Four prises six villages in the Akinyele local government, basic cropping and rotation systems are identified in only 20 km north ofIITA. It is in the transitional zone the area (Fig. 5.22). and its major villages are Alabata, Ijaiye, and Imini. Major agronomic constraints identified and Mean annual rainfall is 1,250 mm distributed in a ranked according to importance include: bimodal pattern (Fig. 5.21). The first season starts in Low inherent soil fertility and fertility in­ March and ends in June; the second is from August adequately maintained through October. The first is more reliable than the Erratic second season rains and non adapted crops and varieties that lead to frequent crop failures in the late season Rainfall (mm/IO days I Build-up of weeds over the years with difficult-to­ control grassy weeds predominating in savanna fields 100 I- 75 percentile Cassava grasshopper Ma ize streak virus ~ Green spider mite and mosaic of cassava Cowpea pest complex Low yield potential of local varieties 801- Suboptimal planting density Soil degradation by tractor till age Farms range from 1.5 to 10 ha of cropped land with either men or women farmers. Up to 40% of the output ~rc.ntile 1\ I .... is marketed to local and urban markets depending on (Median I 'i '" I I the type of produce. Priority is given to producing 60- f'oI the family food supply and cash for nonfood I A : \ expenditures . 1''"\ I I \ Two groups of farmers were identified. One group III' cultivates land mostly in the derived savanna where II ,,A ''v" 'I farms are larger, tractors can be used, and fertilizer is 40- It t. I A V\ : I .. \ applied. The other group farms in the forest. Their rtn! 1\ \ : n ~ fields cannot be tractor plowed, and high land­ /' -Nt \.\ : 1\\ clearing costs limit expansion. Fertilizer is rarely used and transport to market is expensive. For all farmers, family labor is the most important :: aJ. V\" I input, supplemented by hired labor for land clearing I \ I rj I\~ 20 - ~ and weeding. Women are involved in all stages of crop production, processing, and utilization. The explo­ :(\/~ per~n.~ V,.i \h ratory survey indicated that women farmers rely more heavily than men on hired labor. '/ V j' The physical and institutional infrastructure is ,I '~i' , , , , , , 1\ \ deficient because roads are deteriorating and become 31 28 31 30 31 30 31 31 30 31 30 31 worse with increasing distance from the major high­ J F M A M J J A SON 0 way. Agricultural supplies are irregular and often Month (IO-day intervals I not timely. Rural credit institutions are under­ Figure 5.21. Rainfall pattern at Ibadan showing the 25, 50. developed. and 75 percentiles at lO-day intervals (Uni versity ofIbadan, The major constraints are labor at land clearing 20-yr average). and weeding time, lack of credit and cash, and un- 198 Farming Systems Program First Year Second Year Third Year Fourth Year IJIF IMIA M J J A SON o J F MIAIMIJIJ AISIOINID J FIM AIM JIJIA 510 Nil J IF MIAIMIJ JASIDND I E. Maize I /L. Maize I / C Q S !I a v a / / C a 5 5 a v a / Repeat or fallow IL. Maize, IE . Maize I 2 / C a •• a v a / / C a s • a v a / Repeat or fallow LE Maiz.e 7/l.Maize 3 a s s a v a E. Maize/ Cassava / C / LUloize, or fallow I C a • • a v a J IL.Maize /l.MaiZlll'l IE.Ma ize / 4 I Yams I I c a s s a v a I I I,,;ossava Figure 5.22. Major cropping systems and rotat ions in Ija iye (derived savanna). reliable sources of modern supplies. Land is not yet a In the uplands, cropping patterns consist of sor­ constraint in the area. ghum intercr opped with millet, melon, or maize; Major emphasis in tr ial design is on establishing yam/maize intercrop, and cassava (Fig. 5.24). Upland response curves to scarce inputs to derive recom­ fields are cropped for 10 to 15 years and then left mendations for groups of farmers with different fallow a short period. In the lowland (valley bottom) resource endowments. cropping intensity is high with rice in the wet season and upland crops in the dry season. Crops like sweet Gara and Anfani-southern Guinea savanna. cassava, sweet potato, okra, pepper, and green vege­ The research sites are in two inland valleys of Bida, tables use t he res idual soil moisture after rice. Niger state, which were select ed by the Wetland Normally they are planted on big mounds to avoid Utilization Research Project. The Kunka valley standing water in high-seepage a reas. (Gara) east of Gbako river, is approximately 15 km The major technical constraints are location speci­ from Bida township. Villages included are Gara, fic. In the upland, they include erratic rainfall and Patungi , Gbakogi, and Nasuko. The total catchment highly permeable soil (so less drought-tolerant crops area is about 2,000 ha, 5 to 10% valley bottom. The often fail), low soil ferti lity, low-yielding crop va­ Echin Woye valley (Anfani), west of Gbako river, is rieties, mealybugs, spider mites, and mosaic on cas­ approximately 12 km from Bida township. Its total sava resulting in poor growth and low yield. catchment area is about 1,300 ha with 5 to 10% valley Lowland constraints include drought toward the bottom. Ten villages are covered in Echin Woye end of rainy season, weeds competing with crops valley for possible on-farm research_ The valley during crop establishment, low stand densi ties, in­ bottoms of Echin Woye remain wet longer in the dry efficient land preparation and less efficient water and season than those in Kunka valley. weed control in rice, iron toxicity in localized rice Rainfall distribution follows a monomodal pattern areas, and insects, rodents, monkeys, and birds. (Fig. 5.23) and varies from year to year. Annual The major socioeconomic constraint is labor short­ rainfall averages 1,200 mm with more than ]00 mm age for land preparation and weeding. Labor is monthly from May to October. Between early May needed to plant and to weed upland crops at the same and early JUly soil moisture often is not sufficient to time rice should be planted, so it often is planted late. establish upland crops. Other constraints are poor tractor services, poor Soils in the lowland are deep, sandy or loamy sand, input delivery system, and weak extension services. and subject to flooding from surface runoll's during - H. Mutsaers, M. Palada, and W. Vogel the rainy season. In the dry season seepage water keeps the soil-moisture level high. The sandy soils of Socioeconomic Assessments the valley fringes lack fertility, and uncoated sand grains show that leaching losses are high. Alternating The Socioeconomic Unit, previously the Agricultural wet and dry conditions, which prevail in these soils, Economics Unit, was reorganized to institutionalize fa vor high N losses. Soils of the lower slopes are socioeconomic factors in farming systems and crop similar to soil of valley fringes with the same prob­ improvement programs and to assist in providing lems. Soils of the middle slopes are low in most program linkages. The focus is an inter-disciplinary nutrients and contain aluminum toxins. They are strategy in farming systems and on-farm research. planted to sorghum, millet, melon, and groundnut, Particular emphasis is given to establishing a core the crops best adapted to drought and low ferti lity unit at lITA to support expanding outreach teams soils. and programs. Farming Systems Program 199 Rainfall ( mm) donors, both to alert them to the need and to meet the increasing requests for socioeconomic assessments of 160 Probibilily of nol IITA's research programs. This program increases exceeding 80% socioeconomic factors in the crop improvement pro­ 140 gram as recommended by the external program r e­ view of IITA. (CGIAR Technical Advisory 120 Committee, August 1984.) Women and agricultural research. A strategy 100 involving women in farming-systems research, in­ itiated in 1984, includes the design of socioeconomic BO surveys and on-farm research that appropriately considers household and gender factors. Particularly 60 important are designing methods on use offood crops with emphasis on nutrition and improved household 40 food preparation, processing, and storage. Rural appraisals will focus on decision making for market­ ing or household consumption of food crops and 20 variances between female and male farmers in adop­ tion of new varieties and other technology. 0 5 Farm and household equipment for rural Decade number women. The agricultural Engineering Unit is in­ Figure 5.23. Decades of r ainfall totals, Bida , Nigeria. itiating a program to survey types of equipment required by women and to design and test appropriate tools and equipment. The long-term objective is to .:; design, test, and fabricate acceptable small tools, Sorghum ;;- Upland equipment, and machinery for rural women, both in Mi II e! "7 their houses and on their farms. Expansion of IITA library documentation. A L.. Rl c e ::::?-' S Potato ,/ total of 2,000 new entries related to women and Lowland L.. Ric e :;'l- Foll OW ,/ agricultural research h ave been placed in the IlTA library. Linkages were developed to disseminate this Rainfall (mm) information to regional and nationa l institutions, paricularly national women and development units 200 in African ministries, and to Colleges of Agriculture. Food crops utilization program. A new program is being planned to improve the processing and 100 ;j,!~!il!ut utilization of food crops. It is to assist countries in o~I~I~J ~I~'~:i=ll~~=~I~~~~llt~!:! ~;~~~~~l~!: sub-saharan Africa resolve problems related to food utilization by adopting improved crop production practices, increasing food consumption, reducing AMJJA50NDJFM malnutrition and hunger, and providing related Figure 5.24. Cropping systems in rainfed areas of Gara­ economic and social benefits. Anfani Valley. Training women in food crops research, pro­ duction, and utilization. Emphasis is being given to The team approach now includes one scientist women's role in all stages of the food production assigned part-time to each crop improvement pro­ cycle-research, production, and utilization. Research will focus on collecting data indicating gram; networking with socioeconomists in national institut ions and ministries; specific socioeconomic gender differences in decision making and work in studies for maize, roots, and tubers; socioeconomic household and agricultural production. The data will components for on-farm research; and programs for serve as a basis for new emphasis in UTA's training home and small-scale, village-level utilization of program in adding utilization factors to crop re­ lITA's mandated crops. search and production courses, increasing the num­ New areas of emphasis include studies on ber of women in lITA group and individual training household! compound farms as part of the agro­ programs, and designing regional courses for female forestry program. household economu,s, and the extension staff on African Food Crops Research, introduction of lITA technologies to woman farmers. Production, and Utilization. Work by staff and consultants has focused on develop­ Women's role in agricultural development. To ing project proposals to be submitted to international assess women's role in Oyo state of Nigeria, we 200 Farming Systems Program interviewed 62 rural women in five villages in a agreed, were termites, grasscutters, and grass­ sample as representative as possible. At least 15 of hoppers, and a major requirement for capital. them were independent owners of land, primarily inherited from their fathers or a husband. All were Women as independent land owners and man­ involved in marketing farm products (approximately agers. By Yoruba traditional law and custom, both sons and daughters inherit family property. Women 25% of their time) and at least 50"10 of their time was spent in agricultural work in fields. Their lives were do not a lways share equally with men but the right compounded with field. market, and household work enables them to seek protection in court. This can· in addition to processing cassava for extra income. fi rms findings in two ofthe five villages where women The interview days had to be carefully planned were primarily independent owners of land. Earlier research (1976-78) by UTA scientists in the remote because of marketing tasks- no interviews the day village Ayanaba near Osun River (Ibadan East) found before the market or the day of the market. that about one-third of the farmers were women. General trends that were apparent included: (1) joint decisions by men and women concerning crops Family income, decision making, and control. grown; (2) consumption patterns between home and Within Yoruba culture, husbands and wives have market were equal; 50% was for home and 50% was certain individual financial responsibilities. The sold in the market. Several women sa id that the food husband houses and feeds the family (from the farm or from their husband's land should always be used first by providing money to buy staples). provides basic to feed the family . (3) Women used more hired labor clothing items, and pays the children's school fees, than men because of the physical strength required but many of the women said high food costs and and their involvement with other tasks, particularly economic dema nds took the income of both the marketing and processing ; (4) women planted the husband and wife for food a nd clothing. The items same crops as men, regardless of t he size of plots and purchased included milk, tea, coffee, sugar, rice, there seemed to be no variance in years land was left cow peas, and oil. fallow or inputs placed on the farm. Both men and The respondents consistently indicated t hat about women definitely planned according to preference 50% of the food produced on their farms was can· and market prices , which is contrary to studies surned at home and 