Cassava Breeding, Agronomy and Farmer Participatory Research in Asia Proceedings of the Fifth Regional Workshop held in Danzhou, Hainan, China, Nov 3-8, 1996 CATAS CIAT The Nippon Foundation The International Center for Tropical Agriculture (CIAT, its Spanish acronym) is dedicated to the alleviation of hunger and poverty in developing countries of the tropics. CIAT applies science to agriculture to increase food production while sustaining the natural resource base. CIAT is one of 16 international agricultural.research centers sponsored by the Consultative Group on International Agricultural Research (CGIAR). The Center.s core budget is financed by 27 donor countries, international and regional development organizations, and private foundations. In 1996, the donor countries include Australia, Belgium, Brazil, Canada, China, Colombia, Denmark, France, Germany, Japan, Mexico, the Netherlands, Norway, Spain, Sweden, Switzerland, the United Kingdom, and the United States of America. Donor organizations include the European Union (EU), the Ford Foundation, the Inter-American Development Bank (IDB), the International Development Research Centre (IDRC), the International Fund for Agricultural Development (IFAD), the Nippon Foundation, the Rockefeller Foundation, the United Nations Development Programme (UNDP), and the World Bank. Information and conclusions reported in this document do not necessarily reflect the position of any donor agency. Cassava Breeding, Agronomy and Farmer Participatory Research in Asia Proceedings of the Fifth Regional Workshop held at CATAS, Danzhou, Hainan, China, Not 3-8, 1996. Technical Editor: R.H. Howeler Organized by Centro Internacional de Agricultura Tropical (CIAT) and the Chinese Academy of Tropical Agricultural Sciences (CATAS) With financial support from - Nippon Foundation, Tokyo, Japan - Government of Japan ^"•"V-"*, -^s, ]• :• ro 0 3 JUL. 1998 Thl s One 6E2G-CZ4-6342 Cover photo: R.H. Howeler Harvest of Farmer Participatory Research (FPR) trials on erosion control in Bac Thai province of Vietnam. Centro International de Agricultura Tropical (CIAT) Apartado Aereo 67-13, Cali, Colombia CIAT Regional Cassava Programfor Asia Field Crops Research Institute Department ofAgriculture Chatuchak, Bangkok 10900 Thailand April, 1998 Print Order: 400 Copies Reference: Centro International de Agricultura Tropical (CIAT). 1998. Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. Fifth Regional Workshop held in Danzhou, Hainan China. Nov 3-8, 1996. I. Cassava Asia-Breeding-Agronomy-Farmer Participatory Research II. Howeler, R.H. III. Centro Internacional de Agricultura Tropical PREFACE The Centro Internacional de Agricultura Tropical (CIAT), with headquarters in Colombia, has as its mission "to contribute to the alleviation of hunger and poverty in tropical developing countries by applying science to the generation of technology, leading to lasting increases in agricultural outputs while preserving the natural resource base". The ultimate goal is to reduce hunger and poverty through the development of sustainable agriculture, mainly by integrating programs on genetic improvement with those on natural resources management. The two programs are intimately linked, as sustainable land use and resource management will contribute to conserving biodiversity, which is a vital resource for crop improvement. Conversely, more productive germplasm will enable farmers to increase their yield and raise their income, giving them more resources and incentives to conserve the productivity of their soil and prevent erosion. CIAT can not achieve this goal alone, but only in partnership with national and international research institutions. Moreover, in order to enhance the adaption of new technologies, CIAT also collaborates closely with extension organizations and strongly encourages farmer participation in technology development and dissemination. Within the Consultative Group for International Agricultural Research (CGIAR). CIAT has the world mandate for research on cassava production and utilization, while the International Institute for Tropical Agriculture (IITA), located in Nigeria, has responsibility for cassava research in Africa. In order to facilitate communication with national cassava research programs in Asia, CIAT established a Regional Cassava Office in Bangkok, Thailand, in 1983. Through this Regional Office, a network of cassava researchers in national research institutes and universities in Asia was established, with the objective of enhancing communication between researchers, both within and among countries, in order to increase the efficiency of cassava research with the goal of improving the productivity and utilization of the crop. This will ultimately contribute to increased incomes and improvements in the standard of living of cassava producers, processors and consumers. To further enhance communication among cassava researchers in Asia, the CIAT Regional Office has organized Regional Cassava Workshops every three years. These workshops bring together many cassava researchers from different countries in Asia, who present the latest progress in their research, mainly in the areas of cassava breeding and soil/crop management or agronomy. The first workshop was held in Bangkok in 1984 to review the general situation of cassava in Asia, to identify the major constraints and set research priorities. The second workshop was held in Rayong, Thailand in 1987 to review the first year's results in cassava crop improvement and to summarize all agronomic research that had been conducted prior to that date by national programs. From this review it was concluded that future agronomic research in Asia should emphasize mainly the maintenance and improvement of soil fertility and the effective control of soil erosion. The third workshop was held in Malang, Indonesia in 1990. This workshop not only reviewed the recent progress in cassava breeding and agronomy, but also research on cassava utilization, both for human and animal consumption as well as for industrial usage. The fourth workshop was held in Trivandrum, Kerala, India in 1993 to review the latest progress in breeding and agronomy, as well as to discuss more effective ways to transfer the new technologies to farmers in order to achieve adoption and impact. In this workshop extension specialists from several countries were invited to share their experiences and to join the researchers in the cassava network. This will further enhance the adoption of many new cassava varieties that have recently been released, as well as that of new and more sustainable production practices. The workshop was also an opportunity to discuss a proposal, submitted for funding to the Nippon Foundation in Japan, to develop a Farmer Participatory Research (FPR) methodology, mainly to enhance the adoption of more sustainable management practices in cassava-based cropping systems in Asia. The Nippon foundation approved the project in late 1993 for a five-year period. It has funded all activities in the area of cassava soil/crop management research, farmer participatory research and training in FPR methodologies. The fifth workshop was held at CATAS in Danzhou, Hainan, China in November 1996. Besides reviewing further progress in cassava varietal improvement and disemmination as well as agronomy research, it provided an opportunity to present and discuss the first results of the FPR projects conducted in several pilot sites in various countries in Asia. Based on these experiences future activities in the project were discussed. The papers presented in this 5th Regional Workshop are published in this Proceedings. During the fifth Regional Workshop in China the Advisory Committee for the Asian Cassava Research Network met to discuss research priorities, the future outlook for cassava research in Asia, and to decide on the theme, location and time for the Sixth Regional Workshop. During this meeting the following members were elected: Pham Van Bien IAS, Vietnam: Chairman Kazuo Kawano CIAT, Thailand: Secretary Zheng Xueqin CATAS, China G.T. Kurup CTCRI, India Soemarjo Poespodarsono Brawijaya Univ, Indonesia Tan Swee Lian MARDI, Malaysia Fernando Evangelio PRCRTC, Philippines Cham Tiraporn DOA, Thailand Reinhardt Howeler CIAT, Thailand During the meeting it was decided to organize the sixth workshop in Ho Chi Minh city, Vietnam at the end of 1999. CIAT wants to take this opportunity to express its gratitude to the Nippon Foundation and the Government of Japan for their financial support, not only for funding the collaborative research and extension activities reported here, but also the organization of the fifth Workshop in China. Without their financial support the Asian Cassava Research Network would not be sustainable. We also like to thank all the members of the organizing committee, mainly from CATAS, for their hard work and dedication in organizing this event. In spite of unusually wet weather, they managed to organize an interesting field day to Kongba village in Baisha county. All workshop participants also much appreciated the excellent facilities and organization of the workshop, and especially, the delicious Chinese food. Many thanks to all for their hard work. R.H. Howeler CIAT, Bangkok April 1998 CONTENTS Page Preface 1 Trends, Constraints and Opportunities of the Asian Cassava Sector: An Assessment 3 Guy Henry and Clair Hershey Progress in Cassava Breeding at the Chinese Academy of Tropical and Agricultural Sciences (CATAS) 21 Lin Xiong, Li Kaimian, Zhang Weite and Huang Jie Cassava Varietal Improvement and Dissemination in Guangxi 34 Tian Yinong and Lee Jun Breeding and Varietal Improvement in Thailand 48 Jarungsit Limsila, Atchara Limsila, Watana Watananonta and Kazuo Kawano Varietal Improvement and Dissemination by Kasetsart University, the Thai Tapioca Development Institute, and the Dept. of Agricultural Extension 55 Chareinsuk Rojanaridpiched, Somporn Phongvutipraphan, Piyawuth Poolsanguan, Kaival Klakhaeng, Vichan Vichukit and Ed Sarobol Cassava Varietal Improvement in Vietnam 69 Tran Ngoc Ngoan, Ngoc Quyen, Trinh Phuong Loan and Kazuo Kawano Cassava Varietal Dissemination in Vietnam 82 Hoang Kim, Tran Ngoc Quyen, Pham Van Bien and Kazuo Kawano Breeding and Varietal Improvement of Cassava in India 101 S.G Nair, P.G. Rajendran, S.K. Naskar, M.T. Sreekumari, M. Unnikrishnan and M.N. Sheela Cassava Research Program at Tamil Nadu Agricultural University (TNAU) in India 1 1 1 S. Thamburaj Recent Progress in Cassava Varietal Improvement in the Philippines 131 Algerico M. Mariscal and Reynaldo V. Bergantin Recent Progress in Cassava Varietal Improvement in Indonesia 149 Soemarjo Poespodarsono Cassava Varietal Improvement Program at Umas Jaya Farm and its Contribution to Small Farmer Communities in Sumatra, Indonesia 156 Palupi Puspitorini, Usman Kartawijaya and Kazuo Kawano Socio-Economic Contribution of Cassava Varietal Improvement to the Small Farmer Communities in Asia 170 Kazuo Kawano Cassava Agronomy Research in China 191 Zhang Weite, Lin Xiong, Li Kaimian, Huang Jie, Tian Yinong, Lee Jun and Fu Quohui Recent Progress in Cassava Agronomy Research in Thailand 211 A. Tongglum, V. Pornpromprathan, K. Paisarncharoen, C. Wongviwatchai, C. Sittibusaya, S. Jantawat, T. Nual-on and R.H. Howeler Recent Progress in Cassava Agronomy Research in Vietnam 235 Nguyen Huu Hy, Pham Van Bien, Nguyen The Dang and Thai Phien Cassava Soils and Nutrient Management in South Vietnam 257 Cong Doan Sat and Pol Deturck Nutrient Management for Cassava-Based Cropping Systems in Northern Vietnam 268 Thai Phien and Nguyen Cong Vinh Progress in Agronomy Research in India 280 C.R. Mohankumar, V.P. Potty, C.S. Ravindran, S. Kabeerathumma and C.R. Sudharmai Devi Recent Progress in Cassava Agronomy Research in Indonesia 307 J. Wargiono, Kushartoyo, Suyamto H. and B.. Guritno Recent Progress in Cassava Agronomy Research in the Philippines 331 Fernando A. Evangelio and Julieta C. Ladera Varietal Improvement and Agronomy Research in Malaysia 340 Tan Swee Lian Cassava Agronomy Research in Asia - An Overview 1993-1996 355 Reinhardt H. Howeler Farmer Participatory Adaptation and Adoption of Contour 376 Hedgerows for Soil Conservation Sam Fujisaka Farmer Participatory Research in Cassava Soil Management and Varietal Dissemination in China 389 Zhang Weite, Lin Xiong, Li Kaimian and Huang Jie Farmer Participatory Research in Soil Management and Varietal Selection in Thailand 412 V. Vongkasem, K. Klakhaeng, S. Hemvijit, A. Tongglum, S. Katong and D. Suparhan Farmer Participatory Research in Cassava Technology Transfer in India 438 S. Ramanathan and M. Anantharaman Farmer Participatory Research in Cassava Soil Management and Varietal Dissemination in Vietnam 454 Nguyen The Dang, Tran Ngoc Ngoan, Le Sy Loi, Dinh Ngoc Lan and Thai Phien Farmer Participatory Research in Soil Management in Indonesia 471 W.H. Utomo, Suyamto, H. Santoso and A. Sinaga Farmers' Participation in Cassava Technology Transfer in the Philippines 482 Editha A. Gundaya and Fernando A. Evangelio Farmer Participatory Research for Cassava Technology Transfer in Asia - Constraints and Opportunities 497 Reinhardt H. Howeler and Guy Henry Cassava Biotechnology Network 515 Ann Marie Thro Workshop Participants 525 Appendix. Results of Soil Analyses in Asia 1994-1997 530 Reinhardt H. Howeler TRENDS, CONSTRAINTS AND OPPORTUNITIES OF THE ASIAN CASSAVA SECTOR: AN ASSESSMENT1 Guy Henry2 and Clair Hershey3 ABSTRACT This paper attempts to utilize past cassava production, utilization and market trends, integrated with regional macro-economic developments, together with a general summary of current cassava R&D advances, to analyze the principal cassava sector constraints and future opportunities, both at the country and regional level. The first section shows where the major cassava production and utilization areas are and secondly assesses cassava area, yield and production trends from 1980-95 by region and specific countries (cases), explaining past trends due to a variety of factors, including climatological, biological, technological (R&D), political and macro-economic aspects. The second section concentrates on the dynamics of cassava processing, products and markets, analyzing the major cassava product groups, i.e. cassava for fresh direct consumption, flours, dried cassava chips and pellets for animal feeds, and cassava starches. Evidence will be presented to show the significant increases of starch production versus relative declines in pellets. Macro-policy changes, expanded market demand, technology advances and relative price changes are some of the major factors explaining these recent developments. The third section of the paper assesses regional, market (and country) cassava sector constraints which are stratified into four areas: (a) biological/technical aspects, (b) socio-economic aspects, (c) institutional aspects, and (d) political aspects. For some of the aspects quantitative evidence is presented; other aspects are more qualitatively assessed. After this, the most significant future opportunities are analyzed. These are grouped into three areas: (a) technological, (b) market, and (c) institutional/political. The paper ends with the major conclusions and makes several recommendations. CASSAVA PRODUCTION AND SYSTEMS TRENDS The Asian continent currently produces about 46 million metric tons (t) of cassava on 3.5 million hectares (1996). As such, it occupies second place in terms of global cassava production, and first place in yield (13-14 t/ha). Two-thirds of total production originates in Thailand and Indonesia alone. Seven countries account for 99% of the region's production: Thailand, Indonesia, India, China, Vietnam, the Philippines and Malaysia (Table 1). About 40% of cassava is used for direct human consumption, especially in processed form. Most of the remainder is destined for animal feed or processed for starch (FAOSTAT, 1997). 1 This document draws significantly on papers treating the same subject by Henry and Gottret (1996) and Hershey et al. (1997). 2 Economist, CIRAD-PROAMYL, Montpellier, France; formerly Economist, Cassava Program, CIAT, Cali, Colombia. 3 Consultant, Manheim, PA, USA; formerly Plant Breeder, Cassava Program, CIAT, Cali, Colombia. 4Table 1. Cassava area, yield and production in Asia in 1996. Area Yield Production (ha) (t/ha) (t) Brunei 130 11.5 1,500 Cambodia 12,500 7.2 90,000 China 230,080 15.2 3,501,070 India 244,000 24.5 5,979,000 Indonesia 1,300,000 13.1 17,002,500 Laos 5,000 13.7 68,500 Malaysia 39,000 10.3 400,000 Maldives 8 2.5 20 Myanmar 7,300 9.3 68,000 Philippines 215,000 8.8 1,900,000 Singapore 1 10.0 10 Sri Lanka 32,000 9.1 290,000 Thailand 1,265,100 14.3 18,083,600 Vietnam 275,600 7.5 2,067,300 ASIA 3,625,719 13.6 49,451,500 Source: FAOSTAT, 1998 During the past decade, Asian cassava production experienced a growth rate of 0.30% per year. While cassava area experienced a decrease (-0.20%), yields increased at an annual rate of 0.50%. From the 1980s to the present, the main influences on cassava production and commerce were rapid growth in many Asian economies with accompanying changes in food consumption patterns; increased demand from industry for raw products such as starch; and increasing implementation of trade policies that reduced cassava's preferential treatment in European markets. Except for a few products such as krupuk in Indonesia, cassava generally enters markets where other calorie or industrial starch sources may readily be substituted. Future growth, therefore, is largely linked to cost competitiveness, or markets that require specific characteristics that only cassava provides. Thailand, is a cassava processed-product export-led country. Nearly all cassava is grown on small farms of one to five hectares. Chipping and drying is done in nearby commercial drying floors, while processing for starch is generally done in large factories. The pellet export industry is heavily dependent on the middle-men who are usually owners of trucks or drying floors, who consolidate production from these small farms into processing and marketing channels. The various events that have led Thailand's cassava industrialization development are well described by Titapiwatanakun (1996), Henry and Gottret (1996), and Hershey et al. (1997). During the last decade, Thailand has experienced a major diversification of its cassava product and market mix. While up to the early 80's pellets represented 88% of cassava exports (almost entirely to the EU), the pellet share by 1992 was down to 72% and continues to decrease, in favor of starch. Furthermore, Thai starch factories are increasingly producing modified starches rather than lower valued native starch. Total production peaked at 24 million tons in 1989 and decreased to 18 million tons in 1996 (Figure 1). For many years a single variety, Rayong 1, occupied almost 100% of the country's area. This began changing in the mid-1980s as new hybrids gained popularity due to market premium assigned to higher starch content. By 1996, new hybrids from the cassava programs of the Dept. of Agriculture and of Kasetsart University extended over about one-fourth of the total area. Private industry (mainly starch factories) and national extension services play an important role in promotion and distribution of new technology. Mechanized land preparation, fertilizer application, and mechanical or chemical weed control are now common. Despite the adoption of new varieties, Thai cassava yields show only a 0.12% annual increase during the last decade (Henry and Gottret, 1996). This low figure is due to a large extent to a shift of cassava production to more marginal areas, and because of reduced fertilization as a consequence of depressed starch and pellet market prices. Reduced fertilization on soils that have been monocropped for many years will show adverse effects on yield, nothwithstanding improved varieties. In Indonesia, the relatively stable area of planting across years is a function of market diversity and comparative advantage in upland environments not suited for rice. The multi-use characteristics are fully exploited and provide a range of market options to stabilize prices. The traditional products in the internal markets are gaplek (dried cassava chunks used in a variety of local dishes), and krupuk, a crispy snack wafer made from cassava starch. Production systems in Indonesia are in general more complex than elsewhere in Asia. Intercropping is common, especially where there are not severe soil and water constraints. Common intercrops are upland rice, maize and various legumes. On the outer islands rainfall is usually less limiting, but poor soils are sometimes a constraint on the ability to intercrop cereals and legumes. On Java, farms are small and intensively managed, with few purchased inputs. The starch industry in Sumatra is based on large, vertically integrated plantations where moderate input levels are applied, and new high- yielding varieties planted. These industries have joined with the national program in supporting production research, which has benefitted surrounding small independent farmers as well as the plantations. The effects are becoming clear when analyzing production trends. Over the last decade, production grew with a healthy annual 1 .2-1 .4% (Figure 1), this growth being attributed to both increases in area (1.0%) and yield (0.36%) (Henry and Gottret, 1996). Nonetheless, large areas of Indonesia have suffered severe droughts during the past five years, and this is again the case in the 1997/98 season. Although this will much less effect cassava (relative to rice, etc.), production for 1998 is estimated to be down from earlier years (FAO, 1997). Cassava cultivation in India is concentrated in the southern states of Kerala and Tamil Nadu. The country is distinguished by the world's highest average cassava yields, about 24 t/ha. These high yields are accomplished by intensive cultivation, and, in Tamil Nadu, by irrigation. In Kerala, much of the production is consumed as boiled roots, one of the few regions in Asia where this is common. Cooking quality is one of the principal criteria farmers use in selecting varieties for cultivation. In Tamil Nadu nearly all of the production is for starch and sago. Markets for fresh consumption have been slumped during several years, drawing attention to the need for product and/or market diversification. This trend has affected the planted area to cassava, which shows an annual decline of more than 2%. Hence, less area, but with increasing productivity. China produces cassava in the southern provinces of Guangxi, Guangdong, and Hainan (and more recently in Yunnan). Most is planted with few production inputs on hillsides surrounding rice paddies. Historically cassava was a famine reserve crop, planted in marginal areas with high risk of crop failure. Production data from China are not very reliable, but according to FAO, area peaked at 245,000 ha in 1983. Although a minor crop in China as a whole, it is increasingly looked upon as an efficient producer of raw material for starch and on-farm pig feeding. Nonetheless, its area is under severe pressure from competing higher-value crops (fruit trees) and alternative land use. Like Indonesia, the Philippines is a multi-island economy, but differs in that population is spread more uniformly across different islands. Cassava fits well within an agricultural policy that emphasizes self sufficiency in basic foods (except wheat), import substitution, and development of the small farm sector. Cassava is produced throughout the Philippines, but is more concentrated on the southern islands, especially the Visayas region and Mindanao. Most production is on small farms, although there are some large plantations supplying starch factories. Input use is low in comparison to other countries, and this is reflected in some of the lowest yields in Asia. Nonetheless, yields in the past ten years have climbed steadily from about 7 to about 9 t/ha, suggesting adoption of improved varieties and cultural practices. Cassava area has increased steadily, from about 200,000 ha in 1984 to 215,000 ha in 1996. 3 8 5u C a 8 Is ft. 1 C c ex IS "S is § to <3 J I s: (suoj uoiiiiui) uoipnpojd babssbq 8In Vietnam, the area planted to cassava peaked in 1979 at 460,000 ha, but then declined to nearly half that level because of lack of markets and competition from other crops. It occupies the poor soils of mountainous and hilly areas, mainly in monoculture systems. Since many of these soils are highly erodible, their recuperation and conservation is currently a major research thrust. In early years, most of the plantings served as a food security crop, but human consumption now occupies only 10-20% of production. About 30% is used as animal feed, much of it on-farm for chickens and pigs, after chipping, drying and milling the roots. Industrial uses are on the increase as Vietnam undergoes rapid economic development. Industry now absorbs about 30-40% and this is rising. New varieties, mainly introduced from Thailand's breeding programs, and diffused with help from large-scale starch processors, are gaining popularity, especially in the South, for their high yield potential and high starch content. While cassava data do not show this yet, it is expected that these new varieties will increase yields significantly in the future. Cassava areas in Vietnam are shifting away from (semi)urban zones towards "waste lands" and more marginal areas. Due to high industral demand for cassava and changes in "land property" laws, the average cassava area (per farmer) in the South is increasing. PRODUCT AND MARKET TRENDS Diversity is the defining characteristic of cassava products and markets in Asia, both within and across countries. About 40% of cassava in the region is destined for human consumption (FAOSTAT, 1997). In Indonesia, the level is about two-thirds. Most of the remainder is processed for industrial purposes, principally chips for animal feed, and starch. Raw roots are not traded on any significant scale. The initial processing defines to some degree the market sector to which roots can be destined. This is unlike the grains such as maize which are traded as whole, unprocessed grain, to be converted into any number of products in the receiving country. Fresh for Human Consumption Outside of Kerala, India, and some poorer (isolated) districts of China and Vietnam, nearly all cassava for food is first processed; direct consumption of baked or boiled fresh roots is minor. This form of consumption is largely a rural practice, and often by households having their own backyard garden. Fresh consumption will decline with increasing incomes, urbanization, changes in dietary preferences and increasing opportunity cost for cassava from emerging alternative (processed) cassava markets. It must be noted, however, that cassava will remain a crop representing an on-farm reserve emergency crop (for human consumption) in times of rice shortfall, as for example, has been observed in Indonesia. Chips and Pellets for Animal Feed The cassava pellet industry peaked at the end of the 1980s. Since then total volume has decreased, mainly affecting Thailand4 (Table 2, Figure 2). The reduced exports to the European Union (EU) have only partly been offset by a concerted effort to penetrate new, non-EU markets. During the early 90s a further shock was delivered to EU-destined cassava exports, as a result of changes in the Common Agricultural Policy (CAP), lowering internal EU feed grain prices. Nonetheless, although export prices (and profits) have been reduced significantly, Thai pellets still can compete with EU internal (barley) prices. Moreover, due to the 1997/98 baht devaluation, the pellet exports' price edge will gain. It is doubtfull that pellet production will significantly rebound in the future, since domestic competition for roots from the starch factories (who enjoy bigger margins) is very strong. Table 2. World trade of cassava products (chips, pellets and starch). 1992-94 Avg. 1994 1995 1996 World exports 9.8 7.2 5.4 6.4 -Thailand 8.3 5.9 4.1 5.0 -Indonesia 1.1 0.7 0.5 0.6 -China & Taiwan 0.3 0.4 0.4 0.4 -Others 0.1 0.1 0.4 0.4 World imports 9.7 7.2 5.4 6.4 -EU 6.5 5.2 3.2 3.8 -China & Taiwan 0.9 0.6 0.7 0.6 -Japan 0.5 0.3 0.4 0.4 -Korea, Rep. of 0.7 0.4 0.3 0.4 -Other 1.1 0.6 0.8 1.2 Source: FAO Commodity Market Review 1996-97 4 For a complete treatise on this subject, see Titapiwatanakun (1996), Henry and Gottret (1996) and Hershey et al. (1997), among others. 10 c o s oa. X U a 10r ® 8 - "8 > ■J-. -r. u 1970 1975 1980 1985 1990 1995 Figure 2. Exports of cassava products from Thailand between 1970 and 1995. Source: Thai Tapioca Trade Association Starch Starch for industry is classified as native or modified. The technology for modifying starches by physical, chemical or biological processes is highly advanced and evolving rapidly, and modified starches are absorbing an increasing market share. Starch- derived products include sweeteners (high fructose syrup, glucose syrup), dextrins, monosodium glutamate, pharmaceuticals and various chemicals. Starch is used in large quantities in the manufacture of paper, plywood, textiles, and as a filler/stabilizer in processed foods. New products from starch are continually entering the marketplace. Throughout the region, the industry is moving toward larger, more technologically advanced plants, and small, less efficient factories are closing. Furthermore, an increasing trend can be seen in Thailand, Vietnam and China of joint ventures with large international companies from US, Europe, Taiwan, Korea, Japan, etc. Multinational food 11 companies with traditionally corn or potato starch based interest have started to aggressively diversify into cassava starch manufacturing. Thailand is leading the Asian starch boom (Figure 2), surpassing Indonesia in recent years. Both export sales and domestic use have increased significantly. The focus for exports has been on modified starches, partly for the larger profit margins and partly to get around some of the import barriers imposed against native starch. Nonetheless, the increase in starch exports has not nearly kept pace with the decline in pellet exports. Private and public sectors are cooperating to identify and exploit internal growth markets for starch as a complementary strategy to export-orientation. Indonesia's starch production is utilized mainly in internal markets. Nearly two- thirds goes into krupuk. Because of the specific starch characteristics required for this product, maize starch is not a competitor. This gives some insulation from the fluctuations of world starch prices. Growing demand in the near future is envisioned from paper, cardboard and plywood industries. Both China and Vietnam have significantly expanded and modernized their starch industries. Monosodium glutamate and glucose are rapidly growing markets in both countries. Foreign investors are drawn to the "newly opened" countries (Vietnam, Cambodia and Laos) to profit from the relatively cheap factors of production, land and labor, to invest in large-scale starch industries. Flour Cassava flours come in many forms. The most common is gaplek in Indonesia. Roots are peeled, chipped or sliced, and then dried. The dried chunks are ground or milled to a meal, which is used in a wide array of food preparations. It is consumed especially in times of rice scarcity, and partially substitutes for rice in rural daily diets. Cassava flour may partially substitute for wheat flour in bakery and other products. This is still minor in Asia, but is reported unofficially from several countries (Henry and Gottret, 1996). CASSAVA DEMAND PROJECTIONS Thailand's continuing campaign to reduce its dependency on the European animal feed market will dominate directions of the Asian cassava sector for the next decade. This will take several forms: introducing new production technologies to keep prices competitive with alternative energy sources; aggressively seeking new markets outside Europe; development of internal feed markets; and further diversification into starch and flour, with strong support for research on new processes and products. Other countries of the region, once with aspirations to penetrate export markets for pellets, are now recognizing that opportunities will depend very much on increasing production and processing efficiencies. Prospects for starch vary widely depending on the specific market. For native starch, the different sources (maize, cassava, sweet potato, white potato) compete with 12 one another on the basis of price. The markets for specialized starch are rather uncertain. On the one hand there is increasing demand, but on the other, there is a continually evolving technology for modifying starches to meet specific product properties. Ostertag (1996) suggests that most developing countries will use their resources most effectively by first concentrating on developing internal starch markets, in order to reduce the risks inherent in the export sector. In a recent study of the major tropical root crops, Rosegrant and Gerpacio (1997) of the International Food Policy Research Institute (IFPRI) have projected cassava production and utilization in the year 2020, based on a model that takes into account virtually all the world's food production and consumption (International Model for Policy Analysis of Commodities and Trade (IMPACT). They foresee that moderate demand growth for cassava products in Asia through 2020 will sustain viable cassava-based development. The growth sectors vary within the region. In China, growth in feed demand will be among the strongest anywhere, at 2.08% per year, accompanied by a continuing trend for lower direct use as food. Southeast Asia should see healthy growth in all sectors, i.e. 1.4% in food, 0.13% for feed, and a total of 1.25% (including industrial use) (Table 3). The import demand in the non-cassava producing countries of East Asia will rise at 1.0% per year, providing some additional market possibilities. SECTOR CONSTRAINTS AND OPPORTUNITIES Various factors will define cassava's potential as a catalyst for further evelopment in Asia. These factors can be grouped in the domains of policy, market economics, institutional issues and technology. However, the majority of important constraints/opportunities deal with overlapping domains. Policy The policy arena more than any other will set the stage for cassava's role in Asia. Both agricultural policy as well as broader economic policies impact the cassava sector. Distortions in input and output markets, asset ownership, and other institutional and market distortions adverse to the poor must be minimized. Government policy interventions can be divided into direct (impacting on the cassava sector) and indirect policies. The former includes fertilizer subsidies, credits, crop substitution schemes, import tariffs and quotas, etc. The latter includes similar interventions but on other crops (cassava substitutes or complements), that in turn will have an effect on the cassava sector, such as interventions on feedgrains, foodgrains, etc. Furthermore, there exist domestic and foreign policies. For Asian cassava sectors, history has shown that direct and indirect foreign trade policies have had relatively more impact than any other interventions. There is no indication that this will be significantly different in the future. 