50% marketed. indicating that women's crops differ from men's, with more vegetables and food crops produced by women. Food processing. A major activity of women was (5) The women's day was diversified with field work, processing cassava into gari. Harvested fresh. cas· control of the market place, carrying fuel and water, sava does not store well and must be consumed or and looking after children. processed within three days. Well processed gari can be stored up to six months. Yet most of the women sold Cropping patterns. The farmers balanced their their gari within a month-demand was high. In the crops. Those in the field were primarily cassava, November 1984 interviews, women explained that maize, yams, cocoyams. egg plant, igba, okra, pep· they purchased whole cassava fields and then organ· pers, tomatoes, sweet potato, melon, and leafy green ized harvesting and processing for market sales. vegetables, while those in the forest areas were The women undertake all the marketing, not only pineapple. papaya, bananas, guava, plantains. of gari, but all food products, and they use small mills oranges, kola nut, and cocoa. at the market sites to grind cassava. Contrary to some studies of women 's role in ago Recommended areas for further study. The riculture, we found no women's crops as such. Nor did disgnostic study and a review of earlier studies their decision for fallow periods differ from men' s. indicated major gaps that need further study includ· Women participated in more market activities for ing comparing men and women on accepting and extra income, using tree crops and forest goods­ managing new technologies, and on using hired banana leaves, which are carefully bound for wrap· labor. UTA tools and equipment need to he tested by ping moimoi, and selling snails or kolanuts. In [mini , women. particularly those for transporting or pro· a village near Ibadan, Nigeria , a woman entre· cessing food items. And women' s influence on agricul­ preneur hired transport to Kano and other northern tural markets and prices merits further study. Their cities during the peak orange and tangerine season. influence in the rapid adoption of soybeans on a wide Women who owned their land hired more labor scale is detailed elsewhere in the 1984 Annual than men, primarily for land clearing and weeding. Report.- Natalie D. Hahn One price quoted was 160 naira (1 naira = $0.74) for all the weeding, 5 laborers for 12 days. Another woman farmer said, 100 naira for 5 men for 12 days was more in line. Prices for plowing 1 acre ranged OFR Activities ofNCRE from 30 to 50 naira, another task for which women used hired labor. Children's labor is a major factor on Project, Cameroon farms (after school and during the holiduys), parti· Activities of the Testing and Liaison Unit of the cularly for weeding tasks. National Cereals Research and Extension organi­ The problems of the field, both men and women zation (NCRE) expanded in 1984. as both the number Farming Systems Program 201 of farms and the technological packages tried on minerals and reduce yields the first cropping season farms increased. and after long fallow periods, woody residue is We identified seven agro-climatic zones in difficult to bury. It may also increase disease and Cameroon's northwest and west provinces where we insect infestation. Its advantage is conserving and carried out 30 on-farm trials and distributed 600 building up soil organic matter. Burning destroys minikits through the extension service for farmers organic matter but requires much less land­ throughout the two provinces. Most of the farmers preparation labor than burying. Burned plant ma­ used maize with beans, groundnuts, or soybeans. terial provides ash to raise soil pH, which is impor­ Many tested various maize varieties and fertilizer tant in highly acidic soils ofthe region. Burning plant levels_ residue below the surface destroys organic matter, The rice trials were primarily on farms in the humus, and soil microfauna and microftora. Vlhile Menchum Valley and the Ndop Plain of the north­ killing useful microorganisms, it also destroys harm­ west province, 5 irrigated ones in the valley, 21 ful bacterial fungi, nematodes, and insects. The production trials on the plain where the Upper Noun increased ash content reduces soil acidity. Those who Valley Development Authority (UNVDA) operates. burn residue below the soil surface recommend it We also surveyed 172 farmers in the two provinces. especially for colocasia and pumpkin. We studied The on-farm trials show us how success draws on each of the three residue management systems as technological components from research stations, follows: without added fertilizer or nematocide, and that the farmers need to learn cultural practices without fertilizer but 40 kglha of nematocide and ha ve varieties of high genetic potential. (MOCAP), with 100-60-60 fertilizer but no nema­ Scientists at Bambui station have developed se­ tocide, and with both the nematocide and 100-60-60 veral high yielding varieties but little is known about fertilizer. the cultural practices that those varieties need to Fertilizer sources were urea, triple superphos­ reach their potentials. The adaptive maize agronomy phate, and muriate of potash. The phosphate, muriate section was initiated to do research at the station on of potash, nematocide, and 50 kg Nlha were broadcast cultural practices to be tested later on farmers' fields . on ridges and incorporated before planting. The other Available material needs to be tested under local 50 kg of nitrogen was applied six weeks after planting_ climatic and cropping conditions to evaluate crop We used Kasai I maize and the lone colocasia. The rotations, soil amendments, and residue management trial was laid out as a split plot with plant residue that maintain soil productivity while sustaining crop management as the main plot and the fertilizer and yields_ nematocide combinations as subplots. All treatments Researcher-managed Trials were replicated four times. The trial was on land that had been grass fallow at least 15 years. Managing plant residues. This study used crop Both maize and colocasia germinated well, and fields to evaluate plant residue management prac­ differences in maize growth due to plani residue tices and their relationship to long-term land pro­ management were distinct as early as one week after ductivity. Farmers in Cameroon western highlands germination. The maize seedlings where grass was manage plant residue by cutting the vegetation and buried and burned were darker green and more burying it to form ridges ; cutting, burning, and then vigorous than seedlings in other plots. Colocasia forming ridges; and by cutting. drying, burying, and became established much slower than maize, and the finally burning it underground. Each has good and differences persisted most ofche growing season. bad points (Table 0_59). Burying tends to immobilize Highest maize yield (7.072 t!ha) was obtained where Table 5.59. Mean maize and colocasia yields (tJha) under indicated nematocide and fertilizer applications and residue management, Bambui Plain, 1984 Residue Management Fert. & nematocide treatment management mean A l BCD Maize yields Bury grass ........... . 2.49' 1.371 1.477 2.821 4.289 Burn grass at surface .. 3.08b 1.408 1.258 4.892 4.741 Bury grass and burn ... 5.08a 4.546 4.884 7.072 6.889 LSD, 5~{, among management means 0_83 Colocasia yields Bury grass .................. • ..... 5.53a 3.69 3_34 9.08 6.01 Burn grass at surface ............ . 6.55a 4.90 4.63 9.16 7_51 Bury grass and burn ............. . 5.20a 4.43 3_28 7.11 6.00 LSD, 5~{, among management means 1.61 1 A = no fertilizer , no nematocide; B = no fertilizer, with nematocide; C = fert.ilizer, no nematocide; D = fertilizer and nematocide. ~ Means in the same column followed by the same letter do not differ significantly (0.05). 202 Farming Systems Program grass was buried and burned below the surface and and two-thirds six weeks later. At six weeks in warm fertilizer added ; the lowest yield was where residue climates, maize plants are about 75 cm tall and was burned on surface and no fertilizer was added. initiating heads. In cooler climates, a later schedule Colocasia's best yield (9.16 t/ha) was from burning would be appropriate. Proportions applied at plant· residue on the surface and adding fertilizer. Burying ing and later sidedressed should match crop develop­ and burning grass gave significantly highest yields. ment stages and rainfall in addition to soil character· Differences were not significant between burning it istics and ability to resist leaching. on the surface and burying only. To determine the best time to apply nitrogen, we Adding fertilizer increased yields of both maize and used 120 kg K/ha (urea) and Kasai [ variety at three colocasia, but nematocides did not. Nematocides locations: Bambui Plain, Babungo, and Befang. We without fertilizer reduced colocasia yields. High compared applying all the nitrogen at four different yields from burning plant residue below the surface times: at planting, four weeks, six weeks, and eight likely result from quick release of minerals. weeks after planting. Other combinations were 75, 50, Mineralization of the plant residue also likely and 25% at planting and the rest four, six, or eight improved the soil for maize by increasing soil pH. The weeks after planting. We also used the practice trial will be continued four more cropping seasons to researchers commonly use, which is 33% at planting assess changes in organic matter and soil minerals . and 67% six weeks later, and three equal appli· cations: at planting and four and eight weeks later. Plant population and fertilizer for maize. This Each treatment was replicated four times. All plota study was to determine best maize plant populations received phosphorus and potassium at 80 kg P 20,/ha for intercropping, as more than 95% of the farmers and 40 kg K,O/ha a11 at planting and incorporated. served by the Bambui station intercrop maize with Subsequent urea applications were banded on both other crops. sides of the maize row and incorporated. We used coca, a tall, leafy variety, and Kasai I, a The trial was planted at optimal planting dates for relatively short variety, each at five plant popu· each location: 20 March at Bambui Plain, 10 April at lations obtained by changing the in·row spacing and Babungo, and 26 April at Befang. keeping between·row spacing at 75 cm. The plant Germination was good at all locations, but insects populations (26,666 ; 33,333; 44,444; 53,333, and damaged the maize seedlings at Babungo so some 66,666/ha) were obtained by using 50, 40, 30, 25, and 20 replanting was done 10 days later to fill the gaps. At cm in·row spacing. The two fertilizer levels used (60- Bambui Plain the highest yield (5.42 t/ha) was with 40·60 and 120-8·60 kg/ha) were diammonium phos· nitrogen applied equally at planting, and four weeks ph.ate, muriate of potash, and urea. Diammonium and eight weeks later; the lowest yield (3.57 ttha) was phosphate and muriate of potash were applied at with nitrogen applied i5% at planting and 25% four planting; urea, six weeks later. Each treatment was weeks later (Table 5.60) . The common practice of replicated five times. The trial was planted at Bambui applying 33% at planting and 67% six weeks later Plain in a field that h.ad been fallow a year after an unfertilized crop of yam. Coca variety significantly outy ielded Kasai [ Table 5.60. Yields (kg/ha) oC Kasai I maize at three (17%), and applying 120·80·60 kg/ha increased the locations with 100 kg/he of nitrogen yield 14.5% more than 60·40·60 kg/ha. Averaged over applied when indicated. both fertilizer levels and both varieties, plant popu· Timing of lOOkg Nlha lation sign ificantl y affected maize yields ; :>,6,666 applications. weeks Bambui plants/ha yielded significantly less than 44,444 after planting Plain Babungo Befang plantafha. Yield from 33,333 planta per hectare was 0 4 6 8 (1,330m) (1 ,176m ) (700m) 3.55 tfha, significantly below the highest 4.03 t/ha. 100 3,813 6,086 5,779 Coca produced the highest yield, 4.89 t/ha, with 33,333 100 5,312 5,453 6,369 planta/ha and 120·80·60 fertilizers added. Coca and 100 5,094 5,668 5,843 Kasai I showed similar trends in response to plant 100 4.740 5,076 6,023 populations and fertilizer. The 33,333 plant popula· 75 25 3,568 6,400 3,611 tion was optimal with. 12().80·60, but produced lowest 50 50 4.731 6,111 5,727 yields for both varieties when 60-40-60 fertilizer was 25 75 4,934 5,886 5,808 75 25 2,403 5,830 5,928 added. That trend held o"er the range of populations 50 50 4,846 6,086 6,406 used. There was no significant variety x fertilizer 25 75 4,467. 5,582 6,048 interaction so fertilizer affected both varieties 75 25 4,574 3,515 5,410 similarly. The variety x population interaction was 50 50 4,723 6,378 5,917 not significant, so any difference in response to 25 75 4,502 5,827 5,931 population was by chance. Now we need to assess 33 67 5,228 5,635 6,197 effects of interc ropping legumes and maize. 33 33 34 4,518 2,507 5,954 Best time to sidedress with nitrogen. This trial Mean .. . ... ............. . .. 4.677 5.769 5.977 was to determine the best time to apply nitrogen to LSD, 5% . .. ....• . ..•. . 0· · .. 1.517 1.078 0.989 maize. The common practice is one third at planting C.V .. % .......•......... . .. 