13 Table 3. Projected production and utilization of cassava" in 2020. Growth rate for utilization Utilization Production 1993-2020 (%/veart in 2020 in 2020 Food Feed Total (million t) (million t) China -1.27 2.08 1.19 3.9 4.2 India 1.00 0.00 1.00 7.6 7.8 Other East Asia -0.95 1.09 0.63 3.5 0.0 Other South Asia 1.00 0.00 0.83 0.6 0.6 Southeast Asia 1.40 0.13 1.25 27.0 51.1 Latin America 0.26 1.26 0.78 39.3 40.5 Sub-Saharan Africa 2.51 0.29 2.47 166.0 166.0 Developing 2.01 1.18 1.88 248.8 271.1 Liberalized trade has become the economic mantra of the 1990s. The watershed Uruguay round of multilateral trade negotiations, under the auspices of the General Agreement on Tariffs and Trade (GATT), is a fundamental influence on the direction of the global economy. Trade liberalization will bring complex and unpredictable adjustments in the agricultural sector. The implementation of regional trade agreements is well-advanced in Asia. The Asia Pacific Economic Co-operation forum (APEC) has 18 members, which in total comprise half the world economy. Most of the major cassava-producing countries of the region (excepting India) are members. APEC aims to achieve free and open trade and investment by 2010 for its industrialized members and by 2020 for the others. Hence, the future implications include decreased barriers within the APEC trading bloc. This implies for example that protected markets like Japan, would be more accesible for starch exports. On the other hand, EU policies continue to evolve especially regarding Asian pellet exports, and this will be crucial for the future. Careful monitoring, anticipation and timely actions are important. Market Economics Stabilizing or increasing demand in an environment of freer trade will depend on the ability of the industry to respond quickly to shifts in product demand of traditional markets, and analysis and subsequent penetration of new markets. The overwhelming preference for rice as the starchy staple, and the increasing demand for meat, will keep per capita cassava consumption levels low throughout Asia. This growth in meat consumption, however, is the basis for projecting strong potential 14 to use cassava for on-farm feeding, or in balanced rations, for pigs and chickens. Already, a clear increasing trend is identified of on-farm pig feeding (including cassava chips or processing waste) in many sub-urban areas of southern China and Vietnam, to satisfy the rapidly growing demand for pork in urban populations. Cassava's competitive position in national and international markets is closely linked to internal and world supplies and market prices of alternative commodities or products. Because of cassava's versatility, it may compete with a range of products in different markets. In the market for balanced feed rations, cassava in dried chip or pellet form competes mainly with sorghum or maize, and sometimes barley. In world markets, maize is the principal source of starch. In the principal cassava-producing countries of Southeast Asia, rice, maize and cassava production all increased three to five-fold in the past twenty-five years. Even this dramatic success, however, was not adequate for supplying the food needs of growing and somewhat more affluent populations. Grain imports, especially of wheat and maize rose from near zero in 1960 to 17 million tons in 1995. The challenge for cassava products will lie in remaining competitive following decreasing grain prices. This must be accomplished by reducing production and processing costs, but especially the former. The current devaluation of many national currencies in SE-Asia may have severe negative effects for their economic development, but it may be a blessing in disguise for their cassava product exports. Besides price, fortunately, specific intrinsic cassava starch characteristics command a premium, or constitute an edge over potato or maize, in many upstream modified starch-based products. Although these industries are extremely competitive and secretive, they also can be very lucrative. The identification and penetration of these markets will take a concerted technology and marketing R&D effort. Infrastructure Subsistence farming requires virtually no infrastructure, no need for purchased inputs, and no need for highways for reaching markets. Commercial agriculture, on the other hand, depends heavily on infrastructure. The World Bank notes that rapid economic expansion and urbanization have outstripped the capacity of existing infrastructure, and created serious impediments to further investments and growth. Insufficient electricity generation capacity, outdated and inadequate telecommunications facilities, poor roads and inefficient ports are the most crucial infrastructure problems. Purchased inputs for agriculture are for the most part available, but may not be used on cassava because of other constraints. There is little likelihood of major investment in infrastructure aimed solely at supporting cassava development, but the general development of the region will bring benefits to cassava production, processing and marketing. Technology Development Technology development includes the identification, generation and adaptation of 15 technology components suitable for alleviating the principal constraints of cassava production, processing and marketing domains. Besides the alleviation of constraints, technology must target opportunities that are presented through careful analysis. Henry and Gottret (1996) and Van Norel (1997) analyzed and quantified the principal cassava production constraints in Asia. Hershey et al. (1997) adapted and summarized the results, shown in Table 4. Based on these and other findings, Table 5 shows an assessment of expected impact on (a) income, equity and food security, and (b) the environment, by R&D intervention. As such, policy makers, investors and R&D agencies can more specifically target desired technology options. Technology Transfer Technology transfer for cassava is often a bottleneck to achieving impact. Transfer may fail either because the techniques are inappropriate or insufficient, or the technology itself may not be acceptable. Both have had a significant role in Asia; both are amenable to correction and improvement. In a well-functioning system, information flows both ways between the producer and the user of new technology. It is in this sense that technology transfer has an integrative function for the entire production, processing and utilization system. Technology transfer is normally considered an institutional function, the act of moving technology from the hands of science to the fields or factories of end-users, by a public extension service, or by private companies aiming at making a profit. Few of the conventional resources or practices exist at adequate levels for cassava. Virtually all programs in Asia report that extension services for cassava do not function optimally, and some barely at all (CIAT, 1995). In this environment, many programs have used unconventional and creative means to fill the gap left by institutional deficiencies. The Department of Agriculture in Thailand uses three separate channels for its highly successful transfer of new cassava varieties: direct transfer to farmers, mainly by way of on-farm regional trials and demonstration plots; through chip and starch factories, who promote new varieties to upgrade the quality of the raw product available; and the more traditional extension service methods. In Indonesia and especially in Vietnam, starch factories have also complemented the extension service to multiply and distribute new varieties. India has the most structured technology transfer system, and in general a better staffed public service sector than most countries. They have been able to achieve impact both with new varieties and agronomic practices. The cassava sector has the opportunity to be at the forefront of designing and testing unconventional technology transfer systems. The lack of institutional resources has already motivated public and private partnerships, and an array of other means to get technology to end-users. A survey of methods that have succeeded and those that have failed, and brainstorming about future planning, could be a highly useful exercise for Asia, and with implications elsewhere. 16 Table 4. Cassava constraints analysis for Asia, with comparison to global". Yield Total Total Total gain in Area yield yield yield Constraints affected affected gain gain gain as % area(%) (%) (%) ('000 tons) of global2) Soil Management 35 17,067 36 Low soil fertility 32 68 22 10,690 32 Soil erosion 17 60 10 5,039 43 Surface temperature 11 26 3 1,338 62 Crop Management 21 10,291 22 Sub-opt. land preparation 8 33 3 1,262 22 Quality of planting material 17 48 8 3,958 19 Inadequate spacing 8 47 4 1,853 30 Weeds 18 37 7 3,218 23 Intrinsic varietal traits 24 11,384 31 Low yield potential vars. 26 89 24 11,384 31 Climate 11 5,153 25 Drought 16 58 9 4,496 26 Low winter temperature 0 8 1 658 54 Diseases 2 929 3 Root rots 6 5 0 151 5 Bacterial blight 6 19 1 553 6 Anthracnose 2 15 0 132 5 ICMV 6 3 0 92 1 Other leaf/stem pathogens 0 82 2 1,042 30 Pests 3 1,478 7 Green and red spider mites 6 38 2 1,112 9 Mealybug 2 2 0 16 0 Whitefly 4 3 0 51 10 Termites 2 3 0 33 5 Mammalian pests 5 2 0 46 3 Scale insects 5 9 0 211 52 Tipacola plagiata 11 0 0 9 100 Total potential production increase 96 46,301 23 Post-harvest 21 9,923 32 Quality 22 60 13 6,390 31 Processing 15 24 4 1,806 30 Product marketing 20 18 4 1,727 47 Total cassava sector 116 56,224 24 11 Other constraints with near zero potential yield gain are excluded from the table. * Total yield gain in Asia as percent of global yield gain. Source: adaptedfrom Henry and Gottret, 1996. ©0 c Si 0>■ ■- OJo > o 2 0 ia, I £1 •c 3 3 -a s '3 > ■s 60 o 1 V- u O IS>-. JS60 '3o■o > 1 at 8 H § ■s a o 1 ■2 ^ a g § ItJ 1a sp-S-g 3,8 1 •- -a II rchas edto 3 O. ftladapt resistant > 3 f C u -■-■ i u O B .c 3 60 a a o £ 8 ** 1 ~ ~■ s .a a ? .5■ I £ 1 s -g 1 0 o S *Mil pi -J u CL c ? J O. ** <0 3 2 J- o o ■3 -a 5j ...□ "> %£ I I a. -o 33 II u a. i 3 a I-.B. S^ ■8 35 a 13 < 3 ~- 13 -S | 3 I I a! > > -2 o 6.3 fi-g-3 >.| ._ _ T3 3 T3 £ Q. B Q.t—'ao3^-3 CO d £ £ r c w (l) 03 O h- (- N V > o D. < C/3 • O < O o CM d o CM (eiyi) p|9|A jooj ijsajj o o o CM —I— o in n —i— o -tt-Jo 60 I 8 SI - £8 1 1 5 9 5 1 i ^ *> 3 K (%) juajuoo jauBoi Ajp jooy ga 5 33 REFERENCES Chinese Academy of Tropical Agric. Sciences (CATAS). 1995. Research of Tropical Crops. Volum 2, pp. 33-42. (in Chinese) Chinese Academy of Tropical Agric. Sciences (CATAS). 1992. Cassava Research in China. Proc. of 2nd National Cassava Workshop in China, held Oct 19-24, 1992, in CATAS, Hainan, China. pp. 19-46. (in Chinese) Chinese Academy of Tropical Agric. Sciences (CATAS). 1992. Proc. of Hainan Crops Germplasm Resources Investigation. Nov, 1992. pp. 33-38. (in Chinese) Kawano, K. 1987. Inherent and environmental factors related to cassava varietal selection. In: C.H. Hershey (Ed.). Cassava Breeding: A Multidisciplinary Review, Proc. ofa Workshop held in the Philippines, March 4-7, 1985. pp. 207-226. 34 CASSAVA VARIETAL IMPROVEMENT AND DISSEMINATION IN GUANGXI Tian Yinong and Lee Jun1 ABSTRACT The Guangxi Subtropical Crops Research Institute (GSCRI) began their cassava varietal introduction and improvement program in 1981. Since 1985, GSCRI has introduced cassava hybrid seeds as well as promising clones by tissue culture from CIAT/Colombia, from the Thai- CIAT program, as well as from other institutes. Through continuous selection and experiments, several promising clones, such as SM1113-1, SM1600-1, CM5443-1, SM1741-8 etc. have been identified, and these have been tested to a certain extent in on-farm trials. Since their release and dissemination these clones have awakened great interest by both farmers and starch factories. Some of them have now spread to other provinces. In China, high yield and high starch content of cassava varieties are still the most important factors in raising the comprehensive benefits resulting from cassava production. This is also the only way to change the stagnated position of cassava production, which in the past increased mainly by increasing the area planted. At the moment, many local governments of the principal cassava growing areas are conscious of this and have started to spend money on the introduction and dissemination of new higher-yielding varieties. INTRODUCTION In recent years, the cassava planting area and total production in Guangxi province of China have greatly increased (Table 1), the average annual rate of increase from 1991 to 1995 being 6.4 and 17%, respectively. This is the result of the continuous development of cassava processing, the products of which have had a good market. In many areas farmers have adopted new cropping systems and have increased their production investment, resulting in an increase in cassava root yield, as well as the income derived from growing cassava (Tables 2, 3 and 4). The cassava growing area in Guangxi is mainly concentrated in Yulin, Wuzhou, Nanning and Liuzhou districts, which are located in the middle and eastern parts of the province (Figure 1). The cassava area and production in these four districts account for 55 and 50% of the total in Guangxi, respectively, and the cassava processing industry is also mainly established in these areas, except for Yulin district. The natural conditions in these areas are comparatively favorable and communications are well-developed. But they also have a solid industrial foundation, so these areas developed rapidly and all these districts belong to the economically developed areas of Guangxi. But, being a low-value crop, cassava is also partly being replaced by other crops. Thus, the main cassava area in Guangxi seems to be moving to the western and northern parts from the eastern and middle parts (Figure 2). The data of cassava area for the most recent years in different districts of Guangxi have shown this. The western and northern parts of Guangxi include the two districts of Baise and Hechi; these are mainly hilly and mountainous areas, with low fertility soils and poor natural conditions, so they are the two less-developed districts Guangxi Subtropical Crops Research Institute, Nanning, Guangxi, China. 35 of Guangxi. For a long time there were very few processing factories and farmers planted cassava usually on steep slopy land without any investment in production, so that yields were very low. The roots were mostly used for feeding livestock. In recent years, the government has given some "assistance projects" to this area and several starch (or alcohol) factories were built one after another, while some cassava production bases have been set up. They have introduced some new cassava varieties and adopted new production technologies, so the yield of cassava increased markedly. Compared with 1990, the cassava area in these two districts in 1995 had increased more than 20,000 ha, and total production increased 356,106 t. They account for 34% of the total increase in area and 18% of the total increase in cassava production in Guangxi. In 1995, the cassava area and production in these two districts accounted for 17.5 and 15% of the total in Guangxi, respectively, while in 1990 this was only 12 and 12%. Cassava Varietal Improvement at the Guangxi Subtropical Crops Resarch Institute (GSCRI) SC201 and SC205 are the two most important varieties, which are estimated to account for 85% of the total cassava growing area in Guangxi. These two varieties were introduced into Guangxi about 60 years ago. Since cassava has not been an important commodity in the past, farmers did not pay any attention to adopt new varieties or new production technologies. This is the reason that yields in Guangxi were low for a long time. In the 1980s, although several new cassava varieties were introduced, few of these were really adapted to the conditions of Guangxi. Because only 1-2 cassava varieties were used until the beginning of the 1990s, there was little potential to increase farmers' income. Moreover, the harvest period is short and is mainly concentrated in Dec and Jan, so except in these two months, only dried chips can be used for processing. In view of this, many starch factories appealed to agricultural institutes to introduce or improve new cassava varieties with high yield and high starch content, and they have been willing to fund some of the research. en E © 1 £ a. & c o u O 3 I/O .■5 o es 1 Q. C ca c 3 A . a- eg H c o 3 "8 a € In ! 6" 8 i in ! 85 §5 en 00 In §5 £ 1 £> ! On .o Nt--TJ-enr~en — (NO*r-enr~r-OnpN nonoqonOnno — On — r-enm en^-^-enOn^Oesin.'*. ^ am en en O cm en en 83oo — r~ en enen n© r- o oo _ _■* ts r- en o\ tn in r- r- r. r. * r- <* . oooncnoinOnin — ooenenenennono© " - - m JO (N a h m pi 21_,42_ .9,4__ 29,._ 2s__ 229,9._ 29.,42_ _s__ _s__ 46s4_ 4___ _4_._ ____ .s__ i 1 IS ifltOOONO** VO —i ■* (N O o en On (N r- n; r-; in nO oidfid^mi^N ■* tsVi ri iri (N en en cm T+ — © no oo r- no -* oo © on o •-^moooooaooo r- oo — oo no cm no — r> nooocman o CO r~ no — m cm — ^t- en oo r~ Tj-r-^cMinen—.enOTj-r^ CM nO On CM inoo©r~e~oono©enen — — (N — — — o r~ —> " -n oo r- r— cm no in CM" fN oooocMeninVOrn — c^nono-* on r- — en — no ©^ » ©^ cm en ©> en en ©" —<* m" ©" oo" en ■* en" o" cm" «s cm — Ttenooin—■ en oo on —* — * oo o « r~ininmr-enOnr~nOO(NQOnO\t~ N Vn^ On ■* ©^ (N VO^ ©^ — t- In * ■*_ VO^ e^ no" T}-" cm" en" On" no" en" On" m" On" Cm" On" VD -*" ©\ —* rnc-4 enen—. — —i rM ooonor~ — or-inOnf>rninm 't-« T*Cho0O\• >N « o a .s 60 3 .S © *-■ tt -S2 .2 h .2 c •- -C -2 eo .5 3 .3 60 3 ^u ^ o ._ S.s s •o -o p OIsll liil-s s-fl| g>5 cd .2 3 P '53 <9 .2 3 3 3-3 5 .3 S O a Si 65 "8, I 2 37 Table 2. The yield and gross income from planting different cassava varieties in Laibin county of Liuzhou district, Guangxi, China in 1995. Fresh root yield (t/ha) Price roots (yuan/t) Stem yield (t/ha) Price Gross stems income Variety (yuan/t) (yuan/t) SC201 (local) 22.5 360 25.4 200 13,180 SC205*(new) 26.3 370 21.5 430 18,976 SCI 24 (new) 30.0 360 25.6 1100 38,960 * SC205 is a new variety for Laibin county but is common in southern Guangxi Table 3. Comparison between several crops in terms of gross and net income in Laibin county of Liuzhou district Guangxi, China in 1995. Yield Price Gross income Net income (t/ha) (yuan/t) (yuan/t) (yuan/t) Cassava 22.5 360 8,100 3,600 Orange 22.5 1900 42,750 33,000 Sugarcane 60.0 270 16,200 7,200 Mango 7.5 3200 24,000 16,800 38 Table 4. Comparation between several intercropping systems in terms of gross income in Laibin county of Liuzhou district, Guangxi, China in 1995. Cassava fresh Price Intercrop Price Gross root yield roots yield intercrops income (t/ha) (yuan/t) (t/ha) (yuan/t) (yuan/ha) Cassava+Watermelon 22.5 360 30.0 400 20,100 Cassava+ Peanut 24.8 360 1.9 3,200 15,008 Cassava+ Watermelon seeds2 1.8 360 0.5 14,000 14,848 Source: Laibin State Farm The Guangxi Subtropical Crops Research Institute (GSCRI) began their cassava varietal introduction and improvement program in 1981 . At that time, they were mainly working on the introduction of promising clones or new varieties from other Chinese institutes, which were tested in the Institute and then distributed to State Farms. Since 1985, GSCRI has introduced cassava hybrid seeds as well as promising clones by tissue culture from CIAT/Colombia, from the Thai-CIAT program, as well as from other institutes. In the last decade, GSCRI has introduced a total of 23,733 cassava hybrid seeds from 432 parent combinations (Table 5). In consideration of the natural conditions of cassava production in Guangxi and based on the demands from farmers and starch factories, the general objectives of the cassava varietal introduction and improvement program in Guangxi are to develop promising clones with the following characteristics: - High root yield and high harvest index - High root starch content - Non-branching growth habit - Tolerance to wind - Good root shape - Early harvestability During the past ten years, GSCRI has made good progress in developing improved cassava varieties (Table 6). Through continuous experimentation and selection, four promising clones have now been identified, i.e. SMI 113-1, SM1600-1, CM5443-1 and SMI 74 1-8 (Table 7). These four promising clones were higher yielding than the local varieties SC201 or SC205 at three locations (Nanning, Mingyang and Laibin). The yields of clone SMI 1 13-1 at Nanning, Mingyang and Laibin were 23.2, 52.5 and 28.4 t/ha, respectively. The yields of clone SM 1600-1 at Nanning and Laibin were 24.9 and 39 40.7 t/ha, respectively. The yields of clone CM5443-1 at Nanning and Laibin were 29.5 and 34.0 t/ha, respectively; and the yields of clone SMI74 1-8 at Nanning and Laibin were 32.2 and 36.0 t/ha, respectively. The root dry matter content of clone SMI 1 13-1 is also very high (Table 8). Cassava Varietal Dissemination in Guangxi - Channels and Procedures The channels for cassava varietal dissemination in Guangxi are as followed: - Committee of Science and Technology - Extension Station of Agricultural Technology - State Farm (starch factory) - New Technology Development Company - Private enterprises As the department responsible for science and technology, the Committee of Science and Technology is not only in charge of carrying out policies of the government in the countryside, but also plays a very important role in disseminating the achievements of science and technology. It is the key unit for spreading new technologies from scientific institutes to rural areas. The local Extension Station of Agricultural Technologies is the unit responsible for spreading agricultural technologies directly to farmers, but normally they work by introducing new high yielding technologies to agricultural communities. State Farms are separate units with all these functions. There is an Agricultural Research Team (ART), which is composed of technical staff whose members have graduated from an Agricultural School or University (College). State Farms tend to have a lot of land. So, in general, State Farms are not only a good experimental site, but are also an ideal place for scaling-up agricultural production. Trials conducted in State Farms normally conform to the standard. New Technology Development Company is a new organization, which has expanded rapidly in recent years; most of these are operated by the government and are especially engaged in the introduction and dissemination of technologies. The private entrepreneurs are people who achieve success during the rapid development of a modern economy; among them are many who are operating in agricultural production. They are not only businessmen but also producers, and they have not only capital but also flexible management strategies which allows them to easily adopt new things and take advantage of their conditions to quicken the tempo of agricultural development. s c I 1 I I I O Tt Tt\f\ ■■* w+ in o; t~ — © ci to P) <* f* -* i £ F3 2 <- N 85 n m r'tO (S V m oo oo no 8 * " 2 y US § 1 3 8 00 r* tN to cs3 8 N (N 0-J (N s rS i On r- < c o u B o 1 -8 I o 00 o o o o o i o an (IOjjuod jo %) ireaui uoijBpidod Soipasjg I-S; o oo a c ' OS ON SO Os O ON -s oo .oc r- e —r— e oo -1- o © 00 o Tt I o o (Io.nuo3 jo ■%) UF3U1 aopepidod iiuipo.uji 81 REFERENCES Dang Thanh Ha, Le Cong Tru and G. Henry. 1996. Analysis of the current and future cassava starch market in Vietnam. In: R.H. Howeler (Ed.). Cassava Production, Processing and Marketing in Vietnam. Proc. of a Workshop held in Hanoi, Vietnam. Oct 29-31, 1992. pp. 159-172. Henry, C, Le Cong Tru and V. Gottret. 1996. Vietnamese cassava constraints and opportunities: The bottom line. In: R.H. Howeler (Ed.). Cassava. Production, Processing and Marketing in Vietnam. Proc. of a Workshop held in Hanoi, Vietnam. Oct 29-31, 1992. pp.203-220. Howeler, R.H. 1992. Agronomy research in the Asian cassava network. An overview 1987-1990. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Utilization Research in Asia. Proc. 3rd Regional Workshop, held in Malang, Indonesia. Oct 22-27, 1990. pp.260-285. Kawano, K. 1995. Cassava varietal improvement in the developing economy of Vietnam. Paper presented at the National Food Crop Program Meeting, held by MAFI in Hanoi. Sept, 1995. Nguyen Huu Hy, Tran Dai Nghia and Pham Van Bien. 1995. Recent progress in cassava agronomy research in Vietnam. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp.237-252. Nguyen Van Thanh. 1996. Cassava in Vietnam: An overview. In: R.H. Howeler (Ed.). Cassava Production, Processing and Marketing in Vietnam. Proc. of a Workshop held in Hanoi, Vietnam. Oct 29-31, 1992. pp.12-33. Tran Ngoc Quyen, Vo Van Tuan, K. Kawano and Hoang Kim. 1997. Cassava germplasm and results of cassava improvements in South Vietnam. In: Hoang Kim and Nguyen Dang Mai (Eds.). New Progress in Cassava Research and Extension in Vietnam. Proc. Cassava Workshop held at Hung Loc Agric. Research Center. Dongnai, Vietnam. March 5-7, 1996. pp. 24-34. (in Vietnamese) 82 CASSAVA VARIETAL DISSEMINATION IN VIETNAM Hoang Kim1, Tran Ngoc Quyen1., Pham Van Bien2 and Kazuo Kawand* ABSTRACT Before 1985, Gon, H34 and Xanh Vinh Phu were the most popular cassava varieties in Vietnam. From 1986 to 1993, HL20, HL23 and HL24 were selected from the local variety collection by Hung Loc Agricultural Research Center (HARC) and were grown extensively in South Vietnam with annual areas of about 70,000 to 80,000 ha planted to these varieties. More recently (1993-1996) the Vietnam Root Crops Program in cooperation with CIAT, selected and recommended two new cassava varieties, KM60 (Rayong 60) and KM94 (MKUC 28- 77-3); these were recognized and released for production by the Ministry of Agriculture and Rural Development (MARD). The two varieties are now widely grown in an area of about 15,000 ha in 1996. The report presents the linkage between cassava research and extension activities in Vietnam. Experiences and methods of cassava varietal dissemination include ten mutual link-up activities (ten Ts). The most important one was the establishment of the Vietnam Cassava Research and Extension Network (including advanced cassava farmers, researchers, extensionists, managers of cassava research and development projects, cassava trade and processing companies), and the establishment of on-farm research and demonstration fields (Farmer Participatory Research - FPR). The Vietnam Cassava Research and Extension Network obtained good results during the period of 1993-1995. Advanced fanners who obtained high yields and high profits due to the growing of improved cassava varieties, became attractive models for other cassava growers, resulting in the expansion of new varieties. In Tay Ninh province, for example, in 1990 the cassava growing area was 3,350 ha, with an average yield of 10.8 t/ha and a total production of 36,200 tons. With the planting of new high-yielding varieties and new cultivation techniques, in 1995 the cassava growing areas had increased to 18,870 ha with an averaged yield of 20.5 t/ha and a total production of 386,900 tons. Six essential conditions for the successful cassava varietal dissemination in Vietnam include: Materials, Markets, Management, Method, Manpower and Money (six Ms). However, other problems should be taken into account: Crop competition (especially between cassava and sugarcane); soil fertility degradation and erosion; and decreasing varietal diversity. INTRODUCTION Cassava in one of the main crops in Vietnam (Table 1). It plays an important role in the strategy of national food security. It is also a main source of raw material for starch and animal feed factories. In 1995, the total cassava area reached 277,500 ha with an average yield of 8.0 t/ha and a total production of 2,21 1,700 tons (General Statistical Office, 1996). 1 Hung Loc Agric. Research Center (HARC) of IAS, Thong Nhat, Dong Nai, Vietnam. 2 Institute of Agricultural Sciences of South Vietnam (IAS), Ho Chi Minn city, Vietnam. 3 CIAT Asian Cassava Regional Program, Dept. Agriculture, Bangkok, Thailand. 83 Table 1. Area, yield and production of principle crops in Vietnam in 1995. Crop Area Yield Production Food crops COOOha) (t/ha) COOOt) Rice 6,765.6 3.69 24,962.8 Maize 556.8 2.13 1,484.2 Sweet potato 304.6 5.53 1,685.8 Cassava 277.5 7.97 2,221.7 Vegetables 328.2 12.62 4,145.6 Beans 187.5 0.68 126.7 Annual industrial crops Groundnuts 259.9 1.28 334.4 Soybean 121.1 1.03 125.5 Sugarcane 224.8 47.60 10,711.2 Tobacco (leaves) 27.0 1.00 27.0 Mulberry 21.8 6.84 148.9 Cotton 14.6 0.71 10.0 Rushes 9.0 6.49 5.9 Jute 4.2 2.27 9.5 Perennial industrial crops Rubber 278.4 0.84" 122.7 Coffee (beans) 186.4 2.18" 218.1 Tea (dry) 66.7 3.41" 180.9 Cashew nut" 250.0 0.83" 100.0 Fruits2) 310.0 - - Pepper21 7.0 1.28 9.0 " The harvested areas of rubber, coffee, tea and cashew nut in 1995 are: 146,900, 99,900, 53,000 and 120,000 ha, respectively. Source: General Statistical Office, 1996 Cassava areas and production in Vietnam have flucated markedly during the period from 1976 to 1995 (Figure 1). The biggest cassava area was 461,400 ha with a total production of 3,422,000 tons in 1979. Cassava root yield increased from 7.5 t/ha in 1980 to 8.0 t/ha in 1995 (Nguyen Van Thang, 1996). Before 1985, the most popular cassava varieties grown in Vietnam were Gon, H34 and Xanh Vinh Phu. During the period of 1986-1993 (Table 2), three cassava varieties, HL20, HL23 and HL24, were selected from local cassava collections and released by the Hung Loc Agricultural Research Center (HARC). They were grown annually on about 70,000 ha in South Vietnam (Tran Ngoc Ngoan et al., 1995). 84 ID o o o 0) 600 400 I 200 v> c/> a O c o c o .*- o 3 T3 O ra > in ro O i0r 2 8 » (0 > ID V> Tt (Sc4 — •>* Root yield (t/ha) 1 .9. <*- J2 c s OS OS t ttJ I 5 > >* * wi rn r- c> /-> S 00 *-< on r- -H Tt — ~ oo ci 00 >0 On N H oo 8*3 §223 _.- SuufSacacacoac (S >« $s~ % 3 * « "5! -C T3 IS Q T3 I >0-;t*;00*00(»|00 od ^ od ol (^ t»' ci t*\ s *0 VO U"> On f; *" rn t-^ T Nt (*l rt (H TF rn m « « no t t oo OJ J OJ O O r- Q < S aa- o Q3 3.3 CD at O E in q N - no no o c-i N N N (S (<1 * N t-^ o\ ^t r-i — r-l nO VO nd in oo -H i/i On >o oo -h 00 00 ITl O On nd r-l -h — —. r-i -h —i On 8 o\ On !/"> ^ Tt nO r-i nd i/i r4 in in Tt Tt oo 00 on. ,t "1 f- 00 s ■* ffl O K in q S3 $! 0 V 3 o ■3 I 2 On A E - •c 1 o I 1 - c3 Z c c Q 3 fi I* 2 o H 0 C > 5 00 00 O O O r- in -h in r^ to o in On t» r- —. r-i od in -^ Tt ^" W fI d fi no o r- N N N T3 JO X £ 00 On —' * o\ r- m > \6 \d «i ci Nin i o o 2wOUUUUUOS © ci Tt (S] u in o d Q 2 On > eo•a 3 o no O 05 e I o OS H oo a. — 0 N (S N -H d ci ci e-i - -NNci O -H (S —. O O —< oo 0 N N N — r-l ci m ci r*s tn — — - s On 00 On -* (S CI n >« On On On On On On On On On On a. oo o5 O a o £ S « .2- 2- 1: o o « Si! 5" 8.8. | fe <- c <2 <2 o § i - ** ** * * Is O Q O in ITl O V> N m r- in o o 5 >5 — m TJ- —c oo oo r~ 1D © H h N O h Oi -—8 N* OiO 5 ©Q>/"li/->Q©0©0 ioocN©ooo*t--r- oo oo r- m m N d •- »O Q O >« "> « o — — ——8^888° J Je J v-> o O V> IT> O a in r^ o g .S I * DO a 5•o on P ■s gI o ia -«8 95 An outstanding example is Tay Ninh, where the new high-yielding cassava varieties are rapidly replacing the local low-yielding clones. Before 1990, Gon, H34 and Binh Duong varieties comprised 100% of production. In 1995, however, the new varieties covered 80-90%, while the newest clones, i.e. KM60, KM94, SM937-26, and KM95, covered about 20% of the total cassava area of Tay Ninh. According to Table 9, the averaged yield in 1995 was 20.5 t/ha in 18,850 ha, resulting in a total production of 386,900 tons, as compared to an average yield in 1990 of 10.8 t/ha, a planting area of 3,350 ha and production of 36,200 tons (Tran Vien Thong, 1996). An economic analysis shows that in Tay Ninh the planting of KM60 could give a net income of 4.46 million dong/ha with an average yield of 20 t/ha in 1995 (Table 10). Table 11 shows the inputs and economic benefits obtained by three advanced cassava farmers who planted the new variety KM60. Private companies that participated in the cassava R&D system also promoted greatly the varietal dissemination and the change in cropping systems. In 1995, KM60 and KM94 were released as national varieties. Table 9. Cassava growing area, yield and production in Tay Ninh province from 1990 to 1995. Year Area(ha) Yield (t/ha) Production (t) 1990 3,355 10.8 36,210 1991 7,366 11.5 85,142 1992 7,173 11.7 83,652 1993 9,337 13.7 129,624 1994 15,846 18.7 196,598 1995 18,849 20.5 386,904 Source: General Statistical Office, 1996. Tran Vien Thong, 1996. 96 Table 10. Cost and return analysis per hectare for a new cassava production scheme using cultivar KM60 in Tay Ninh province in 1995, assuming a yield of 20 t/ha fresh roots. Unit Amount Unit price COOOd) Total price COOOd) 1.Total cost: 3,937 Plowing buffalo-day 3 140 420 Other land preparation buffalo-day 1 150 150 Planting man-day 12 12 144 Weeding man-day 10 12 120 Harvesting man-day 30 12 360 Transportation pick-up 4 200 800 Planting stakes bundle 50 10 500 Chemical 72 Herbicide liter 2 170 340 Urea bag of 50 kg 1 148 148 Ammounium sulfate bag of 50 kg 3 75 225 Single superphosphate bag of 50 kg 6 48 288 Potassium chloride bag of 50 kg 3 90 270 Farm yard manure t 2 50 100 2.Gross income: 8,400 Root harvest t 20 300 6,000 Planting stakes bundle 70%) of cassava is used for direct human consumption, whereas in Tamil Nadu and Andhra Pradesh 70% of production is utilized as industrial raw material for the manufacture of starch and sago. There are about 1000 small-scale starch factories in Tamil Nadu and about 45 in Andhra Pradesh, but in Kerala there are only 20 units. With the wide-spread availability of cereals like rice and wheat resulting in a reduced consumption of cassava as a cereal substitute, the crop's diversification to industrial processing is the only alternative to sustain and increase the current level of production. With the cassava-based industries picking up in Tamil Nadu and Andhra Pradesh, it is no surprise that cassava hybrids have become popular in those states. Even in Kerala, which has low industrial utilization of cassava, the yield of cassava has remarkably increased from 7-8 t/ha in the 1950s to 18-19 t/ha in the 90s. This was achieved mostly by low external-input improved production practices, along with a limited coverage of high-yielding varieties. There is a distinct variation in the preference for cassava varieties in different regions of Kerala. A survey conducted by CTCRI revealed that the local varieties are still very popular in Kerala state (68%) and their concentration is highest in central and northern regions with a coverage of 80-90% of the area in Kottayam and Malappuram districts (Table 2). In Trivandrum district, local varieties occupied about 53% of the area and in Quilon around 36% (Ramanathan et al., 1989). The improved variety M-4 is very popular throughout the state and its dominance is especially noticeable in the southern region. 104 Table 2. Varietal coverage of cassava in various districts of Kerala. Varietal coverage in each district ( % of area planted) Varieties Trivan drum Quilon Pathanama Kottayam thitta Malap puram Overall Local varieties 52.76 35.55 12.65 81.65 90.51 68.78 Improved variety M-4 40.63 49.26 53.94 18.35 9.21 25.13 High-yielding varieties H-165 4.68 2.25 5.45 - 1.65 H-226 0.15 10.80 22.02 - 0.15 3.27 Sree Visakham 1.22 1.07 - - 0.13 0.44 Sree Sahya 0.56 1.07 5.94 - - 0.73 Source: Ramanathan et al. , 1989 The utilization pattern of cassava in various districts of Tamil Nadu was observed to have an influence on the varieties preferred for cultivation by the farmers. As most of the production is used for direct human consumption in Kanyakumari district, good tasting local varieties are preferred and are presently grown in about 80% of the area (Table 3). The improved cassava variety M-4 had a coverage of only 6% of the area. The high-yielding varieties H-165 and Sree Visakham (H-1687) were also rather popular in this district and they accounted for about 13% of the area. With the concentration of cassava-based starch and sago factories in and around Salem, there was a preference for growing high-yielding varieties in the districts of Salem, South Arcot and Dharmapuri. About 75% of the cassava area in each of these districts was found to be planted with the hybrids H-165 and H-226. The remaining 25% of the area in Salem and Dharmapuri was planted to the popular local variety "Burma" (Ramanathan et al., 1990). Table 3. Varietal coverage of cassava in various districts of Tamil Nadu. Varieties Varietal coverage in each district (% of area planted) Kanyakumari Salem South Arcot Dharmapuri Overall Local varieties 80.21 27.83 7.65 Improved varities M-4 6.40 - 11.33 High-yielding variteties H-165 4.76 5.15 57.03 H-226 - 67.02 23.99 Sree Visakham 8.63 - - 27.58 72.42 23.10 4.41 22.33 49.73 0.43 Source: Ramanathan et al. , 1990. 105 VARIETAL IMPROVEMENT Germplasm Collection and Breeding In India breeding research in cassava is mainly centered at the Central Tuber Crops Research Institute (CTCRI), in Thiruvananthapuram, Kerala and at its regional Centre in Bhubaneswar. The All India Co-ordinated Research Project on Tuber Crops, whose headquarters is located at CTCRI, is also engaged in a limited way in cassava improvement. Out of the eleven centers located at different Agricultural Universities and at the ICAR Research Complex for the Northeastern Hill Region in Shillong, eight centers are conducting cassava research. Genetic variability is the essence of any plant breeding program. An assembly of diverse genetic stocks of any crop is the raw material from which a new variety can be moulded to suit the requirement of the farmers. A germplasm bank, including wild relatives from within and outside the country, has been built up. The CTCRI germplasm bank presently maintains 781 exotic and 806 indigenous acessions of cassava, as well as eight wild species. The All India Co-ordinated Research centres maintain 765 indigenous accessions. Many of the local varieties of cassava under cultivation in Kerala are either chance seedlings or are bud mutations selected for desirable characteristics and maintained by farmers through vegetative propagation. Varieties best suited to the requirements imposed by the local conditions are generally adopted and popularized in various cassava growing areas. Moreover, the tendency of farmers to clonally multiply the self-sown seedlings, if these are bestowed with any desirable attributes, particularly better root characteristics, led to the addition of numerous varieties (Nair and Pillai, 1995). The majority of these types have local names, which generally indicate one of the striking features of the plant. A systematic evaluation of the genetic stocks has resulted in the identification of several promising clones, which were later utilized either as parents in the intervarietal hybridization or were released as varieties. As such, six varieties, i.e. H-165, H-226, H-97, Sree Visakham, Sree Sahya and Sree Prakash, have been released from the Central Tuber Crops Research Institute (Magoon et al., 1970; Jos et al., 1981). The Kerala Agricultural University released a short duration cultivar called "Nidhi". Attention is currently given for the development of early-maturing, good cooking quality varieties, which can be harvested at six months, so that they can effectively be utilized in the rice- cassava crop rotation program now in vogue in Kerala state. Research in this direction has resulted in the identification from the germplasm bank of three short-duration indigenous clones, i.e. CI-649, CI-731 and CI-732, which are now in the pre-release stage in Kerala. These varieties are presently under initial evaluation in Andhra Pradesh, Tamil Nadu and Madhya Pradesh under the All India Co-ordinated Research Project on Tuber Crops. 106 Evaluation of CIAT Germplasm Cassava seeds received from CIAT were initially screened at the Regional Centre in Bhubaneswar. During the period from 1989-96, five sets of cassava botanical seeds comprising 1 19 accessions were introduced from CIAT. Out of five sets, four sets of seeds were evaluated preliminarily for yield characters and their reaction to diseases. Germination percentage of seeds ranged from 0-100% . The highest average germination was observed in the third batch (63.2%). Seedlings were evaluated in the field. Harvesting was done at 7-10 months. The fresh root yield, along with the number of selections made in each batch, are presented in Table 4. Table 4. Fresh root yield obtained and the number of selections made in four batches of cassava seeds received from CIAT. Range in Average No.of fresh root root yield selections Batches yield (t/ha) (t/ha) 1" batch 0.00-24.69 5.43 1 2nd batch 4.94-43.02 8.11 1 3rd batch 8.64-43.20 19.45 9 4th batch 0.00-37.65 12.86 3 The root yield ranged from zero to 43.2 t/ha. The highest average yield of 19.45 t/ha was recorded in the third batch. A total of 14 high-yielding lines were selected (Table 5). The highest fresh root yield (43.2 t/ha) was obtained with CMR36-32, which was not significantly different from that of CMR33-67 (43.02 t/ha). SM2077 and SM2090 were the next highest yielders, which recorded 37.65 and 35.18 t/ha, respectively. Cassava Mosaic Disease (CMD) was not noticed at the Regional Center in Bhubaneswar, but when these promising selections were transferred to the germplasm bank at CTCRI, clear CMD symptoms were noticed in all the clones. The most important disease noticed at the regional center was witches broom, which infected six accessions, i.e. CMR38-58, CMR36-12, CMR36-38, CMR36-1 16, CMR36-71 and CMR36-73. Angular spot disease and leaf spot disease were also noticed. The fifth set of seeds received from CIAT was distributed to the Regional Center in Bhubaneswar as well as to the Co-ordinating Centers in Madhya Pradesh and Andhra Pradesh for further evaluation and selection. 107 Table 5. Performance of 14 selections from cassava botanical seeds received from CIAT. Fresh root yield Root Root Selections (t/ha) color" taste2) OMR32-02 24.69 W NB CMR33-67 43.02 w S CMR36-32 43.20 w S CMR36-34 22.21 w NB CMR36-38 22.21 w NB CMR36-116 25.91 w S CMR36-123 28.50 w B CMR36-159 22.21 w S OMR36-28 23.44 w S OMR36-31 28.38 w B OMR36-73 23.45 w B CM8619 25.43 w S SM2077 37.65 w S SM2090 35.18 w NB »W = white 2>NB = not bitter; B= bitter; S1 = sweet Production of Chromosomal Lines Polyploidy breeding has great potential in crops where the economic produce is a vegetative part, especially when it can be clonally multiplied. The somatic chromosome number in cassava is relatively low (2n = 36), and hence the plant can tolerate higher ploidy. So the production of induced tetraploids (4n = 72) and triploids (3n = 54) constitute yet another basic approach to cassava improvement. Auto- tetraploidy is being induced with colchicine in a wide range of genotypes, so as to provide considerable variability at the polyploid level and allow a large-scale hybridization program, followed by selection among polyploid types (Magoon et al., 1969). The induced tetraploids so far produced have also been successfully utilized for the production of triploids. Triploids are obtained by crossing induced tetraploids with some of the cultivated cassava types; they were found to be superior to colchicine- induced tetraploids in yield and sometimes outyielded diploids (Jos et al., 1987). The triploid plants consistently showed higher root dry matter (DM) contents. The triploid 2/14 produced a higher root yield, dry matter (>45%) and recoverable starch content (>35%) under multi-locational trials than the standard cassava check varieties. This promising triploid was proposed for release to the State Variety Release Committee under the name "Sree Harsha". 108 Enhancement of Carotene in Cassava In most cassava clones, the flesh or the edible portion of the root is white and devoid of any carotene. Yellow pigmented cassava varieties are cultivated in a limited way in Colombia, the Philippines, Jamaica and in some south African countries (Oduro, 1981). Among a total of 654 accessions screened for carotene, 21 clones had yellow flesh (Moorthy et al., 1990). The frequency of high-carotene clones were found to be higher among the exotic collections. The carotene content ranged from 65 IU/100 g to 670 IU/100 g. An attempt was made to elevate the carotene levels through genepool development from the existing genetic resources (Jos et al., 1990). By simple recurrent selection the carotene content could be elevatd to 1500 IU in the first cycle, to 2200 IU in the second, to 3217 IU in the third, and to 3985 IU/100 g in the fourth cycle (Table 6). Table 6. Carotene content of cassava roots in the fourth cycle of recurrent selection. Carotene content Color of flesh No. of clones (IU/lOOg) Orange 12 2000-3985 Yellow 9 2880-3925 Yellow orange 12 2200-3400 Interspecific Hybridization Interspecific hybridization and genome analyses carried out on different crops have opened up new avenues of improvement of crop plants and have successfully contributed to the development of radically new and better types (Magoon, 1967). By virtue of sharing a common genepool, cassava is easily crossable with a number of wild species, and the occurrence of desirable genes makes interspecific hybridization one of the most significant approaches in cassava germplasm development. However, the absence of any wild species in Asia makes it difficult to procure the materials and to maintain the species under our conditions; the lack of clonal propagation in some of them, as well as their poor flowering, make the interspecific hybridization also rather difficult. Though half a dozen species were used in the program, success could be achieved only in four combinations, i.e. cassava x M. glaziovii, cassava x M. tristis, cassava x M. flabellifolia and cassava x M. caerulescence. The hybrid, cassava x M. caerulescence was found to be completely free of CMD symptoms. However, the transfer of tolerance to the cultivated varieties was hindered by its erratic flowering and female sterility. Interspecific back-cross (BC1 and BC2) populations were grown for CMD screening and evaluation. Fifty seven per cent of the back cross population was 109 completely free of CMD symptoms. Twenty eight hybrids recorded yields ranging from 2.0-3.5 kg per plant. True Cassava Seed Program In the traditional production system it is difficult to bring large new areas under cassava cultivation because of the low multiplication rate, bulkiness of planting material, difficulty in transportation, rapid loss of viability of the planting material and the high risk of introducing disease and pest problems (Rajendran and Ravindran, 1993). Studies using sexual seeds have revealed that in the true seed program the multiplication rate could be raised to above 1:150, as compared to 1:10 in the conventional method. Promising cultivars have been identified which can produce more than 200 kg of sexual seeds per hectare. At the time of transplantation of the seedlings in the field, the tap root is cut and removed, which contributes to a high clonal yield at the seedling stage. The first clonal progeny lines of promising parents had comparable root yield potential and root dry matter production as those of the popular clonal varieties. These studies indicate that cassava propagation from true seed has considerable potential in some new industrial areas due to their high multiplication rate, ease in covering extensive areas with lesser expense and comparable root yield and dry matter content. The incidence of CMD is considerably reduced in these progeny lines. Rapid Multiplication of CMD-free Plants A quick method was developed at CTCRI for the multiplication of mosaic free planting material. Symptom-free plants are selected from the field and stems are collected. The stems are cut into small stakes having only 3 to 4 nodes. In the normal practice, the setts are planted directly in the field. But in this new system, a nursery stage is introduced before transplantation to field. The stakes are planted on nursery beds and left to sprout. Screening for CMD symptoms is done as soon as leaf emergence has started. Only symptom-free settlings are transplanted to the field at 25 days after nursery growth. Roguing is also continued in the main field. At harvest, symptom-free stems are again collected and setts are screened through this nursery technique before transplanting to the field. REFERENCES Food and Agricultural Organization (FAO). 1994. Production Yearbook. Vol. 48 Jos, J.S., N. Hrishi, S.B. Maini and R.G..Nair. 1981. Two high yielding hybrids of cassava. J. Root Crops 6:1-6. Jos, J.S., K. Vijaya Bai and M.T. Sreekumari. 1987. Triploidy as a tool in cassava improvement. Proc. Nat. Symp. Production and Utilization of Trop. Tuber Crops. CTCRI, Thiruvananthapuram, India. pp. 7-19. Jos, J.S., S.G. Nair, S.N. Moorthy and R.B. Nair. 1990. Carotene enhancement in cassava. J. Root Crops: ISRC Nat. Symp. Special Edition: pp. 5-11. Magoon, M.L. 1967. Recent trends in cassava breeding in India. Proc. 1st Symp. Int. Soc. Trop. 110 Root Crops. Trinidad, West Indies, pp. 110-117. Magoon, M.L., J.S. Jos, K.N. Vasudevan and S.G. Nair. 1969. Cytomorphological studies on induced polyploids of cassava. Genet. Iberica 21:27-47. Magoon, M.L., S.G. Appan, R. Krishnan and R.C. Mandal. 1970. Some promising high yielding hybrids and selections of cassava. SABRAO Newsletter 2: 19-26. Moorthy, S.N., J.S. Jos, R.B. Nair and M.T. Sreekumari. 1990. Variability of/? carotene content in cassava germplasm. Food Chemistry 36:233-236. Nair, S.G. and Santha V. Pillai. 1995. Cassava breeding in Kerala. Proc. Seminar on Crop Breeding in Kerala. Dept. of Botany, University of Kerala. pp. 49-59. Oduro, K.A. 1981. Some characteristics of yellow-pigmented cassava. In: Tropical Root Crops: Research Strategies for the 1980s. Proc. IDRC. Ottawa, Canada. 163e. pp. 42-44. Rajendran, P.G. and C.S. Ravindran. 1993. Cassava cultivation through true seed propagation. J. Root Crops 19(2):81-88. Ramanathan, S., M. Anantharaman and T.K. Pal. 1989. Cassava in Kerala: A study on the present status and varietal distribution. J. Root Crops 15(2): 127-131. Ramanathan, S., M. Anantharaman and T.K. Pal. 1990. Cassava in Tamil Nadu: A study on the present status and varietal distribution. J. Root Crops 16(2): 83-86. Ill CASSAVA RESEARCH PROGRAM AT TAMIL NADU AGRICULTURAL UNIVERSITY (TNAU) IN INDIA S. Thamburaf ABSTRACT Cassava is cultivated in Tamil Nadu in an area of 85,983 ha with an annual production of about 2.5 million t of roots (1993-94). At the world level, Tamil Nadu ranks first in yield, with 28.7 t/ha, as compared to the Indian average of 22.8 and a world average of 9.6 t/ha. In Tamil Nadu cassava is grown mainly in Salem and Kanyakumari districts located in the dry western zone (60% irrigated and 40% rainfall) and the high rainfall southern zone (mostly rainfed), respectively. The area under cassava is increasing in other districts of Tamil Nadu as well. With respect to varieties, about 50% of the total area is grown with high-yielding varieties, like H-226 and MVD-1 in most areas of Salem district, and with Co-1, Co-2, Co-3 in different parts of Tamil Nadu. Locally adapted and traditional varieties, like Burma and Malabar, are still under cultivation, but they are low yielding and susceptible to cassava mosaic disease. The cassava germplasm collection has now increased to 480 accessions, including some CIAT clones. Two early maturing clones were identified, i.e., H-1 19 from CTCRI, Trivandrum, and Co-3 developed by Tamil Nadu Agr. Univ. in Coimbatore. The crop duration is 7 1/2 to 8 months with a high yield of 32 and 35 t/ha, respectively, and a starch content of about 30 % . For the high rainfall zone of Kanyakumari district, the local clone Adukkumuttan performs well, giving a 15 to 28% increase in yield compared to the released varieties. For coastal areas of South Arcot district, two clones, i.e. ME-46 and ME-10, were shown to be better, with a yield of 44 t/ha; these are now under further evaluation. Under the Asian cassava breeding network a total number of 85 hybrid progenies were received from CIAT/Colombia and the Thai-CIAT program. Preliminary evaluation has indicated that a few clones have high root yields of more than 5 kg/plant, high starch content of over 35%, field tolerance to mosaic virus and a shorter crop duration of 7 to 8 months. The optimum fertilizer rate for Co-3 cassava under irrigated conditions was found to be 60:60:120 kg N-P2O5-K:O/ha, which increased the yields by 36% over the control. For the recently developed short-duration clones, like H-1 19, CI-590and S-856, under irrigated conditions the optimum spacing and fertilizer requirements are 75 x 75 cm and 75:25:75kg N-P2O5-K2O/ha, respectively. Application of Azospirillum inoculum at 2 kg/ha as basal dressing with recommended fertilizers, combined with three foliar sprays of zinc sulfate (0.5%) and iron sulfate (1.0%) at 60, 75 and 90 days after planting, increased the root yield and starch content of Co-1 and Co-2 varieties. Under abundant water supply, cassava intercropped with groundnut can be irrigated at 0.6 IW/CPE ratio with an interval of 12 days and the application at 10 t/ha of coconut husk waste. Under limited water supply, irrigation may be reduced to 0.45 IW/CPE ratio with an interval of 16 days. INTRODUCTION In India, the major cassava growing states are Kerala and Tamil Nadu. In Tamil Nadu cassava is an industrial crop grown in an area of about 86 thousand ha (1994/95) with a production of 3.24 million tonnes of roots per year (Figure 1). Tamil Nadu ' Head, Horticultural College and Research Institution, TNAU, Coimbatore - 641 003, Tamil Nadu, India. 112 accounts for 33% of the area and 46% of cassava production in India. The area has increased considerably during the last ten years because of the industrial exploitation of the crop. In 1986/87 there were only 48 thousand ha with a production of 1.8 million tonnes. The crop has now acquired a status of one of the important commercial crops in the state. Tamil Nadu is blessed with varied soil and climatic conditions that are well suited for cassava growing. Already Tamil Nadu has the highest cassava yield, with a range from 14.76 to 60.33 t/ha and a mean of 37.72 t/ha. This could be attributed to the growing of high-yielding varieties released by CTCRI in Trivandrum and by Tamil Nadu Agricultural University, and the adoption of recommended crop production technologies, both under irrigated and rainfed conditions. The Indian average cassava yield is 22.87 t/ha, while it ranges from 5.1' to 14.4 t/ha in most other cassava growing countries. Tamil Nadu state is divided into seven agroclimatic zones and the major traditional cassava growing areas in the western part of the state are the districts of Salem, South Arcot Vallalar and Dharmapuri (60% irrigated and 40% rainfed) and the southern district of Kanyakumari (mostly rainfed) (Figure 2). There are about 800 sago and starch factories in and around Salem, South Arcot Vallalar and Dharmapuri districts which depend on cassava roots. The number of factories in Salem district alone is 650. It is estimated that 60% of the cassava starch produced in India is from Salem district. This district accounts for 44% of area and 47% of production of cassava in Tamil Nadu. Because of ease in cultivation, few pests and diseases problems, drought tolerance and the increase in root prices, the area under cassava is also increasing in other districts of Tamil Nadu, i.e. in Periyar, Tiruchirapalli and Coimbatore districts. Even though the area is increasing, the production is not sufficient to meet the demand of the factories, which operate only about six months of the year. The industries operate at only 50% of capacity. Hence, there is a need to increase the cassava area and production in Tamil Nadu. There is also scope for further increases in yield through the development of new varieties with high yield and higher starch content and the recommendation of zone- specific packages of practices, especially suited for rainfed culture of cassava. (mj/J) p|3iA e o © t 1 r (suo; uoiihui) uoipnpojj o -I r -—~—— Illllllllllllllll llllllllllllilllllllllllll- NON ON ON Os ON mmmmmmm o -i- o 00 pSO 1^ o o («M 000.) B3JV '_ ea 2■33 c o V- u 3 -a o a; u < %> § So Os I "3 | I e2 ! Q 1 t CO u CO "a to LO CD CD CO C3 co O a o u 3 O E °° CD O C/' CO o ^t 1 3 1 'I I (/) 115 RESEARCH ACCOMPLISHMENTS Earlier Research on Cassava As far as the varietal situation is concerned, about 50% of the total area is grown with high-yielding varieties like H-226, MVD-1 (in most areas of Salem district), and Co-1 and Co-2 and Co-3 in different parts of Tamil Nadu. The salient features of these varieties are shown in Table 1. Another improved variety MVD-1 (Mulluvadi 1) was released during 1983 by the State Department of Horticulture through clonal selection. It yields 34.5 t/ha in a duration of nine months. The roots contain 35% starch. It is a non-branching type and exhibits field tolerance to mosaic virus. Locally acclimatized and traditional varieties like "Burma" and "Malabar", are still under cultivation, but they are low-yielding and susceptible to CMD. The recommendations on agronomic requirements and utilization of cassava, based on earlier field experiments at TNAU are: 1) use of single-bud setts for rapid multiplication, 2) dipping the setts in a superphosphate slurry, 3) use of ridges and furrows (75 x 90 cm) for irrigated crop; bed system for rainfed crop (60 x 60 cm); and mound system (90 x 90 cm) for high-rainfall zones, 4) application of 25 t/ha of FYM with 50:50:100 kg/ha of N-P205-K20 for irrigated crop, 5) growing of intercrops like onion and blackgram, 6) use of growth hormones TIBA at 1000 ppm, Alar at 1000 ppm or bioregulator Ethrel at 500 ppm, 7) irrigating cassava at 60% moisture with a consumption of water of 1607 mm and witholding water one month before harvest, 8) allowing two shoots per sett, 9) adoption of no-tillage method with application of organic mulch of banana pseudostem or cumbu stalks, 10) pre-emergence herbicide application of pendemethalin at 1.0 kg a.i./ha, followed by two hand weedings, 1 1) utilizing cassava leaves for rearing silkworm, 12) using dry flour of cassava leaves in cattle or pig feed, 13) utilization of seed for extraction of edible oil, 14) use dry cassava chips or pellets as cattle feed, 15) development of recipes for Indian foods like pergath, bonda, stuffed parota, sweet balls, susiam, munthirirotu, idly, halva, idiappam, vermicelli, ready mixes of soji and bakery products, 16) protein enrichment of cassava vermicelli by the incorporation of green gram, bengal gram or soybean flour and 17) development of animal feed pelletization. Earlier work on mutation breeding indicated the possibility of inducing variability for higher yield, compact growth habit, early maturity, higher starch content and lower HCN. A protocol for in-vitro propagation of cassava has been developed. Research on the use of true cassava seed indicated that soaking the seeds in potassium nitrate solution (0.5%) for 24 hours prior to planting enhanced the germination as well as the seedling vigor. The performance of plants raised from true cassava seeds is being studied. 116 Table 1. Salient features of cassava varieties developed by Tamil Nadu Agric. University. Particulars Co-1 Co-2 Co-3 Year of release 1974 1984 1993 Root yield (t/ha) 30.11 37.65 42.58 Crop duration (days) 255-270 255-270 240 Starch content (%) 35.00 34.35 35.60 Taste Sweet Sweet Sweet Color of skin Brownish white Brownish white Dark brown Color of flesh White White White Incidence of mosaic disease Low Very low Very low Branching habit Non-branching Branching at later stage Branching Current Research on Cassava 1. Crop Improvement For any breeding program to succeed, the availability of diverse genetic material is important. With that in view, attempts were made to enrich the cassava collections. The germplasm bank at TNAU has presently the following number of accessions: Indigenous 115 From CTCRI, Trivandrum 30 From CIAT/Colombia and the Thai-CIAT program 525 Total 670 Polyploidy breeding has been reported to have many advantages in cassava. The two promising triploids developed at CTCRI, Trivandrum, were tested for two years at Coimbatore. The results have indicated that the triploid variety 2/14 was similar to the check Co-2 with respect to root yield, DM and starch content. Results of the triploids being tested under Salem conditions are shown in Table 2. Developing varieties with a shorter crop duration is another important objective in cassava breeding. Four short duration clones, i.e., H-1 19, CI-590, S-856 and H-5/78, were evaluated for three seasons. The results indicate that the clones H-1 19, CI-590, S- 856 and H-5/78 have a shorter crop duration of 7 1/2 to 8 months. Among these four, H-1 19 recorded a significantly higher average root yield of 36.43 t/ha. It is a non- branching clone. However, the clone H-1 19 was found to be susceptible to CMD under Coimbatore conditions (Table 3). Good performance of H-1 19 has also been reported by Nanda et al. (1996) under Madhya Pradesh conditions, a non-traditional area of cassava cultivation in India. 117 Table 2. Performance of triploid cassava varieties in Salem, Tamil Nadu, India. Root dry Root Root yield (t/ha) matter starch content (%) content Clones 1993 1994 Average (%) 76/9 (CP.4(2x) x S-300 (4x) 19.5 18.3 18.9 45.8 34.2 2/14 (CP.2 (2x) x H-2304 (4x) 28.6 25.3 27.0 48.2 35.6 H-1687 30.0 23.8 26.9 39.1 30.2 H-2304 18.5 20.6 19.6 41.6 31.5 M-4 22.4 24.5 23.5 38.1 30.0 Co-2 28.4 27.1 27.8 48.6 35.0 SED 0.887 0.957 0.922 CD (P=0.05) 2.14 2.310 2.230 Table 3. Performance of short-duration cassava clones in Coimbatore, Tamil Nadu, India. Root yield (t/ha) Crop Root duration (days) starch contentClones 1987/88 1988/89 1989/90 Average (%) HI 19 34.3 38.3 36.7 36.43 230 32.0 CI-590 20.4 28.7 26.9 25.33 240 33.0 S-856 29.5 30.7 30.7 32.96 230 30.8 H-5/78 13.9 27.1 30.6 23.06 240 30.6 SED 1.43 2.27 1.80 CD(P = 0.05) 4.02 6.35 3.78 Based on these results, the best two clones, H-119 and S-856, were tested in multi-locational trials in farmers' fields in Coimbatore using Co-2 as a check. Again, the clone H-119 had a significantly higher yield of 30.56 t/ha, a 13.0% increase over Co- 2. The root yield of S-856 was 27.80 t/ha, which was not significantly different from that of Co-2. However, with regard to quality, Co-2 registered better consumers' preference (80.2%) than H-l 19 (78.5%) and S-856 (78.0%). The mosaic incidence was 118 also higher in H-119 and S-856 (65.0 and 61.2%, respectively) as compared to that of Co-2 (5.2%). For the high-rainfall zone of Kanyakumari district, the local clone, Adukkumuttan performs well with 20-28% higher yields than the released varieties. For coastal areas of South Arcot Vallalar district, two clones, i.e., ME-46 and ME-10, were found to be the best with a mean root yield of 44 t/ha; they are now under further evaluation. Cassava is grown on a large-scale in Salem district of Tamil Nadu. Sofar, varieties developed at CTCRI, Trivandrum, and at TNAU, Coimbatore, have been introduced. However, there is a need to develop or identify varieties more suited to the conditions that exists in this altogether different environment. With this objective, a total of 13 clones were tested during 1995 in a farmer's field in Salem. Among the clones, S-1315 performed best with a root yield of 4.5 kg per plant, followed by CE-22 (3.5 kg) and H-2304 (3.0 kg). There was no virus incidence in the clones H-97, CI- 167, H-1 19, H-2304 and Co-2 under field conditions. There exists a possibility of introducing true cassava seeds for commercial cultivation in Tamil Nadu where CMD is noticed on a large scale. To elucit basic information, research was conducted at the Department of Seed Technology of Tamil Nadu Agricultural University. The seed maturation studies conducted in true cassava seed have indicated that seeds attained physiological maturity 60 days after flowering under Coimbatore conditions. The physical index of maturity was the color change from green to yellow. Delayed harvest resulted in the shattering of the seeds (Nepolian, 1995). Germination studies on true cassava seeds have brought out that soaking of seeds in 0.5% KN03 solution for 24 hours prior to planting resulted in 94% germination. Storage studies revealed that cassava seeds can be stored at 6.26% moisture for ten months under ambient conditions without a decrease in germination. Rajendran et al. (1993) reported that the seeds could be stored for four to six months with good germination, while the seed viability reduced thereafter under Kerala conditions. Sowing at 2 cm depth in soil with 50% water holding capacity was observed to be optimum (Nepolian, 1995). Studies on the physiology of root development in plants raised from true cassava seed is in progress at the Horticultural College and Research Institute, Tamil Nadu Agricultural University, Coimbatore. 2. Evaluation of CIAT Hybrid Clones Under the International Evaluation Network Programme, seeds of 109 cassava hybrids were received from CIAT/Colombia and from the Thai-CIAT program during 1993. The seeds were sown in polybags in the nursery and the seedling progenies were transplanted. Among 109 hybrids only 91 established in the field. A total number of 525 clonal progenies were forwarded to further clonal generations. From the seedling progenies, stakes were prepared and clonal progenies were evaluated for two clonal 119 generations during 1995 and 1996. The results on the variability of the progenies for morphological traits, root yield, harvest index, starch content and reaction to cassava mosaic disease are presented in Tables 4 and 5. Based on the performance of these clonal progenies, a total number of 20 and 38 clonal progenies were selected during 1995 and 1996, respectively. The data on root yield along with other traits are presented in Tables 6 and 7. The following 21 CIAT hybrid clones were selected for non-branching habit, which will be tested later for high density planting. The figures in parenthesis indicate the number of clones in each hybrid exhibiting non-branching habit. CM8487 (3), SM1450(2), SM1454(8), SM1520(2), SM1542(13), SM1557(1), SM1636 (1), SM1673 (2), SM1679 (2), SM1784 (8), SM1789 (10), SM1791 (4), SM1858 (1), SM1864 (3), SM1969 (3), OMR36-05 (12), SM1780 (2), SM1883 (3), SM1890 (3), SM1891 (6) and SM 8208 (4). Based on the performance of two seasons' clonal generations, the following 16 clones were selected for short-duration (early harvestability), ranging from 205 to 225 days. The figures in parenthesis indicate the clone number (ME) of each hybrid. SM1475 (359), SM1858 (578), SM1470 (356), SM1544 (426), SM1665 (476), SM1475 (358), SM1454 (323), SM1774 (521), SM1432 (310), SM1454 (321), SM1447 (314), SM1460 (349), SM1523 (398), OMR36-4-1 (633), SM1787 (245), and SMI670 (200) Since flowering is an important trait for breeding programs, the CIAT hybrid clones were also evaluated for their flowering habit. Among the clonal progenies, a total number of 33 and 48 clones flowered during 1995 and 1996, respectively. Out of these, the following 12 clones flowered uniformly during both years. The figures in parenthesis indicate the clone number (ME) of each hybrid. SM1437 (175), SM1765 (208), SM1787 (242), SM1788 (249), SM1789 (250), SM1789 (255), SM1447 (312), SM1519 (375), SM1521 (383), SM1564 (445), SMI789 (566) and SMI 858 (578) The third clonal generation is in the field. Based on the performance of the clonal progenies, a few selected ones will be forwarded for regional evaluation trials in order to identify promising ones and to release as a new variety suited for cultivation in Tamil Nadu. 3. Crop Production Research Nutrient management is an important aspect in cassava and studies on this aspect have provided useful results. Application of 10 tonnes of FYM + 60:75:100 kg/ha of N-P2O5-K20, 2 kg/ha of Azospirillum, combined with three foliar sprays of 0.5% zinc sulfate and 1 % ferrous sulfate (60, 75 and 90 days after planting) resulted in root yields of Co-1 and Co-2 varieties of 34.57 and 35.43 t/ha, respectively (Table 8). The starch content was also the highest (39.98%) in Co-2 in this treatment. There was no 120 significant difference in the HCN content of the root parenchyma. A quantity of 2 kg of Azospirillum inoculum is to be mixed with one liter of water, made into a slurry and the basal 1/3 of the stake is to be dipped for 20 min in the slurry before planting. Table 4. Range in plant characteristics of hybrid clones received from CIAT 1995 1996 Characters Range Mean CV (%) Range Mean CV (%) Plant height (cm) 121.3-291.6 210.75 13.1 118.8-296.2 205.86 9.1 Stem girth (cm) 2.0-11.6 7.21 15.2 2.1-17.2 8.06 14.8 Root length (cm) 6.8-45.2 18.18 14.0 6.1-61.0 23.06 16.2 Length of peduncle (cm) 0.0-4.6 1.20 8.5 0.0-5.9 1.36 9.6 Dry matter content (%) 18.2-37.5 26.52 16.2 17.5-40.2 29.22 17.2 Starch content (%) 21.0-36.8 31.20 10.6 22.0-37.2 30.51 11.2 HCN content (mg/kg) 12.2-40.3 22.56 11.2 14.5-41.6 23.19 10.4 Harvest index (%) 32.0-66.7 47.07 12.3 28.1-59.0 43.71 10.2 Fresh root yield (kg/plant) 1.3-3.5 2.41 19.9 1.2-4.9 2.27 28.5 Foliage weight (kg/plant) 1.3-5.1 2.63 26.2 1.6-5.2 2.77 26.7 From another study, the optimum fertilizer dose for the newly released cassava variety Co-3 under irrigated conditions was determined to be 60:60: 120 kg/ha of N-P2O5- K2O, which increased the root yield by 36% over the control. A trial was conducted for three seasons to assess the fertilizer and spacing requirements for the short-duration clones H-1 19, CI-590 and S-856. The results indicate that all three short-duration clones produced higher yields at a closer spacing of 75 x 75 cm and a lower dose of fertilizers of 75:25:75 kg/ha of N-P2O5-K20. Besides the root yield, the dry matter and harvest index were also increased. The interaction effects were significant, and irrespective of the fertilizer dose, closer spacing resulted in higher root yields. Nitrogen use efficiency by crops is of major interest in the tropics. Nitrogen is subject to leaching, denitrification and volatilization, which makes it unavailable to plants. Hence, studies were conducted under irrigated conditions on the use of nitrification inhibitors with nitrogenous fertilizers. The results of the trial on the use of slow-release fertilizers conducted for three seasons has indicated that the root yield in cassava was increased by all the four slow-release fertilizers (Table 9). However, a significantly higher average yield of 34.8 t/ha was obtained with the application of neem- coated urea; the next best treatment was sulfur-coated urea (32.3 t/ha). 