22.7 13.08 11.58 Farming Systems Program 203 produced the second highest, but not s ignificantly thieves often destroy parts of trials. Weeding time different, yield. and cropping histories vary as do other fa ctors. Still Maize responses to application date and propor­ we present maize yield resul ts for on-farm trials in tions of nitrogen applications varied across the three three of the six zones (Table 5.61). All trials in the locations. Higher yields came at lower altitudes, so high altitude zone of the Northwest province were in Kasai I seems best suited to lower altitudes. the Bui division, a savanna area with many herds­ Best yields a t the h ighest altitude were from men . The local maize, Kenya White, significantly nit rogen applied in three equal amounts; the lowest, out yields all improved varieties except MLC. Yields when a ll or most of the nitrogen was applied at ranged from 1.4 to 2.69 t lha. Local bean varieties planting. The cooler climate showed plant growth, yielded better than introduced varieties under local and rainfall , highest at Bambui Plain, leached some maize but not significantly so. Bean yields were 217 to of the nitrogen that was applied early. 495 kg/ha. Fertilizer had no effect on maize yields, but At Babungo 75% of the nitrogen at planting and increased bean yields 125 kg/ha (31% ). The local 25% four weeks later produced the highest yield; all maize without fertilizer returned the highest net nitrogen applied eight weeks after planting, the benefit to farmers and bad the lowest variable cost. lowest. Babungo is much warmer than Bambui, so the crop grows much faster there. A fast-growing crop Ndop Plain, NWP yields. Maize trial yields in the uses nutrients early and high rainfall with leaching Ndop Plain were relatively low in 1984 (Table 5.61). may increase the need . Eight weeks after planting, Contributing factors included poor germination and the fast growing crop had already exhausted the soil soil infertility. MLC with fertilizer yielded most (2.27 nitrogen and suffered irreparable damage. t/ha) ; Bacoa with no fertilizer, tbe least (0.67 t lha). The only explanation we have for why the crop at MLC outy ielded the local variety 700 kglha when Befang did not respond to applied nitrogen is that the both received fertilizer. Uncontrolled variation site is fer tile, so nitrogen added any time would only again was high (C.V. = 38%) . . add to existing supplies. That remains to be verified. Associated bean yields were not significantly affec· The data showed that good results can be obtained ted by maize variety but responded positively to at Bambui Plain by three applications of N, the fertilizer (mean yield increase, 78 kg/ha or 39% ). smallest at planting. In Babungo more should be Mean top revenue was 129,039 CFA or $287Iha for applied early . At Befang the response to nitrogen was fertilized MLC with 38% from the associated beans. mixed but equal applications, at planting and six Beans sold in Ndop for almost five times as much as weeks later gave good r esults . More trials are needed maize in September 1984 (one month after harvest) ; to verify all the results. groundnuts , nine times as much as maize. A farmer is expected to require a risk premium On-Farm Trials before adopting a new agricultural practice. Perrin et al. define the min imum risk premium asa MRRof40% Mix-cropped maize variety x fertilizer trials, or more . Moving from local maize with fertilizer Using rainfall data, alti tude, general soil fert ility , (LOC+F) to MLC with fertilizer (MLC+F), yields a common maize-based cropping patterns, maize· MRR of645%, so the optimal technology is MLC with disease incidence, and market factors (primarily fertilizer. maize price) as criteria, we broadly defined recom­ mendation domains in the Northwest (NWP) and Table 5.61. On-farm variety x f'ertilizer trials in West provinces (WP) as follows: the Eastern indicated zones (Maize yields, kg/hal Highlands, NWP; the Ndop Plain , NWP ; the Bamenda Plain, NWP : Bali Subdivision, NWP ; Haut· Ndop Plateau Bui Plain Menchum Valley, NWP ; Haut·Platea de l'Ouest, WP ; Maize Fertilized (WP) (NWP) (NWP) and Valle de Noun, WP. "A re commendation domain" is a group offarmers PH 290 .... " ... yes l 4,164 PH 290 . " . " . " " " . . no 5,020 in an agro-climatic zone who follow similar practices Saw ...... ....... . yes 4,266 on similar farms so a recommendation is applicable to Saw ... . ............ . no 4,278 all (Perrin et a!. , CIMMYT; 1979). Thirty·three maize Coca ............. . . . . yes 1,417 1,548 trials were sent out by the TLU to farms in the six Coca ......... . .. .... . no 1.713 991 zones. These trials were to compare improved maize Bacoa . . . . . . . . . . . . . . . yes 1.438 1.374 varieties with locally grown maize, with and without Baeoa . . . . . . . . . . . . . . . no 1,602 669 fertilizer, across locations 1 in mixed culture, on MLC ....... . . .. . . . . . yes 3,914 2,176 2.273 farmers' fields. and under farmers' management and MLC . . . . . . . .. . . . . . . . no 3,976 1,574 1.128 under joint management with researchers. Controls Kasai . ... " . ......... yes 4,764 1,628 2,108 were local maize varieties provided by farmers. Maize Kasai. , .... . . . ..... . . no 4,317 1,538 1,303 Local. . " ........ , " . yes 4,058 2,573 1,527 was associated with beans or ground nuts as the Local ........ . . •. . . . . no 2,860 2,693 1,014 farmers preferred. Management of farm plots. despite the good in· LSD, 5'}'. . . . . . . . . . . . . . . . . . . 1,035 770 776 tentions of cooperating farmers, vary too much to be c.v., % . . . . . . . . . . . . . . . . . . . . 25 36 38 considered uniform. Domestic and wild animals and 1250 kg (2o.1()']O)!ha at p1anting. 204 Farming Systems Program Haut-Plateau de l'Ouest, WP yields. Fertile creases were obtained for the new entries over IRB, ffi- soils produced high maize yields for TL U trials on the 42, IR 288-3, or Tainan I, introduced from Taiwan in Haut·Plateau of the West province. Mean yields were 1968-71. from 2.86 (Local) to 5.02 t/ha (PH 290, Table 5.6l. Yield increases over local varieties were highly Maize varieties differed significantly (5%), but the significant for all new entries. Random yield varia­ fertilizer effect was not significant, and the interac­ tion was much lower than for the on-farm maize trials tion effect was significant only at 10~~. Random C.V. = 13.9%). IR 3273-339-3-5 was the high yielder at variation was relatively low for on·farm trials (C.V. 7.3 t/ha. High solar radiation the last two months of = 24.9%). the growth cycle accounted for high paddy yields, and Shading by vigorous maize apparently reduced disease incidence was low. The participating farmers groundnut yields, and fertilizer reduced them especially appreciated the earliness of Cisadane and slightly. Because groundnut yields were so low (an lITA 222's grain type and short stature. average of 2.4% of maize yields), total revenue paralleled maize yields. The highest returns included Exploratory survey of rice farmers. The TLU only $17.40 of the total (272,705 CFA or $606/ha) for staff interviewed 22 rice farmers in the Menchum groundnuts. Valley of the Northwest province in June and July. The MRR of quitting local maize with no fertilizer, The Tai wanese in traduced rice into the area in 1968. and adopting PH 290 variety, with no fertilizer is Since the Taiwanese left, the farmers have been on 2,459%. MRR's of adopting Saw·F, Kasai I·F, or MLC· their own in regard to rice. The Menchum Valley rice farmers differ somewhat from the UNVDA farmers of Fare 1,764%; 1,280%, and 1,045%, all of which are attractive. And the MRR of applying fertilizer to a the Ndop Plain. About a third of those in the farmer's local variety is 164%. Menchum Valley (32%) have off-farm employment; 68% are members of farmers' groups, and 41% grow Minikit maize trials (NWP & WP). The TLU can rice twice a year; 86% apply fertilizer, and a surpris­ carry out only 30 to 40 on-farm maize trials a year. A ing 59% use other chemicals (e.g. insecticides) on larger sample across more locations is needed for their rice. better estimates of improved maize performance Kearty all (96%) h ire occasional labor to assist (yields, yield stability, and farmers' preferences). For with clearing, tilling, and/or transplanting. But only that reason, TLU initiated a minikit maize trial in 9% have access to credit; 59% sell their rice to the 1983, and distributed it to 150 extension agents in rice cooperative at Tingoh, for an average 1983 price two provinces for farmers' fields. In 1984, 600 were of 6,253 CFA/100 kg, which compares with 5,194 distributed. CAF/1oo kg paid by UNVDA in 1983. A minikit trial consisted of four 6· X 6·meter plots During the off season, 73% of the farmers grow for one improved maize variety, with and without other crops in the rice paddies, mainly maize, ground­ fertilizer (400 kg of20-10-10/ha) and the farmer's local nuts, and vegetables.- D. McHugh, J. Kikafunda­ variety, with and without fertilizer. The improved Twine, Pauline Zeberg, M. &matana, and E. Atayi maize varieties tested included Coca, MLC, Bacoa, Kasai I, Polyhybrid 290, and Saw for the mid and high altitude zones; and Ekona White, Ekona Yellow, MLC, and Kasai I for low altitudes. The 600 minikits Maize Agronomy NCRE distributed brought 248 (41.3%) report forms to the TLU, up from 39% in 1983. Project, Cameroon Sixty percent of the participants were men; 39% We carried out 56 trials with maize in North mix-cropped their minikit maize, most commonly with beans (74%); 60% planted on ridges ; and 92% Cameroon during 1984, 24 were in IRA antennas at incorporated weeds and crop residues during land Soucoundou, Bere, Sanguere, Fignole, Tchollire, preparation. Bird damage was cited oftenest among Ndjock, Touboro, and "Mbang Mboum; 3 at specific problems (by 36% of the farmers) followed by Wassande; and 29 with la Societe pour le Developpe­ ment du Coton (SODECOTON). We planted in April stem borers, 31 %; animals, 25%; soil infertility, 18%; and May in the highland plateau to Adamaoua pro­ lodging, 17%; and diseases, 7%. PH 290 with fertilizer was the top yielder (4.78 t /ha) followed by Kasai I with vince, and in May and June in the lowland savanna of fertilizer (4.72 t/ha). Saw without fertilizer produced North provinces. We used a RCB design with four 2.81 t/ha. There was a clear response to fertilizer by replications in all trials except those on farmers' all nine varieties. Mean fertilizer response for local fields. Besides the two main varieties (Shaba 1 in varieties was 869kg/ha; for improved varieties, it was Adamaoua and TZPB in lowland savanna), we used 922 kg/ha. TZESR, Mexican 17 early, PozaRica 7843, anda local. Table 5.62 gives the soil and rainfall characteristics On-farm irrigated rice variety trials. This was of the test sites. t he t hird consecutive year that TLU conducted Objectives of NCRE research on maize are to irrigated rice variety trials in the Menchum Valley in develop improved, adapted packages of cultural collaboration with the NCRE Rice Research Unit. practices and to identify the main constraints and But it was the first year that significant yield in- test agronomic practices to alleviate the constraints. Farming Systems Program 205 Table 5.62. Effects of applied fertilizer on grain yield ofTZPB maize at indicated lowland sa vanna locations in North Cameroon 19&0& Sanguere Touboro Tchollire2 Fignole After AftCT After After After After After After Fertilizer added cotton, groundnut, cotton, gToundnut. cotton, groundnut, cotton, groundnut, N P,O. . K, O kg/ha kgfha kg/ha kgfha kgfha kg/ha kg/ha kg/h. 0 0 0 2.81 3.94 2.28 2.99 2.90 3.20 2.95 2.83 0 0 0 2.45 4.04 2.58 3.25 3.77 3.73 3.65 3.53 0 60 0 3.17 5.22 2.78 3.67 4.20 4.22 3.42 3.20 0 60 60 2.43 5.28 2.93 4.60 4.20 4.74 3.28 3.51 50 0 0 3.19 5.18 5.02 5.41 5.73 3.97 4.74 3.79 50 0 60 3.27 4.32 4.57 5.47 5.35 4. 25 4.60 3.51 50 60 0 3.69 6.88 4.84 5.61 5.28 5.04 4.71 3.70 50 60 60 4.76 6.75 4.86 5.61 5.47 5.64 4.78 3.98 100 0 0 3.75 5.28 5.26 6.57 6.65 4.66 4.76 3.91 100 0 60 3.86 4.30 5.35 6.30 6.46 4.62 4.67 4.10 100 60 0 4.61 6.79 5.60 6.45 7.07 4.40 6.10 4.17 100 60 60 5.10 7.49 5.62 5.85 6.60 4.30 5.37 4.03 150 0 0 4.17 5.43 5.73 6.06 6.29 4.60 4.89 3.61 150 0 60 4.57 4.61 6.16 5.88 6.39 4.34 4.69 3.53 150 60 0 4.97 6.65 5,49 7.23 6.81 5,45 5.49 3.74 150 60 60 4.42 7.32 6.09 5.91 6.72 4.34 4.60 4.03 X' + S 4.78 7.11 6.11 6.00 6.88 5.17 4.31 4.17 X + Zn 4.70 6.56 5.98 6.21 6.84 4.85 4.04 4.19 X + Mg 4.93 7.38 5.08 6.51 6.74 4.87 5.33 4.55 F. Big. F. sig. F.sig. F. sig. F. eig F. sig F. sig. F . 51g. ' X = 10060-60 CV=19% CV =12% CV= 14% CV=9% CV=13% CV =l3% CV= 16% CV=14% LSD,5% LSD,5% =74 LSD,5% LSD,5% =87 LSD, 5% LSD, 5% =-687 LSD. 5% LSD, 5%=588 =855 ~82 =819 ~812 At Tchollirf:, trial after cotton was planted 6 June; trial after ground nut, 12 July 1984. In our second year in northern Cameroon, we two cropping systems: cotton·maize and groundnut· evaluated yield potentials of improved varieties at maize . Ten trials were established in five t\orth different sites, responses to applied fertilizer under province locations: Sanguere, Soucoundou, Fignole, different cropping systems, planting dates, plant Tchollire, and Touboro. Treatments and results are densities. and relay intercropping maize/cowpea. We shown in Table 5.62. also tested lime and P fertilizer in Adarnaoua's As in 1983, applied N significantly increased yields ; ferralitic soils and some results of our fertilizer P did in some cases but not K or micronutrients (S, Zn, research in farmers' fields. Mg). In most cases 100 to 150 kg Nlha gave maximum Our 1984 data confirm those of 1983. The subhumid yields. But responses varied with sites, soil types, low land sa vanna of northern Cameroon has good rainfall patterns , and the preceding crop. After potential for profitable maize production . TZPB cotton and in zones with predominantly sandy soil reached 8. 