121 Table 5. Evaluation during 1995 and 1996 of plant characteristics of cassava hybrid clones received from CIAT. Character Variations 1995 season (%) 1996 season (%) Vigor Very poor 2.80 2.50 Poor 9.21 8.00 Intermediate 50.50 55.80 Good 26.20 27.20 Very good 11.29 6.50 Flowering Nil 85.21 89.87 Few 2.50 2.12 Intermediate 8.18 5.21 Abundant 3.11 2.80 Plant height Dwarf (< 200 cm) 49.64 44.85 Medium (200-300cm) 46.04 49.47 Tall (> 300cm) 4.32 5.68 Stem girth Thin (<5cm) 5.60 55.47 Medium (5- 10cm) 84.20 85.05 Thick (> 10cm) 10.20 9.48 Branching habit Non-branching 69.20 68.14 Early-branching 18.50 20.04 Late-branching 12.30 11.82 Lodging None or light 40.80 42.55 Intermediate 55.20 52.25 Severe 4.00 5.20 Petiole color Pink 58.00 58.65 Green 37.50 36.71 Pink + Green 4.50 4.64 Emerging leaf color Green 61.20 61.18 Pink 38.80 38.82 Root length Short ( < 20cm) 36.20 39.08 Intermediate (20-30cm) 45.10 42.36 Long ( > 30cm) 18.70 18.76 Length of peduncle Nil 85.20 66.65 Short (< 3cm) 5.60 25.25 Intermediate (3-5cm) 5.26 5.04 Long (>5cm) 3.94 3.06 Root color White 10.00 10.08 Light brown 24.60 23.84 Dark brown 65.40 66.08 Flesh color White 69.00 69.15 Cream 28.90 28.66 Yellow 2.10 2.19 122 Table 5. (continued) Character Variations 1995 season (%) 1996 season (%) Root shape Conical 2.50 2.41 Conical cylindrical 38.00 38.51 Cylindrical 56.00 55.14 Irregular 3.50 3.94 Constrictions None or few 80.00 82.05 Intermediate 16.80 13.57 Several 3.20 4.38 Foliage and root Very good 45.00 42.80 evaluation Good 25.00 20.60 Regular 15.20 18.20 Bad 7.58 8.00 Very bad 7.22 10.40 Dry matter High (> 30%) 26.10 22.15 Medium (20-30%) 40.80 42.63 Low (<20%) 33.10 35.22 Starch content High (> 30%) 23.20 24.10 Mekium (25-30%) 52.10 53.26 Low (<25%) 24.70 22.64 HCN content High(>30mg/kg) 30.00 31.65 Medium (20-30mg/kg) 45.28 40.15 Low (<20mg/kg) 24.72 28.20 Harvest index Low (<0.40) 20.95 38.35 Medium (0.40-0.50) 54.36 30.40 High (> 0.50) 24.69 31.25 Fresh root yield <2 55.67 62.88 (kg/plant) 2-3 33.49 19.00 3-4 8.13 9.17 >4 2.71 8.95 Incidence of CMD 0 (no CMD) 2.60 2.56 (score) 1 0.50 0.24 2 1.21 1.73 3 7.90 8.64 4 34.50 31.75 5 53.29 55.08 123 Table 6. Yield data of selected CIAT hybrid cassava progenies" grown at TNAU in 1995. CIAT Clone Plant Root yield Foliage Harvest Starch hybrid No. No. height (kg/plant) weight index content (ME) (cm) (kg/plant) (%) OMR36-42 641 297 5.6 5.20 0.52 36.25 OMR36-05 671 288 4.4 8.30 0.35 35.20 OMR36-67 645 238 4.0 5.10 0.44 35.15 SM7564 656 304 4.9 4.00 0.55 36.00 CM8487 296 175 3.5 2.70 0.69 33.50 SMI 432 306 127 3.0 1.90 0.68 33.10 SM1432 310 231 3.2 3.30 0.49 34.10 SM1456 332 205 3.0 3.25 0.48 33.90 SM1460 347 270 3.2 3.20 0.50 33.90 SM1470 356 270 3.1 3.20 0.49 33.20 SM1673 397 151 3.1 2.25 0.58 33.30 SMI 789 564 185 3.7 3.20 0.53 35.30 SMI 855 572 210 3.1 5.20 0.37 34.30 SMI 858 578 245 4.0 3.00 0.57 36.80 OMR36-05 618 258 3.8 4.10 0.48 35.00 OMR36-05 625 259 3.9 2.60 0.60 35.12 OMR36-42 654 248 3.6 6.00 0.37 34.86 OMR36-42 637 266 3.5 4.20 0.45 34.90 OMR36-42 640 240 3.0 2.60 0.54 33.85 SM1564 657 240 3.2 4.10 0.44 33.90 "clonal progenies obtained from sexual seed received from CIAT in 1993. 124 Table 7. Yield data of selected CIAT hybrid cassava progenies" grown at TNAU in 1996. CIAT Clone Plant Root yield Foliage Harvest Starch hybrid No. No. height (kg/plant) weight index content (ME) (cm) (kg/plant) (%) SM1444 183 270 6.0 3.65 0.62 35.00 SM1414 186 350 7.3 9.00 0.45 35.62 SM1432 302 217 4.1 4.60 0.47 34.80 SM1432 310 250 5.2 6.20 0.46 36.10 SMI 447 314 267 5.1 7.50 0.40 35.50 SM1454 321 255 5.2 4.20 0.55 35.90 SM1454 343 205 5.5 3.40 0.55 35.96 SM1460 346 225 5.1 4.85 0.46 35.90 SMI 460 344 256 5.1 2.70 0.57 30.90 SMI 470 356 226 7.0 6.00 0.54 35.50 SM1475 358 266 5.6 7.50 0.57 36.20 SM1475 359 323 9.0 10.70 0.46 36.80 SM1521 384 185 4.3 5.70 0.43 34.80 SM1523 398 240 5.5 4.30 0.56 35.90 SMI 544 426 195 5.8 1.80 0.76 35.30 SM1545 430 215 5.0 7.00 0.42 36.20 SM1573 464 110 4.0 2.50 0.62 34.80 SM1665 476 225 5.8 4.50 0.56 35.20 SM1670 200 168 4.9 3.40 0.59 35.60 SM1670 201 240 4.1 2.00 0.67 34.60 SM1718 218 260 4.3 5.60 0.43 35.00 SM1774 269 285 9.8 5.20 0.65 36.40 SMI 774 521 204 5.5 4.80 0.53 35.00 SM1783 222 228 4.8 4.70 0.51 35.50 SM1784 539 295 4.5 4.00 0.53 34.90 SM1784 541 180 4.0 0.85 0.82 34.00 SMI 784 551 220 4.6 4.70 0.50 35.00 SM1787 240 277 4.0 4.50 0.41 34.00 SM1787 243 200 4.2 3.70 0.53 34.25 SM1787 245 240 4.8 3.70 0.36 35.00 SM1787 256 295 4.1 5.00 0.45 34.60 SM1789 251 290 4.1 5.20 0.44 34.60 SMI 855 573 220 4.6 2.20 0.68 34.80 SM1858 578 245 8.5 5.60 0.60 36.20 SM1883 265 267 8.0 6.00 0.57 36.00 SMI 890 270 280 7.0 3.90 0.64 35.00 SM1891 278 268 4.1 4.90 0.46 34.82 OMR36-41 635 178 5.0 2.80 0.63 35.10 "clonal progenies obtained from sexual seed received from CIAT in 1993. 125 Table 8. Effect of micronutrients (Zn and Fe) and biofertilizer (Azospirillum) on the yield and starch content of two cassava cultivars grown at TNAU. Fresh root yield (t/ha) Starch content (%) Treatments Co-1 Co-2 Co-1 Co-2 Control (fertilizers only) 16.50 22.40 32.60 34.14 Fertilizers + Zn + ¥e+ Azospirillum 34.57 35.43 35.16 39.98 Earlier studies on intercropping indicated that Bellary onion (Allium cepa var. cepa) and blackgram were suitable intercrops for cassava. In a separate study by Mohammed Yassin (1995) conducted for two years on the N management and intercropping in cassava, it was found that growing vegetable cowpea or sunhemp as intercrops and incorporating the haulms into the soil increased the root yield of cassava from 18.87 to 45.0 t/ha in the first experiment, and from 13.45 to 35.63 t/ha in the second experiment. The cost benefit ratio was higher, i.e. 1:3.8 and 1:2.8 during the first and second years, respectively, as compared to monocropped cassava. The starch and DM content of the root was also increased. The effect was more pronounced when neem-coated urea or prilled urea + Azospirillum at the rate 2 kg/ha were applied. Ayyaswamy (1994) conducted experiments on irrigation scheduling and inter cropping systems in cassava. The results have indicated that under abundant water supply, cassava grown as an intercrop in groundnut and irrigated at 0.6 IW/CPE ratio with an interval of 12 days and with coconut fibre waste (as soil amendment) applied at 10 t/ha increased the root yield and net returns per hectare. Under limited water supply, the same treatment with a change in irrigation scheduling at 0.45 IW/CPE ratio, applied at an interval of 16 days, was found optimum. There was improvement in both dry matter and starch content of the roots. 4. Post Harvest Technology Research programs are in progress at Tamil Nadu Agricultural University on the processing of cassava roots. Susheela Thirumaran and Aruna Seralathan (1996) conducted experiments on the production of cassava-based defatted soyflour noodles. The results showed that the noodles had desirable properties, such as a more than six months shelf life and significantly high acceptability scores, while providing 340 kilo-calories of energy and having 16% protein. They have suggested that extrusion of nutritious noodles can be initiated as a commercial venture for school lunch programs. 126 Table 9. Effect of the application of slow-release nitrogen fertilizers on the yield of cassava grown in TNAU. Root yield (t/ha) Treatments 1992/93 1993/94 1994/95 Average Prilled urea 30.3 31.4 33.1 31.6 Urea super-granule 29.6 30.8 31.2 30.5 Neem-coated urea 33.5 34.2 36.7 34.8 Sulfur-coated urea 30.5 32.7 33.6 32.3 Tar-coated urea 28.8 27.6 30.8 29.1 Control 24.7 26.1 27.9 26.2 SED 0.688 0.456 0.513 CD(P= 0.05) 1.480 0.970 1.262 At Tamil Nadu Agricultural University, the College of Agricultural Engineering is involved in the development of machinery for cassava harvesting and processing. A root puller (harvester), chipper, peeler, mechanical stirrer for starch settling tanks, sago roaster and pelletizer are some of the pieces of equipment developed for use in small- scale industries. THRUST AREAS AND SUGGESTIONS 1 . Cassava in Tamil Nadu is affected by mosaic virus disease as a result of which the yield is considerably reduced. The incidence is greater in Salem, South Arcot Vallalar and Dharmapuri districts. Resistant varieties are to be developed. It is observed that there are variations in the susceptibility of clones to cassava mosaic disease in different agro-climatic zones. Some of the clones susceptible at Coimbatore are free from CMD at Salem or Kanyakumari in Tamil Nadu. It is suspected that there may be strains or races of the virus which needs to be investigated. 2. In Salem district, there are location-specific problems in cassava fields, like twisting, knotting and splitting of roots, which affect the root yield and industrial quality. The reasons are to be investigated and suitable remedial measures suggested. 3. Occurrence of root rot caused by Sclerotium rolfsii has recently been noticed in some locations of Salem district and resistant varieties and suitable management practices are to be developed. 4. In most of the locations cassava is grown as a rainfed crop. Unlike Kerala, the annual amount of rainfall is only 800-900 mm. There is a need to develop varieties adapted to rainfed cultivation and a package of cultural practices are to be developed. 5. In some parts of Salem and South Arcot Vallalar districts, cassava is grown in hilly areas at an altitude of 500 masl. There is a need to develop varieties for these zones along with agronomic practices. 6. Much emphasis is now given for production of organic foods. There is a need to 127 intensify research on organic gardening with special reference to the use of organic 8- manure, biofertilizers and biocontrol of pests and diseases. 7. Crop improvement programs are therefore to be planned with the objective of developing varieties with shorter crop duration (< 7 months), high root yield (> 35 t/ha), high starch content in roots (>35%) and resistance to biotic (mosaic virus, root rot, tip drying, leafspot and mites), and abiotic (drought and salt) stresses. Due emphasis is to be given to developing crop production technologies suited to varied agroclimatic zones of Tamil Nadu. It is also necessary to develop technologies for the manufacture of value- added products utilizing cassava as the raw material. 8. The major reason for slow adoption rate of improved varieties is the lack of planting material. There is a need for the development of large-scale rapid multiplication of elite clones including in-vitro propagation. 9. It is estimated that 60% of the starch produced in India is from cassava roots and the entire quantity is used for domestic consumption. At present the machinery used in the sago and starch factories is old, which results in lower productivity and it takes a longer time for extraction; these need to be modernized. Further, starch produced in these factories do not reach the standards for export. The starch recovery is only 60% and 40% of the starch left in the fibrous waste material is used in cattle feed. 10. There exists the possibility of manufacturing starch and animal feed pellets for export purpose. In Dharmapuri district, one fully-automated factory is being established for the manufacture of starch, exclusively for export. A few factories have been set up for manufacture of other value-added products, like industrial alcohol, sorbitol, liquid glucose, dextrin, high-frutose syrup, glue, cattle feed etc. This is a good beginning. There is need to develop varieties suitable for these various end-products and optimum growing and processing technologies. 1 1 . The major constraint in the cassava industry is the high degree of price fluctuation, both for fresh roots and for products like starch and sago (Figures 3 and 4). There is a need to stabilize the market prices of roots and cassava-based products. 12. The starch and sago factories require a large quantity of water for starch extraction and the effluent discharge pose a health hazard to the people and animals, as well as for the environment. Hence there is a need to modernize factories and to develop new low- cost effluent treatment technologies. 00 8>J06 jo SBq/sfl ui aouj 1 ON s 2 I I | I I |JBlAI 130 REFERENCES Anonymous. 1995. Horticultural Statistics. Departm. of Horticulture, Government of Tamil Nadu. Madras, Tamil Nadu, India. Ayyasamy, M. 1994. Irrigation scheduling for cassava comrade cropping system. PhD thesis. Dept. Agronomy, TNAU, Tamil Nadu, India. Mohammed Yassin. 1995. Studies on intercropping and nitrogen management in cassava. PhD thesis. Dept. Agronomy, TNAU, Tamil Nadu, India. Nanda, H.C., S.S. Rao and B.S. Chandrasekar. 1996. Stability analysis in short duration cultivafs of cassava (Manihot esculenta Crantz). Indian J. Agric. Sc. 66(3): 165-169. Nepolean, K. 1995. Studies on true seeds in cassava (Manihot esculenta Crantz) MSc (Ag) thesis. Seed Technology, TNAU, Tamil Nadu, India. Rajendran, P.G., S.G. Nair, C.S. Easwari Amma, K. Vasudevan and M.T. Shreekumari. 1993. Recent progress in cassava varietal improvement in India. In: R. Howeler (Ed.). Cassava Breeding, Agronomy Research and Transfer of Technology. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov. 2-6, 1993. pp. 61-83. Susheela Thirumaran, A. and M. Aruna Seralathan. 1996. Process of cassava-based defatted soy flower noodles suitable for school lunch programmes. Proc. 2nd Int. Soybean Processing and Utilization Conference. Bangkok, Thailand. Jan 8-13, 1996. pp. 257-261. 131 RECENT PROGRESS IN CASSAVA VARIETAL IMPROVEMENT IN THE PHILIPPINES Algerico M. Mariscal and Reynaldo V. Bergantin1 ABSTRACT In recent years cassava has become an important raw material for various domestic needs (food, feed, alcohol and high fructose sugar). To cope with high demand for cassava roots the breeding and selection program has been intensified to identify varieties that are high yielding and having high starch and dry matter contents. The continuous introduction of improved cassava populations from the Thai-CIAT program and from CIAT/Colombia contributed to the identification of good varieties, and, consequently, the establishment of cassava genepools having a wide genetic base. This has led to the establishment of a breeding program using the best parents of local and foreign sources. Progress in selection for the last three years was observed but was not phenomenal. In fact, from the regional trials jointly conducted by PRCRTC and its cooperating stations, two new varieties of cassava from CIAT/Colombia selected by the University of the Philippines at Los Banos, were released by the National Seed Industry Council. These are CM34 19-2A as PSB CV- 1 1 and SM972-20 as PSB CV-12. Both are intended for food and feed because of their low levels of HCN. Using the popular variety Lakan as check for the selection and evaluation among introduced materials from CIAT/Colombia and Thailand, gradual progress is made in yield and dry matter content. Superior varieties have been identified but are still in the pipeline for further testing. Varietal dissemination of new varieties has been intensified through on-farm trials and demonstration farms in strategic areas where farmers have the opportunity to observe and select. The involvement of the private sector, specifically the San Miguel Corporation, in the creation of high demand for cassava has paved the way for the rapid increase of production in terms of area and varietal dissemination. For the last two years the Agribusiness Division of San Miguel Corporation has supported 3,000 ha of cassava in Mindanao. Support given is in terms of planting material, land preparation and fertilizer input. All produce will go to San Miguel with a price profitable to the farmers. The projected cassava area needed to meet the needs of San Miguel Corporation is 55,000 ha. To this effect, variety trials are conducted in farmers. fields using recommended varieties (Rayong 60, Rayong 90, Rayong 5, KU-50 and the VC series) in Mindanao. Furthermore, rapid propagation is planned to back-up the need for more planting material. Aside from the San Miguel Corporation, starch millers throughout the country continue to expand their production area due to the high demand for starch. High HCN varieties are preferable. VC-5, which is high in HCN, is now planted in more than 3,000 ha in Lanao, while more than 1,000 ha of Lakan are planted in Negros Occidental to support Unistarch. Further selection will focus on identification of superior varieties with high starch content and high yield while having low to high levels of HCN. INTRODUCTION In recent years, domestic utilization of cassava in the Philippines has markedly increased. This is basically because cassava is recognized as a cheap raw material for 1 Cassava Breeders, Philippine Root Crop Research and Training Center (PRCRTC), ViSCA, Baybay, Leyte, Philippines. 132 production of food, feed and other industrial products. This recent development has finally realized the projection of the 1984 Asian Cassava Workshop in Thailand about the future potential of cassava in Asia. In fact, Walters (1983), Lynam (1986) and Singh (1986), emphasized that cassava will play a major role in satisfying the domestic needs of the country and that any future increase in the output used by the cassava producing countries in Asia should be aimed primarily at their domestic markets, such as for 1. animal feed; 2. starch production (food processing, textiles, paper and board, modified sugar); and 3. manufacture of ethanol. True to the projection 12 years ago, cassava became an important raw material for production of animal feeds, starch and its derivatives, as well as alcohol. The demand for cassava roots is hastened by the involvement of the giant business conglomerate San Miguel Corporation, which needs around 55,000 ha of cassava to supply their various needs for feed, modified starch and alcohol (Bacusmo, personal communication). The overall effect is the massive expansion of cassava production in terms of area and yield. Superior varieties are, therefore, needed to satisfy the existing demand. In support of this development, the cassava varietal improvement program of the Philippine Root Crops Research and Training Center.(PRCRTC), and that at the Institute of Plant Breeding (IPB) of the Univ. of the Philippines at Los Banos, will continue to keep pace with the needs of the cassava industry. The introduction of elite materials from the Thai-CIAT program and CIAT/Colombia has greatly improved the selection and identification of best varieties for outright release and for utilization as cross parents in the breeding program. VARIETAL IMPROVEMENT Breeding Objectives The breeding objectives for cassava in the Philippines have not changed since 1987 (Mariscal and Bacusmo, 1995). Generally, the breeding activities aim to satisfy the needs of the cassava farmers and processors who grow cassava in diverse agro- climatic conditions and utilize the storage roots in a variety of ways. Specifically, the breeding objectives are as follows: 1. High yield 2. High dry matter and starch content 3. Early harvestability 4. Resistance to pests and diseases 5. Tolerance to environmental stresses 6. Good plant type (root formation, root shape and branching habit) 7. Level of HCN (low for staple food, high for starch millers) Recent Developments (1993-1996) 1. Germplasm collection 133 To back up the cassava improvement activities at PRCRTC, a germplasm collection with wide genetic base is needed. Thus, PRCRTC maintained and upgraded the genetic stocks of cassava. To date PRCRTC maintains in the field 300 cassava accessions. These include 53 elite materials selected from Advanced Yield Trials as well as from tissue culture materials introduced from Thailand (Table 1). With support from the Cassava Biotechnology Network (CBN) of CIAT, germplasm materials from PRCRTC will be sent to CIAT Headquarters in Colombia for global maintenance. 2. Advances in selection Since the later part of 1993, 14,962 cassava hybrid seeds coming from 246 crosses were introduced to PRCRTC (Table 2). A total of 4,405 hybrid seedlings have been evaluated. Most of these hybrid seedlings came from the seed introduction from CIAT/Colombia and the Thai-CIAT program at Rayong. Summaries of the number of genotypes at different testing stages are shown in Table 3. Since 1993, all trials conducted included materials from both Thailand and Colombia, with Lakan, a variety with a very stable yield and adaptability, used as check variety. From the replicated trials of all entries from the Thai-CIAT program, specifically the MT series, there was no progress in terms of yield and harvest index. However, it seems that there is positive improvement in terms of dry matter content (Figure 1). This shows that the check variety Lakan is actually quite good. Table 1. Cassava germplasm collection at PRCRTC, ViSCA, Baybay, Leyte, Philippines, 1993-1996. Source 1993/94 No. of Accesions 1994/95 1995/96 Local 86 86 Foreign 131 131 Tissue culture (Thailand) 30 30 Elite clones 26 41 Total 273 288 86 131 30 53 300 134 Table 2. CIAT cassava hybrid seeds supplied to PRCRTC, ViSCA, Baybay, Leyte. 1993-1996. Number of Number of Date seeds crosses Sources March 1993 2,361 35 CIAT/Colombia June 1993 1,050 29 Thai-CIAT program July 1994 1,250 26 Thai-CIAT program July 1994 2,038 35 CIAT/Colombia March 1995 2,043 35 CIAT/Colombia May 1995 1,190 17 Thai-CIAT program January 1996 2,230 31 CIAT/Colombia April 1996 1,350 19 Thai-CIAT program July 1996 1,450 19 Thai-CIAT program Total 14,962 246 Table 3. Number of entries in different stages of cassava selection by PRCRTC, ViSCA, Baybay, Leyte, Philippines. 1993-1996. Year Selection stage 1993/94 1994/95 1995/96 Total Observational Trial (Fl seedlings) 1,193 1,850 1,362 4,405 Single-row Trial 392 695 410 1,497 Preliminary Yield Trial 72 136 53 261 General Yield Trial 25 73 22 120 Advanced Yield Trial 19 20 23 62 Regional Trial 10 10 10 10 Varietal Release 2 1 1 4 Promotional Trial - - 2 2 On-farm Trial/Verification 4 5 5 14 Total 1,717 2,789 1,886 6,375 Considering the OMR series from Thailand, a similar trend as that of the MT series was observed. Not much improvement was observed, except for dry matter content, which was similar to that of the check (Figure 2). These results are also shown 135 in Tables 7 and 8. From the MT series, no entry was able to surpass the check variety in terms of yield. Lakan had a yield of 38.0 t/ha, while the highest yield among entries was only 34.5 t/ha (MT4-111). However, with respect to dry matter, several entries surpassed the check. Some of these entries are MT4-52 and MT4-251 with dry matter contents of 37.4 and 37.0%, respectively (Table 4). The check variety had only 34.0% dry matter. From the OMR series (Table 5) only one entry, CMR33-13-14, produced a high yield of 60.4 t/ha, which surpassed the check. Nevertheless, many of the entries had high dry matter contents that are comparable to, or far better, than that of the check variety. These are CMR33-13-14, OMR33-12-3, OMR33-12-7 and OMR33-15-3 with dry matter contents of 34.9, 35.8, 36.2 and 35.2%, respectively. Lakan had 34.9% dry matter. Table 4. Yield parameters of six cassava hybrids from the Thai-CIAT program evaluated in the Advanced Yield Trial conducted at PRCRTC in 1994/95. Entry Fresh root Dry matter Starch Harvest yield content content index (t/ha)" (%) (%) MT12-44 MT4-111 MT4-52 MT9-21 19.0 34.5 23.7 26.5 34.6 32.9 37.4 33.2 22.5 20.2 26.4 20.6 0.46 0.56 0.46 0.62 MT4-251 MT 15-32 Lakan(check) VC-4(check) Average 14.6 26.5 38.0 25.2 24.1 37.0 25.8 0.59 35.9 24.4 0.50 34.0 21.7 0.55 31.1 17.6 0.60 35.2 23.3 0.52 " Average over 4 replications Source: PRCRTC, 1995 E ^ c p .2 £ oJ o a 3 Q. i- O U °- § 60 .S 6^ Lakan (Check) 92-93 93-94 94-95 Figure 1. Change ofmean frequency of all entries in Replicated Yield trials in terms of fresh root yield, dry matter content and harvest index of materials (MT series) from the Thai-CIATprogram. s § 3 O o Sa- > 60 ° 41*^ —- o u s 5 c m 140- 130 120 110 100 90 80 70 Lakan (Check) ~• root dry matter ~~A harvest index fresh root yield 93-94 94-95 95-96 Figure 3. Change ofmean frequency of all entries in Replicated Yield Trials in terms of fresh root yield, dry matter content and harvest index of materialsfrom CIAT/Colombia. a 120oo > 110 c £ 100 S u 90 B 1 80 70 harvest index 93-94 94-95 95-96 Figure 4. Change ofmean frequency of all entries in Replicated Yield Trials in terms of fresh root yield, dry matter content and harvest index oftissue culture- derived materialsfrom CIAT/Colombia. 140 Table 7. Yield parameters of ten cassava hybrids from the Thai-CIAT program introduced through tissue culture, and evaluated in the Advanced Yield Trial at PRCRTC in 1994/95. Fresh root Dry matter Starch Harvest Entry yield content content index (t/ha)" (%) (%) CM3299-4 29.2 36.1 24.5 0.38 CM4054-40 24.7 36.5 25.1 0.60 MCol 2215 23.5 37.2 26.1 0.56 CM2772-3 27.3 37.4 26.4 0.52 Rayong 60 36.8 36.5 24.2 0.63 CM4231-32 28.3 32.9 20.1 0.58 CM3306-4 25.8 38.2 27.5 0.48 MCol 1505 33.8 36.5 25.2 0.48 MMal 2 50.0 35.5 23.7 0.46 CMR29-56-101 15.4 33.7 21.2 0.49 Lakan(check) 36.5 38.4 27.8 0.49 VC-4(check) 40.8 34.9 22.9 0.62 Average 29.5 36.0 24.5 0.56 "Average over 4 replications Source: PRCRTC, 1995. 3. Utilization of CIAT germplasm for local hybridization Aside from screening and field evaluation of introduced hybrid seeds from CIAT, the PRCRTC utilized elite clones for its hybridization work to improve the local cultivars. A hybridization block was established and selected parents of local and foreign origin were planted for subsequent crossing. Open pollinated seeds in the germplasm collections were gathered for possible screening and evaluation. Controlled crossing work is still to be undertaken upon flowering of parental lines. From the hybrid population previously developed, five promising local hybrids are included in the Regional Trial. Potential varieties for recommendation are the following: CMP62-15 and CMP21-15, each with average yields of 24.0 t/ha (Table 8). These two possible varieties outyielded the two check varieties. 141 Table 8. Average cassava root yields (t/ha) in Regional Trials conducted in eight locations in the Philippines from 1990 to 1995/96. Year General averageEntry 1990-93 1993/94 1994/95 1995/96 1. CMP21-6 . 17.5 25.3 - 21.4 2. CMP62-15 - 20.6 22.4 20.3 24.0 3. CMP32-10 - 14.9 31.2 19.2 21.8 4. CMP26-1 - 21.3 22.7 - 22.0 5. CMP21-15 - 24.9 26.1 21.0 24.0 6. VC-2(check) 26.4 17.7 26.1 17.8 22.0 7. CM3419-2A** 27.4 19.7 24.9 - 24.0 8. CM3283-4 24.8 14.5 21.8 - 20.4 9. CM3422-1 25.9 12.6 25.4 28.9 23.2 10. SM928-1 - 20.3 19.9 25.0 21.7 11. SM972-20** - 13.6 39.1 24.1 25.6 12. UPLCa-5(check) - 14.6 19.9 19.7 18.1 13. SM577-8* - - - 16.2 16.2 14. SM466-6* - - - 32.3 32.3 15. CG90-4-1* - - - 32.4 32.4 16. CG87-3-11* - - - 29.0 29.0 ** Approved by the National Seed Industry Council for release * New entry for Regional Trial 4. Evaluation for tolerance to shade and red spider mites Considering that large areas intended for cassava in Mindanao and the Visayas are presently under coconut, selection of cassava genotypes grown under this condition was undertaken to identify cassava genotypes that will tolerate partial shading. Results of the General Yield Yrial involving the introduced materials from the Thai-CIAT program are very promising (Table 9). Several hybrids had a superior performance under coconut compared to the check variety Lakan. Some of the superior clones are CMR33- 13-11, OMR33-12-3, OMR33-12-7 and OMR33-05-2 having average yields of 24.2, 20. 1 , 20.0 and 17. 1 t/ha, respectively. Aside from yield, the CIAT materials have high dry matter contents that are comparable and even higher than that of the check variety. It is expected that from this population improved and superior genotypes suited for growing under coconut will be identified. The screening for tolerance to red spider mites involving locally developed hybrids tested in the Advanced Yield Trial has resulted in the identification of two clones that outyielded the check variety (Table 10). These clones are MOP24-2-13 and MOP24-2-40 with average yields of 12.5 and 12.7 t/ha, respectively. Generally, all 142 selections have shown field resistance to mites. However, the yields obtained were not high because the plants were affected by a strong typhoon during their critical growth period. Several trials, both for shade and mite tolerance, are still on-going. 5. Varietal release Regional trials conducted during the period 1993-1996 have resulted in the release of two new cassava varieties by the National Seed Industry Council (NSIC), formerly known as the Philippine Seed Board (PSB). The two varieties of CIAT origin are CM3419-2A released as PSB CV-1 1, and SM972-20 released as PSB CV-12. These two new varieties were selected by the breeding program of the Institute of Plant Breeding at the University of the Philippines at Los Bafios from the materials introduced from CIAT/Colombia. Both varieties have a low level of HCN and can be used both for human food and feed. Tables 8, II and 12 show the performance of these cultivars in the Regional Trial. It is interesting to note that some progress has been made in the selection for high yield among high HCN entries. However, the selection for high dry matter content among these high HCN varieties has not yet produced positive results (Figure 5). On the other hand, for low HCN entries (such as the two released varieties), remarkable results have been obtained in the selection for high yield, and some progress has been made in selecting for high dry matter content (Figure 6). 6. Varietal dissemination Nowadays, the quest for planting materials of high-yielding cassava varieties continues to rise. Expansion of production areas in Mindanao, specifically in Lanao and Bukidnon provinces is the outcome of the San Miguel Corporation.s Agribusiness Divisions' campaign for production of cassava to satisfy their current demand. No less than 55,000 ha are needed by San Miguel to supply their needs for production of animal feeds, modified starch and alcohol. For the last two years, this company has supported 3,000 ha of cassava. Support given was in terms of planting material, land preparation and fertilizer inputs. Farmers are made to sign a contract with San Miguel stipulating that all their produce will be absorbed by the company at an assured market price. Lately San Miguel is buying cassava dried chips at P2.50 per kilo. For initial expansion, the readily available Golden Yellow variety was used by the farmers. The need for cassava raw materials was also felt by the starch processors and other feed millers in Mindanao. As such, cassava became a high demand commodity. 143 Table 9. Yield parameters of ten cassava hybrids from the Thai-CIAT program evaluated in a General Yield Trial under coconut at PRCRTC in 1994/95. Fresh root Dry matter Harvest HCN Entry yield (t/ha)" content index rating2' CMR33-13-8 9.3 34.3 0.48 6 CMR33-13-10 13.8 31.7 0.53 6 CMR33-13-11 24.2 35.8 0.54 7 CMR33-13-14 14.3 34.5 0.52 5 OMR33-12-3 20.1 36.6 0.61 7 OMR33-12-7 20.0 37.4 0.60 6 OMR33-11-7 15.3 34.8 0.66 6 OMR33-62-9 14.1 34.9 0.63 6 OMR33-05-1 16.2 36.2 0.62 6 OMR33-05-2 17.1 35.8 0.60 7 Lakan(check) 16.3 35.8 0.49 5 VC-3(check) 16.4 32.6 0.54 6 Average 16.4 35.2 0.58 "Average over 3 replications :.Based on picrate test rating scale of 1 to 9: 1 = low, 9 = high Source: PRCRTC, 1996. 144 c z — U> -J £ 140 130 120 110 100 90 80 70 + 93-94 fresh root yield UPL Ca-5 (Control) root dry matter + + 94-95 95-96 Figure 5. Change in mean frequency of all entries in Regional Yield Trials in terms of root yield and dry matter content of high-HCN cassava accessions. fresh root yield = z■-> t_ > z %J £ c UJ 93-94 94-95 95-96 Figure 6. Change in mean frequency of all entries in Regional Yield Trials in terms of fresh root yield and dry matter content oflow-HCN cassava accessions. 145 Table 10. Yield parameters of seven locally developed hybrids for mite resistance evaluated in the Advanced Yield Trial at PRCRTC in 1995/96. Fresh root Dry matter Harvest Mite Entry yield (t/ha)" content (%) index reaction2' MOP24-2-34 9.3 36.2 0.42 R MOP24-2-13 12.5 35.1 0.37 R MOP24-2-19 9.0 35.1 0.42 R MOP24-2-40 12.7 32.7 0.45 R MOP24-2-8 7.4 39.4 0.48 R MOP24-2-46 7.2 32.5 0.38 R MOP24-2-4 8.4 38.1 0.49 R Lakan(Check) 12.4 39.6 0.49 R Avarage 9.8 35.6 0.38 "Average over 4 replications ^R means resistant Source: PRCRTC 1996 As far as varietal preference is concerned, feed millers prefer low-HCN varieties, while starch millers prefer high-HCN varieties. In effect, VC-5, which has high HCN is now planted in more than 3,000 ha in Lanao for starch production. Lakan, on the other hand, is now planted in more than 1,000 ha in Negros Occidental for starch production by Unistarch. Similarly, Lakan is also widely grown in Bohol to support Philstarch. With this development, farmers clamor for planting material of new high- yielding varieties. In their support, PRCRTC has undertaken several activities as follows: Promotional trials These trials are conducted in areas where cassava is a dominant crop. Farmers like to see for themselves the performance of new varieties before adopting them. Thus, promotional trials enable farmers to select which varieties they wish to plant. A farmer in southern Leyte who supports a cooperative cassava-based feedmill- project participated in a promotional trial, which was established using four recommended varieties planted on big plots on sloping land. Vetiver grass was planted as contour hedgerows to reduce erosion. Results of the trial, which was jointly observed by farmers 146 and researchers, were quite good (Table 13). VC-1 gave the highest yield of 34.0 t/ha. Farmers liked all of the varieties tested. In fact, all stakes were subsequently planted, resulting in a 40 ha expansion for the varieties tested. Presently, an On-farm Trial including several varieties is being conducted at Baloi, Lanao del Norte in cooperation with San Miguel Corporation. The varieties planted are: Rayong 60, Rayong 90, Rayong 5, KU-50, VC-1, VC-2, VC-3, VC-4, VC-5 and Lakan. Establishment (4 Model Farms and Nurseries To facilitate further distribution of new cassava varieties, PRCRTC initiated the establishment of one-hectare model farms in strategic locations in Mindanao where the bulk of cassava production is located. The model farms all intended to showcase the recommended varieties plus the necessary cultural management practices. These will also serve as the source of planting material for the farmers in the area. One model farm in Bukidnon, which was jointly supported by PRCRTC and San Miguel Corporation, produced yields of 40 t/ha. Table 11. Summary data of the newly released cassava line CM 3419-2 and the check variety VC-2" Parameters CM3419-2 as PSB CV-1 1 Check VC-2 25.0 32.1 19.0 low MR HR HR "Average over 14 trials in 6 locations, 1991-1995. Average fresh root yield (t/ha) 25.5 Dry matter content(%) 30.2 Starch content(%) 16.4 HCN content low Reaction to pests and diseases: a. Red spider mite HR b. CBB HR c. Scale insects HR 147 Table 12. Summary data of the newly released cassava line SM 972-20 and the check variety VC-2" Parameters Average fresh root yield (t/ha) Dry matter content(%) Starch content(%) HCN content Reaction to pests and diseases: a. Red spider mite b. CBB c. Scale insects "Average over 11 trials in 6 locations, 1993-1996. SM972-20as Check PSBCV-12 VC-2 23.0 21.4 33.9 32.1 21.5 19.6 low low HR MR HR HR MR HR Table 13. Resuls" of On-farm Trial of four cassava varieties planted on sloping land in Bontoc, southern Leyte, Philippines, in 1993/94. Fresh root Dry matter HCN Entry yield (t/ha) content (%)rating2> VC-1 34.0 33.26 VC-2 19.6 30.33 VC-3 25.6 27.03 Lakan 26.1 34.32 Average 26.3 31.2 "Average over 4 replications, plot size 60 sq. meters 2>Based on picrate test rating scale of 1 to 9 : 1 =low, 9 = high Source: PRCRTC, 1994. 