7 t /ha in our experiments and around 7 t /ha texture and higher rainfall, more N was necessary for in farmers' fields. But each yield comes only by maximum yields. In general, maize, even tbe control, applying improved, adapted cultural practices. We performed better after groundnut than after cotton. need to know more about cultural practices that Groundnuts apparently contribute N for the next reduce soil erosion, favor water infiltration, and maize crop. At Mbang Mboum rnA Antenna, in the maintain soil organic matter and fertility. Crop Adamaoua highlands (about 100 mm). the fertilizer rotations, including legume green·manure crops. trial with Shaba I was in a representative ferralitic deserve concentrated study. soil after fallow. Yields responded significantly to N, Maize performed relatively well at most sites this P, and S. Maximum yield by Shaba I (7 t/ha) was with year, despite 20 to 30% less rain at many sites. 135 N, 60 P,05' 60 K,O, and 25 Mg SO. kg/ha. Replanting was necessary in some locations because After cotton at Soucounda, no fertilizer com· of rainfall difficulties and soil erosion. Streak virus bination on maize was economically justified; results was a serious problem in some locations, ]ike Fignole. after groundnuts were similar. The big surprise at And a long dry spell during grain formation reduced Mbang Mboum was a highly significant increase (to yields at other locations. 7,029 kg/hal when Mg was added to the 1(J().6().60 kg/ ha applied fertilizer. Maize Response to Fertilizer Response to planting dates. To determine best These trials were to determine response by maize planting dates for maize in north Cameroon 10' TZPB to N. P, K, and micronutrient fertilizers under cations , we selected sites at Bere. Sanguere, 206 Farming Systems Program Tchollir., Fignole, Ndock, and Touboro in the low­ land savanna and Mbang Mboum in the highlands. OFR Activities, Rwanda Maximum yields were around 8. it/ha for TZPB, 6 tfha for TZESR, and 8.7 tlha for Shaba I or Kasai I at Mbang Mboum. The 1984 results confirmed that The farming Systems/On-Farm Research Project in the Bugesera-Gisaka-Migongo (BGM) region of timely planting is necessary for improved maize varieties to achieve their yield potentials. The best Rwanda is an applied research program of the Hlnstitut des Sciences Agronomiques du Rwanda" time to plant in the lowlands was from the last week of (ISAR). It is funded through the World Bank with May to the end of June. In the highlands, it was between mid-April and mid-May. For late planting, technical assistance by lIT A. TZESR, an early maturing, streak-resistant maize The project started in 1982; a summary of the was best. r esearch activities undertaken during 1983- 84 is given below. Adding P and lime to Adamaoua's ferralitic soil. We studied short- and long-term effects of Crop Improvement applying lime and P on the ferralitic, acid soils of Adamaoua's highlands in three experiments with Crops included in the testing program are cassava, Shaba I maize. Yield response to P was dramatic, but sweet potatoes, groundnuts, cowpeas, beans, soy· not to lime. This soil seems to have a high P-fixing beans, guar, sorghum, millet, and maize_ Several lines capacity. There was no interaction between P and of each crop are tested on the poor plateau soils of the lime. The experiment will be continued to observe Kagasa (Karama) site for two or three seasons. long-term effects of P and lime, and to confirm 1984 Selected promising lines from the tests are evaluated results. in representative agroclimatic zones of the BGM region by multilocation on-farm testing. Disease Effects of plant densities on maize. These ex­ nurseries are maintained at Kayovu to screen lines periments were to determine optimal plant popu­ against major diseases_ lations for improved maize varieties at locations in the region. The tes t varieties were TZPB. Poza Rica Cowpeas. Several promising high yielding cowpea 7843, TZESR, or Mexican 17 early. The four locations lines have been identified that resist both cowpea selected are Sanguere, Fignole, Tcho11ire, and scab and ascochyta blight, the two most serious Touboro. Three plant densities were used, 66,666; diseases of cowpeas in Rwanda. Additionally, some 50,000 and 41,666/ha, with the results shown in Table lines seem to have tolerance to thrips. A combination 5_62_ of thrips, cowpea scab, and ascochyta blight some­ times completely destroyed crops and caused many Relay intercropping maize/cowpea. SAFGRAD farmers to stop growing cowpeas. results show that a variety like TVx 3236 should be relay planted about 30 to 40 days before maize har­ Soybeans. Yield of beans was low because rains vest; three to six weeks after planting is too soon and ceased early. The five highest yielding varieties were will depress both maize and cowpea yields. But we RITP·707 SI (454 kg/hal, RITP-1599 SI (387), AM-13 tested planting three and six weeks after maize, using (383), RITP-1476 SI (300), and RITP·1336 SI (290) at TVx cowpea at Sanguere and Bere. The maize va­ Kagasa_ In multilocation testing at two sites, four rieties were TZPB and TZESR at Bere and TZPB and high yielding varieties were RITP-1599 S 1 (441 kg/hal, Mexican 17 early at Sanguere. Our results showed Ikinimba blanc (372), Nyiragahini (363), and Rubona- intercropping more profitable than monocropping at 5 (324)_ Bere but the opposite at Sanguere_ The best in­ Sorghum. Emphasis is on developing early­ tercropping at Bere was cowpeas plan ted three weeks maturing, high-yielding varieties that give flexibility after maize TZPB. in planting time and permit sorghum planting in both On-farm maize variety x fertilizer test. These seasons. During the October 1983 season, the highest 29 trials conducted with SOD E COTO N on farms were yielders were Serena and Tegemeo with 1,671 and 972 to determine optimum fertilizer rates for maize TZPB kg/ha, respectively. During the February 1984 sea­ and Mexican 17 early, and to compare results from son, top yielders were Dobbs Bora. Serena, and our station trials with results from farmers' fields. Tegemeo with averages of 2,230; 1,752, and 1,746 The same fertilizer treatments were used in 25 far· kg/ha, respectively. All were introduced from Taro mers' fields with two replications per site and four (l1onga), Tanzania. additional fertilizer trials per site. Results have not Millet. A highly drought-resistant, short-season yet been analyzed. crop, millet can be successfully grown in the semi· Taste tests with maize. Taste tests were con· arid regions of Rwanda. Efforts are being made to ducted early in 1984 in four north province towns select high -yielding millets with long awns to minim­ with TZPB, Poza Rica 7843, and Samaru 123. More ize loss of grain to birds_ Best yielders during the 1983 than 85% of 150 families who participated preferred October season were Nigeria composite, IKMV-8021, the white maize TZPB or Poza Rica over the yellow Botswana 1, and Serere composite with yields of Samaru 123.-H. Ta/leyrand 1,177,1,072,1,056, and 932 kg/ha, respectively. Farming Systems Program 207 Soil Fertility Management Planted Fallows Excessive cultivation of arable land and overgrazing Owing to severe land shortage, fallows have been pastures, along with uncontrolled cu tting of trees reduced to one year or less in the BGM region. To and shrubs for firewood, have seriously degraded soil improve soil fer ti lity during such short fallow per­ throughout the region. Improving and maintaining iods. we evaluated sunn hemp (Crotalaria sp.), mu­ soil fertility are vital for food production and survival cuna (Mucuna pruriens), and pigeon pea (Cajanus of the region's population. Rwanda cannot afford cajan) as fallow species. Where fall ow land is not costly inputs, so the FSR Project is testing soil available, we intercrop with food crops. In one 7- improvement technologies based mostly on locally month trial, mucuna and sunn hemp. respectively, available biological inputs. produced 37.5 and 9.0t/ha when planted pure, and 22.3 Involved are using locally available organic ma­ and 7.1 t/ha planted with sweet potatoes. nures (farm manure and composts), rational, in­ Intercropping with fall ow species reduced sweet tensive use of legumes in cropping systems, using potato tuber yield more than 50%. To reduce that shrubs and tree legumes in association with food adverse effect, we shall use the fallow species in relay crops, and mixing crops and livestock. planting. Effects of green matter incorporated into the soil also will be monitored . Alley Cropping On-Farm Trials In alley cropping trials started at Karama and In 19 farmer-managed, on-farm sweet potato trials Kibungo, four leguminous shrubs, Leucaena leuco­ started in October 1983, mean tuber yields over the cephala, Calliandra cllothyrsus, Cassia spectabilies, Bugesera region ranged from 10.2 to 12.9 t /ha (Table and Sesbania sesban, are being evaluated mainly for 5.63). Based on yield and tuber characteristics, far· their ability to improve soil fertility in a semi-arid mers chose Rusenya (most preferred for taste) and TIS environment with poor soils and severe t ermite 2498 as the best varieties followed by Nyiramujuna, problems. In initial tests, Sesbania was best for early local, and Caroline Lee. growth and green matter production , followed by In 19 cassava trials planted in October 1983, root Cassia and Leucaena. Calliandra was planted too late yields 10 months later ranged from 18.0 to 25.9 t /ha for comparisons_ Measured nine months after plant­ (Table 5.63). Among the varieties, Eala 07 and PYT ing at Karama, mean heights were 2.15,1.06, and 1.01 Bulk 77/69 gave consistently high yields at many m for Sesbania, Leucaena, and Cassia, respectively. sites. Root yields of harvests 12, 15, and 18 months Cassia seems to res ist drought and termites better after planting, which we shall report next year, are t han the others. The trial continues so conclusive needed to assess cassava varieties.- A.f. Price, V. results will come later. Balasubramanian, D. Ci.shahayo, L. Sekayange Table ~.63. Yield and farmer preference of sweet potato and cassava varieties grown in the Bugesera region during 1983-84 Sweet Yield. Farmers' Yield, Farmers' potsto t/ha choice I Cassava t/ha choice ' Rusenya ............ . ......• ... . 12.4 3.4 PYT Bulk 77/69.. . . . . .. . .. . . 25.0 2.5 TIS-2498 . ... .. ....... . ...... . .. . 10.8 3.4 Kiryumukwe . . . . . . . . .. . ..• . . . . .. 23.6 2.4 Nyiramujuna . .. .......... . . • .... 12.9 3.2 Maguruyinkware . . . . .• . . . . . .. . . . 21.4 2.2 Local. .. . .. ... . ........ .. .. .... . U.6 2.7 Eala . .... ... . ... . .... __ .. _ . . .. . . . 25.9 2.2 Caroline Lee .......... . ........ . 10.2 2.3 Kibombwe·l. . . . . . . . . .. . .. .. . .. .. 20.3 1.0 Local. .. . ..... ........ .......... 18.0 1.6 lRanked from most to Jesst preferred. International Cooperation and Training Program The Institute consolidated its training and agronomic research should be continued at present International Programs in 1984 to serve national or increased levels of funding. If IDRC discontinues research programs better. Strengthening the pto­ support in Burkina Faso for the UTA cowpea breed­ grams was recommended by an external review panel ing activities, they should be funded under that visited the Institute during September and SAFGRAD by USAID. October 1983. The reorganization was followed by the Maize. Modest progress is being made toward high appointment of a principal training officer to manage all IITA's training activities, and a principal plan­ yields with moderate inputs, but the program's objec· tive was to develop high yielding maize with low ning officer, responsible for developing cooperative research proj ects, inputs for small farms in the semi·arid areas. The new arrangement will better ensure that tech­ The project is now on an interim one·year exten· sion preceding phase II. nology developed at UTA reaches farmers through collaborative work with nat ional institutions. It also makes training a major instrument for validating and Training transferring technology . UTA's training program is to increase the capacity In 1984, 463 people received training at UTA, includ· ing for the first time participants from Burma and of developing countries to solve their food-pro· duction problems. lIT A also hopes to multiply its Italy. The numbers and countries of those who have received training since the program began in 1970 are contacts with countries in the humid and subhumid listed in Table 6.1. The types of training offered by the tropics and, through the contacts, to introduce new technology that improves agricultural research and Institute are described in the following sections. production. Degree-related Training Project Evaluations In the degree. related training program, postgraduate students from universities throughout the world One condition of all UTA's cooperative and core· conduct the field work portion of their degrees at supplementing projects is that they be reviewed UTA, supervised by Institute scientists. The program periodically to assess progress in delivering inputs, is designed primarily to give students from the humid creating outputs, and accomplish;ng project objec· and subhumid tropics an opportunity to conduct their tives. The Semi-Arid Food Grains Research and field work in the tropics with the crops, soils, and Development (SAFGRAD) Project was evaluated in farming systems they will be faced with when they 1984. complete their degree programs. The project is to conduct regionally oriented re­ During 1984, 31 M.Sc. degree students (research search in cooperation with national, regional, and scholars) from 15 countries and 27 doctoral students other international research agencies to develop (research fellows) from 13 countries did research improved varieties of maize and cowpeas and im­ under the supervision of UTA scientists. Of the 28 proved cultural practices compatible in African small­ countries represented by the students, 20 are in farm, semi·arid farming systems. Although the head· Africa, as are 12 of the 29 universities involved. The quarters of the project is in Burkina Faso, project students and their topics are listed in Table 6.2. activities extend to 27 countries in semi-arid Africa. The major conclusions of the evaluation team Vacation Student-research Scholarships follow. The vacation student scholarship program was crea­ Cowpeas. The cowpea research program imple­ ted to establish cooperative relationships with agri· mented by UTA at Kamboinse in Burkina Faso is culture faculties in Africa, some of which have no progressing satisfactorily. The cowpea breeding and postgraduate programs. It helps identify promising 209 210 International Cooperation and Training Program Table 6.1. Training program participants by citizenship and category. October 1970 through 31 December, 1984. Citizenship Total RS' RF2 VSRS3 RTA4 Courses African Countries Angola ......... . ......... ....... ....... .... ...... ...... ......... 