148 Aside from the model farms, nurseries of the new cassava varieties are initiated under the management of farmers cooperatives. Planting material produced will be distributed to other farmers at cost. Considering the urgency of the need for more planting material of improved varieties, the Agribusiness Division of San Miguel Corporation will adapt a rapid- propagation technique for multiplication of varieties supplied by PRCRTC. In this way a more rapid dissemination of new cassava varieties can be attained. FUTURE PLANS With the expansion of the area planted to cassava, PRCRTC will have to double its efforts to monitor the performance of cassava varieties planted in these plantations. The Center will also conduct intensive selection for superior varieties for the food, feed and alcohol industry. The competitive advantage of cassava over other crops as raw material for domestic utilization is very high. Thus, in line with the Medium-Term Philippine Development Plan, priorities are geared toward domestic utilization of cassava. PRCRTC will cooperate in the effective transfer of mature technologies to cassava farmers and processors. It is envisioned that in the next five years, average yields will increase and dissemination and adoption of new recommended varieties will have markedly increased. Accordingly, the breeding objectives for cassava will remain the same, but more emphasis will be on the identification of superior varieties with high yield, high starch content and low to high HCN levels that will suit the various needs of the food, feed, flour and starch industries. For effective technology transfer, closer cooperation among the government institutes, non-government organizations (NGOs) and the private sector will be sought. REFERENCES Lynam, J.K. 1986. A comparative analysis of cassava production and utilization in tropical Asia. In: Future Potential of Cassava in Asia. Proc. of Regional Workshop, held in Bangkok, Thailand. June 4-8, 1984. pp. 171-196. Mariscal, A.M. and J.L. Bacusmo. 1995. Recent progress in cassava varietal improvement in the Philippines. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandum, Kerala, India. Nov 2-6, 1993. pp. 306-321. Shingh, R.B. 1986. Past performance and future prospects of cassava products in Asia and the Pacific Region. In: Future Potential of Cassava in Asia. Proc. of Regional Workshop, held in Bangkok, Thailand. June 4-8, 1984. p. 404. Walters, P.R. 1983. Cassava's markets - where next? World Crops, pp. 130-133. 149 RECENT PROGRESS IN CASSAVA VARIETAL IMPROVEMENT IN INDONESIA Soemarjo Poespodarsono1 ABSTRACT In Indonesia, cassava is used mostly for human consumption, while a smaller part is used for animal feed and as raw material in industry. This crop is cultivated in almost all provinces of Indonesia. However, because the amount of rainfall among the regions varies and is not evenly distributed, some regions have a wet climate while others have a dry climate. The type of climate affects the suitability of a particular variety and its productivity. The area planted to cassava in Indonesia is about 1.3 million ha producing approximately 16.3 million tons fresh cassava roots; this means that the average production per hectare is still low at about 12.2 t/ha. This low yield is due to the use of local varieties, traditional cropping patterns and simple cultural practices. According to the government, cassava production is projected to increase to about 16.5 million tons in the year 2000. This production can be achieved by intensifying the cultural practices as well as by the use of higher yielding cassava varieties, either sweet or bitter ones. Up to now, a very limited number of high yielding varieties have been released, i.e. only ten new varieties from 1969 to 1990 for all cassava areas in Indonesia, compared to 74 rice, 28 maize, 25 soybean, and 5 sweet potato varieties. Some new cassava varieties were released, such as Adira 1 (sweet) and Adira 2 (bitter) in 1978, and Adira 4 (bitter) in 1986. In 1993 MARIF (Malang Research Institute for Food Crops) released two new varieties named Malang 1 (CM4049-2) and Malang 2 (CM4031-10). Since 1984, Brawijaya University, in cooperation with CIAT, has been conducting cassava breeding. From F, cassava seed that had been provided, four promising clones were selected, i.e. UB 1/2 (CM3962-2), UB 15/10 (CM3380-10), UB 881-5 (SM881-5) and UB477-2 (SM477-2). Multi-location trials have been conducted several times with these clones, especially in East Java. Concerning governmental regulations about the realease of new varieties, one must conduct multi- locational trials of these promising lines or clones in at least five provinces. In 1995, those promising cassava clones were planted in East Java, Central Java, Lampung, North Sulawesi and Lombok. The result of these trials could suggest the release of these lines as new cassava varieties. From the results it was concluded that clones UB 1/2 and UB 15/10 (both bitter) are more suitable for regions with a wet climate, while clones UB 881-5 (bitter) and UB 477-2 (sweet) are more suitable for those with a dry climate. INTRODUCTION In Indonesia, cassava is cultivated in almost all provinces, particularly by small farmers but also by some large plantations. Recently, the cassava area has been approximately 1.33 million hectares, producing 16.3 million tons of fresh roots with an average yield of 12.2 t/ha (CBS, 1992). Most cassava produced by small farmers is used for food, either as occasional dishes or as a staple food; the rest is processed into dried root pieces (gaplek) or into starch. Cassava produced by big farmers or plantations is processed almost entirely into starch. Due to the usage of cassava roots mainly as human food, sweet varieties are needed for food or starch production, while bitter varieties can Brawijaya University, Malang, E. Java, Indonesia. 150 be used for production of dried roots, starch or animal feed. In general, cassava has a low social status, which is a factor that affects the price and government policy. According to government planning, the projected increase in cassava production during the next five years is only 0.17% per year, compared with 2.01% for rice, 3.77% for maize, 3.17% for soybean and 1.96 for sweet potato (NBP, 1993). Cassava is planted in various types of soils and climates. Generally, the planting area of cassava consists of uplands, while a few areas are marginal land. As a result, cassava productivity and planting time depend on rainfall and its distribution. Based on Oldeman's climatic classification, climate types A up to type E are found in Indonesia. According to these climate types, Indonesia is generally divided into two main regions: one having a wet climate (A and B types) and another having a dry climate (C, D and E types). Roughly, the wet climate is found in the western and northern part and the dry climate in the southeastern part of Indonesia, including most of East Java. In the wet region cassava can be planted at any time of the year, but this is not the case in the drier regions, where cassava is planted at the beginning of the wet season. This affects the availability of cassava roots as raw material for the starch industry. Since cassava can be planted throughout the year in Sumatra most of the large cassava plantations are found there. Rainfall distribution also determines the yield of cassava (Wargiono, 1990). Up to now farmers still plant many local varieties. These varieties are well adapted to the local environment, but they generally have a low starch content. On the other hand, a limited number of high-yielding varieties have been released, but not all of these were distributed to farmers. From 1969 to 1993 only ten new cassava varieties have been released, compared with 74 varieties of rice, 28 of maize, 21 of soybean and 5 of sweet potato (Baihaki, 1996). Two cassava varieties, i.e. Adira 1 (sweet) and Adira 2 (bitter), were released in 1978. Adira 4 (another bitter variety) was released in 1986 and is now widely grown. In 1993, MARIF (Malang Research Institute for Food Crops), now renamed as RILET (Research Institute for Legumes and Tuber Crops) released another two varieties, i.e. Malang 1 (CM4049-2) and Malang 2 (CM4O31-10). Considering that a wide range of ecological conditions exist in Indonesia, there is a need for many high-yielding varieties that are adapted to special local environments. That is why Indonesia needs more than one institution for handling the cassava breeding program. CASSAVA BREEDING IN INDONESIA Cassava breeding and the release of new varieties dates back to Dutch colonial times. During that era, some high-yielding varieties were probably released since many local varieties still have Dutch names, such as Faroka and Vandrum. These varieties, as well as others released at that time, are bitter varieties and were cultivated in areas surrounding starch factories. Sweet varieties were planted mainly near the farmers' homes. 151 Since 1984, Brawijaya University has been involved in cassava breeding, along with the Research Institutes for Food Crops in Bogor and Malang, and in close collaboration with CIAT. At the same time, CIAT also cooperated with private institutions such as the Umas Jaya Farm, which is located in Lampung province of Sumatra. Thus, cassava breeding is conducted in the two main regions, namely the wet region, represented by Bogor (West Java) and the Umas Jaya Farm, and the dry region, represented by Malang in East Java. It is expected that several high-yielding varieties will be released in the future, with each one adapted to a given environment. The main objective of the cassava breeding program in Indonesia is to produce both sweet and bitter cassava varieties with the following characteristics: 1. High root yield 2. High harvest index 3. High root starch content 4. Tolerance to major pests and diseases 5. Non-branching growth habit 6. Good root shape High root yield has the highest priority. Because many farmers are using high inputs, so they expect a variety to respond well to fertilizers and to good management. A high harvest index is needed in order to minimize total nutrient uptake and prevent nutrient exhaustion of poor upland soils. Varieties with high root starch content are preferred by farmers and cassava plantations, because the price of cassava is calculated from its starch content. Since most farmers do not control pests and diseases they need varieties that are tolerant to the major pests and diseases. A non-branching growth habit is preferred by farmers, because cassava is often intercropped with other annual crops; non-branching varieties cause less competition to those intercrops. The last characteristic, i.e. good root shape, is of particular interest to large cassava plantations, because it facilitates the harvest. Besides the general objectives mentioned above, there are also more specific objectives which aim to solve various adverse conditions. The specific objectives are: 1. Tolerance to certain adverse soil and climatic conditions 2. Early harvestability As mentioned before, cassava is mainly planted in uplands having different soil and climatic conditions, including some adverse conditions, such as long dry seasons or high levels of soil acidity; so, in order to obtain relatively high yields, varieties well adapted to those conditions are needed. A variety with early harvestability aims to spread the time of harvest, particularly to allow small farmers to increase their income. Breeding Activities Cassava breeding activities at Brawijaya University start with single plant 152 selections from F, seeds. Initially these seeds were introduced only from CIAT/Colombia and about 2,000-3,000 seeds were sent each year. Since 1993/94, however, also F, seeds from the Thai-CIAT program, with clone codes of OMR or CMR, have been received. At the start, each seed is germinated in polybags and after 2-3 months the seedling are transplanted to the experimental field. At harvest time about 10-20% of plants are selected, based on root size and shape as well as other morphological characteristics. The second generation is planted in a Single Row Trial with stakes from each selected clone, generally five stakes for one row of each line and with two rows of control varieties, i.e. Adira 4 (a national high-yielding variety) and Faroka (a local variety). These two control varieties are used for all stages of selection. The selection criterion for this second step is that selected clones must yield higher than or equal to the control varieties. This criterion is similar to that used in the third generation, i.e. the Unreplicated Yield Trial, in which each line is planted in a single plot. The next selection step is called the Advanced Yield Trial, which uses a randomized block design with three replications. In this generation, usually 15-20 clones are selected. Fresh root yield, root dry matter content (RDMC), biomass and harvest index are determined at harvest time. Usually, from these trials, sufficient planting material is obtained for planting in three sites of the Multi-location Trial. In this paper, the results of these Multi-location Trials, conducted in Jatikerto (Central Malang), in Kalipare (South Malang) and in Tarokan (Kediri), are presented in Tables 1, 2 and 3, respectively. All these sites are located in East Java. Table 1. Results of a Multi-location Trial conducted at Jatikerto Experiment Station in Malang, East Java, Indonesia in 1994/95. Total Harvest Fresh root RDMC biomass index Clones Parents yield (t/ha) (%) (t/ha) OMR33-05-4 CMC76 37.2 33.1 46.7 0.80 OMR33-10-3 Rayong 5 30.0 34.8 50.7 0.54 OMR33-10-8 Rayong 5 33.0 34.8 57.7 0.57 CMR33-56-5 - 27.4 35.7 41.6 0.65 OMR33-10-11 Rayong 5 37.0 32.4 56.7 0.65 CMR33-12-14 CMR26-14-9xR90 30.3 34.0 46.3 0.65 CMR33-52-9 CMR26-14-9xR90 29.8 35.5 38.8 0.77 CMR33-56-8 CMR26-14-9)cR90 25.7 34.4 34.3 0.75 CMR33-52-8 CMR26-11-9)cR90 21.3 34.7 30.2 0.71 Adira 4(control) 25.4 34.0 38.6 0.66 Faroka (control) 34.6 33.6 49.1 0.70 153 Table 2. Results of a Multi-location Trial conducted in Kali pa re in South Malang, East Java, Indonesia in 1994/95. Total Harvest Fresh root RDMC" biomass index Clones Parents yield (t/ha) (%) (t/ha) OMR33-05-4 CMC76 24.0 33.5 36.8 0.65 OMR33-10-3 Rayong 5 24.9 33.4 36.7 0.68 OMR33-10-8 Rayong 5 23.1 32.6 36.3 0.64 CMR33-56-5 - 21.5 35.6 32.8 0.66 OMR33-10-11 Rayong 5 25.5 32.1 36.6 0.64 CMR33-12-14 CMR26-14-9xR90 20.5 34.7 31.7 0.65 CMR33-52-9 CMR26-14-9xR90 20.1 34.9 30.9 0.65 CMR33-56-8 CMR26-14-9xR90 30.7 35.0 33.0 0.63 CMR33-52-8 CMR26-1l-9xR90 19.2 34.5 29.0 0.66 Adira 4(controi;1 24.4 35.7 38.5 0.63 RDMC = root dry matter content Table 3. Results of a Multi-location Trial conducted at Tarokan in Kedire, Java, Indonesia in 1994/95. East Total Harvest Fresh root RDMC" biomass index Clones Parents yield (t/ha) (%) (t/ha) OMR33-05-4 CMC76 23.5 32.6 34.1 0.69 OMR33-10-3 Rayong 5 24.1 31.3 33.9 0.71 OMR33-10-8 Rayong 5 24.1 31.2 37.4 0.64 CMR33-56-5 - 19.8 35.0 29.8 0.66 OMR33-10-11 Rayong 5 21.2 31.4 33.5 0.63 CMR33-12-14 CMR26-14-9xR90 18.9 34.5 28.1 0.67 CMR33-52-9 CMR26-14-9xR90 17.7 34.9 26.0 0.68 CMR33-56-8 CMR26-14-9xR90 15.1 35.2 22.0 0.69 CMR33-52-8 CMR26-1l-9xR90 13.2 34.8 21.1 0.66 Adira 4(controi;1 25.3 34.8 37.7 0.67 " RDMC = root dry matter content Among these three locations, the soil and rainfall conditions of Jatikerto are better than those of Kalipare and Tarokan. This affects the performance of the selected clones. From these tables it is clear that OMR33-05-4 produced a high root yield and 154 had a relatively high RDMC and harvest index. But OMR33-10-3 seems to be more suitable for poor conditions, such as in Kalipare and Tarokan. Similar previous trials have produced several highly promising clones, such as UB 1/2 (CM3962-2, a selection from a cross of MCol 22 with CM849-1), UB 15/10 (CM3380-10, a selection from a cross of CM568-1 with CM523-7), UB 881-5 (SM881-5, selected from open-pollinated seeds of CM849-1) and UB 477-2 (SM479-2, selected from open-pollinated seeds of CM 1022-4). According to government regulations, promising lines or clones can be released as new varieties only if they have been tested in at least five provinces (Multi-provincial Trials). Therefore, these highly promising clones have also been tested for this purpose in Lampung (Sumatra), Central Java, East Java, North Sulawesi, and Lombok. These trials have been harvested in early 1996, the result of which are presented in Table 4. It seems that UB 1/2 (CM3962-2) and UB 15/10 (CM3380-10) produced high fresh root yields in most sites. In Central Java all clones had relatively low yields, because cassava was intercropped with rice and maize, followed by mungbean. From these results, the two promising clones, UB 1/2 and UB 15/10, will be suggested to the Indonesian government for release as new cassava varieties. Table 4. Results of Multi-provincial Trials conducted in Lampung, Central Java, East Java, Lombok and North Sulawesi in 1995/96. Fresh root yield (t/ha) Clones Lampung Central Java East Java Lombok North Sulawesi Average UB 1/2 33.3 22.6 23.7 37.5 46.2 32.7 UB 15/10 32.9 25.5 31.8 34.0 25.5 29.9 Kasetsart 50 31.9 - - - - - UB 477-2 20.9 22.8 27.6 33.6 34.0 27.8 UB 881-5 18.0 21.4 29.1 29.6 29.5 25.5 UB 1-2/15 - 23.3 - 20.5 20.9 21.6 UB 1-2/20 24.6 - 32.1 31.4 32.7 30.2 CMR33-10/3 28.3 - - - - • OMR30-10-11 - - 19.9 - - - Adira 1 . 17.2 - . 21.1 19.1 Adira 4 29.9 - 25.6 23.1 22.1 25.2 Malang 1 23.5 21.8 28.3 23.2 22.9 23.9 Malang 2 - - - 25.1 24.5 24.8 155 REFERENCES Baihaki, A. 1996. Prospek penerapan breeder right di Indonesia. Simposium Pemuliaan Tanaman IV. PERIPI Komda, Jawa Timur. 16 p. Central Bureau of Statistics (CBS). 1992. Indonesian Statistics. National Bureau of Planning (NBP). 1993. Five Years of Indonesian Planning. Poespodarsono, S. and Y. Widodo. 1995. Recent progress in cassava varietal improvement in Indonesia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp. 175-182. Wargiono J., 1990. Pengaruh distribusi curah hujan terhadap penampian klon ubikayu (Effect of rainfall distribution on cassava performance). Penelitian Pertanian. Balai Penelitian Tanaman Pangan. Bogor, Indonesia. No 10, Vol I: 156 CASSAVA VARIETAL IMPROVEMENT PROGRAM AT UMAS JAYA FARM AND ITS CONTRIBUTION TO SMALL FARMER COMMUNITIES IN SUMATRA, INDONESIA Palupi Puspitorini1, Usman Kartawijaya1 and Kazuo Kawano2 ABSTRACT A three-way collaboration in cassava varietal improvement and dissemination was established in 1982, in which a development-oriented private corporation (Umas Jaya Farm in Lampung, Sumatra of Great Giant Pineapple Coy of Gunung Sewu Group) provided ample facilities for varietal selection, multiplication and dissemination to a resource-limited national cassava breeding program (CRIFC), while an international organization (CIAT) contributed the basic training of research personnel and additional breeding materials. The program continued even after the CRIFC breeder left for abroad in 1989. This program is now regarded to be the best functioning cassava varietal development program in Indonesia. We selected a clone with distinctly high yield and root starch content from the original CRIFC breeding stocks, multiplied it and released it. We donated the planting stakes of this cultivar to many small farmers. cooperatives and shipped out millions of stakes to large plantations in and outside Sumatra. This new clone is now widely known as M-31 in the production fields and Adira 4 in the official registration. The economic benefit generated by the new cultivar started with the additional root harvests and additional starch produced within UJF, but it soon spread into the production fields of thousands of small farmers. Hard data, and a few extrapolations from these, indicate that Adira 4 is now planted in more than 100,000 ha and the additional economic effects caused by its adoption is more than 194 million US dollars, of which at least one half has directly entered the household income of innumerable small farmers. We continued selections based on CIAT materials and are identifying new clones that are better than Adira 4. From this work, two new cultivars, Malang 1 and Malang 2, were officially released recently and one clone, Kasetsart 50, is being pre-released. Our collaborative program offers an example that the private sector can efficiently and effectively handle cassava varietal development, especially the multiplication and dissemination, when government research and extension institutions cannot afford a functional research and dissemination program. INTRODUCTION Cassava has been a traditional food crop in the upland farming areas of Indonesia and its role, especially as a famine relief crop, should not be underestimated. Yet, with the continuous improvement of the diet of the rural population, especially through the increased availability of rice, the form of cassava utilization has been shifting steadily from fresh human consumption to processed foods, animal feed and industrial starch. While the importance of cassava for direct human consumption is gradually decreasing, its role as a major source of cash income for small farmers and as raw material for agro- industry is rapidly increasing. Several large-scale cassava starch processing plants have been established in southern Sumatra province of Lampung during the past fifteen years, in order to supply starch to the national food industry utilizing existing small farmers' 1 Umas Jaya Farm, Great Giant Pineapple Coy, Lampung, Indonesia. 2 CIAT Asian Cassava Program, Dept. of Agriculture, Chatuchak, Bangkok, Thailand. 157 production, as well as that of the large areas of previously under-utilized land now used for cassava plantations. This is in accordance with the national policy of establishing agro-industries and creating employment using under-utilized lands on the outer islands. Umas Jaya Farm (UJF) was a pioneer of such enterprises, starting as a cassava production and starch processing plantation. We established a collaborative cassava varietal development program at UJF in 1982 and the initial development was highly successful. At the 3rd Asian Cassava Research Workshop in 1990, we reported this initial success and concluded the presentation by stating "How much of this success is going to be extended to the mass of small farm communities is yet to be seen" (Hardono et al., 1992). Six years later, we therefore report how the initial technical success has been transferred and spread into the multitude of small farmers in Sumatra of Indonesia. Umas Jaya Farm A license to lease 13,000 ha of underutilized land located at Terbanggi Besar, Lampung, Sumatra, was conceded by the Indonesian government to the Gunung Sewu Group with the condition of developing the area for value-added agro-industrial production. Cassava roots produced in their plantation and the starch processed from these in their own factory were the major activities during the early years. The major operation has gradually shifted to the production and canning of pineapple, which proved to be more profitable for UJF. The pineapple operation, called Great Giant Pineapple Company, now employs more than 10,000 workers. The UJF planted cassava annually at their own plantation in areas between 0 and 3,000 ha during the past 15 years; presently, it is no longer planting cassava within the UJF plantation, as the company has moved cassava production to their new satellite operations outside UJF. The starch processing plant of UJF, which was processing only the cassava roots produced in UJF in the early years, is now processing roots that are nearly 100% produced by small farmers outside UJF. The capacity of the starch factory is being upgraded to absorb more cassava produced by small farmers outside. Thus, the production of fresh cassava roots has shifted to the hands of small farmers while UJF has shifted their emphasis of operation to factory processing during the past 15 years. This is a general pattern in all of Lampung where more than 20 large- scale cassava starch factories are in operation. UJF has invested heavily, not only in the factories and production fields, but also to improve the local infrastructure like roads, housing for 8,000 laborers, electricity, water supply, schools, clinics, mosques, grocery shops, cafeterias, sport and entertainment facilities, and last but not least, research facilities. UJF now maintains one of the best managed field experimental plots in Indonesia, which are basically open to national research organizations. The University of Lampung, the Agricultural University of Bogor (IPB), the Technical University of Bandung 0TB), and Brawijaya University, 158 among others, have been using the UJF field for their research. While establishing a commercially viable agro-industrial business is the principal goal of the UJF/GGPC operation, rural development and better social welfare are equally important long-term objectives. Three Way Cooperation During the 1970s, a modest root crop research program of the Bogor Research Institute for Food Crops (BORIF) under the Central Research Institute of Food Crops (CRIFC), was the only institution where research on cassava varietal improvement was conducted. Based on the national cassava germplasm collection, small-scale hybridization and selection at Bogor produced a series of improved hybrid clones. While there was a certain progress in their breeding work, their success was largely limited to the small scale dual purpose production of West Java. BORIF/CRIFC did not have sufficient resources to evaluate the advanced breeding materials for large-scale industrial production multi-locationally or to multiply planting stakes on a large scale. We established a three-way collaboration in cassava varietal improvement and dissemination at UJF in 1982, in which UJF provided ample facilities for varietal selection, multiplication and dissemination to the BORIF/CRIFC cassava program, while CIAT contributed the basic training of research personnel and additional breeding materials. Both the breeder of the BORIF program (Roberto Soenarjo) and the UJF research and development manager (Hardono Nugroho) in the initial years of the collaborative program, had been participants in a CIAT Cassava Training Course in Cali, Colombia. From 1983, advanced breeding materials in the form of hybrid seeds have been introduced to BORIF from CIAT/Colombia and since 1985 from the Thai-CIAT program. These have been evaluated and selected jointly by the three parties at Umas Jaya Farm. Multi-location evaluations of the most advanced materials have also been conducted jointly. It was a new experience for the national program breeder to be able to evaluate his breeding materials without worrying about the size of experimental plots, availability of field labor or other expenses. The collaborative program continued even after the BORIF breeder left for abroad in 1989 and the program is now regarded as the best functioning cassava varietal development program in Indonesia. Selection of M-31 (Adira 4) A clone identified as M-3 1 was one of the breeding materials produced at BORIF that was taken to UJF for large-scale evaluations. At the second cycle of evaluation, it became clear that M-31 was outstanding, not only among the 40 clones brought from Bogor, but also in comparison with the then leading traditional cultivar of the plantation and the region, Kretek (Table 1). M-31 possessed high yielding capacity, high root dry matter content, and a good combination of vigorous vegetative growth with comparatively high harvest index; it also had a good plant type, and uniform and well-shaped roots with 159 white root flesh color, an important trait for starch production. By 1986, the official release of M-31 was recommended at BORIF and the name Adira 4 was given. Adira 4 showed its superior capacity for large-scale industrial starch production, giving a yield up to 80% higher than Kretek (Table 2); moreover, the adaptability of Adira 4 to drier climates of Central and East Java was also confirmed. By 1988, large quantities of planting stakes of Adira 4 were available for distribution outside the UJF plantation. Varietal Dissemination We started the distribution of Adira 4 planting stakes in 1987 by donating packages of long stakes (2500 long stakes each) to 40 cooperatives of small farmers in Lampung. This activity of giving the stakes of improved cultivars to any interested small farmer community continues up to the present. A massive distribution followed in 1988 and 1989. A total of approximately 6 million long stakes, enough to plant more than 3,000 ha, have been sold at a nominal price to large-scale plantations in Lampung, South Sumatra, West Sumatra and Palembang, and to extension units in South Sulawesi, South Kalimantan, and West-, Central-, and East Java. The area planted with Adira 4 is now estimated to be more than 100,000 ha in Lampung province alone (Tables 2 and 3; Figure 1). Adira 4 is still widely known as M-31, the UJF code name, in Lampung rather than as Adira 4, the BORIF/CRIFC official name. Since Adira 4 has spread to other provinces of Sumatra, as well as to other outer islands, the total Adira 4 acreage may now be much more than these figures. The large-scale adoption of Adira 4 caused substantial economic gains first to the large plantations and factories; yet, it is now resulting in wide-spread economic benefits to tens of thousands of small farmers. We shall analyze these effect more in detail in the subsequent sections. Analysis of Economic Benefits 1. Production data used We used the sales value of the products as the basis for measuring economic effects. Value of sales includes production costs and net profits. The gross production costs are made up of expenditures for labor, equipment, supplies, maintenance, and depreciation. At the company management level, net profit rather than the value of sales may be more critical. However, in the national development context, value of sales may be a more important indicator of the socio-economic benefits, as it represents employment, purchasing power and the availability of useful products. ca 3 c E 5 u ea § 1 p g 3 ea ea 8u o 8 c ca E u us. o £2 s < n a _o •3 u os oo u. b -^ ^ £ >- o Z 3 2 oo O 00 OS PL, >- 00 O oo E oo q oo NO r-. cm nO On en oo •8 > < 00 On nO r-. 00 nO CM (N no r- m 00 m i/-> r- Tt Tt ci nq 00 _^ iri ON oo IT> On On 00 "8 u CM d VO —. oo nO 00 I I E 06 2 « 2 .2 e £ >> 8 «- 2 •*= o o * !3 <3 *-. O II 06 oo >- II II 06 U >* tt. 00 00 2 I ■a c VO :s E•■3 2 < •/> ■- S CS OV. £ »H 11 w £ 3 E e « J! " « !3 I* si w ex a © U S rvj V) cm r- no en r~ r- en oo in Tt in —■ (NoomotOTtTf-Hi/-lo N cm oo in in — oo no cm no —* io Tj- in m n r> m ■a "* SIM ■* ^; VO 00 N © © © © no Tj:odl-.6vdooOQ.o.»3-*_ 00 00 (N VO On ^OOlftTt © en oq ifl © © in cn on oo dd—Tt.iflodCowndo CM cn On o In en »m q io m © en in r~. t-. © CM VO VO r>. en oo in oo r~' •*.■*. en en oo cm r- no *vOr- - en en © © ©■* -i CM m w (N ID tO © 00CN ITl 00 —i VO © CM o m in - cm N O O On r- ■* © —" r-. Tj-. nb no en oo oo CM r- © en no oo o d vo d •*. vi o no *o r- —« N CM N VO oo ^r^ooon© — in m Tt in vo r- «3oooooog\OnOnonO\OnO>On On no r- oo on — 0O 00 00 00 © — N en Tj- in no W 9> 0\ C\ 0> Q\ On oo oo VO 00 3 0) > 3 o ■a c .-*—< 2 no ° On D On•S *7 D On > On o — ^ o <*5 >D en 1- < *> C O e3 c 4) —* h On IT "3 In 2£ .22 —. >> i- >5 < ** >n In (N 00 fc* VO On On In On On ■9 J3 5 CM •a On On On On £ 8 i.■•-« 9) c•5 ■g 3 3 3 os V3 •o J3 t/3 2 £ 3 .■ * ,u " « i1 s § i a c s t/5 ft_ CMP->QQIOiOQQQQQ — innor-oor-cNin — on -*cNmcN-Hr~r^cn — N * ^ r~ N in — in —<* cn no" S Tt r~* cn" —1 — (N Tj- 00 00 © en oo en -* —* oo Tt a - N 6d-Hinodt^MnoNVi^ -o■a o ia .2 ** ** « 5 c-> 3 "a o >< H S3 = 3 en i 12 fl = *-* — w o w S 1 S .S * oowqpininpoooq en On in nO ' n in - on © - 3 O 13 c o ■a •5 La 53 > o 3 .J C o c 0) L. i'•5 S nq r-' m m -o ■8 CM inO cn *r On (N l On On — -a a c2 >- <5 8 (N tn ^8 SO — is On CTJ « «S 4- ** m tn cn oo tn C c o o M 2 aS m H 00i2 On u. -"" c5<2 o 00o m in v» c c o o c © c c % 3 S3 S3 OQ CQ CQ CQ a w 163 t 1 1 1 1 1 r 1985/86 86/87 87/88 88/89 89/90 90/91 91/92 92/93 93/94 94/95 95/96 Year Figure 1. Area planted with new cassava cultivars (mostly Adira 4) in Lampung, Sumatra, Indonesia from 1985 to 1996. 164 To concentrate on the most direct monetary effects and avoid possible duplicates and exaggerations, we used only two measures for the economic gain: additional field production of fresh cassava roots due to the higher yield of Adira 4, and additional factory starch production due to the higher starch content of Adira 4. The additional factory profit generated by the gross additional availability of raw materials is not accounted for, let alone the additional production of numerous secondary products. Thus, the real gross additional economic effects are much more than described in this simple analysis. The following hard data were available: 1. Area planted with Adira 4 within the UJF campus (Table 2). 2. Yield and starch content data from numerous trials as well as from large commercial plots (Table 1). 3. UJF survey on cassava planting at 62 municipalities in three regions of Lampung province during every month from Feb to Sept 1995. 4. Data about cassava acreage in Lampung province from 1985 to 1994 supplied by the Agriculture Department of Food Crops of Lampung province. 5. On-farm price of fresh cassava roots and starch price for the factory from 1985 to 1996. We used the low price in each 2-5 year period for the analyses. We used the following information as semi-hard data: 6. UJF estimates of the percentage of total area planted to Adira 4 in three regions of Lampung. 2. Economic effects within UJF The first analysis is on the additional monetary gain within the UJF campus in the past ten years (Table 2). This is strictly based on hard data, and probably the most reliable economic analysis of this kind, even though the scope is limited to a relatively small, closed scheme. The new cultivar has been planted in between 0 and 3000 ha depending on years; it produced an additional 128,000 tons of fresh cassava and 23,000 tons of starch, which led to the additional values of US$ 3.8 million and US$ 4.6 million, respectively, or a total of US$ 8.5 million. This analysis is well in line with the observations of the last author. The sales value includes a considerable proportion of production costs; thus, the net profit to the company is smaller than the values presented above. Nevertheless, it leaves little doubt that the new cultivar has brought in a profit vastly and positively disproportional to the research investment. Economic effects in Lampung province Our field survey estimates the proportion of Adira 4 in the total cassava planted area to be 75% in North Lampung, where UJF is located, 60% in Central Lampung, and 165 25% in South Lampung. More than 90% of the cassava roots brought by small farmers to the UJF factory are of Adira 4. In North Lampung, nearly all of the cassava planted near large starch factories are Adira 4. There are 21 large starch factories in Lampung. Thus, the projection above may be a reasonable estimate. Extrapolating from data sources 3, 4 and 6 mentioned above, the total area of Adira 4 is estimated to be presently about 1 18,000 ha in Lampung (Table 3, Figure 1). The total additional money received by the producers for the additional fresh root yield due to the adoption of Adira 4 is estimated to be US$ 85 million (Table 3). The total additional starch value generated by the higher starch content of Adira 4 is estimated to be US$ 109 million (Table 3). As of now, monetary benefits generated through the improved starch content of the new variety are even greater than that generated by the higher fresh root yield of the new variety (Figure 2). Yet, the total economic effects of US$ 194 million is the figure for Lampung province alone. The total effects for all over Indonesia must be significantly larger than this. 4. Estimating direct economic effects to small farmers Since the great majority of the fresh root producers are now small farmers, we can happily and safely assume that the US$ 85 million additional profit for the higher fresh root production went directly to the pockets of small farmers in its near entirety. How much of the additional profit generated by the higher starch content of the new cultivar was returned to the farmers depends on what differential price the starch factory pays to the farmers. If the factory pays the same price for roots of different starch content (in other words, does not honor or recognize the breeders' and farmers' efforts to produce higher quality products), all the additional profit would be pocketed by the factory. At UJF, the factory is paying differential prices to the farmers; thus, a significant portion of the profit resulting from the higher starch content is being returned to the small farmers. Yet, there are many , mostly small, factories which do not yet pay according to starch content. Under any circumstances, larger operators have more access to larger profit making opportunities. Yet, in cassava as a cash crop for small farmers, a significant portion of the profit generated by new varieties is surely going directly to the multitude of small farm families. Further Progress in Varietal Selection Varietal selection continues and lines CM4049-2 and CM4031-10 were selected from the hybrid seed introductions from CIAT/Colombia. Through the CRIFC varietal naming and release scheme, the Research Institute for Legumes and Tuber Crops (RILET), located in Malang, East Java, and functioning under CRIFC, named them Malang 1 and Malang 2, respectively. These are equally high yielding as Adira 4 but not significantly superior. UJF did not multiply them in a big manner. They are now still planted in only small areas here and there. Kasetsart 50, a clonal introduction from 166 Thailand, showed excellent results (Table 1) and is currently being multiplied by UJF for planting in other sites outside UJF and for distribution to interested farmers. Several new selections are passing through the breeding pipeline (Tables 4 and 5) and they may eventually prove superior to Kasetsart 50. While Kasetsart 50 has become a universal standard of selection in many parts of Asia, no breeding program outside Thailand has produced yet anything clearly superior to Kasetsart 50. Our selection at UJF may be the next program to attain this honor. Aside from the two highly functional cassava breeding programs in Thailand, one at Rayong Field Crop Research Center and the other at the site of the Thai Tapioca Development Institute conducted by Kasetsart University, our program at UJF may be the best functioning cassava breeding program in Asia. Pineapple replaced cassava as the most profitable crop at UJF and we may discontinue large scale cassava plantings within UJF. If we view the on-going cassava breeding program at UJF as a short-term money making scheme, it might have accomplished very profitable functions already. On the other hand, since long-term contribution to the welfare of local communities is one of the important objectives of the UJF operation, we may continue the program. After all, this is virtually the only cassava varietal improvement and dissemination program that is producing results in Indonesia at present. CIAT collaboration here will continue as long as the opportunity for selecting superior genotypes for small farmers exists. CONCLUSIONS Our thirteen years of collaboration offers the following clear messages: 1 . Indisputable success of cassava varietal improvement and dissemination took place through a private corporation (UJF), which fully utilized the materials and expertise offered by an international research organization (CIAT), i.e. breeding materials, training and technical guidance. 2. The contribution of varietal improvement can be quantitatively measured in terms of additional income generated to small farmers. A hundred million US dollars shared by hundreds of thousands of poor farmers as additional income has quite different social effects compared with the same amount of money made by a multi-national conglomerate. 3. Cassava varietal improvement proves to be one sure vehicle with which to improve the plight of small farmers. 4. The private sector can efficiently and effectively handle cassava varietal development, especially the multiplication and dissemination, when government research and extension institutions need additional research and dissemination capabilities. 167 Table 4. Results of an Advanced Yield Trial at Umas Jaya Farm, Lampung, Sumatra in 1994/95. Dry Fresh Root dry Total Harvest root root matter biomass index yield yield content yield Clone Parents (t/ha) (t/ha) (%) (t/ha) SM 1650-2 CM3306-4 15.9 41.5 38.4 66.6 0.62 Malang 1 CM1015-19xCM849-l 14.4 41.7 34.5 69.5 0.60 Kasetsart 50 Rl x R90 14.3 38.9 36.8 80.6 0.48 Rayong 90 CMC76 x V43 13.9 39.3 35.5 70.1 0.56 CMR30-56-1 CMR23- 17-251 x Rl 13.6 36.5 37.2 57.2 0.63 Mean of all 22 entries 11.2 30.8 36.3 55.0 0.56 Adira 4 (control) = BORIF 528 Kretek (local control) 13.5 36.9 36.6 68.5 0.54 10.1 33.5 30.0 71.5 0.47 Table 5. Results of a Preliminary Yield Trial at Umas Jaya Farm, Lampung, Sumatra in 1994/95. Clone Parents Dry Fresh Root dry Total root root matter biomass yield yield content yield (t/ha) (t/ha) (%) (t/ha) Harvest index Thai-CIAT clones CMR35-42-4 CMR29-60-15 x KU50 18.5 48.0 38.6 90.8 0.53 CMR35-20-1 R5 x CMR28-72-131 18.0 50.0 36.0 76.6 0.65 CMR35-118-7 R60xCMR28-72-131 17.0 46.2 36.8 72.4 0.64 OMR35-30-3 OMR29-20-118 15.1 42.3 35.7 89.5 0.47 CMR35-110-3 Rl x R90 13.7 42.2 32.5 71.1 0.59 Mean of all 14 entries 12.7 34.9 36.5 64.8 0.54 CIAT/Colombia clones SMI 8 12-6 SG804-5 20.0 52.8 37.8 92.1 0.57 SMI 853-2 CG1141-1 17.2 50.0 34.4 81.8 0.61 SM 1778-2 CG1320-10 16.3 42.7 38.3 77.0 0.55 SMI 879-1 MPar 159 16.3 46.9 34.8 76.2 0.62 SM 1787-1 HMC-1 15.6 46.7 33.5 88.5 0.53 Mean of all 20 entries Adira 4 (control) = BORIF 528 12.3 13.1 36.1 35.9 34.1 36.5 66.7 67.8 0.54 0.53 168 2(KH u J2 "5■o e o en la > 3 U is c c o Q. O o 3 Total gross additional sales caused by adoption of new cultivar 150- s ins 1 ■« B O 73■o « 13 "5 E 3 u u < 100- 50- Additional starch sale Additional fresh root sale 0J —i 1 P~ r 1 1 1 r— —i 1 r 85/86 86/87 87/88 88/89 89/90 90/91 91/92 92/93 93/94 94/95 95/96 Year Figure 2. Additional economic effects generated by the adoption of new cassava cultivars in Lampung province of Sumatra, Indonesiafrom 1985 to 1996. 169 REFERENCES Hardono Nugroho, J., R. Soenarjo and K. Kawano. 1992. Umas Jaya project - An example of successful cooperation between the private sector, a national institution and an international organization. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Utilization Research in Asia. Proc. 3rd. Regional Workshop, held in Malang, Indonesia. Oct 22-27, 1990. pp. 162-169. 170 SOCIO-ECONOMIC CONTRIBUTION OF CASSAVA VARIETAL IMPROVEMENT TO THE SMALL FARMER COMMUNITIES IN ASIA Kazuo Kawano ABSTRACT The success or failure of any large-scale crop breeding program should be measured by achievement of the following step-by-step goals: 1 ) Establishment of a breeding program 2) Building of effective research capability 3) A large number of genotypes produced and evaluated 4) Selection of superior genotypes 5) A large number of cultivars released 6) A considerable area planted with new cultivars 7) Additional yield and quality as a result of the planting of new cultivars 8) Additional economic benefits generated by the adoption of new cultivars 9) The less privileged strata of society receive the greatest share of the profits 10) Beneficial social changes brought about by the planting of new cultivars Accomplishing all of these was our wildest dream when we established a major cassava breeding program at CIAT headquarters in Colombia more than 20 years ago. The first step towards achieving these goals came in the form of a distinctly higher yielding capacity of the breeding population within CIAT. When a major part of the CIAT cassava program extended to the collaborative program in Thailand, a significant improvement followed in root dry matter content and in adaptation to drier climates. Then, this progress was transferred to other major national cassava programs in Asia. Many new cultivars were developed and released, and thanks to the effective involvement of national and private research, extension and development institutions, the new cultivars are now planted in more than half a million ha in Asia. The additional economic benefits generated by the additional fresh root yield and higher starch content of these new cultivars are surpassing 500 million US dollars. More importantly, at least half of this additional benefit is going directly to the family income of hundreds of thousands of small farmers. It is fortunate that we have been able to work with the same basic strategy and the same fundamental objectives throughout the history of nearly a quarter century of our cassava varietal improvement program. Now, we are beginning to see many of the original goals being attained. INTRODUCTION The spectacular success of new high-yielding varieties of rice (initiated by IRRI) and wheat (initiated by Rockefeller/CIMMYT wheat program), adapted to high-input cultural conditions not only increased the production of major food commodities in many tropical countries, but also convinced administrators of both developed and developing countries of the benefit of investing in tropical agricultural research. The "Green Revolution", as it was called, swept the Asian continent in the late-sixties and seventies, but was limited to relatively well-off farmers in the irrigated areas. The next major challenge was to raise the productivity and incomes of small farmers in the upland areas. Cassava, being one of the major crops for small upland farmers in the tropics, but at the CIAT Asian Cassava Program, Dept. of Agriculture, Chatuchak, Bangkok, Thailand. 171 same time the least researched crop, was yet to be taken as a next target of international research attention. With this background and a clear expectation of repeating the success of IRRI's rice varieties with this little-known crop cassava, CIAT established its Cassava Program in 1972, and the CIAT cassava breeding efforts started in 1973. Peter R. Jennings, the undisputed mastermind of the Rice Green Revolution of IRRI earlier in the 1960s, and the pundit of CIAT during the 1970s, challenged every novice breeder by indicating that any major crop breeding program of an International Agriculture Research Center should aim at a 100% yield increase, and the success of the program should be measured only by the area planted with the new varieties and the economic benefits that were generated by them. Both James H. Cock, then leader of the CIAT Cassava Program, and myself, who started the CIAT cassava breeding program, were young and ambitious with a measure of naivety, idealism and audacity. Borrowing a passage from the Gabriel Garcia Marques' master-piece "Hundred Years of Solitude": "The world was so new that many things even lacked a name", so free was our program in defining the basic research structure and so flexible was the administration of CIAT with a minimum of institutional strait jacketing. Our determination was to double the physiological productivity of cassava through breeding as a means to contribute to the improvement of small cassava farmers' income. In retrospect and with a little bit of hindsight, our overall objective and ambition were to be measured by the achievement of the following ten intermediate goals, and our wildest dream was to accomplish all of these: 1) Establishment of a breeding program 2) Building of effective research capability 3) A large number of genotypes produced and evaluated 4) Selection of superior genotypes 5) A large number of cultivars released 6) A considerable area planted with new cultivars 7) Additional yield and quality as a result of the planting of new cultivars 8) Additional economic benefits generated by the adoption of new cultivars 9) The less privileged strata of society receive the greatest share of the profits 10) Beneficial social changes brought about by the planting of new cultivars We were full of enthusiasm and optimism. We felt that we, the field workers, were the principal players of the whole game. Quixotic as it might have looked, this was the spirit of the young CIAT with which I started the cassava breeding program. Organizing the CIAT Cassava Breeding Program and Enhancing National Programs I have been writing on the history of CIAT headquarter breeding program and our collaborative breeding program with the Department of Agriculture, Thailand on many occasions, including in the CIAT Annual Reports and in the past Proceedings of this Asian Cassava Research Workshop. From these, at least the following points 172 deserve special mention. We started the breeding program with a vast genetic diversity, i.e. a collection of 2,200 local cultivars from the center of origin and diversification, instead of basing the program on a small number of preselected elite germplasm. After many cycles of intensive hybridizations and selections, our breeding populations show no sign yet of reaching a plateau of yield selection. This may be a result of the inclusion of a large genetic variability from the beginning. As the big "bang" of organic evolution in the Cambrian period suggests, the condition in the beginning may hold a lasting influence up to the very late stages of development. Then, modeling after the development of the CIAT breeding program, many national cassava breeding programs were established and have been upgraded. Generation and Distribution of Breeding Materials Some 450 thousand hybrid seeds from CIAT/Colombia have been distributed to the national breeding programs in Asia (Table 1). Thanks to the cooperation of the Rayong Field Crops Research Center of the Department of Agriculture (DOA) of Thailand, another 100 thousand hybrid seeds have been distributed from the Thai-CIAT program to other Asian national programs as well as to CIAT/Colombia through the CIAT Asian Cassava Program (Table 2). These breeding materials have been offering immediate opportunities for varietal selection, as well as the means to organize and improve the breeding operations of the national programs. Upgrading Yield Capacity and Adaptation of Breeding Populations We continue our joint efforts by the collaborative program in Rayong, by CIAT headquarters in Colombia and by the cassava breeders of national programs. In the Thai- CIAT program, we have been using the mean yield data of all clonal entries in the Regional Trial network relative to control as a way to measure breeding progress, because this represent the value of immediately available materials, as well as the potential of on-going hybrid populations. These advanced clonal entries form the basis of the hybridization program at Rayong Center. Since the hybrid seeds from the Rayong program are now a vital source of selection in many national programs in Asia, the advance in clonal selection in Thailand also effects the potential of yield selection in other national cassava programs in the near future. During the past years satisfactory progress was made in selecting for high fresh root yield (Figure 1), root dry matter content (Figure 2), and total biomass productivity (Figure 3). The harvest index has been stable at a relatively high level during the past ten years (Figure 4). Coming from the level of 1982 when data-taking started, we have made considerable progress. Adding to this the very substantial improvements made during the earliest years of CIAT headquarters breeding, which constituted the basis for the subsequent progress in Asia, the mean yield capacity of the present breeding population, i.e. the average yielding capacity of the hybrid seed population distributed 173 to national programs, is 100% higher than of the population during the early 1970s. Progress in each national program is reported by each program elsewhere in this Workshop Proceedings. Varietal Dissemination The number of CIAT-related cultivars officially released in Asian national programs has now passed 25. The number of clones pre-released and waiting for official release or left to the plantings by farmers, a very typical manner of varietal dissemination in many parts of the world even now, far exceeds this number. In Thailand where statistics are available on the area planted with each cultivar from data collected by the Department of Agricultural Extension, the total area planted with five new cultivars is given as 376,250 ha in the 1995/96 planting season (Figure 5). In Indonesia, hard data and an extrapolation from these indicate that new cultivars are now planted in more than 1 10,000 ha (Palupi etal., this Proceeding). In Vietnam, where the CIAT collaboration started much later than in other national programs, progress has been highly satisfactory, and good observation data show that the area planted with the new cultivars was more than 6,500 ha in the 1995/96 season, and is now estimated to be 15,000 ha in the middle of the 1996 planting season (Figure 6 and also see other reports in this Proceeding). In the Philippines where no detailed data are available, the area with new cultivars is estimated to be at least 3,000 ha (see Mariscal and Bergantin, in this Proceedings). Thus, the total area planted to CIAT-related cassava cultivars in Asia appears to have just passed the half a million ha line this year. Economic Effects I used the sales value of the products as the basis for measuring economic effects. Value of sales includes production costs and net profits. The gross production costs are made up of expenditures for labor, equipment, supplies, maintenance and depreciation. For regular commercial operations, net profit rather than the value of sales may be more critical. However, in the national development context, value of sales may be a more important indicator of the socio-economic benefits, as it represents employment, purchasing power and the availability of useful products. To concentrate mainly on direct monetary effects and avoid possible duplicates, I used only two measures for the economic gain; additional (or reduced) field production of fresh cassava roots due to the higher (or lower) yield of new cultivars, and the additional factory starch (or chips) production due to the higher starch (or dry matter) content of new cultivars compared with the traditional cultivars. The additional factory profit generated by the gross additional availability of raw materials is not accounted for, let alone the additional production of numerous secondary products. Thus, the real gross additional economic benefits are much greater than described in these simple analyses. N© in o E Q. 1 3 .■5 cs 15 E © © U H < E .2 v. t/3 CB o i_ XI E o no On On On On £ £ rm On On £ On ID On C 3 Ou ^gooooi/ienr^oooorpno — m — ooTtgg - °i "*. °1 ^ "1 "\ K r- & cn" —" cn" no" oo" cn" ^ ^' ^S t- r- ** >T> —' —■ —i O en ITl On oo r— en — oo r- r- On On CM On" i/">" en" UO CO 7± ON -rt tr> V m CM nO r- - *"" ® ^ ""* ~ m" no" m" en" cm C m uo ^o r» ^ - "1 °i. R. °.VO £J *»" "* •* —* £": cm —< © r- o © EJ in cm o on r- © - T R. T ^i _„ °i ^ On" \o" T* T*" cm" —" CO On — 00 —* — O On 00 in r~ — r» VO On o* ^t m en r- cm ~ O On O VO rr O On i/O CO "^ 00 On 00 r-- s §S8 oo i/-> en o —i O —T —T CM en —< en — — •o -2 ™ c « [Si 8 CTJ .n, .c/5 5 c S J= 2 2 -J £ 22 -s "3 ^ — u > £ S £ c/5 uo o p- On" 5 00 00 VO On" CM CM 00 cm" CO nO CO p- s CO nO r* On" CO rn CO nO CO 5 CM £ 3 £ I/-N Mm ■* o On X) on S 3z en » o On c 3 O 0 00 CM On O —i On © in r-^ c^ en en —i cnT rn "-T ?gto oo enoo m r-- 00 (N 2g ITl O no r- © ts On en © © —< © no —< en un Tt © oo r- r- oo — oo r- #n *. r. r. ft en Tj- -j- —i -- N m m —• On On -J- en r-^ ■* 3 on - m _ „ t» "* u-> • Es io ■* _ en © © r- no u-i r~ es e-4 O O O -H ITl I/"> in £l On t- r- r^ VO 2 .B E S J .9 83 £ ^ u -■ S3 s J3 .2 U>J u^■o 2 .2 .? « -a >> — g<§ s oo en o Ons CI 00 r- nO r- i/-f en oo On 0O en en 176 190 -. 180- j-. ne .eu I 160- Q i I50" I 14°"'?. I 130- u "I 120- u Cb 110- 100- 90- Breeding population mean Control (Hanatee) 1982 -T— 83 84 85 86 -r- 87 —I- 88 89 90 -I- 91 -J— 92 —r— 93 94 95 Year of planting Figure 1. Change in the mean fresh root yield of the breeding population (all entry mean yield trials) in relation to that of Rayong 1 from 1982 to 1995 at the Thai- CIAT collaborative cassava breeding program in Rayong, Thailand. 177 o 120'o u C s O 110- 1 E o o en 100- 90" Breeding population mean —I 1 1 1 1 1 1 1 1982 83 84 85 86 87 88 89 Year of planting —T" 91 —I— 92 -r- 93 -1— 94 —i 9590 Figure 2. Change in the mean dry matter content of the breeding population (all entry mean in yield trials) in relation to those of the control varieties from 1982 to 1995 at the Thai-CIAT collaborative cassava breeding program in Rayong, Thailand. 178 140 n 130- o 120 O U■*-< B 8 no es E .2 100 90" 80 J Control (Hanatee) Breeding population mean —I— 1982 -T— 83 84 85 86 -1 1 1 1——1— 91 1——1— 93 —1 9587 88 89 90 92 94 Year of planting Figure 3. Change in the mean total biomass production of the breeding population (all entry mean in yield trials) in relation to Rayong 1 from 1982 to 1995 at the Thai-CIATcollaborative cassava breedingprogram in Rayong, Thailand. 179 >■ u 150 n 140- 130- 120- 110- 100- Breeding population mean Rayong 1 Control (Hanatee) —T 1 1 1 r~ 1982 83 84 85 86 "T" 87 "T" 88 —|— 89 92 -T- 93 -1 9590 i 91 94 Year of planting Figure 4. Change in the mean harvest index of the breeding population (all entry mean in yield trials) in relation to Rayong 1 at the Thai-CIAT collaborative cassava breeding program in Rayong, Thailandfrom 1982 to 1995. 180 400n 300- I 200 I 100 0J 1 r 85/86 86/87 87/88 88/89 89/90 90/91 91/92 92/93 93/94 Year 94/95 95/96 Figure 5. Change in area planted with various new cassava cultivars in Thailandfrom 1985/86 to 1995/96. Source: Department ofAgricultural Extension. 00 a - 2^ O On I- 5; On (Pq 0001) SJBAI)|IU> BABSSK) M3U qjIM pdlUR)d B3jy 182 For the yield and starch content advantage (or disadvantage) resulting from the adoption of new cultivars, I used the means of many regional trials and on-farm trials (Tables 3 and 4). The absolute yield figures represent what advanced farmers would be able to obtain in their fields and the yield difference from the traditional cultivar, which actually matter more, would represent what could be obtained under a wide range of farming conditions. For on-farm price of fresh cassava roots and factory starch price, I used the low price in each 2-5 year period from 1985 to 1996. One nagging question that always arises when discussing cassava production is the issue of marketing. Such questions as "Can the farmers always find the market?" or "Can you assume that all the producers sold their fresh cassava at the market price?" are valid and these are real concerns, especially for small farmers who grow cassava for fresh human consumption. We have been dealing mainly with the increasing commercialization of cassava, i.e. the processing of starch or chips, and we have been developing new cultivars mainly suited for factory processing. Most farmers who adopt new cultivars are growing cassava as a cash crop for factory processing. The cassava market for starch processing is now rapidly expanding in Asia. It must be reasonably safe to assume that all the additional productions herein reported have led to additional cash income almost in their entirety, and the unabating adoption of new high-yielding cultivars is indirect proof for this. The subsequent analyses show that cassava is clearly an attractive means for small farmers to improve their cash income. In Thailand the varietal adoption started with Rayong 3, which has a higher starch content that the traditional variety Rayong 1 but may produce lower fresh root yields. Farmers who adopted Rayong 3 have been actually losing money by the sales of fresh roots but gained money from its higher start content. Only recently, when new cultivars that have both high fresh root yield and high starch content started to replace Rayong 3, did benefits result from both higher starch and higher root yields (Figure 7). The additional economic effects caused by the higher starch content has been highly significant. Until recently, virtually all the economic benefits in Thailand have been due to the higher starch content of new cultivars. In Indonesia, the advantage of new cultivars is clearly in both fresh yield and starch content; thus, the additional fresh root yields in the fields and the additional starch production in the factories both produced substantial economic gains (Palupi et al., in this Proceedings). -C i a r=C/3 <5 O *»* JS c 3.2P <0 > 1 2 S 2 '3 X) 0 3 60 0 0 >•i .a8 c 0 2 c Uh rlSt JS60 '3 2 0 <« c a JS a^o 0 60 8 JS c C > Uh cs TD 'c? S OS 0 <§ 1i u. •J? H-■ S3 8 "— s JS0 '3 >. S3 4-*c a>■■-■ c 0 C/5 +-■ 0 0 0 V OSe e E -0 c 3 H-■ £ 0 0 ^ 3 Uh -C «5 1> >• -■—■3 hi O u- « > ca<*> <4- O 0 S3O 1> 2 3 > D •0 Uh ■a5 •« >- en 3 V > s •■-( OS 3 H U P cUm 3■■8 NO Os Os E 0 IT) > r~ no 0 00 N 00 m CMca 1 u-> O so en' ci Os Os s© I-. Os i {/$ 0 •g -c § fe§ NO Os 4= t> 0 1 , NONrtrt i*n os «/-> Tt nn C/3 1 > 1 ? tj 60 a C3 c «n Uh 3i-• Os X T3 D Os Os 0> £2 8•s—• © © r- 00 © — 0 Uh so iri 0> 3 en> 4> «<^ M >• <*- «s r- CD u •* —i■a lyi c 0 E c •J3 Tt NO — CM Tj- 13 3 ^ 0Tradil 9TO 9TO 99 99 99 c e lowes ithRaj g '5b a> OS OS D >• > Jh # 0 E X) -«> ■s — en so Os w IT) 60 60 60 60 3 60 Q O ""> 0 ■8 60 3>Basedon comparison CA 3 Oc c c c is c CS ■3 O O O O S O rti Cv c co cd cc ccj w ca JS OS OS OS OS ix. OS oo c ! t "5 u I c I a. o at "8 S c .3 ex c o E ■a c e OS 3 CS o 3 es■O JO CT3 60 c o s J3 S u 2 M e 5 8 3 eo — *;— - ^8 E p c sa> o 00 o o o <-l —. 8 8~ a: 8 s i_ >^ -—• O 4> S3 S3 _>■*—. 3 u OO in o oo m q oo — •* r-4 O 13 c o ■a c o C E 3 oo •a < 5 ooin o o oo in (N ro q rn q q oo o o o CM q (N 13 c o o 1 S18 I ■s 3 O oo >6 CO On i s 3 q q On o q On o o 13 c o ■s H o no in u > si S£ oo o m q q On CM o o 13 c o ■3 q in ■a o x: £ gc 1/3 .EL O .9* 60 IE Z a. E 3 E o o « .2 I > g .5 c On On c° 3 E35 on ■s 3 ^ s s 185 In Vietnam, more money has been made from the sale of planting stakes of new cultivars than from fresh root harvests and higher starch content, although the benefits caused by the additional fresh root production and the additional starch production will probably outstrip the stake sale as of the 1996/97 season (Figure 8). Throughout all this process, the advantage of higher starch content was highly significant. It is the higher starch content rather than the higher fresh yield of new cultivars that accelerated the adoption of new cultivars. The total economic benefits due to the superior yield and quality of the new cultivars, accumulated up to now, is estimated to be 479 million US dollars. In this article I am analyzing primarily the economic effects caused by the additional root yield and additional starch content of the CIAT-related new cultivars. Yet, when we consider the history of the whole cassava processing industry in Asia, the establishment and development of large export markets for cassava products by Thai entrepreneurs, and the selection of cultivar Rayong 1, which has been the backbone of the Thai cassava industry, are by far the most significant factors. The monetary contributions of these must be in the billions of US dollars. The greatness of Rayong 1 is not limited to the immensely successful field production in the past two or three decades, but also extends to its role as an effective cross parents, as it produced Rayong 60 and Kasetsart 50, among others, through hybridization with the introduced parents from CIAT/Colombia. The contribution of Thai farmers who selected Rayong 1 should always be remembered. Profit to Small Farmers Virtually all the cassava production takes place in small farmers' fields and all the roots in Thailand are sold to cassava processors. In Vietnam also, all the cassava is produced by small farmers, and at present, those advanced cassava farmers who have adopted the new cultivars sell virtually all the harvested roots to the processors. In Indonesia and the Philippines, some cassava is produced in large plantations; yet, the majority of production takes place in small farmers' fields. Thus, we can assume that virtually all the additional economic benefits generated by the higher fresh root yield of new cultivars are entering directly into the pockets of small farmers. On the other hand, how much of the additional profit generated by the higher starch content of new cultivars is shared by the farmers depends on what differential price starch factories (or chipping plants) pay to the farmers. If the factory pays the same price for roots of different starch content (in other words, does not honor or recognize the breeders' and farmers' efforts to produce higher quality products), all the additional profit will be absorbed by the factory. Analysis of a price scheme used at a starch factory in Rayong, Thailand indicates that more than half of the cost of starch production is the cost of fresh cassava, even using cassava roots of high starch content (Table 5), suggesting that in general a fair 186 price is paid to the producers. More interesting is the question on what proportion of the additional starch produced due to the higher starch content of the new varieties is returned to the farmers (Table 5). Since the factory can benefit from the increased amount of final product, its business can still be viable even if the factory fully (100%) compensates the farmers for the higher starch content. In effect, most factories return between 55 and 100 % of the value of additional starch production caused by higher starch content of the raw material to the farmers. All in all, the scheme is not entirely unfair to the farmers. The initial adoption of Rayong 3 testifies this (Figure 5). While the farmers were losing money by the reduced sales of fresh roots, due to the lower fresh root yield of Rayong 3 compared with that of the traditional cultivar Rayong 1, they must have been receiving reasonable compensation from the factories for the higher starch content of their products. If the factories had not given a fair compensation, the planting of Rayong 3 would have halted immediately. We can safely assume that a substantial portion, more than a half, of the 480 million US dollars so far generated by the adoption of new cassava cultivars has entered the household income of small farmers. Social Effects In addition to the very basic expectation of strengthening the cassava status as an efficient crop for increasing the cash income of small farmers, our work is expected to contribute to the improvement of the efficiency of factory processing as well. Higher production and processing efficiency could lead to a higher competitiveness of the products domestically and on international markets. All of these are taking place already although the magnitude of the effect differs among countries. Cassava is often cited as a crop conducive to soil degradation. Much efforts are directed to identify and enhance adoption of soil management practices that lessen soil erosion or mineral nutrient exhaustion. While individual methods are important and indispensable components of soil management, a more fundamental requirement is to first upgrade the economic situation of farmers, so that farmers can consider the long term future of their farming operation and start adopting better soil management practices. As cassava processing becomes more lucrative, there will be stronger concerns for making the whole cassava production and processing industry more sustainable. It is yet to be seen whether the additional production efficiency generated by the new cassava cultivars would eventually add up to cutting the vicious cycle of poverty and environmental degradation. Epilogue It is fortunate that I have been able to work with the same basic strategy and the same fundamental objectives throughout the nearly a quarter century history of the CIAT Cassava Program. Generating useful breeding materials, offering them to national 187 programs and participating in their selection have been the basic style of the CIAT cassava breeding program. Many of the national breeding programs represented in this Workshop have been highly responsive to utilizing our materials and expertise. They are now highly successful in selecting truly useful cultivars and contributing to the farmers. We are proud that, among so many so-called research networks, our Asian cassava breeding network is one that is producing truly meaningful results. Now that we are beginning to see the results of our collaborative work, it is all the more fortunate that we can measure the achievement of our long time goals by the same original criteria defined when I was full of boyish naivety and idealism 25 years ago. Acknowledgement None of the good results quoted in this report is the work of CIAT alone. They all are the result of our collaborative work with national programs. My personal involvement in producing research results ranges from nearly total, such as the development of Rayong 60 and Rayong 5 in Thailand and KM-60 in Vietnam, to marginal, such as the release of SC8002 in China. The breeding materials and technical data thus produced in our collaboration with national programs primarily belong to the national programs. I appreciate the good collaboration of national cassava breeding programs in Asia. Among these, the contribution of the Rayong Field Crops Research Center, Department of Agriculture, Thailand, has been particularly significant in jointly producing advanced breeding materials to be contributed to other Asian programs, as well as to enrich the germplasm and breeding populations back at CIAT Headquarters, Colombia. We are all greatly indebted to the collaboration from the Thai cassava research community. 