9 1 8 Benin . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . .. . . . . .. . . . . . . ..... .. 232 34 12 7 179 Botswana .. , . , ... .. . .. .... . . . . . . . . . . .. . ..... . . . . . .. . . .. ... ... . , . 8 4 4 Burkina Faso . ...... .. . . . . .. . .... ... . .. . ... . . . . , . . . . . . . . . . . . . . . . . 53 1 1 1 3 47 Burundi . ..... . . ... . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1 12 Cameroon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . 120 11 8 1 17 83 Central African Republic . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1 16 Chad. . . . . . . . . . . . . . . . . . . . . . . . . . .• . . . . . . . . . . . . . . . . . . .• . . . . . .• . . . . . 13 1 1 11 Comores .. . .. . . ... . ... . .. . . ....... . . . . . . . .. . . , . . . . . . . . . . . . . . . . . 1 1 Congo ....... . . . ... ... . .. . . .... . .. . . . . . ....... ...... .. . . . . ... .. 27 3 24 Egypt. . .. .. ... .. ...... .. ...... .. .... .... . .. . .. ...... .... ....... 5 1 4 Ethiopia ..... ... . .. ...... .. .. . .. . . . .. . .. . .. . . . .. . .. . . . .. . .. . . . . 32 1 1 2 28 Gabon .. . .. .. .. . .. . . . .. . . . . . .. . . ... .. ... . ..... .... . . . . .. . .. .. _. 7 7 Gambia .... . . . .. . .. . .. ...... . . ...... . . . . . ..... .. . ..... .. . .. . ... . 18 18 Ghana .......... . .. . .. . . .. . . .... ...... .. . .. ... " . . . . . . .. . .. . 153 12 8 21 21 91 Guinea.. . .. . . . . . . ... . . .. . . .. . . . . . . . . . . . . . . . . . . .. . ... . .. . . . . . 46 6 40 Guinea-Bissau. . . . . . •. . . . . . . . . . . . . . . . . . .. . . . . . . •. . .. . . 12 12 Ivory Coast. . . . . . . . .................... . ......... . _ . . . . . . . . . . 34 3 31 Kenya.. . . . . . . . . . .. .. . .. . . . . . . .. ... . . . . . . .. .. ... . . . . .... . .. . . • .. 61 2 3 56 Liberia . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . . . . . . .. . . • . . . . . . . . . . . . . . . . . . 54 3 12 39 Libya . .. . . . . . . . . . . . . .. . . . . . . .... . . • . .. .. .. . . .. . . . • . . .. .. ... . . • . . 1 1 Malagasy Republic .. . . .... .. . . .. . . .. . .. .. . .. . . . . .. .. . .. .. .. .. .. . . 7 7 Malawi. . . .......... .. .. . .. .. ... ... ......... . ... . .. ... ..... . .... 28 1 1 26 Mali. . .. . . . .. . . ...... . ... .. .. .. .. ..... .. . .. . .. .. ........ .. . .. . .. 55 9 46 ~auritania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 10 ~auritius . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1 1 2 Mozambique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 2 Niger . .. . . . .. . . . . . . ..... . .. . . . . . ... .•. ..... . . . . . . . .. . .. . . . . . 17 2 15 ~igeria . . . . .. . . .. . . .. . .. . .. . . . . . . . .. .. .. . .. .. .. .. .. . . 1.159 19 34 129 53 924 Rwanda ....... ....... . ....•.... ... .... .. . ...... .. .. 16 1 1 2 12 Sao Tome . . . . ......... . . ..... . . .......... . ...... . . ... . ........... 10 5 5 Senegal . . . .... ... . .. .. .... . . .. ...... .. . .. . ....... ........ 36 2 34 Sierra I...eone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. III 1 2 19 24 65 Somali Republic . . ... .. ... .. . .......... . . .. .. ..... ..... . .......... 18 1 17 Sudan . .... ..... . . . . ... . .. .. . .. ...... . . .. . . . .. . ....... . .. . . . . . ... 24 1 23 Swaziland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 1 Tanzania . ... .... .... . . .. . ....... .. , . . , . . . . . . . . . . . . . . . . . . . . . . . . . . 130 4 2 26 98 Togo.... ......... ... ....... .......... ... . ... . . .. ........ ........ 69 3 2 5 8 51 Uganda. . . . . . . . . . . . . . . .•. . . . . . .. . .. . . . .•. . . .. . .. . . . . . . . • . .•. . . .. 65 4 2 6 8 45 Zaire.. . . . .. ...... .... . .. . .. ........................... .. . .. ..... 133 18 11 16 88 Zambia. . . . . . . . . . . . . . . . .. . . • . . . • . . . . . . . . . . . . . . . • . . . • . . . . . . .• . . . . . 29 3 26 Zimbabwe.. . . . . . . .. . . .. .. .. .. .. . . . . . .. .. .. . .. . .. . .. . . . .. . .. . . . .. 18 13 5 Totals . 42 countries .. ..... . .. .... .. ...... ...... . ... .. ......... . ... 2.860 115 78 211 242 2.215 Asia Bangladesh . .. . . . . . . . . . .. . .. .. . .. .. . .. . .. .. . .. . . • .. . .. . . . . . . • . 3 1 2 Burma ... . .. ..... . . .. .. . , .... . . . . ........ . . . . . ... . . .. . . . . ..... 2 2 Japan ........ .. ...... .... .... . . .... . . . . ... . ... . .... .... . .. . . .. 3 2 1 India. . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . •. . . . . . . .. . . . . . . . . . . . 21 2 2 2 15 Indonesia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 4 13 Malaysia .. .. . .. ........ . . ... . . . . .. ... . . ... . . . . . . .. .. ... . . . . .. . . . 6 6 Pakistan .. . . . . . . . .... . .. . . . . . . . . . .. . .. . . ... . . . .. .. .. . . . . . . . .... . 3 1 2 Philippines. . . . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . . . . . . 10 1 9 South Korea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . 2 1 1 Sri Lanka ...... . .. .. . .. . . . . .. .. . .. .. .. .. .. . . . . .. .. .. .. .. .. . 38 1 9 28 Thailand .. . . . . . . . .. . . . . . . . . . . . . . . . ... . .. . . . . . . . . .. .. . . . . . .. .. .. . 4 1 3 Vietnam. .... . ... . ............. . . .. .. .. . .. .. ........... . .. . ... . .. 2 2 Totals, 12 countries . . .. . .•. . •.. . • ... • •..•... . . .. . . .. • ...•.. . .. . . . . 111 2 2 24 82 Australia Australia . . . . .. . . . . . . ... ..... . . . . . . . . . . . . . . .. . . . . . . . . . . . ... . . . . . . 3 1 2 Papua New Guinea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 I 4 Totals, 2 countries . . .. ....... . . ... . ... , . . . . . . . .. . .. . .. . . . . . . . . . . . . 8 2 1 4 continued overleaf International Cooperation and Training Program 211 Table 6.1. continued Training program participants by citizenship and category, October 1970 through 31 December. 1984. Citizenship Total RS' RF' VSRS' RTA' Courses Europe Belgium ................. . .. . . . . . . .. . .... . .. . . . .. . ....... . . .. ... . 42 39 2 1 England. . . . . . . . .... . . .. .. ... .. . . .. .... . . ....... .......... . ... . 15 8 1 2 2 2 West Germany ....... .... ....... . . ...... . ... . . . ........... . . . 19 6 9 4 Italy ............. .. .. .. ..... .. .. ... . .. . .. . ...... ... .. .. .. . .... . 1 1 N etherlands. . . . . ... . .. . ... , . ... . , ...... . ...... . . . . 12 7 1 3 1 Switzerland. . . . . . . . . . . . . . .. .... .. ... ....... . 3 1 1 1 Totals , 6 countries. 92 61 13 2 6 10 North America Canada . . .... . .... .. ..... ... . . .... . 5 2 2 1 United States ................ .. ..... . 54 4 5 1 14 33 Totals , 2 countries . ... .... . ... ... . . .. . 62 6 7 1 14 34 South/Central America Antigua. . . . . .. . .. . . . . . ... . . . . . . .. . . . . . . .. .. . . . . . . . . . . . . ... . 1 1 Belize. .. . .. ......... .. ....... . .... .... .... . ..... . ... ...... .. .. . . 1 1 Brazil .... . ....... ....... ..... ................ .. . .... . .. . ... ... 18 3 15 Colombia........... .. ......... .... ....... .. ................... . . 2 1 1 Costa Rica .. . ..• .. • .......... • .... .. . . .... . .. • .. • . , . . . . . . . ,. ... .. 1 1 Cuba.... . .. . ....... ...... . ..... .. . .. .. ... .... . ... . .. ... .. . ...... 1 1 Dominica. . . . . . . . . . . . .. . ..... .. . .. . .. . . . . . . . ... . . . . . . . . . .. .... ... 2 1 1 Dominican Republic. . . . . .. . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . 1 1 Guyana ............... . .. .... .... ........ . .. . .... .. . ............ 3 3 Haiti. ................ . . .. . .. ......... .. ... .. ....... .. .. ...... .. . 2 2 Jamaica . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . 4 1 1 2 Nicaragua. . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 Peru.............. .... . .. ......... .. .... ........ ... ... .. ... .. ... 6 1 5 Saint Lucia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 Saint Vincent. . . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 Trinidad... ............ .. . .. ... . .. .. . .. . ....... .. ... . ..... ... .. . 6 6 Vene-.uela.. .. . .. . . ... . ....... .. . .. ..... . . .. . . . .. . . . .•. . . . ..•. . 2 2 Totals, 17 countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 1 1 1 6 46 Pacific Region Fiji Islands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. . . .. . .. . . . . . . 2 2 Solomon Islands. . . . .. . .. .. .. . . .. . . .. .. .. . .. . . . .. . .. . . .. . . .. .. . . 2 2 Tonga..... .. .. .. . . .. . ... . . . ..... . ... .. .. . . . . . .. .. ... .. . • ... . ... 1 1 Western Samoa ............. .... . .................. ...... ... ..... 1 1 Kiribati. . . . . . . . . . . . . . . . . . . . . . • . . . • . . . . .. . . . .. . . . . . . . . . . . . . . 1 1 Totals, 5 countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 7 Grand totals. 85 countries . . .. .... ....... . ....... .... ....... . ...... 3, 195 186 102 216 293 2.398 IRS = Research Scholar (Candidate for M.Sc. or equivalent degree). 2RF = Research Fellow (Candidate for Ph.D. or equivalent degree). 3VSRS = Vacation Student-research scholars. 'RTA = Research training associate. students and often motivates them to pursue careers ing for employees of departments and ministries of as agricultural scientists. Scholarships are awarded agriculture, universities, agricultural experiment to students before the final year of their first degree stations, international organizations, and private programs. During their last long vacation as under· agencies of countries in Africa and elsewhere. graduates, they carry out field and laboratory studies Training courses are organized to fit individual needs supervised by IITA scientists. of participants. Because their needs vary widely. the Since it began in 1971, the student program has nondegree training varies from two weeks to nine attracted nominations from faculties in francophone months. Forty-two participants representing 17 and anglophone African countries. During 1984. nations (12 in Africa) received nondegree training at vacation scholarships were awarded to 12 students the Institute during 1984. from four universities in four African countries. Group Courses Nondegree-related Training Group courses that deal with various aspects of The nondegree·related training program is the agricultural production in the humid and subhumid Institute's response to requests for agricultural train- tropics are organized to help large numbers of agri· 212 International Cooperation and Train ing Program Table 6.2. Research fellows who pursued Ph.D degrees and research scholars for M.Sc. and other post-graduate degrees at UTA during 1984 with their fields of study and other indicated information Name Country Univers lt.y Field of study Research fellow. S.B.C. Wanki Cameroon University of [badan. Mineral nutrition and soil fertility of Nigeria upland and lowland rice on the Plain of Ndop, Cameroon F.N. Anno·Nyako Ghana Univers ity of Science Characterization of viruses & Technology Kumasi. Ghana K. Atta Krah Ghana Un iversity ofIbadan. Management of Leucaena fO T production Nigeria of stakes, fuel. wood, and leaf nitrogen and its potential in food crop production M.E.Ogula Zaire Faculty Institute of Soil-water extraction and root system Ag Sciences, development of rice in relation to Yangambi. Zaire cultural practices C.A. Mba Nigeria Technical University Intercropping of cassava and cowpeas of Berlin, West Germany N.O.lwuafor Nigeria Ahmadu Bello University, Crop residue management Nigeria N.Sanginga Zaire Faculty Institute of Nodulation of Leucaena and its Ag Sciences, contribution to t he nitrogen status of Yangambl, Zaire the soil E.T. Eshett Nigeria University of [badan, Production potential and fertility of l\ igeria soils from northern Cross River state, Nigeria M.B. Niyungeko Zaire Faculty Institute of Micronutrient chemistry in tropical Ag Sciences, soils Yangambi, Zaire B. Kayombo Tanzania University of Dar-eg­ Effects of soil compaction and crop Salaam~ Morogoro response under r educed ti llage systems N.B. Lutaladio Zaire University oflbadan. Effect of differen t planting periods Nigeria and associated agronomic practices on cassava performance in Zaire WongT. Fook Mauritius University of Reading, Nitrogen and water balance of maize and England upland rice in the high rainfall tropics H.R.K. Franzen West Germany University of Gottingen, Influence of tillage and traffic on yield West Germany in an Alfisol in Nigeria U. Sabel-Koschella West Germany University of Munich~ Soil erosion West Germany A.B. Salifu Ghana Wye College. University The biology and behavior of Hower thrips of London and potential for host r esistance V.A.O.Okoth Uganda University of Reading, Biology, distribution, bionomics, and England role of the Cicadulina species in the