188 Table 5. Fresh cassava and starch price scheme and distribution of additional profit caused by higher starch content of cassava roots between farmers and starch factories in Thailand. Raw material cost as proportion of Share (%) of additional profit caused Root starch Fresh cassava value of starch by 1 % higher root starch content content" price produced3) (%) (Baht/t)2) (%) Fresh root seller Starch factory 30 920 51 65 35 29 900 52 65 35 28 880 52 64 36 27 860 53 63 37 26 840 54 62 38 25 820 55 61 39 24 800 56 61 39 23 780 57 59 41 22 760 58 58 42 21 740 59 57 43 20 720 60 56 44 19 690 61 83 17 18 660 61 82 18 17 630 62 81 19 16 600 63 80 20 15 570 63 79 21 "Determined by Reihamnn scale. 2)Price data for different starch content at Rayong in Aug 1996; US$ 1 .00 = Bant 25.00. 3)Based on starch price of US$ 240/t. 189 300- Total gross additional sales caused by adoption of new cultivars c ^o o. o•a a .e a > se loo 's 100- -o 1 1 1 1 1 1 I 1 I 1 1 1- 1985/86 86/87 87/88 88/89 89/90 90/91 91/92 92/93 93/94 94/95 95/96 96/97 Year Figure 7. Additional economic effects generated by the adoption of new cassava in Thailand between 1985/86 and 1995/96. 1 (je||op sri tI0!II!1") sjB,\i||nj A\au jo ucMjdopE ai|i oj anp s.>[i:s |euoi)ippe pajBinuinaay 5 3 1 3 3 2 c I ■S 2 e c s is 2J •2P 191 CASSAVA AGRONOMY RESEARCH IN CHINA Zhang Weite1, Lin Xiong\ Li Kaimian1, Huang Jie\ Tian Yinong2, Lee Jun2 and Fu QuohuP ABSTRACT Through cooperation with CIAT, national programs in China have paid more and more attention to cassava agronomy research and a lot of progress has been made. This paper describes briefly the most common cropping systems and production practices presently used by farmers. It also reviews the results of research on soil/crop management, with emphasis on soil fertility maintenance and soil erosion control, during the past decade. In China, cassava is usually planted from Jan-May, while it is harvested from Nov-Mar. The plant population is about 10,000-12,000 plants/ha under normal conditions, while the population could increase to 15,000-18,000 plants/ha in poor soils. Results of soil erosion control trials showed that two treatments, i.e. no-tillage but making a planting hole (30x30 cm) by hoe, or complete land preparation followed by contour ridging, not only increased cassava yields, but also decreased soil erosion. However, soil loss was very serious with complete land preparation but without ridging. Long-term fertilization trials conducted in CATAS, GSCRI and UCRI, indicate that cassava yields increased significantly with the application of N and K. Increasing the N application from 50 to 200 kg/ha while maintaining a constant rate of 100 kg K2O and 50 kg P2Os/ha, cassava yields increased significantly, but the root starch content decreased. In contrast, both cassava yield and root starch content increased with an increase in K application from 50 kg to 200 kg K2O/ha, while maintaining a constant rate of 100 kg N and 50 kg P2Os/ha. Cassava intercropped with watermelon produced the highest economic returns, but soil erosion was controlled most efficiently by intercropping with peanut. INTRODUCTION The Status of Cassava Production and Development in China In China, cassava could be cultivated in the areas south of the Qinling mountain and Huaihe river, with mean annual temperatures above 18°C and a frost-free period of more than eight months of the year. Therefore, there is a tremendous potential for further expansion of the cassava production area. The total cassava area in China has now reached 450,000 ha. Recently, cassava processing and utilization have developed more in-depth, and the production of cassava-based products has increased from 30 to 70% . Cassava is turning into a very important upland crop in the southern part of China. The production of cassava has changed from a scattered and backward crop into one that is farmed intensively. According to statistics in Hainan, there are now ten enterprises cultivating 100-300 ha of cassava, and the number of farmers who plant 3-10 1 Chinese Academy ofTropical and Agricultural Sciences (CATAS), Danzhou, Hainan, China. 2 Guangxi Subtropical Crops Research Institute (GSCRI), Nanning, Guangxi, China. 3 Upland Crops Research Institute (UCRI) of the Guangdong Acad. Agric. Sciences, Guangzhou, Guangdong, China. 192 ha has been increasing year after year. Thus, cassava production is moving towards more mechanization and more intensive farming, with greater use of chemical fertilizers and other farm chemicals. According to the statistics of Hainan, about 30% of the total farm work is mechanized, which is achieved by working in commodity production units; this has reached 70% in a few units. Over 60% of these units use herbicides and chemical fertilizers. Yunnan province is a new area for cassava production. According to the statistics, in 1990 the total cassava cultivated area was 7,000 ha. However, the area under cassava has recently increased as more and more planting material was introduced from Guangxi. Presently, in Honghe district of Yunnan alone, the area under cassava has reached 16,000 ha. It is estimated that the cassava area in Yunnan will be 20,000- 30,000 ha. Improved Varieties Research on cassava breeding and cultivation in China began in 1914. Most of the cassava varieties have been introduced from abroad, like those presently known as SC201 and SC205; these were both introduced to China in the 1940s and 50s. At that time the number of cultivars was very limited. Since 1958 China has been collecting and evaluating cassava germplasm, and has been doing research on cassava breeding and agronomy. The edible cultivar SC6068 has been released and popularized in the early 1980s. In recent years, with the accelerated development of cassava production and processing, new varieties with high yield potential and starch content are urgently needed. In Guangdong, Guangxi and Hainan provinces, local institutes started research on cassava breeding. With the cooperation and support of CIAT, and after many years of hard work, CATAS released SC124, SC8013 and SC8002, and has selected some advanced clones, such as OMR33-10-4; Guangdong has selected Nanzhi 188, SC8002, and more recently 901531; while Guangxi selected SMI 113-1. CASSAVA AGRONOMY RESEARCH IN CHINA Cultural Practices for Erosion Control During spring and summer soil erosion is serious in southern China, due to the monsoon climate, characterized by high temperatures and heavy and concentrated rainfall . Farmers usually build terraced fields or practice contour ridging to prevent soil erosion. This method has been mostly adopted in Guangdong and Guangxi provinces. But in the plains or in areas with gentle slopes, which usually have a high population density and thus less available land, farmers generally plant cassava intercropped with early-maturing and short-statured crops. This is not only very effective in reducing soil loss, but also increases the multiple crop index and has economic benefits. From 1990 to 1992, researchers at GSCRI in Nanning, Guangxi, conducted an 193 erosion control trial on the effect of different cultural practices on cassava yields and soil loss due to erosion. It can be seen from Table 1 that contour ridging with fertilizer application was most effective in reducing erosion, followed by intercropping with peanut and planting at closer spacing. Method of planting (vertical or horizontal) had no significant effect on erosion. Researchers at CATAS in Hainan also conducted erosion control trials and obtained similar results. It can be seen in Table 2 that twice plowing, followed by twice discing and contour ridging increased the yield by 5% and reduced soil loss by 38% compared to the same tillage method but without ridging. Preparation of either big or small planting holes decreased yields by 0.8 and 8.7% and reduced soil loss by 40 and 36%, respectively, compared with twice plowing and discing but without ridging. Plowing and discing increased yields compared with minimum or zero tillage but also caused more soil erosion. As farmers in Hainan tend to have more land but less labor, they usually plant cassava with zero tillage. This practice results in somewhat lower yields, but is quite effective in reducing erosion. In Guangdong and Guangxi, there are more people but there is less available land. On flat or gently sloping land farmers normally use tractors or cattle to plow and disc, and then plant cassava intercropped with other crops; this can increase yields and economic benefits, reduce costs and erosion, while maintaining soil fertility. Table 3 shows the results of erosion control trials conducted at CATAS. Fertilizer application with either contour ridging or peanut intercropping increased cassava yields by 37 and 26% and reduced soil losses by 42 and 29%, respectively, compared with the same non-fertilized treatments. Similar data from a soil erosion control trial conducted for three years at CSCRI in Nanning (Table 4) indicate that contour ridging, peanut intercropping and vetiver grass contour hedgerows were very effective in reducing erosion, while also increasing cassava yields. In recent years, in the gently sloping and highly populated regions of Guangdong and Guangxi, farmers plant cassava intercropped with early-maturing and short-statured crops, such as watermelon, peanut, soybean etc. This not only increases yields and gross income, but also reduces soil losses due to erosion. However, with the development of a market economy, in order to increase profits farmers tend to plant cassava with intensive land preparation and without any cultural practices to control erosion. Soil erosion may become the major restriction for cassava production in the future. 3 Z u OS•— uc (•2 « (J 8a c u — V j= S3 H I u 60 2 u> < 8! On o On On U 00 2 > < On On On On O On On I m On r- r-i o od 00 r-l nO On. r-i cn q in o r- o nO m NO On f> c*i o r-i is o (N in On nO On od m o\ On O O f r-i O r-i nO in m in On in od On NO in r-i O NO On C-l r- r-i r-i VO o in O (N I 8 3 _o 60 .9.9u 1 w M 00 3 60 c C c CO V 1 1a 1 B, 3 .5 1 00B 1d. 3 3 3 Q. 11 E B •c 0 13 os 3 .a D. s ■c c E t2 1*o o a u 00 XI 43 > JS J3 a. « , ^ 2 £ £ £ .9 J J .s € '€ -C '•E .€ c & £ '? "£ <£ & •£ B ■S ■5 B % * 60 73 i % in 1) i 8 C 8 ■si M> 00 60 3 60 60T3 -a TJ 0 T3 T3•c ■c ■c a ■c •c 0 c 0 0 0 0 z z z u Z z I } a s in I s 1 * 1 ■o .2 • siO ON 00 B ON o — E.g 2 o I CO =5 O Q to 3 60 2 > < g On On O On On On 00 u 60 2 > < o\ On On On O On On On 00 On C I nq ""> 00 (S (S ci oo © s nO fl nO 00 00 On O^ <* d in On On q IT> m °i nO 00 NO nO c-i O O o ■* 00 "* q r-^ On i On 0000 5 o oo r- On On r-i O Tt 00 On (N) oo m nO m ci 00 (N o^ O (N — nO ir> [-! oo On. d nO nO nO li"N, nO (NI On O On * *is c! c! t^ in ^H On _i C ** 00 t> ~^ o (N (N> *-H ■M <"H -h -h < ts SO oo 1 ""' * On On 4) r-3 ■* (N — (N SO Tt > <—< cN (N cs (N SO i i *w* ^H —* tN f*H ■■■) 2 13 On 00 m ■>» 00 ON no r" >* On ri d On 00 r-3 ^ ci,-H -* (S "* Cnl cn I o ri oo O cN t- (N oo oi On "i d r4 piOn ^H On r-* W Tt nO q N On nO ci J3 -o 60 60 60 60 60 60 1 C .9 .3 ca 3 O oo .5b 60 60 00 5b T3 -o -a •a -a •a C •c .C C ■B ■c -C o o o o o o o u a c c a c a « „ ^ at „ p £2 B E B E X is u S « S S 8 1a ■ ■* ■ 2 ■-* ■■* — r .€ € .c € r ^ ■8 •8 JJ JS J2 1*4 ■fl 5 •5 •S 5 5 c ^ ^ £ ^ ^ ^z 8. 9 c o c J! c © 3 ©n IS *;? ll 2 E ii © E 1 c U o 1* c* 3 2 y u ed H V 60 2 on cs < GO on On 00 rn 5S on ci -J n 00 —* 9 On rt ■* >/■> d cs s~\ < * "9 o XI 8 ^H CO l 00 t"« "* OC rn o\ a O en 00 ■*On On 00 pi d~* cs m- (N O (N s od 1 i (N od nq nq (N cn 2 - 2 J? i■c J o a a | 60 5 •9 a, 8 .3 + "Ieo a o On cn Xi On 00 o NO (N (N On 00 (N (N (N On o\ cn 00 nO n 5 nO (N n 00 1 (N Z On C~* nO 00 i E g00•o c 8 60 .9 •3 + 60 _o 0- .S E — V 1 B . a o 60 •T3 ■c o a eo _o a. o o > > I o 2- * Q 3 >u 198 Cropping Systems Cassava is an annual crop in China. The cropping systems can be classified as either monoculture, intercropping, interplanting, continuous cropping or rotational cropping. Monoculture is the major cropping system, and is the common practice in the mountainous or semi-mountainous areas, where there are less people but there is more land. The common method of land preparation is either reduced tillage or zero tillage. Land preparation consisting of plowing and discing increases root yields but also increases the costs and may cause serious erosion. Therefore, in sloping areas, planting cassava in a monoculture system with intensive tillage should be done only with contour ridging, in order to reduce soil losses by erosion. Intercropping and interplanting is usually carried out in the plains and on gentle slopes, where the population density tends to be high. Land preparation is intensive and results in a good income. According to an investigation done by GSCRI, when cassava was intercropped with watermelon, the income was increased by 30,000-75,000 yuan/ha, cassava intercropped with watermelon for seed increased income by 7,500-10,500 yuan/ha, while cassava intercropped with peanut increased income 4,500-7,500 yuan/ha. Table 5 shows that when cassava was intercropped with peanut or watermelon for seed, both cassava yields and income were increased, while this system will also reduce soil loss. This cropping system should be popularized. A cassava rotation cropping system is usually adopted in the mountainous regions, where the slopes are steep and soil loss is serious. Farmers rarely apply fertilizers, so, after planting cassava for 2-3 years, the soil nutrients are exhausted and yield have come down. In Yunnan and Hainan, farmers usually leave the land in fallow until the soil fertility has recovered. In the flat areas or gentle hills, farmers usually rotate with sugarcane or other crops. In sugarcane production areas, cassava is rotated with sugarcane after the latter has been planted for three years. This rotation can give good yields of both sugarcane and cassava. Time of Planting and Harvesting The climatical conditions in southern China are characterized by high temperatures and abundance of rainfall in the spring and summer, but rather cold and dry, sometimes with frost, in the autumn and winter. It is difficult for cassava to live through the winter. After many years of experience with cassava, farmers have determined the time of planting and harvesting that is most suitable for the climatical conditions in China. Cassava is usually planted in the spring, and when the temperature is high and there is abundance of rainfall in the summer, cassava grows rapidly. During the autumn the temperature and rainfall drop; this is a good time for cassava root production and the accumulation of starch. This is followed by the root harvest in winter. From 1990 to 1994, an expertiment was conducted at CATAS in Hainan to 199 determine the optimum time for planting and harvesting cassava. In this trial, cassava was planted monthly and was harvested at either 8 or 12 months, using two varieties and four replications. The result (Figure 1) indicated that when cassava was harvested at 8 months, the highest yield was obtained when cassava was planted in Feb-May. When cassava was harvested at 12 months, the highest yield was obtained when cassava was planted in May-June, but in two out of three years cassava yields were not greatly affected by date of planting. The highest starch content was obtained by harvesting in Dec-March, inrespective of whether cassava was harvested at 8 or 12 months. The root yield at 8 months after planting was positively correlated both with the average temperature and rainfall during the 3d, 4th and 5th months after planting and this relationship was highly significant (Figure 2). The starch content was significantly and negatively correlated with the mean temperature during the last month before harvest: r = -0.770~~ and r = -0.732" for harvests of SC205 at 8 and 12 months, respectively (Zhang Weite, 1996). Root starch content was also negatively correlated with rainfall during the month prior to harvest, but the correlation was barely significant (r = 0.48~ for SC205 harvested at 8 months). These research results correspond well with the actual farmers practice. It made it clear that the time of planting and harvesting in China as practiced by farmers is very suitable for the existing climatical conditions. However, the research results indicate that in Hainan province cassava can be planted almost any time of the year as long as roots are harvested after a full 12-month cycle. Table 5. Yield and gross income of two intercropping systems as compared to monoculture cassava in Qujiang county, Guangdong, China. Cassava Cassava Intercrop Intercrop Total gross root yield income yield income income (t/ha) (Y/ha) (t/ha) (Y/ha) (Y/ha) Cassava 13.50 4500 - - 4.500 Cassava+ Peanut 15.12 5040 2.5 5.000 10.040 Cassava+ watermelon seeds 18.00 6000 0.6 8.400 14.400 "Prices: cassava fresh roots: Y 0.33/kg peanut dry pods: 2.00/kg watermelon seeds: 14.00/kg Source: Qujiang Agric. Bureau of Guangdong, 1988 (personal communication) 200 32 h # S 28 c o o B 24 10 o o 20 * - harvest at 8 months harvest at 12 months j i_ M M N 25r 20 15 2 *-■ o o ■ 10 > to If) u> to O J L J_ _L -l_ J L J I MAMJJASOND Month of harvest Figure 1. Cassava root starch content (top) and root yield (bottom) averaged over three varieties and three cropping cycles, when planted during different months of the year at CATAS, Danzhou, Hainan, China, and harvested after either 8 or 12 months.- .c c o E CO 2 0) 25 20 15 *-* 10 o o CO > (0 w co U SC 205 0= 1st cycle 2d cycle 3d cycle Y = -29.507+ 1.785X r = 0.857** OW/- 18 20 22 24 26 28 30 Average temperature (°C) during 3d, 4th and 5th month after planting 1205 + 0.0641X 50 100 150 200 250 300 Average rainfall (mm) during the 3d. 4th and 5th month after planting Figure 2. Relation between root yield of cassava cultivar SC205, harvested at 8 months, and the average mean temperature (top) or rainfall (bottom) during the 3d 4th and 5th month after planting in SCATC, Hainan, China. Data are for 36 monthly plantingsfrom 1990 to 1993. 202 Planting Methods In China, the method of cassava planting is different in different areas. Horizontal planting is normally practiced in steep land with zero tillage, while inclined planting is mostly used in the soft soil after tillage; vertical planting is sometimes adopted in regions with strong winds. Horizontal planting is easy to do and requires less labor, but sprouting is slow and percent germination may be low, resulting in lower yields. Vertical and inclined planting requires more labor, but the sprouting of stakes is fast and the percent germination is high; the root yield is also quiet good. Table 6 shows the results of two planting method trials conducted at GSCRI from 1990 to 1992, and in CATAS in 1994. In both sites inclined planting produced highest yields, while horizontal planting produced the lowest. Vertical and inclined planting resulted in a higher percent germination at one month after planting than did horizontal planting. Ridging resulted in a lower percent germination than planting on the flat. Table 6. Effect of stake planting position and ridging on cassava yield and germination at 1 month in GSCRI, Nanning, Guangxi, and in CATAS, Danzhou, Hainan, China. Data are the average for SC201 and SC205 in CSCRI, and for SC205 and SC124 at CATAS. Planting Position GSCRI (1990-1992) CATAS (1994) Germination" (%) Root yield2) (t/ha) Root yield (t/ha) Horizontal -ridging 61.5 11.7 20.0 -no ridging 67.4 10.9 18.6 Inclined -ridging 66.4 13.0 25.3 -no ridging 78.1 11.5 16.9 Vertical -ridging 82.8 11.1 19.4 -no ridging 85.8 11.2 18.5 l) Average of 1991 and 1992 (no data taken in 1990) 5 Average of 1990 and 1992 (no harvest in 1991 due to drought) 203 Field Management of Cassava As cassava is more and more produced under intensive farming practices, the old way of field management is no longer suitable. Cassava field management will move towards mechanized farming and more use of chemicals. In China weeds are usually controlled 2-3 times per year. Hand weeding costs too much labor, the work efficiency is low, one person weeding in one day on average 250-330 m2 (30-40 mandays/ha). Due to the development of a market economy, the cassava planting area per farmer has been increasing year after year, now often being around 3-5 ha; in some areas more than 10 ha. Thus, hand weeding is no longer suitable for this type of cassava production. The use of pre-emergence herbicides is now in common use in Hainan; mechanized farming has also been introduced and has been practiced by some enterprises and companies, such as the Siyueten farm in Changjiang county of Hainan. Plowing, harrowing, stake cutting, contour ridging, planting, fertilizing, weeding, harvesting and root chipping are almost all done by machinery. In 1995 twenty ha and in 1996 one hundred ha of cassava were planted by machinery in this farm. The cost of production was reduced 30-40% and work efficiency has been raised 7-8 times. This is the way forward for intensive cassava farming in China. Fertilization The long-term fertility trials conducted in Guangdong, Guangxi and Hainan showed that when cassava was grown continuously in the same field, the combination of 100 kg N, 25 kg P205 and 100 kg K20/ha generally produced the highest yields and profits. In all three locations cassava responded mainly to the application of N, followed by K, while the response to P was generally not statistically significant, except at UCRI in Guangzhou (Figure 3). Application of 5-10 t/ha of pig manure in addition to modest rates of N, P and K further increased cassava yields in Nanning, but applications of "burned soil", i.e. a mixture of leaves and twigs slowly burned together with soil, had no significant effects on yield even at high application rates of 30-60 t/ha. Applications of K tended to increase the starch content of roots, applications of N tended to decrease starch while that of P had no significant effect (Figure 4). In a rate of fertilizer application trial, conducted in CATAS, cassava yields increased with increasing amounts of fertilizer applied; the application of 400-900 kg/ha of 15-15-15 compound fertilizer gave the highest profits, as shown in Table 7. Table 8 shows the effect of N, P and K, applied singly or combined, on cassava yields in a trial conducted at CATAS from 1989 to 1990. Combined application of N, P and K was better than that of any single nutrient, and the application of N alone or NK were better than that of P or K alone or in combination. Table 9 shows the results of a time-of-fertilizer-application trial conducted at CATAS in 1988. A single application at 30 days after planting or a split application at 204 30 and 90 days resulted in higher yields than later applications. When fertilizer application was postponed, the yield and the number of roots decreased. There was no significant difference between a single application and a split application using the same total amount of fertilizer. Table 7. Effect of different amounts of applied 15-15-15 compound fertilizer on cassava root yield and net income when cassava, SC205,was planted at CATAS, Danzhou, Hainan, China in 1988 and 1989. Amount of Cassava root Gross Fertilizer Net fertilizer yield(t/ha) income (Y/ha)" costs (Y/ha) income (kg/ha) 1988 1989 Average (Y/ha)2) 0 18.6 21.8 20.2 2828 0 2828 150 23.0 31.6 27.3 3822 75 3747 300 27.2 27.3 27.2 3808 150 3658 400 28.2 38.9 33.5 4690 200 4490 900 35.6 34.1 34.8 4872 450 4422 LSD (0.05) 7.2 4.2 (0.01) 10.1 5.9 1) Prices: Cassava fresh roots: Y140/ton 15-15-15 fertilizers: 0.5/kg 2) Net income= gross income minus fertilizer costs. 205 Table 8. Effect of N, P and K application, either single or in combination, on the fresh root yield (t/ha) of cassava, SC205, planted in CATAS, Danzhou, Hainan, China from 1988 to 1990. Treatments 1988 1989 1990 Average No NPK 15.0 23.1 17.5 18.5 N 16.3 29.5 28.0 24.6 P 20.0 25.3 21.7 22.3 K 19.3 28.6 19.7 22.5 NP 16.8 27.7 22.8 22.4 NK 21.8 31.1 33.7 28.9 PK 22.7 28.5 22.7 24.6 NPK 24.8 34.7 30.2 29.9 Table 9. Effect of time of application of fertilizers on cassava root numbers and root yield at CATAS, Danzhou, Hainan, China, in 1988. Root Root yield numbers/plant (t/ha) Check without fertilizers 8.5 14.5 Fertilizers applied at: 30 days after planting 60 days after planting 90 days after planting 120 days after planting Fertilizers applied at: 30 and 90 days 60 and 120 days LSD (0.05) (0.01) 11.8 27.2 9.0 24.8 8.5 24.2 7.9 22.0 11.1 27.5 9.7 23.7 2.3 4.9 3.1 7.5 206 30 Guangzhou 1992 □ = SC 201 • = j= ■S- ^° X* T, 20 ' P—tf mrp f ast^-CL >• fr //o A/ //o -/ //1* M > 10 4 50 P205 100 N 100 N u n 100 KjO III I 100IC,O iii i 50 P205 III 1 i i i 0 50 100 200 kg N/ha 0 25 50 100 kg PA/ha 0 50 100 200 o-o-o 100-50-100 200-100-200 kg KA'ha kg N-PA-K20/ha c c S3 u 30 r 20 Nanninp 1995 10 50 PA 100 K20 _l I D = SC 201 • = SC 205 J L 100 N 100 KjO J J_ 100 N 50 PA I L -L 50 N 25 PA 50K2O J_ -i_ 0 50 100 200 kg N/ha 0 25 50 100 kg PAs/ha 0 50 100 200 0-0-0 100-50 100 200 100-200 0 5 10 kg KA'ha kg N-PA-KAha t pig manure/ha 30 r 20 u CATAS 1995 10 50 PA J L -1_ _L SC 205 J L 100 N 100 K,0 _1— —1_ SC 124 J L 100 N 50 PA JL X. J_ J_ J L 50 N 25 PA 50K2O _i_ _i_ 0 50 100 200 kg N/ha 0 25 50 100 kg PA/ha 0 50 100 200 0-0-0 100-50 100 200-100-200 0 15 30 60 kg K2Q/ha kg N-PA-KAha t burned soil/ha Figure 3. Effect ofannual applications of various levels ofN, P and K on the root yields of two cassava cultivars grown for the 4th consecutive year at UCRI in Guangzhou, Guangdong,for the 7th year at GSCRI in Nanning, Guangxi, and for the 4th year at CATAS in Danzhou, Hainan. o o Z Q. * o w o in cm in in * o CM CM i— o O r^ Vl Vl o II II fN • < o N Z * o o o o o o Vl CO Q> c in 5 «- 2 (0 - z o> o .»: o o O O CM «j ,r O O O V O ni.* o o T— ro £ LO o o in oT m D> CM -a: ra £ "5 w T3 O c k. 3 .O o CM Id O a. .C o a .C o 0. 0) r (BL|/J) p|9jA 100J EAESSBO (o/,; juaiuoo ljojbis 208 Harvesting Methods Cassava harvesting methods in China can be divided into hand harvesting and mechanical harvesting. Harvesting by hand is the most common. Hand harvesting requires heavy work and has low efficiency, usually requiring 30 mandays/ha. Mechanical harvesting is rather light work and has high efficiency; on average one machine can harvest 0.4-0.5 ha per day. In the 1970s and 80s, Xijiang State Farm in Guangxi was a special cassava farm; over 70% of cassava farming work was done by machinery. But later the farm changed to cultivate other crops, and the machinery was left unused. In 1995, Siyueten Farm in Changjiang county, Hainan, began to develop cassava machinery. Now 80% of cassava work on that farm is done by machinery. Production costs have been reduced by one-third, and productivity has increased markedly. Stem Storage Due to the cold and dry weather in winter, sometimes with frost, the cassava harvesting time in China is normally from Nov to March, and planting time from Jan to April. Harvesting must be done before the onset of frost to guarantee that good quality stakes can be stored for use next year. In the north, the cold weather arrives earlier and lasts longer. During the winter cassava may suffer seriously from frost. Thus, in the north the cassava harvest time is from Nov to Jan and planting time from March to April. Thus, the period of stake storage is 4-5 months. In the south, the cold weather arrives later and lasts only a short time, so cassava rarely suffers from frost. Therefore, the harvest is usually done from Dec to March and planting from Jan to April. Thus, the storage method of stakes is different in the south and the north. In the south stems are usually stacked under the shade of trees, and then the stems are covered with dry straw. If the weather is very dry, water is splashed on the stems. Stem storage is light work and easy to do. But in the north, the stems are normally stored in soil trenches or pits. Trenches or pits are dug in a southern exposure and in more elevated areas. The stems are bundled and placed in the trenches or pits, and are then covered with straw and soil, allowing some air ventilation. It is important to assure that the trenches do not get waterlogged. Stem storage in the northern regions is heavy work, and if something is done wrong, it can cause stake damage. This is a principal sector constraint for cassava production in the northern part of China. General Recommendations Table 10 summarizes some general recommendations for cassava production in China based on results obtained from research conducted at various institutes during the past ten years. 209 Table 10. Recommended cultural practices for sustainable cassava production in China. 1. Variety: use high yielding and vigorous varieties: SC 201: tolerant to poor soils and cold climate SC 205: high yield in better soils SC 124: high yield and cold tolerant SC 8002: high yield and cold tolerant SC 8013: high yield and typhoon tolerant 2. Planting time: Guangdong/Guangxi: March-April Hainan: Feb-May or year-round if harvested at 12 months 3. Land preparation: plow once or twice at 15-20 cm depth with tractor or oxen; on gentle slopes: make contour ridges after plowing on steep slopes: plow with oxen or prepare planting holes with hoe 4. Planting material: select healthy 10-1 1 month old plants; store stems in frost- free location; cut 15-25 cm long stakes after eliminating dry portion of stem 5. Planting method: inclined or vertical with buds facing up, 5-10 cm deep; plant on ridges if soil is wet, on flat if very dry 5-10 t/ha of manure incorporated before planting 50 kg N/ha as urea 20 kg P205/ha as simple or triple superphosphate (or thermo-phosphate if soil is low in Mg) 50-80 kg K20/ha as KC1 6. Fertilization: N, P and K applied at first weeding at 1-2 months after planting. 7. Weeding: 2-3 times manually, at 30-45 days and 2-3 months later; or pre-emergence herbicide right after planting and post-emergence herbicide (using shield) at 2-3 months after planting. 8. Harvest: at 10-12 months; incorporate leaves and stems back into the soil. 9. Intercropping: cassava may be intercropped with two rows of peanut or mung- bean, with one row of maize or with watermelon, squash etc. 10. Erosion control: plant 1 row of vetiver grass (10-15 cm between plants) along the contour, with 1 m vertical distance between contour lines. Cut leaves back to 30 cm above ground at planting of cassava and at 4-6 months; spread the cut leaves on soil surface between cassava plants. 210 REFERENCES Liu Jianpin. 1994. Cassava production research in Hong He district, Yunnan province of China. (in Chinese) Qujiang Agricultural Bureau. 1988. Investigation of cassava intercropping systems. (personal com.) Tian Yinong and Lee Jun. 1992. Production and development of cassava in Guangxi. Paper presented at 2nd Chinese Cassava Workshop held Oct 19-24, 1992 at CATAS, Danzhou, Hainan, China. Zhang Weite. 1992. Progress in research on cassava agronomy and utilization in China. Paper presented at 2nd Chinese Cassava Workshop held Oct 19-24, 1992 at CATAS, Danzhou, Hainan, China. Zhang Weite. 1996. Summary of experiments on time of planting and harvesting of cassava conducted at CATAS from 1990-1994. Research on Tropical Crops No. 66:22-27. CATAS, Danzhou, Hainan, China. (in Chinese) Zhao Qiguo. 1987. Soil resources in upland areas of the tropical and subtropical zones of China. Soil: 19:1. (in Chinese) 211 RECENT PROGRESS IN CASSAVA AGRONOMY RESEARCH IN THAILAND A. Tongglum1, V. Pornpromprathan1 , K. Paisarncharoen2, C. Wongviwatchai2, C. Sittibusayd, S. Jantawat*, T. Nual-ons and R.H. Howeler* ABSTRACT Cassava agronomy research in Thailand during 1994-1996 emphasized mainly erosion control, soil fertility maintenance and weed control. Regarding soil erosion control, the planting of cassava in April, at the start of the rainy season, and harvesting in March resulted in the highest soil loss compared with other planting dates. Planting at the start of the dry season in Dec and harvesting in Nov produced by far the highest cassava yield, while soil loss due to erosion was relatively low. Intercropping cassava with either peanut or pumpkin were found to be the best cropping systems to reduce erosion while also giving high gross incomes. Cassava planted on contour ridges at closer spacing (1.0x0.6 m) and with fertilizer application was the most promising package of cultural practices for reducing erosion and increasing yields. Soil fertility maintenance through the use of legumes grown for in-situ production of mulch was studied at Rayong Research Center. The highest yield of cassava, 46.17 t/ha, was obtained when Crotalaria juncea was planted as a green manure and mulched, followed by planting cassava, which was then harvested after 18 months. When Canavalia ensiformis was intercropped with cassava, cut at 2 months and left as a mulch, the yield of cassava, harvested at 12 months, was as high as that obtained with a high rate of chemical fertilizers. The sequential planting of fertilized Rayong 60 after unfertilized peanut, produced the highest yield at Kalasin in the Northeast. The cultivar Rayong 5 planted in either Sattahip or Banbung soils in the East produced a relatively high yield with application of 312 kg/ha of 15-15- 15 together with 100 kg of urea and 78 of KCl/ha. In the Northeast the application of 25-25-25 kg/ha of N, P2Os and K2O produced a significantly higher yield of Kasetsart 50, Rayong 5, Rayong 60 and Rayong 90 than without fertilizer application. Research on the long-term effect of soil management on cassava planted continuously for 15 years in Khon Kaen in the Northeast, showed that when cassava was rotated yearly with sequentially planted peanut and pigeonpea, this could maintain a relative cassava yield of 87 % of that obtained in the first year. Similarly, the application of soil amendments (lime, rock phosphates and compost), as well as that of soil amendments with chemical fertilizers, could reduce the rate of yield decrease over time. However, after 15 years of continuous cropping the cassava yields in all treatments were lower than those obtained in the first year. Research on the optimum period of weed control for Rayong 60 and Rayong 90, planted in both the early and late rainy seasons in the Northeast, indicate that both cultivars need to be free of weeds at least three months after planting in order to produce high yields. 