transmission of virus and occurrence of various host plants in Nigeria. B. Duguma Ethiopia University oflbadan, Factors affecting the performance of Nigeria selected species of potential in alley cropping M.Owu8u·Akyaw Ghana University ofSdence Resistance of different varieties of & Technology, Kumasi CO\"'peas to attack by storage pests, Callosobruchus macuiatus M.Gumedzoe Togo Laval University, Canada Cowpea-aphid borne mosaic virus C.E. Gyan98.Ameyaw Ghana University of Ghana, Flowering (sex expression) and fruit Legon. Ghana set in yams K. McGinnis Canada University of Guelph, Resistance to cowpea aphid. Aphis Canada cracciuora Koch in cowpeas E.N.Obazee Nigeria University oflbadan. Stalk and ear rot of maize Nigeria continued ouerleaf International Cooperation and Training Program 213 Table 6.2. continued Research fellows who pursued Ph.D degrees and research scholars for M.Sc. and other post­ graduate degrees at UTA during 1984 with their fields of study and other indicated information Name Country University Field of study B. Herren Gemmill U.S.A. University of California, Ecological dynamics and human use of Davis, U.S.A. forest fallow in southwestern Nigeria J .M. Ngeve Cameroon University of Maryland. Factors affecting yield stability among U.S.A. sweet potato genotypes C.F. Yamoah Ghana Univers ity oflbadan, Field studies of Gliricidia sepium Nigeria J . Ambe Tumanteh Cameroon University ofIbadan. Cassava agronomy Nigeria H.O. Ns;ama She Zaire University oflbadan, Biological control of cassava mealybug Nigeria Research Scholars A.S. Kumwenda Malawi Wye College, University Breeding behavior of resistance to rice of London yellow mottle virus disease of selected populations J .P.C. Koroma Sierra Leone Wye College, University Development and tes ting of synthetic maize of London varieties from selected high combining maize inbred lines in the tropics Muyolo Gilumbu Zaire University of Thadan, Synergistic effect of Xanthomonas Nigeria Campestris PV manihotis and Colletotrichum manihotis on casss va and assessment of potential yield loss B. Ban Canada University of Guelph, Canada Relationship between the root systems of fallow plants and nutrient recycling T. Kasu Zaire University oflbadan. Effect of pubescence on the cassava Nigeria mealybug and cassava green spider mite behavior and reproduction D. Ndombo Zaire University ofIbadan. Improvement of cassava by clonal mass ~igeria selection H. Adu.Dapah Ghana University of Ibadan, Extent of heterosis for yield and other Nigeria quant itative characters in cowpea J . Adabasi Benin Nationa l University of In herita nce of resistance to brown Benin, Benin blotch, septoria. and scab in cowpea A.R.S. Ibrahim Ghana University of Ibadan, An economic analysis of rice response Nigeria to fertilizer application in northern Ghana J.M. Van Den Henkel Netherlands University ofWageningen, Baseline studies of traditional The Netherlands farming in the Onne area J .C. Tossa Benin National University of Response of Gliricidia sepium to pruning Benin, Benin during and after dry sesson as well as the litter fall and its quality C.T. Njoh Cameroon University oflbadan, Integration of extra early cowpea Nigeria varieties, planting dates. and minimum insecticide application in the control ofpod sucking bugs in cow peas C. Nkwiine Uganda Makerere University, Resistance of introduced strains of Kampala, Uganda Rhizobiumjaponicum in a maize/soybean cropping system M. Kayibigi Burundi Laval University, Canada CicadulinB rearing and streak eva luation M. Makame Tanzania University oflbadan, Effect of reciprocal grafting on the Nigeria cyanide content of cassava M.J. Dalton England University of Reading, Farm management England A. Carter England University of Reading, Farm management England K.Y. Fosu Ghana University of Ghana, Consumer choice of cowpeas Legon, Ghana D.M.C. Deeoene Belgium Catholic University, Plantain agronomy Louvain. Belgium contin ued o(Jerlea{ 214 International Cooperation and Training Program Table 6.2. contin~d Research fellows who pursued Ph.D degrees and research scholars for M.Sc. and other post· graduate degrees at lIT A during 1984 with their field. of study and other indicated information Name Country University Field of study Research Scholars P.E. De Preter Belgium Catholic University. Soil chemistry Louvain, Belgium J. K. Vannoppen Belgium Catholic University, Alley cropping Louvein, Belgium P. Pussemier Belgium State University of Agric., Striga Sciences, Gembloux, Belgium F. Dumortier Belgium Catholic University. Effeet of tillage and compaction on Louvain, BeLgium soybean nodulation, microbial biomass. soil respiration and nitrogen mineralization A.M. Maponoko Zaire Catholic University, Genetic variability of root Louvain, Belgium characteristics in cowpea and its effect on drought tolerance C.M. Mayona Tanzania Sokoine Agrie. University, Factors affecting seedling establishment Morogoro. TanUlnia in the tropics B.D. Asonibare Nigeria Catholic University. Case study of the Itokin rice Louvain, Belgium project G. Akpokodje Nigeria University of Ihadan, The effects of temperature and humidity Nigeria on the pod·borer , Maruca K.P. Atropo Benin National University of Relationship between pod sucking bug Benin, Benin (Clavigrallo tomentusicollis) and RiptotU8 dentipes population and cowpea damage K.N.A. Ahouandjinou Benin National University of In vitro and in L'iuo studies for the Benin, Benin determination of a screening method for cassava anthracnose disease G. Atuahene-Amankwa Ghana University of Guelph, Canada Host plant resistance to insects in cowpea breeding A.A. Younous Chad Univer'Bity of Ibadan, Grain legumes pathology Nigeria cultural research and extension workers meet goals regions. Conducted jointly by IITA and the Centre for and cope with problems as they arise. The length of a International Programs of the University of Guelph, course depends upon the time required to achieve its Canada, the course was funded primarily by the goals. The 15 group courses offered during 1984 for 349 Canadian International Development Agency people are descr ibed. (CIDA). It attracted 24 participants from 14 countries. Cowpea and soybean production (9 January-2 Rapid multiplication of cassava (23 January-3 March). This course was to give research workers February). This course acquainted 17 Nigerian field the theoretical knowledge and practical skills needed and extension workers from Anambra. Bendel, to establish and manage cowpea and soybean crops, Kwara, Imo, Ondo, and Oyo states with the latest to conduct trials, and make the best use of improved techniques for rapid multiplication of cassava. lines distributed through international trials and nurseries. The course was attended by 29 people from Hybrid maize seed production (6-25 February). 11 countries. Sponsored by the Nigerian Seed Services Organi· Biological control of cassa va pests (9 January- zation, this course familiarized 17 extension agents 3 February), This course showed persons working to from several states with various aspects of hybrid improve root crops the latest techniques for bio­ seed production. logical control of cassava mealybugs and green spider mites. Microbiology (19 March-26 April). In this course research fellows and training officers working on soil Soil and plant analysis (16 January-I7 fertility learned how to design and draw up research February). Senior laboratory technicians were programs to clarify the role of microorganisms in trained in management, methodology, instrumenta­ tropical soils and how to increase the beneficial tion, and techniques of soil and plant analyses in this effects of the organisms. The course was conducted in course, with emphasis on soils and crops of ~ropical French for 10 people from 10 African countries. International Cooperation and Training Program 215 Weed control (26 March-l4 April). This course Ministry of Agriculture and World Bank Agricul­ emphasized improved weed control and integrated tural Development Projects. Participants were pest management. It was attended by 37 persons from brought up to date on techniques of rice research, the Nigerian Federal Department of Agriculture. extension, and production. Tropical root and tuber crop production (30 April~ July). Identifying production problems and applying improved technologies were the fo cus of this course for research workers and extension super­ visors. Working alone and in teams, 31 participants from 14 countries carried out applied research pro· jects, which they presented during an IITA farmers' field day. The course was funded mainly by the United Nations Development Program (UNDP) under a joint grant, entitled Technology Transfer on Root and Tuber Cropa. to the International Center for Tropical Agriculture (CIAT), the International Potato Center (CIP), and IITA. Also included in root and tuber training was a one· week, intensive course in root crop production for 32 extension workers. Use of farm analysis package (14 May-8 June). UTA and FAD jointly organized a four·week training workshop on the use of a farm analysis package (F ARMAP), a computer package for processing farming-systems survey data_ Fourteen people from five countries attended_ Intensive course on root and tuber crop pro­ duction (~29 June). Participants in the course on tropical root and tuber crop production planned and conducted this course for 32 field and extension workers selected from Nigerian federal and state departments of agricuLture. Tissue culture for root and tuber crops (6-24 August). This course acquainted research workers with tissue culture techniques and trained them to handle tissue culture material safely when the ma­ terial arrives in their home countries. The 13 parti­ cipants, representing 11 countries, also learned guide­ lines for testing material received. in tissue culture form_ Engineering design and manufacturing (13 August- 13 October). In this course 16 African mechanical engineers studied machine design and manufacturing techniques or industrial and agricul­ tural equipment in Africa. The course was sponsored and conducted by the Africa Regional Centre for Engineering Design and Manufacturing and was funded with a UNDP grant. Rice production (3 September-7 December). This course taught research workers and extension supervisors improved methods of cultivating irri­ gated, hydromorphic, and upland rice in tropical Africa and acquainted them with applied research for rice_ Thirty people from 18 countries attended_ Intensive course in rice production (26-30 November). This course was planned and conducted for 10 field extension workers of the Nigerian Federal Documentation, Information, and Library Reorganization were given in U.K. (Ardingly, Sussex); Italy (Bari, Florence, and Rome) and Nigeria (lbadan (NISER), The Managemen t Review of IIT A recommended that Akure, Jos, and Benin). all information units be consolidated. The recom­ mendation was accepted by the Board of Trustees, Field days. As an information and public r elations and a new Program, Documentation, Informat.ion activity, DIL will organize regular field days in l and Library (DIL), was established late in 1983. In collaboration with the research programs and the Februa ry 1984, it got a new leader and was r e­ Farm Management Unit. Various groups of con­ structured into six units: Library; Public Affairs; cerned persons, farmers , extension personnel, re­ Publications; Conference and Visitors ' Center; search staff of other agricultural inst itutions, Interpretations and Translations: and Printing, government representatives, and businessmen and Binding and Reprography. The new structure is women will be invited to such field days. Highly sustaining and improving on·going activities while successful field days were held on both seed-yam initiating new ones and planning improvements. production and hybrid-maize production during 1984. Highlights of the new activities, whose objective is to Influential persons program_ Through this pub­ make UTA's work better known and applied around lic relations activity, w e invite such persons to visit the world, are: UTA as ambassadors and agricultural attaches accre­ Workshop and briefing for mass media per­ di ted to African countries, African ministers and sonneL The idea is to invite feature editors of permanent secretaries of agriculture, science and newspa pers and magazines as well as radio and TV technology, economic and rural development, also journalists from Africa and donor countries to !ITA prominent businessmen and women, key officials of for workshops and briefing sessions on agricultural chambers of commerce and industry and high ly research and development issues. Familia r with distinguished scientists from outside Africa. As a n UTA's work and the context in which the work is example, Nigeria 's former Head of State, General done, they can explain it to their varied audiences Olusegun Obasanjo, visited !ITA during the year. through channels acceptable to the audiences. Deposit library scheme. Libraries of selected key The first such activity, OIConference on the mass institutions in Africa and donor countries will be media and food production in Africa," was held designated as IITA Deposit Libraries to receive all during 9-11 October 1984. A science writing work­ UTA publications gratis and will be requested to shop for West African mass media personnel is make them readily avai lable to anyone seeking scheduled for July 1985. It will be held in collabora­ information about our work. Where advantageous, tion with the Department of Mass Communication of the llTA Library will request publication exchanges. the University of Lagos and the International Develop­ Audiovisual presentations on UTA_ The first mentResearch Center. We hope to use materials from appointment made under the new DIL leadership was [ITA that can be worked on and then taken home to be that of an audiovisual specialist. He has made an published. excellent start in improving the quality of our audio­ Outreach seminar program_ This program is to visual presentations and developing new displays ensure that the work of liTA is presented in a hol istic and exhibits. A new concept for a slide-tape pre­ fashion to broadly based scientific gatherings to sentation on [ITA has been developed and is being complement the necessarily specialized presen­ carried out. A series of video recordings is also tations by scientists in our research programs. The planned. seminars will, in general , be based on standardized Press releases_ A program of regular press re­ texts and sets of appropriate slides and will be leases will be instituted . Written by the staff of our presented by selected staff of DIL and the Institute's public affairs unit, the releases will be distributed to top management_ Another facet of the program, newspapers and radio and television sta tions in aimed at various public audiences, provides seminars Mriea and donor countries. As appropria te, some that leave UTA's work better understood in donor press releases will be aimed at the news sections of countries as well as in ~igeria. In 1984 such seminars major scientific journa]s and magazines. 217 218 Documentation, Information and Library The Library Table 7.1. Nigerian press coverage of lIT A, January-December 1984 At the "nd of D<>cember 1984, the library consisted of Newspaper Number of articles 28,700 volumes of periodicals, 26,200 books, 4,500 pamphlets, 3,900 microform items, 1,860 slides, and 83 Business Concord. . . . . . . . . . • . . . . . . . . . . 2 audio-cassettes. In addition, the library has about Business Times. . . . . . . . . . . . • . . . . . . . . . . . . . . . . . 2 Daily Sketch .......... .. _ .. .. .. .. _ . . .. . . .. . . to 7,000 reprints on specialized subjects such as cow­ Daily Time •.. _. ... .... . .. . .... . .. _.......... 6 peas, yaros. and plantains. Use of tbe library by lIT A The Guardian. . . . . . . . . . . . . . • . . . . . . • . . . • . . . . . . . . . 4 staft'and traineees remained heavy and has continued National Concord . . . . . . .. . .• . . . . . . . . . . . . . . . . . . . . 2 to increase. The number or users not associated with New Nigerian. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 4 IITA also increased from 2,900 in 1983 to 3,350 in 1984. Nigerian Tribune, .. . .. , . . . • . . . . . . . . . . . . . . . . 8 The major activity in 1984 was automation. Data­ Sunday Concord. . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . 1 base management software especially suited to lib­ Sunday Sketch. . . . . . . . . . . . . . . . . . . . . . . • . . . . . . 1 rary operations was installed in July. Called BASIS, Sunday Tribune .... . ...... • ........ ... , .. , . . 8 it was developed and is maintained by Battelle Total. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Institute. Our objective is to develop a wholly in­ tegrated, library automated system. Access to library holdings and acquisition and circulation records are Ta.ble 7.2. External newspaper coverage of UTA, on-line. Also searchable on-line are authors, subject January-December 1984 descriptors, t itles, publication dates, publishers, Country of corporate sources, and abstracts. Newspaper publication The first priority was t o input the catalog records, Corriere de la Serra 1 •••••••• •• • • • ••.• Italy circulation records, and tloutstanding order" files . Ehuzu .. . .. ........... . . .......... . Benin Republic Named ALISTRA, our bibliographical data base by Financial Times ..... . .... . . U.K. the end oftbe year contained 25,112 records including International Herald Tribune . ..... . France about 60% of a ll our catalog records. When the II Messagero . .... . . . . .. .... . .. . . .. . Italy catalog records are complete, emphasis will shift to Mwetil ... ....... ....... . .. . ..... . Congo putting UTA staff publications into the data base. The New York Times . . ...... .. ..... . .. . . U.S.A. Washington Post . ....... . .•. .. •... U.S.A. International Grain Legume Information Center has started inputting current information on cowpeas. lTwo articles du ring 1984. Four terminals and one printer were installed in the library dur ing 1984. To facilitate access by library users and a rapid build-up of the data base, additional Shows. The Unit represented the Institute at two terminals will be installed in 1985. international agricultural shows- TASACONEX 84 The fourth volume of abstracts of the world litera· in Ardingly at the South of England Showground in ture on cowpeas was published and distributed duro July and at the Agro-pastoral Show in Bamenda. ing 1984. Much progress was made in preparing the Cameroon in December. About 60,000 people visited second volume of an annotated bibliography on yams IITA's stand on each occasion and more people knew and the genus Dioscorea. It will be completed and about the Institute's participation at the Cameroon published in 1985. show through radio and television. Mass media conference. An internationa1 con­ Public Affairs ference on "The mass media and food production in Mrica" was held at UTA, 9-11 October. Fifty·eight The activities of the Public Affairs Unit assumed new participants representing newspapers, radio, and dimensions during 1984. and use ofmedia to publicize television stations attended. The nature of the food the Institute's activities were intensified. crisis in Africa and the efforts being made to find long· Media. UTA staff and research activities were term solutions by orga nizations such as IITA reo featured 22 times on television stations in Nigeria ceived considerable publicity. and five known times outside the country. Included Visitors. The unit counseled about 355 indio was a nationwide, hour·long interview program (X­ viduals. They were mostly farmers-to-be, and the r ay) featuring our director general. Radio programs positive action that most of them took to s<>cure land featuring IITA and its staff within Nigeria numbered and set up farms during the year was most encourag­ 45 while the Institute's staff participated in nine ing. Many of them are influential Nigerians; several interviews on Britain's BBC and Voice of America. are professionals in different fields. An arrangement was made with BBC whereby senior scientists and top management of IITA visiting Farm visits. To keep up with the thinking, prob­ London would be interviewed for the BEe programs lems, and progress offarmers, the Unit's staff made 22 " Farming World" and " Blueprint Africa." The dra· visits to nearby farms during 1984. matic increase in press coverage of UTA's work Meeting with governor. An official visit between during 1984 is illustrated in Tables 7.1 and 7.2. the chairman of our Board of Trustees and the Documentation, Information and Library 219 military governor of Oyo state was arranged on the The staff of the unit assisted with speeches for the occasion of the annual meeting of the Board. director general and wrote aricles for publication in Cooperation has been close between the state's magazines, as shown in Table 7.3. Ministry of Agriculture and IITA. Training. Two students. one a candidate for a Conference and Visitors' Center Master's degree in communication arts and the other for a Bachelor's degree in the same field did the Conferences. The staff of the Conference and research portion of their degree requirements in the Visitors' Center we re involved, for the first time in Public Affairs Unit. One staff member of a Nigerian 1984, in organizing workshops held outside :- Nsukka M.A. Aluko, acting head of printing, binding, and F.O. Ochiobi, executive secretary reprography J.A. Okogun, research assistant A .A . Azubuike, M.L.S .. bibliographer S.O. Olubocle, research associate S.O.T. Babaleye, M.e.A., communications associate Collaborators and Trainees 229 B. Bakare, A.A., head of conference and visitors center Research Fellows: G.A. Cambier, Lie., head of interpretations/translations unit E.N. Obazee (Nigeria), University ofIbadan, Nigeria D. T. De Grandsaigne, Lie. , interpreter/translator V.A.O. Okoth (Uganda), University of Reading. England C. Duval, Lie., interpreter/translator E.B. Ojurongbe, B.S., graphic designer Research Scholars: B. Fadare, senior photographer F .M. Gatmaitan. B.S., senior graphic designer M. Kayibigi (Burundi), Laval University, Canada G.O. lbekwe, B.A., principal librarian J.P.C. Koroma (Sierra Leone), University of London J.C.G. lsoba, M.S., writer/editor P. Pussemier (Belgium) , State University of Agricultural J.E. Keyser, B.S., head of publications unit Sciences of Gembloux, Belgium E. Molinero, Lic., interpreter/translator E.F. Nwajei, B.A., acquisitions librarian Rice Improvement Program M.O. Oduhanjo, B.S., cataloger* E.V. Oro, B.S., audiovisuals specialist Collaborators J.O. Oyekan, B.S. , head of public affairs unit A.T. Akpan, chief agricultural officer , Food Crops N.C. Russell, M.A. , editor/writer Division, Ministry of Agriculture, Cross River State, A.K. Shiwoku, senior executive secretary Nigeria E. Tordeur, Lie., translator M.O. Aluko, director, Plant Quarantine Centre, Moor Plantation , Ibadan, Nigeria Physical Plant Services K.A. Ayotade, chief research officer. NCRI. Badeggi. Nigeria J.G.H. Craig, assistant director for physical plant services S.O. Fagade, rice coordinator. NCRI, Moor Plantation, E.O.A. Akintokun, research vehicles service officer Ibadan. Nigeria A. Amrani, heavy equipment service officer A.E.A. Lebo, deputy chief agricultural officer, Food Crops A. Bhatnagar, r efrigeration/air conditioning service Division, Ministry of Agriculture, Cross River State, officer Nigeria A.C. Butler, buildings and site service officer T. Moerdyk, Sudan United Mission, Abakaliki. Nigeria O.O.A . Fawole, automotive service officer F. Ndabuka, farm manager, Imo State Rice Project, J.M. Ferguson, fabrication/water utility service officer Nigeria p, Ghosh, scientific(electronics service officer C.O. Obasola, director, NCRI, Moor Plantation, [badan, J . Lukowski, electrical service officer Nigeria M.O. Yusul, construction(site engineering service officer P.U. Ohunyon, agronomist, Shell B.P., Warri. Nigeria "Left during the yeaT Z. Russom, plant breeder, Institute of Agricultural Research, University of Port Harcourt. Nigeria Moussa Sie, rice breeder , Centre d 'Experimentation du Collaborators and Riz et des Cultures Irriguees, Bobo·Dioulasso, Burkina Faso Trainees M.N. Ukwungwu, rice entomologist, NCR!. Badeggi, Nigeria Grain Legume Improvement Program Research Fellow: Research Fellows: S.B.C. Wanki (Cameroon), University ofIbadan. Nigeria K. McGinnis (Canada). University of Guelph, Canada Research Scholar: M. Owusu.Akyaw (Ghana), University of Science and Technology, Ghana A.S. Kumwenda (Malawi), University of London A.B. SaliCu (Ghana). University of London Root and Tuber Improvement Program Research Scholars Collaborators J. Adabasi (Benin), National University of Benin H. Adu.Dapaah (Ghana), University ofIbadan, Nigeria J.P.J. easier, professor , Catholic University of Louvain, G. Akpokodje (Nigeria). University ofIbadan. Nigeria Belgium K.P. Atropo (Benin), National University of Benin L.K. Djokoto, senior research agronomist, Volta Region G. Atuahene-Amankwa (Ghana), University of Guelph. Agricultural Development Project, Ministry of Canada Agriculture, Ghana A.M. Maponoko (Zaire), Catholic University of Louvain, Belgium Research Fellows: C. T. Njoh (Cameroon), University ofIbadan, Nigeria A.A. Younous (Chad), University ofIbadan. Nigeria J. Ambe Tumanteh (Cameroon). University of Ibadan, Nigeria C.E. Gyansa-Ameyaw (Ghana), University of Ghana Maize Improvement Program N.B. Lutaladio (Zaire), University oflbadan. Nigeria Collaborator: J.M. Ngeve (Cameroon), University of Maryland, United States K.I. Nwosu, entomologist, NCRI, Amakama, Nigeria H.O. Nsiama She (Zaire), University ofIbadan. Nigeria 230 Collaborators and Trainees Research Scholars: M.O. Omidiji, coordinator of on·farm adaptive research, IAR&T, Ibadan, Nigeria K.N.A. Ahouandjinou (Benin), National University of J.A.I. Omueti, senior lecturer, Department of Agronomy, Benin University ofIbadan, Nigeria Muyolo Gilumbu (Zaire), University of [badan. Nigeria C.O. Oputa, senior agricultural research and extension T. Kasu (Zaire), University of Thad an , Nigeria coordinator. FACU, Ibadan, Nigeria M. MBkame (Tanzania), University oflbadan, Nigeria Y.O.K. Osikanlu, plant pathologist, IAR&T. Moor D. Ndombo (Zaire), University or Thedan, Nigeria Plantation, Ibadan. Nigeria P. Reid. senior agronomist, Ilorin ADP, Nigeria Farming Systems Program J. Rice, project manager, Bida ADP, Nigeria U. Schwertman, professor, Institute for Soil Science, Collaborators Technical University of Munich. West Germany G.O.!. Abalu, coordinator, WAFSN and Nigerian W. Shearer, professor, Development Studies Division, Farming Systems Research Network, [AR , Zaria, Nigeria United Nations University, Tokyo, Japan E. Adewoye, senior evaluation officer, UorinADP, Nigeria K. W. SmiJde, soil scientist, Institute for Soil Fertility, The A.G. Agbahungba, microbiologist, Department de 1a Netherlands Recherche Agronomique. Cotonou. Benin K.G. Steiner, head, GTZ lntercropping Project, M.O. Arigbede. farmer, Oyo State, Kigeria Heidelberg, West Ge['many P. Awelewa, sma)) tools fabricator, Ibadan, Nigeria R.P.A. Unamma, coordinator, Farming Systems M. Clairon, soil scientist, INRA Station d'Agl'onomie, Resea['ch Program, NRCRI, Umudike, Nigeria Petit Bourg. Guadeloupe O. Uthman, farmer, Ibadan, Nigeria R. Delogu, director, Okomu Oil Palm Plantation, Okornu, J. van der Heide, agronomist, Institute for Soil Fertility, Nigeria The Netherlands J. de Wolf, water management engineer, ILRI. C . Van Hove, professor. Universite de Louvain·la-Neuve, Wageningen, The Netherlands Belgium M. Diomande, coordinator, OFRIC, Abidjan, Ivory COBst K. V1assak, professor. University ofLeuven, Belgium W. Ehlers, professor, Institute for Agronomy and Plant P.L.G. Vlek, soil scientist, IFDC, Muscle Shoals, Breeding, University ofGottingen, West Germany Alabama, U.S.A. D. Gabriels, professor, Department of Soil Physics, University of Ghent, Coupre Links, Belgium O. Gordon, director, LWDD, Ministry of Agriculture and Research Fellows: Natural Resources, Sierra Leone K. Attab Krah (Ghana), University ofIbadan, Nigeria H. Grimme, soil science, Buntehof Agricultural B. Duguma (Ethiopia), University ofIbadan , Nigeria Experiment Station, Hannover, West Germany E.T. EBhett (Nigeria), University ofIbadan, Nigeria F.S. Idachaba, professor. Department of Agricultural H.R.K. Franzen (West Germany), University of Economics. University ofIbadan. Nigeria Gottenge,n, West Germany Y. Isola, senior research officer, Ilorin ADP, Nigeria B. Herren Gemmill (United States), University of A. Jibrin, senior agronomist, Bida ADP, Nigeria California at Davis, United States O.T.A. Kadri, Shadlas Metal Engineering Works, Ibadan, N.O. Iwuafor (Nigeria). Abamadu Bello University, Nigeria Nigeria B. Kalabari , assistant head , Community Development B. Kayombo (Tanzania), University of Dar-es-Salaam at Program, Shell Nigeria, Warri Morogoro, Tanzania A. Kogblevi, director, UNDP/FAO Project, Cotonou. M.M. Mafuka (Zaire). Uni .... ersitk de Louvain-la-Neuve, Benin Belgium W. Lummins, Diocesan Development Services, Benue C.A. Mba (Nigeria). Tecbnical University of Berlin State, Nigeria M.B. Niyungeko (Zaire), Faculty Institute of Agricultural A.F. Maba wonku, professor, Department of Agricultural Sciences, Zaire Economics, University ofIbadan, Nigeria M.E. Ogula (Zaire), Faculty Institute of Agricultural J.A. Meyer, professor. Universite de Louvain·la-Neuve, Sciences, Zaire Belgium U. Sabal-Koschella (West Germany), University of U. Mokwunye, agronomist, IFDC, Muscle Shoals, Munich. West Germany Alabama, U.S.A. N .. Sanginga (Zaire), Faculty Institute of Agricultural F.B. Ndubuka, project officer, Imo State/World Bank Sciences, Zaire Small· holder Rice Production Scheme, Nigeria Wong T. Fook (Mauritius), University of Reading. O. Nduaka, project manager, ISADAP, Owerri England L.A. Nnadi. soil scientist. Ahmadu Beno Universi ty, C.F. Yamoah (Ghana), University ofIbadan. Nigeria Zaria, Nigeria M. Obaton, head, Laboratoire de Recherche sur les Symbiotes des Racines, [NRA, MontpelHer, France C.T.1. Odu, professor, Department of Agronomy. Research Scholars: University oflbadan, Nigeria B.O. A80nibare (Nigeria), Catholic University of P. Ohuyon, head, Community Development Program, Louvain, Belgium Shell Nigeria, Warri B. BaIl (Canada), University of Guelph, Canada D.U.U. Okali, professor. Department of Forestry, D.M.C. Deeoene (Belgium), Catholic University of University of Ibadan. Nigeria Louvain, Belgium B.A. 01unuga, leader. Farming Systems Research P.E. De Preter(Belgium}, Catholic University of Lou vain, Program. NCRI, Ibadan, Nigeria Belgium Collaboralors and Trainees 231 F. Dumortier (Belgium), Catholic University of Louvain, Belgium K. Y. Fosu (Ghana), University of Ghana A.R.S. Ibrahim (Ghana), University ofIbadan, Nigeria C.M. Mayona (Tanzania), Sokoine Agricultural University, Tanzania C. Nkwiine (Uganda), Makerere University, Uganda C.1. Okonkwo (Nigeria), University of Ibadan, Nigeria J.e. Tossa (Benin), National University of Benin J.M. Van Den Henkel (The Netherlands), University of Wageningen, The Netherlands J.K. Vannoppen (Belgium), Cathollc University of Louvain, Belgium Vacation students B.O. Djoleto (Ghana), University of Ghana, Legan, Ghana N. Kusina (Zimbabwe), University of Zimbabwe, Zimbabwe E.K. T. Ouma (Uganda), Makerere University , U gands Special training B.K. Landu (Zaire), Pronam·M'Vuazi, Zaire Research Support Units Collaborators: n.M. Okioga, director, Plant Quarantine Station, Kenya Agricultural Research Station, Muguga, Kenya F. Quak, head of virology, Research Institute for Plant Protection, Wageningen, The Netherlands Research Fellows: F.N. Anno-Nyako (Ghana), UTA Virology Unit, University of Science and Technology, Ghana M. Gumedooo (Togo), !ITA Virology Unit, Laval University, Canada Research Scholars: A. Carter (England), !ITA Farm Management, University of Reading, England M.J. Dalton (England), UTA Farm Management, University of Reading, England 232 Pub lications Publications Alam, M.S. , and Y. Efron. 1984. Bionomics of rice stem borers in Nigeria (abstract). In Abstracts of the S even· Grain Legume Improvement Program teenth lnternatu:mal Congress of Enwmology, 2G-26 August, Hamburg, West Germany. Aggarwal, V.n.,N. Muleba.I. Drabo, J . Souma, andM. John, V .T., G . Thottappilly, and V .A. Awoderu. 1984. Mbewe. 1984. Inheritance of Striga gesnerioides reo Occurrence of rice yellow mottle virus in some Sahelian sistance in cowpea, In Proceedings of the Third Inter­ countries in West Africa. FAO Plant Protection Bulletin national Symposium on Parasitic Weeds. 143-147. Aleppo. 32: 86-87. Syria: ICARDA. Ukwungwu, M.N., M.S. Alam, and Kaung Zan. 1984. Kueneman, E .A., W.R. Root, K.E . Dashiell, and J. Incidence of gall midge (aM) Orseolia oryziuora H & a in Hohenberg. 1984. Breeding soybeans for the t ropics Edozhigi, ~igeri a. International Rice Research News­ capable of nodulat ing effectively with indigenous letter 9(3): 13. Rhizobium spp. Crops and Soils. In press. Ntare, B.R .• M.E. Aken 'ova, R.J. Redden, and B.B. Singh. 1984. The effectiveness of early generation tes ting Root and Tuber Improvement Program and the single seed descent procedures in two cowpea Afolami, S.O.,andF.E. Caveness. 1984. The frequency of (Vigna unguiculata (L.) Wa lp.) crosses. Euphytica 33: occurrence and geographical distribut ion of plant para­ 53!t-548. sitic nematodes associated wi th Theobroma cacao in Ntare. B.R., R .J . Redden, and B.B. Singh. 1984. Nigeria. Turrialba 33: 97-100. Evaluation of early generation selection procedures for Caveness, F.E . 1984. Update on the UTA root a nd tuber yield in cowpea (Vigna unguiculata (L.) Walp.). Field improvement program activities. N ewsletter, Inter­ Crops Research 9: 91- 100. national Society for Tropical Root Crops, Africa Branch 2: Pandey, R.K. , W.A.T. Herrera. and J.W. P endleton. 4-6. 1984. Drought response of grain legumes under irrigation Theberge, R.L . 1984. Phytopathology of cassava diseasesl gr adient ; I. Yield and yie ld components. Agronomy pests. Nigerian Journal of Plant Protection. In press. Journal 76: 54!t-553. Pandey, R .K., W .A .T. Herre ra, and J.W. Pendleto n. L984. Drought response of grain legumes under irrigation Farming Systems Program gradient; II . Plant water status and canopy temperature. Ahis8ou. A., and 1.0. Akobundu. 1984. Weed interfer· Agronomy Journal 76: 554- 557. ence in hydr omorphic rice. In Proceedings of the Second Pandey, R.K., W.A.T. Herrera, A .N. Villegas, and Biennial Conference of the West A frican Weed Science J .W . Pendleton. 1984. Drought response of grain Society, 3+-39, Eds. M. Deat and P. Marnotte. Bouake. legumes under irrigation gradient : III. Plant growth. Ivory Coast: WAWSS. Agronomy Journal 76 : 55!t-560. Akobundu, 1.0. 1983. No·ti ll age weed control in the Pulver, E.L., E.A. Kueneman, and V.R. Rao. 1984. tropics. ]n No· Tillage Crop Production in the Tropics , Ident ification of promiscuous nodulating soybeans ef­ 32-44. Eds. LO. Akobundu and A.E. Deutsch. Corvallis, ficient in N2 fixation . Crop Science. In press. U.S.A.: IPPC. Singh, B.B. 1984. Recent research advances at UTA Akobundu, 1.0. 1983. Weed control in no-tillage cassava relevant to SADCC research programmes and agrieul· in the subhumid and humid tropics. In No-Tillage Crop ture of the Semi·Arid Tropi cs. In Proceedings of the Production in the Tropics. ll!t-l26, Ed •. l.0 . Akobundu SA DCC Agricultu re Research Conference. 88-99. and A.E. Deutsch. Cor vall is, U.S.A.: IPPC. Botswana. Akobundu, 1.0. 1984. Advances in hve mulch crop pro­ Maize Improvement Program duction in the tropics. Proceedings of the Western. Society of Weed Science 37: 51-57. Dabrowski, Z. T . 1984. Handling new Cicadulina colonies. Akobundu, 1.0. 1984. Response of cowpea cultivars to IITA Research Briefs 5(1): 2-4. pre·emer gence herbicides. Nigeria JourfUll of Plant Dabrowski. Z.T. 1984. Rearing CicadulilUl : technical Protection. [n press. methods, equipment needed. IITA Research Briefs 5(2) : Akobundu, ].0.1984. Advancing weed. control strategies 2-3. for developing countries. In Advancing Agricultural Dabrowski, Z.T. 1984. Releasing leafhoppers for maize Production in Africa. 193- 199. Ed. D.L. Hawksworth. streak resistance screening. IITA Research Briefs 5(3): Slough: Commonwealth Agricultural Bureau. 4- 5. Akobundu, 1.0.1984. Weed science technology in relation Dabrowski, Z.T., and Y. Efron. 1984. Entomological to food production in West Africa. Proceedings of the 9th aspects of a maize improvement program in Africa. East African Weed Science Society conference, pp. 22-60. Annual Plant Resistance to Insects Newsletter 10: 67-68. Nairobi, Kenya. Efron, Y., J.M. Fajemisin, S.K. Kim, M. Bjarnason, Akobundu, 1.0., and B.N. Okigbo. 1984. Preliminary H.N. Pham, and D. 1Ilakonnen. 1984. A summary of evaluation of ground cover for use as live mulch in maize maize resistance breeding. Annual Plant Resistance to production. F ield Crops Research 8: 177- 186. Insects Newsletter 10: ~9. Ayeni, A.O. , 1.0. Akobundu, and W.B. Duke. 1984. Hammes, J.L. and J .W. Pendleton. 1984. Photoperiod. Weed interference in maize, cowpea. maize/cowpea inter­ sensitive tropical com as a potential source of biomass or c rop in a subhumid tropical environment. ll. Early paper. Agronomy Journal 76 : 15!t-160. growth and nutrient content of crops and weeds. Weed Rice Improvement Program Research 24: 281-290. Ayeni, A.O., W.B. Duke, and 1.0. Akobundu. 1984. Alam. M.S. 1984. Incidence of brown planthopper and Weed interference in maize, cowpea, maize/cowpea inter­ white fly in Nigeria. International Rice Research News­ crop in a subhumid tropical environment. I. InRuence of letter 9(4): 13. cropping season. Weed Research 24 : 269-279. Publications 233 Ayeni, A.O., W.B. Duke, and 1.0. Akobundu. 1984. LaI, R. 1984. Soil erosion from tropical arable lands and its Weed interference in maize, cowpea, maize/cowpea inter­ control. Advances in Agronomy 37 : 183-248. crop in a 8ubhumid tropical environment. In. Influenceo! Lal, R. 1984. Soil erosion research and extension needs in land preparation. Weed Research 24: 439-448. the tropics. Erosion Research Newsletter, Australia 7: Baryeh, E., C. Garman, N. Navasero, and B.T. Kang. 2- 3. 1984. Fertilizer band appllcator. [n Fertilizer Application Lal. R. 1984. Tillage in lowland rice based c ropping Equipment lor Small Farml!rs, 35-42. Rome: FAO. systems . ]n Physical Aspects of Soil Management in Rice· Ezumah, H.C., and T.L. Lawson. 1984. Intercropping of Based Cropping Systems. Los Banos. Phllippines: IRRI. morphologically different cassava and maize genotypes Maduakor, H.O., R. La1, and O.A. Opara.Nadi. 1984. (abstract). In Abstracts of the Annual Meeting of the Effects of methods of seedbed preparation and mulching American Society of Agronomy, 20--25 November. Las on the growth and yield of white yam on an Ultiso] in Vegas, Nevada, U.S.A. southeast Nigeria. Field Crops Research 9: 119-130. Ghuman, B.S., and R. La!. 1984. Effects of soil moisture Mbagwu, J., R. Lal, and T.W. Scott. 1984. Effects of regime on plant-water relations and yield of tropical root desurfacing of Alfisols and Ultisols in southern Nigeria. crops. In Twelfth Congress, International Commission on L Crop performance. Soil Science Society of America Irrigation and Drainage. QR38·R72, 1.141- 1,156. Journal 48: 828-833. Ghuman. B.S., and R. Lal. 1984. Water percolation in a Mbagwu, J., R. Lal, and T.W. Scott. 1984. Effects of tropical Alfisol under conventional plowing and no-till desurfacing of Alfisols and Ultisols in southern Nigeria. systems of management. Soil and Tillage Research 4: II. Changes in soil physical properties . Soil Science 263- 276. Society of America Journal 48 : 834-Jl38. Gowman, M.A. , and 1.0. Akobundu. 1984. Herbicide Ngatunga, E.L.N., R. Lal, and A.P. Uriyo. 1984. Effects screening for preplant vegetation control in no-tillage of surface management on runoff and soil erosion from crop product ion in southern Nigeria. In Proceedings of some plots at Mlingano, Tanunia . Geoderma 33: 1- 12. Ihe Second Biennial Conlerence 01 the West Alrican Weed Poku, J.A., and 1.0 . Akobundu. 1984. Chemical weed Science Society, 42- 49, Eds. M. Deat and P. Marnotte. control in cowpea (Vigna unguiculata (L .) Walp.) in the Bouake, Ivory Coast. subbumid and Guinea savanna zones of Nigeria. In Hulugalle, N.R., R. Lal, and C.H.H. ter KuiJe. 1984. Soil Proceedings 01 the Second Biennial Conference of the West physical cbanges and crop root growth following dif· Weed Science Society, 216-225. Eds. M. Deat and P. ferent methods of land clearing in western Nigeria. Soil Marnotte. Bouake. Ivory Coast. Science 138: 172- 179. . Poku, J .A., and 1.0. Akobundu. 1984. Effect of t illage Ikeorgu, J.E.G., T.A.T. Wahua, and H.C. Ezumah. and herbicide on cowpea weed control and yield in the 1984. Crop performance in complex mixtures: melon and subhumid tropics (abstract). In Abstracts of the 1984 okra in cassava-maize mixture. In Proceedings of the Meeting 01 the Weed Science Society of America, Ed. J.F. Second Triennial Symposium on Tropical Root Crops­ Ahrens. Champaign: WSSA. Alrica Branch, !l3-