1 Rayong Field Crops Research Center, Huai Pong, Rayong, Thailand. 2 Khon Kaen Field Crops Research Center, Khon Kaen, Thailand. 3 Soil Science Division, Dept. of Agric, Chatuchak, Bangkok, Thailand. 4 Kasetsart University, Chatuchak, Bangkok, Thailand. 5 Kalasin Field Crops Research Station, Kalasin, Thailand. 6 CIAT Asian Cassava Program, Dept. of Agriculture, Chatuchak, Bangkok, Thailand. 212 The most appropriate weeding method for cassava intercropped with either mungbean or peanut, was the application at planting time of the pre-emergence herbicide Metolachlor at the rate of 1.50 kg ai/ha, followed by spot application of the post-emergence herbicide Paraquat at the rate of 0.50 kg ai/ha whenever necessary. INTRODUCTION The Thai cassava breeding programs have recently released several new cultivars with high yield potential and high starch content, such as Rayong 90 and Rayong 5 and Kasetsart 50. Thus, the lack of good cultivars is now less of a problem to cassava farmers. But, appropriate cultural practices are still needed to attain the high yield potential of these cassava cultivars. Most research on cassava cultural practices has focused on soil fertility maintenance/improvement in those regions where the crop is grown continuously, with somewhat less emphasis on appropriate cultural practices to obtain high yield or income. Regarding the decline in soil productivity by continuous production of cassava, research has focussed on erosion control and soil fertility maintenance, while research on cultural practices for the newly recommended cultivars, such as optimum rate of fertilizer application and weed control for both monoculture and intercropped cassava, had also high priority. Soil Erosion Control Research on cassava soil conservation has continued since the last Workshop in 1993. Several erosion control experiments were conducted at Rayong Field Crops Research Center and in Pluakdaeng town of Rayong province, where cassava experiments were planted on 4-6% slope with highly erodable sandy loam soils. The objective of these experiments was to quantify the effects of different planting dates, intercropping and various cultural practices on both soil erosion and cassava yield. Eroded sediments were collected from plastic covered channels dug along the lower side of each plot. At monthly intervals the sediments from each plot were weighed and samples were taken to be dried, in order to determine soil losses on a dry weight basis. Results of some experiments conducted from 1994 to 1996 are as follows: 1. Effect ofplanting date on cassava yield and soil loss Table 1 shows the effect of cassava planting date on the root yield and soil loss in an experiment conducted on 4.2% slope at Rayong Field Crops Research Center. The cultivar Rayong 90 was planted at six different times of the year; there were no replications. Rainfall received in the different planting date treatments varied from 777 to 1893 mm, while the total dry soil loss ranged from 7.5 to 12.8 t/ha, and the cassava yield varied from 15.5 to 46.4 t/ha. 213 Table 1. Effect of cassava planting date on total dry soil loss due to erosion, and on root yield and starch content of Rayong 90 planted at Rayong Field Crops Research Center in Rayong, Thailand in 1994/95/96. Planting date Rainfall Total dry Plants Fresh Root (Planting-Harvest) received soil loss harvested root yield starch (mm) (t/ha) (#/ha) (t/ha) (%) June'94-May'95 777 7.68 9906 15.47 22.3 Aug'94-July'95 997 7.47 9906 24.68 20.0 Oct'94-Sept'95 1265 7.98 9806 38.09 29.2 Dec'94-Nov'95 1749 8.14 8175" 46.44" 28.0 Febr'95-Jan'96 1731 9.65 9038" 39.04" 33.0 April'95-March'96 1893 12.76 9806 37.52 28.6 "Some irrigation was used to ensure establishment during the dry season. Source: Rayong Field Crops Research Center, Annual Report 1996. The greater the amount of effective rainfall the higher the soil loss. Cassava planted in April and harvested in March received the highest amount of rain, which also caused the greatest soil loss of 12.8 t/ha and resulted in a root yield of 37.5 t/ha. But the highest root yield of 46.4 t/ha was obtained from planting in Dec and harvesting in Nov, which also received a lot of rain and caused a considerable amount of soil loss of 8.1 t/ha. Root starch content of Rayong 90 planted at different dates varied from a low of 20.0 to a high of 33.0% due to different months of harvest. The highest root starch content of 33% was obtained when Rayong 90 was harvested in the dry month of January. 2. Use of economic covercrops to increase cassava profitability and reduce erosion Table 2 shows the results of an experiment on the use of economic covercrops to increase cassava profitability and reduce erosion, conducted in 1994 and 1995, on sandy loam soil with 5% slope in Pluakdaeng of Rayong province. Cultivar Rayong 90 was planted in ten different cropping systems without replication. In the 1994/95 experiment, the effective rainfall was only 589 mm, compared with 1592 mm in 1995/96. Nevertheless, soil loss due to erosion was high, ranging from 9.1-35.6 t/ha in 1994/95, but only 10.6-19.9 t/ha in 1995/96. This is due to very heavy rainfall of 188 mm during Oct 1994, which caused serious run-off in all plots, but particularly in two plots of sole cassava and cassava intercropped with watermelon. In both years the results show the same trend: intercropping cassava with peanut was the most effective in reducing soil loss due to erosion, while intercropping with cucumber or pumpkin was intermediately effective. When the intercrops were planted one month after cassava, they were less effective in reducing erosion. CI 8 nO ON On en "-*■* ON On 3 II o -S *> W5 —) 2 ~ St ©< ^ E T3 « 2 u a £ M/ cO O 3 t£ WO.C £ E -2- o £ .c a 2 o H a 00 3 a E o c .— co u .-3 c 3 S? u O ,>r p ^* _o J, E s Si .5 a u » H ^ — m 00 nO O tr> Cn| — (S — *• -h on ^ m i"■ **• OnOnopv>rNio~*CNiwoo t--On©On00 — nOOnnOOn rt (Sl —I .H *o in -hOft fj © — Cn| ci O ' . On •—" — 00 On rn| r- in N oo^ooonOTt^^^ci -HonodM « — o ' oo m — On no\-r^l^ooonOfNiTt (NlN(N|6*~*(li(pi m (N —■ m tr> no — « Tt n a>ftnv>»ohoan On TJ- — Or4 — OOn. — On NCr-'t^t^tinoo-^ton CNjOnOOnOnr-nq-hint- rionwiNo'*oooi--cl6 ci rr, ,-*.-.„ „ „_-* ■S -s c c o E 0 c E o c ,_, — E F./- CO -~' c c c r: es c n .5 0 n u CO u u .2 u u u _c E .s ! J*> E 1 cEo 3.J 3 U E 3U1 3 3 I 3 3 3> E E u (A + + + + + + + + + C/) a co CO a 31 CO CO CO U > ;> > ;> > > ^ > > a co co co CO ecl CO CO■_t V, M yi V> sfl t^ :r, VI Ifl VI Ifl ^ cfl ifl a 01 co a M CO CO CO CO C/) U u P u U u G u 0 — NmtwonOi— ooonO > ^r 03 Onw2 yn «1 E « p . . ■si <3 > -55 « § "§2 g X E c u- w_^ JZ NNjinfall ainfal ablis ca i- *-* ^ ■«; <-" . « a> q 5* 15 & lO oo c/5 t/3 Ol On o °S d "O T3 oa CO 3 2> c c nO ^5On ON O On in no r3 ON On > ofinti ayong On On On On CTJ .J 2 CK- C C •^^ C i> •. ii u >, h ^J _ »- O a. 5 a Z 215 Table 2 shows that some cropping systems produced no yield of intercrops due to either poor establishment or due to competition from cassava. In both years, most intercrops failed to grow well when planted one month after cassava, but nearly all established well and yielded better when planted simultaneously with cassava, i.e. peanut, muskmelon, cucumber and pumpkin yielded 875, 369, 1 19 and 2900 kg/ha, respectively, while all failed when planted one month after cassava, except pumpkin; but in that case the yield was only 331 kg/ha. Cassava yields ranged from 4.9-19.1 t/ha. Due to rather dry weather in 1994/95, with only 589 mm of rainfall in the experimental period, there was strong competition between crops for soil water. Intercropping cassava with peanut resulted in the lowest cassava yield of 4.9 t/ha, but this treatment was effective in reducing soil loss. Nevertheless, total gross income from the various intercropping systems was generally higher than that of sole cassava. The highest gross income was obtained from intercropping cassava with pumpkin, both crops planted at the same time. In the 1995/96 experiment the same trends were observed. When the intercrops were planted simultaneously with cassava, the intercrops grew well and yielded much better than when they were planted one month after cassava. Peanut, watermelon, muskmelon, cucumber and pumpkin yielded 906, 2119, 7125, 5831 and 2900 kg/ha, respectively, but no intercrops planted one month after cassava could yield anything due to shading from cassava. Cassava yields ranged from 16.4 to 25.6 t/ha. Intercropping with peanut reduced cassava yields only about 15% due to less competition for soil moisture compared with the previous year. Total gross income for the intercropping systems with simultaneous planting was always higher than that of cassava monoculture. The highest gross income was obtained from cassava intercropped simultaneously with muskmelon. The gross income varied from year to year due to variations in the price of each crop. Considering both the effectiveness in reducing soil loss and the gross income obtained, the systems of intercropping cassava with peanut or pumpkin, both planted at the same time as cassava, appeared the most promising. 3. Cultural practices for erosion control on farmer 's fields Table 3 shows the results of experiments on different cultural practices to control erosion in cassava, cv. Rayong 1, conducted on farmers' fields in four locations in Rayong province in 1994/95 and 1995/96. Each trial compared five "improved practices" with the "farmer's practice" without replication. In all locations the soil had a light sandy loam texture with slopes ranging from 4.3 to 6.7%. The overall average soil loss due to erosion in the different locations in 1994/95 ranged from 5. 1 to 43. 1 t/ha, and in 1995/96 from 7.2 to 18.0 t/ha. In both years the soil loss depended on the rainfall and the slope at each location, except that in Nongrai soil loss was very low because of natural terrace formation as a result of land preparation in the same trial conducted during two previous years. 216 Table 3. Slope and rainfall as well as the average dry soil loss due to erosion in on-farm erosion control trials conducted in four locations in Rayong province of Thailand in 1994/95 and 1995/96. Slope Rainfall #Rainy Dry soil loss Location (%) (mm) days (t/ha) 1994/95 1 . DLD Center, Rayong 6.05 1071 80 43.10 2. Nongrai, Rayong 4.80 890 58 5.10 3. Nongbua I, Rayong 5.10 832 58 30.51 4. Nongbua II, Rayong 4.30 831 58 27.24 1995/96 1 . DLD Center, Rayong 6.05 1585 107 17.88 2. Nongrai, Rayong 4.80 1325 96 7.19 3. Mabka I, Rayong 6.70 1475 97 16.98 4. Mabka II, Rayong 5.50 1475 97 17.99 Source: Rayong Field Crops Research Center, Annual Reports 1995 and 1996. The actual "farmer's practice" varied from site to site each year. In the 1994/95 experiments, three of the four farmers planted cassava on contour ridges and all applied 15-15-15 fertilizers at a rate of about 206 kg/ha. In the 1995/96 experiment all farmers planted cassava on contour ridges and again applied 15-15-15 fertilizer at a rate of 206 kg/ha. Table 4 shows the effect of different cultural practices on cassava yield and soil loss in the 1994/95 and 1995/96 experiments. In the 1994/95 experiments, the average dry soil loss caused by different cultural practices ranged from 20.5 to 38.6 t/ha, while the cassava yields ranged from 11.7 to 18.8 t/ha. Best results were obtained from cassava planted at 0.8 x 0.8 m on contour ridges and with fertilizer application; this produced the highest cassava yield of 18.8 t/ha with the least soil loss of 20.5 t/ha. There were some questionable data on soil erosion at Nongbua II during the first month (July) with rainfall as high as 319 mm in 14 days; this heavy rain broke some contour ridges in treatment 3, thus causing higher than normal levels of erosion. To solve the problems of runoff water entering the plots from fields above, it was suggested to partially dig into the soil narrow strips of zinc sheet along the upper side of the plots. L. 3.8 • .S | "8 1 *8 £ .2 g is © E"S i_ ex; 5 S8fa- !~ s >o ■ C 1) — fa. c « .2 asi 0 S c "2 .2& ^^ 2£ 1 e c e e ,£8 5 jC o "5 ox> it 93 fa s. e t~ X)C3 w oj 9o i/"> nO — C"> On m r- in m (N nO oo nO oo m o o * On CO—-■ (N —■* -^ —■* O m o m o r- o o 00 r-- r- On o o oo Z r- X o3 m x> (N Tj- in On —' in no On r— -^- oo tn od h * 6 m (N fO (N (N IN (N o in oo in oo o (N no oo N m ,■" < ."2o + -a Zz u C » 'C ao 0X> u- D g O B 3O o !2 "G c .£ C tj u- o c o c 0 ed O Z U Z Q. Z *. n r. » E E E oo" ^3 E nO nO nO d D — c> d d X E X X X X 00 — — —< *— d u. VO Z XI X ed X XI S3 ca cd 00—- nO m in 00 00 nO oo m r-- r- I- m # ■*"^ •* r- *~* 00 in X H3 as CT3 CTJ 00 0O s 8 nO00rO m e*i H ro O <— nO nO OO ■* ■* "* T* S l£ 8 > U 218 In the 1995/96 experiment, soil loss from different cultural practices ranged from 8.6 to 26.7 t/ha and cassava yields ranged from 1 1.5 to 22.8 t/ha. The best results were again obtained from cassava planted at either 0.8x0.8 m or 1 .0x0.6 m on contour ridges and with fertilizer application; those treatments resulted in the highest cassava yields of 22.8 and 21 .7 t/ha, with relatively low soil loss of 10.2 and 8.6 t/ha, respectively. When cassava was planted without contour ridges the soil loss was higher, ranging from 15.3 to 26.7 t/ha. The results of these two years of experiments show clearly that contour ridge planting can reduce soil erosion in cassava fields while increasing cassava yields; they also indicate that fertilizer application will generally increase yields, but contrary to previous results obtained, it also increased erosion. Soil Fertility Maintenance Several studies on the maintenance of soil fertility in cassava fields through the use of green manures, chemical fertilizers or soil amendments were conducted in the East and Northeast of Thailand, where cassava is the major upland crop. The overall objective was to maintain or improve the soil fertility in fields of continuously grown cassava in order to sustain a high long-term productivity. Results of a study on soil fertility maintenance through the use of green manures conducted from 1988 to 1994 at Pluakdaeng in Rayong province have shown that legume species such as Crotalaria juncea, Mucuna fospeada and Canavalia ensiformis, when planted as green manures, could improve soil fertility and increase cassava yields (Tongglum et at., 1992; Sittibusaya et at., 1995). In the 1991/92 experiment these legume species produced 7.31, 4.83 and 5.17 t/ha of above-ground dry matter, respectively, which contained 123, 157 and 136 kg/ha of total N. The yield of eassava increased from 3.6 t/ha without green manure to 7.7, 7.0 and 6.0 t/ha when these green manures were used. It was found that incorporating the green manures into the soil had no significant effect on cassava yield compared to leaving the cut green manures as a mulch on the soil surface, resulting in average yields of 8.3 and 7.4 t/ha, respectively (RFCRC, 1994). /. Effect of green manures grown as in-situ production of mulch on cassava yield In 1994/95 a new study was conducted at Rayong Field Crops Research Center with the objective of determining the most appropriate management of green manures grown for in-situ production of mulch, which would produce high yields of cassava and maintain soil productivity. Rayong 90 was used as the test variety in the experiment and four species of green manures, i.e. Crotalaria juncea, Canavalia ensiformis, pigeonpea ICP 8094 and Mucuna fospeada, were planted, in three systems: 1. simultaneously intercropped in cassava and cut off and mulched at 2 MAP; 2. interplanted into 6-7 month old cassava and mulched before the next cassava planting; or 3. planted as a normal green manure, cut and incorporated before planting of cassava, which would then 219 be harvested at 18 months. In addition, in two treatments cassava was intercropped with cassava which was either pulled up or cut off at 2 months and mulched. These various green manure treatments were compared with two treatments of sole cassava with fertilizer 13-13-21 applied at rates of either 156.25 or 468.75 kg/ha. Table 5 shows that among the green manures Crotalariajuncea always produced the highest above-ground dry matter, ranging from 1.44 to 9.89 t/ha, which contained 39.9 to 262. 1 kg/ha of total N. Pigeonpea was the next most productive green manure, followed by Canavalia ensiformis, while Mucuna fospeada failed to germinate. Crotalaria juncea, Canavalia ensiformis and pigeon pea planted as intercrops in a 6-7 month old cassava stand, produced higher above-ground dry matter, which contained also more N than other management treatments, since these green manures were left to grow and were cut only after 4.5 months (at cassava harvest). Cassava yields ranged from 8.75 to 46.17 t/ha. Very high cassava yields, ranging from 38.8 to 46.2 t/ha were obtained when cassava was harvested at 18 months for a 2-year cycle. The highest cassava yield of 46.2 t/ha was obtained when Crotalaria juncea was planted as a green manure and then cut and mulched before planting cassava, which was then harvested at 18 months. Of those treatments harvested at 12 months, lowest yields were obtained when cassava was intercropped with cassava, which was cut at 2 months at 30 cm above ground level to serve as mulch, as well as the interplanting of Crotalariajuncea into a 6-7 month old cassava stand, due to the strong competition from the intercropped cassava or Crotalaria. The other treatments produced cassava yields ranging from 15.9 to 23.8 t/ha, which were not significantly different from the treatment without green manure but with high fertilizer application. The intercropped Crotalaria juncea planted at 6-7 MAP cassava exerted a strong competitive effect on cassava, resulting in a yield reduction; intercropping cassava with cassava also reduced yields, not only by the competitive effect but also because the cassava intercrop produced the lowest amount of mulching material. According to the objective, the intercrop residues were left as mulch for the next cassava crop. However, the results in Table 5 show only the competitive effect on the yield of cassava in the first year, but the residues of these green manures are to be mulched for the next cassava crop, which may show a positive effect on cassava yield in the second year. With regard to the usual harvest of cassava at 12 months, when Canavalia ensiformis was planted as an intercrop at the same time as cassava, cut at 2 months and left as mulch, this increased the cassava yield to 26.9 t/ha, nearly as high as that of sole cassava with a much higher fertilizer rate of 469 kg/ha of 13-13-21. 220 Table 5. The effect of green manures grown as in-situ production of mulch on the yield of cassava grown at RFCRC in Rayong, Thailand in 1994/95/96. Treatment 1. Cassava+Fert. 13-13-21 (156 kg/ha) 2. Cassava+Fert. 13-13-21 (469 kg/ha) 3. Cassava + Crotalaria juncea (cut at 2 months) 4. Cassava + Canavalia ensiformis (cut at 2 months) 5. Cassava + Pigeon pea ICP 8094 (cut at 2 months) 6. Cassava +Mucuna fospeada (cut at 2 months) 7. Cassava+cassava (pulled out at 2 months) 8. Cassava +cassava (cut at 2 months) 9. Cassava + Crotalaria juncea (6-7 months) 10. Cassava + Canavalia ensiformis (6-7 months) 11. Cassava + Pigeon pea ICP 8094 (6-7 months) 12. Cassava+Mucuna fospeada (6-7 months) 13. Crotalaria juncea-Cassava(18 months) 14. Canavalia ensiformis-Cassava(\8 months) 15. Pigeon pea ICP 8094-Cassava( 1 8 months) 16. Mucuna fospeada-C&ssa\&(\8 months) LSD(O.Ol) - - - 13.45 F-test - - NS ** CV(%) - - 7.28 23.88 Note: Treatments 9-12: green manures were cut at 4.5 months (at harvest of cassava) Treatments 6, 12 and 16: Mucuna fospeada failed due to poor germination and stem rot Treatments 3-16: 156 kg/ha of 13-13-21 were applied to cassava Treatments 1-12: cassava was harvested at 12 months Treatments 13- 16: cassava was harvested at 18 months Green Total Plants Fresh manures N harvested root (t/ha) (kg/ha) (#/ha) yield (t/ha) - . 8613 17.56 - - 9169 29.78 1.92 44.75 8888 23.75 0.94 20.13 9025 26.94 1.09 27.00 8613 21.39 - - 9444 20.28 0.36 11.75 8750 18.25 0.09 1.69 8613 12.00 9.89 262.13 8888 8.75 1.54 36.63 9306 22.83 8.92 221.69 9725 15.86 - - 9025 17.25 1.44 39.94 8888 46.17 0.93 18.38 9725 42.98 1.05 25.63 8056 38.81 - . 9306 38.86 Source: Rayong Field Crops Research Center, Annual Report 1996. 221 Table 6. Soil pH and organic matter(OM), before planting and after harvest of cassava in different treatments of green manures at RFCRC, Rayong, Thailand in 1994/95/96. Treatments' ' Before plantingAfter harvest pH %OM pH %OM 1. Cassava + Fert. 13- 13-21 (156 kg/ha) 2. Cassava + Fert. 13-13-21 (469 kg/ha) 3. Cassava + Crotalaria juncea (cut at 2 months) 4. Cassava + Canavalia ensiformis (cut at 2 months) 5. Cassava + Pigeon pea ICP 8094 (cut at 2 months) 6. Cassava + Mucuna fospeada (cut at 2 monthsa) 7. Cassava + cassava (pulled out at 2 months) 8. Cassava + cassava (cut at 2 months) 9. Cassava + Crotalaria juncea (planted at 6-7 months) 10. Cassava + Canavalia ensiformis (planted at 6-7 months) 11. Cassava + Pigeon pea ICP8094 (planted at 6-7 months) 12. Cassava + Mucuna fospeada (planted at 6-7 months) 13. Crotalaria juncea green manure-Cassava (18 months) 14. Canavalia ensiformis green manure-Cassava (18 months) 15. Pigeon pea ICP 8094 green manure-Cassava (18 months) 16. Mucuna fospeada green manure-Cassava (18 months) 5.60 0.51 5.20 0.54 5.60 0.46 5.00 0.52 5.90 0.49 5.20 0.49 5.70 0.53 5.40 0.56 5.60 0.54 5.20 0.61 5.50 0.59 5.00 0.63 5.60 0.44 5.30 0.52 5.60 0.45 5.30 0.45 5.70 0.48 5.20 0.47 5.80 0.43 5.20 0.50 5.70 0.46 5.10 0.63 5.60 0.56 5.10 0.65 5.60 0.44 4.90 0.41 5.50 0.48 5.20 0.59 5.80 0.56 5.10 0.59 5.60 0.57 5.00 0.47 "Treatments 3-16: 156 kg/ha of 13-13-21 were applied to cassava Source: Rayong Field Crops Research Center, Annual Report 1996. Soil chemical analysis data, shown in Table 6, indicate a decreasing trend in soil pH both with and without green manures, as compared to the soil analysis before planting. Although Crotalaria juncea had the highest above-ground dry matter production in all patterns of green manure management, this did not result in an increase in soil OM. However, these results are still very preliminary and it will take a few years to confirm the effect of green manures on the yield of cassava and on soil fertility. 2. Effect ofplant spacing and fertilization on the yield of Rayong 5 In 1995/96, a study on the effect of plant spacing and rate of fertilizer application was conducted in Banbung and Sattahip soil series of Rayong province in the eastern part of Thailand. The objective was to determine the optimum spacing and rate of fertilizer application for the new Rayong 5 cultivar. Table 7 shows that plant spacing at 0.8x0.8 or 1.0x1.0 m had no significant effect on either yield or starch content of Rayong 5 in both Banbung and Sattahip soil series. This suggests that Rayong 5 could be planted at any plant spacings corresponding to a plant population ranging from 10,000 to 15,625 plants/ha in both soils. 222 Table 7. The effect of plant spacings and rate of fertilizer application on the yield and starch content of Rayong 5 planted in Banbung and Sattahip soil series, Rayong, Thailand in 1995/96. Banbung soil series Fresh root Starch yield content (t/ha) (%) Sattahip soil series Fresh rootStarch yield content (t/ha) (%) Spacings(S) 0.8x0.8 m 1.0x1.0 m 28.50 29.00 18.03 17.85 33.25 29.94 22.04 21.35 F-test(S) NS NS NS NS Rates of fertilizer(R) N-PX)<-K,Q(kg/ha) 0-0-0 46-23-46 92-46-92 138-69-138 184-92-184 F-test(R) F-test(SxR) CV(%) 18.00c 14.50b 23.06c 21.13 26.06bc 17.23ab 27.25c 22.00 30.06b 19.32a 35.75b 22.57 32.69ab 18.98a 34.19b 21.20 37.00a 19.67a 40.56a 21.57 * * * NS NS NS NS NS 25.92 15.26 14.71 5.47 Source: Rayong Field Crops Research Center, Annual Report 1996. The effect of chemical fertilizers applied in the ratio 2:1:2 of N:P205:K2O by using combinations of compound fertilizer 15-15-15 with urea and KC1 at four different rates, in comparison with the non-fertilized check, is shown in Table 7. In Banbung soil series, there was a significant response to fertilizer rates both in terms of cassava yield and starch content. Application of fertilizers at the rate of 92 kg N, 46 P2O5 and 92 K2O/ha produced a significantly higher root yield of 30. 1 t/ha than did the check without fertilizer (18.0 t/ha). The highest yield of Rayong 5 was obtained with the application of 184 kg N, 92 P2O5 and 184 K2O/ha applied as 625 kg/ha of 15-15-15 supplimented with 200 kg of urea and 156 kg of KCl/ha, which produced a yield of 37.0 t/ha. The treatment without fertilizer application also resulted in the lowest starch content of Rayong 5 at only 14.5%, compared to 19.7% when the highest rate of fertilizers was 223 applied. The results indicate that Rayong 5 cultivar, planted in Banking soil series needs application of fertilizer 15-15-15 at least at the rate of 312 kg/ha with 100 kg of urea and 78 kg of KCl/ha to produce a high root yield and starch content. In the Sattahip soil series, the response to fertilizers was also significant in terms of root yield. The intermediate rate of 92 kg N, 46 P205 and 92 K20/ha, applied as 312 kg/ha of 15-15-15 with 100 kg of urea and 78 kg of KCl/ha, produced a significantly higher root yield of 35.8 t/ha than the check without fertilizer (23. 1 t/ha). The highest root yield of Rayong 5 was again obtained with the highest rate of applied fertilizers, i.e. 625 kg/ha of 15-15-15 with 200 kg of urea and 156 kg of KCl/ha, which produced a yield of 40.56 t/ha. But, fertilizer application did not result in any significant difference in starch content of Rayong 5, which ranged from 21.1 to 22.6%. The results indicate that Rayong 5 planted in Sattahip soil series also needs the same application of fertilizer 15-15-15 at least at the rate of 312 kg/ha supplimented with 100 kg/ha of urea and 78 kg/ha of KC1 to produce a high root yield and starch content. 3. Use of legumes as green manures to sustain the productivity of cassava at Kalasin] A study on the use of grain legumes as green manures to sustain the productivity of Rayong 60 was conducted in 1994/95 in Kalasin province in the northeast of Thailand, where cassava has been a major field crop. The experiment consisted of eight cassava-based cropping systems, in which three grain legume species, i.e. cowpea for grain, cowpea for green pods and peanut were planted in rotation with cassava, cv. Rayong 60. Leaving the land in fallow before planting cassava was another treatment used for comparison; in this case fertilizers were not applied or were applied as 12-24-12 at a rate of 156 kg/ha. In the legume rotation treatments either 15-15-15 fertilizer at the rate of 125 kg/ha was applied to cassava but without fertilizer for the legume, or 12-24-12 fertilizer at the rate of 156 kg/ha was applied to the legumes without fertilization of cassava. Legumes were planted in June and harvested in Sept. Cassava was planted in Oct and harvested in May of each year on Korat soil series with pH 5.54, 0.60% OM, 110 ppm available P and 46 ppm exchangeable K. The results of the first year experiment, shown in Table 8, indicate that with fertilizer application, all grain legumes produced higher yields (863-2300 kg/ha) than without fertilizer (663-1938 kg/ha). Cassava yields differed significantly depending on the various cropping systems. Cassava yields tended to be significantly higher when fertilizers were applied to cassava rather than to the preceding legume. When fertilizers were applied to cassava there was no significant effect of the preceding grain legume, but when no fertilizers were applied to cassava, yields were significantly higher after the grain legumes than after bare fallow. This indicates that the residual effect of fertilized legumes could help maintain soil fertility for cassava production. The highest yield of 22.5 t/ha was obtained from planting fertilized cassava after unfertilized peanut. The 224 root starch contents of cassava in the various cropping systems were not significantly different among treatments. The experiment needs to be repeated several years with soil analysis data to be able to conclude whether any of these cropping systems improves the long-term sustainability of cassava production in these poor northeastern soils. Table 8. The effect of several cassava-grain legume rotation systems on the yield and starch content of cassava, Rayong 60, as well as the yields of the grain legumes grown at Kalasin, Thailand in 1994/95. Cropping systems" Grain Cassava Cassava legumes plants fresh root Starch yield harvested yield content (kg/ha) (#/ha) (t/ha) (%) 863 9256 16.38b 20.53 663 8775 19.13ab 22.68 1750 8938 18.06b 21.14 1000 9063 22.50a 23.70 2300 9125 18.56b 20.80 1938 8875 19.38ab 22.80 - 8750 15.38c 20.40 - 9125 18.88ab 20.55 - NS * NS - 10.50 17.50 12.60 1. Cowpea (Seed) + F -Cassava-F 2. Cowpea (Seed)-F -Cassava+F 3. Peanut + F -Cassava-F 4. Peanut-F -Cassava + F 5. Green Cowpea+F -Cassava-F 6. Green Cowpea-F -Cassava + F 7. Fallow -Cassava-F 8. Fallow -Cassava + F F-test CV(%) " +F = with fertilizer ;-F = without fertilizer 156 kg/ha of 12-24-12 were applied to legumes in treatments 1, 3 and 5 125 kg/ha of 15-15-15 were applied to cassava in treatments 2, 4, 6 and 8 Source: Khon Kaen Field Crops Research Center, Annual Report 1995. 4. The fertilization of new high-yielding cassava varieties In 1994/95 a study on the effect of chemical fertilizers on the yields and starch contents of newly released cassava varieties was conducted at Udon Thani and Khon Kaen provinces in the northeast of Thailand. Three cultivars were tested in each location, i.e. Kasetsart 50, Rayong 5 and Rayong 60 tested in Udon Thani, and Kasetsart 50, Rayong 90 and Rayong 60 in Khon Kaen. The results shown in Table 9 indicate that in Udon Thani the yields of the three varieties were not significantly different, ranging from 28.5 to 34. 1 t/ha. The root starch content of Rayong 60 was significantly lower than that of Kasetsart 50, and slightly but not significantly lower than that of Rayong 5. There were also no significant effects of 225 rates of chemical fertilizer in terms of yield and starch content. Nevertheless, there was an increasing trend in root yield with increasing rates of chemical fertilizers applied: at higher rates all cassava cultivars tended to produce higher yields. Table 9. The effect of chemical fertilizers on the yield and starch content of three cassava cultivars grown in Udon Thani and Khon Kaen, Thailand in 1994/95. Udon Thani Khon Kaen Fresh root Starch Fresh root Starch yield content yield content Cultivars(C) (t/ha) (%) (t/ha) (%) Kasetsart 50 33.05 25.0a 16.10a 22.6 Rayong 90 - - 10.49b 22.6 Rayong 5 34.07 23.0ab - Rayong 60 28.51 21.0b 8.36b 17.4 F-test(C) NS * * NS Rates of fertilizer(R):kg/ha 0-0-0 24.44 23.0 8.36b 22.4 25-25-25 30.24 22.0 12.74a 21.9 50-50-50 35.78 23.0 13.33a 20.6 75-75-75 37.03 23.0 12.16a 18.4 F-test(R) NS NS * NS F-test(CxR) NS NS NS NS CV(%) 24.8 8.03 28.50 13.20 Source: Khon Kaen Field Crops Research Center, Annual Report 1995. At Khon Kaen, the yields were much lower, but the yield of Kasetsart 50 was significantly higher those of Rayong 90 and Rayong 60. There were no significant differences in root starch content between these cultivars. On average, chemical fertilizer, applied at the rate of 25-25-25 kg/ha of N, P205 and K2O, produced significantly higher yields than without fertilizer application, which produced only 8.36 t/ha. At higher rates of fertilizer application the effect on cassava yields was not significant. Although there was a trend of decreasing starch content with increasing rates of fertilizers, differences were not statistically significant. 226 There were no significant interactions between cultivars and rates of chemical fertilizer on yield and starch content at both Udon Thani and Khon Kaen. 5. Long-term effect of cropping system on cassava productivity in the Northeast Since 1980, a semi-demonstration study on the long-term effect of various soil management treatments on cassava production has been conducted at Khon Kaen Field Crops Research Center in the Northeast of Thailand. The cropping systems of sole cassava, cassava intercropped with peanut and cassava rotated with sequentially planted peanut and pigeonpea were tested with and without fertilization and with and without soil amendments. The objective of the trial was to determine the most appropriate soil management system to maintain soil fertility and sustain high cassava yields. Rayong 1 was used as the test cultivar in this long-term experiment. Table 10 shows the cassava yields obtained in the different cropping systems with various soil management treatments. In the first year (1980) sole cassava produced the highest average yield of 27.4 t/ha, compared to 24.0 t/ha for cassava rotated with peanut-pigeon pea, and 25.5 t/ha for cassava intercropped with peanut. During the first year, cassava could still produce a rather high average yield of 25.6 t/ha without any fertilizers or soil amendments. When only fertilizers were applied, cassava produced the highest yield of 28.7 t/ha, or 12% higher than without fertilizers, while soil amendments slightly decreased cassava yields, both in the absence and presence of chemical fertilizers. In the 15th year (1994), the highest average cassava yield of 20.9 t/ha was obtained when cassava had been rotated with sequentially planted peanut and pigeon pea, while the continuous planting of sole cassava produced a lower yield of 16.8 t/ha. The lowest cassava yield of 12.9 t/ha was obtained when cassava had been intercropped with peanut during the past 15 years, probably due to competition from the intercropped peanut. Considering the soil management effect on cassava yields in the 15th year, the lowest average yield at 11.2 t/ha was obtained in the check plots without fertilizers or soil amendments. When only chemical fertilizers had been applied, cassava produced the highest yield of 20.0 t/ha, or 78% higher than without fertilizers. Soil amendments alone or in combination with chemical fertilizers resulted in yields of 17.5 and 18.7 t/ha, or 56 and 67%, respectively, higher than the check plots. When the average yields in the 15th year are compared to those of the first year, it is clear that neither cropping systems nor soil management treatments could maintain or increase cassava yields. The overall average yield of all cropping systems and soil management treatments in the 15th year was only 66% of that of the the first year. Only the rotation of cassava with sequentially planted peanut and pigeon pea could maintain a relatively high yield of cassava, corresponding to 87% of the yield obtained with the same cropping system in the first year. Among soil management treatments, the check plots without any fertilizers or soil amendments produced the lowest relative yield of 227 44% of that obtained in the first year. Only the treatments of cassava rotated with peanut-pigeonpea and with either chemical fertilizers alone or in combination with soil amendments could produce-high cassava yields in the 15th year that were comparable to those obtained in the first year. Table 10. A semi-demonstration study on the long-term effect of soil management on the yield (t/ha) of cassava grown at Khon Kaen, Thailand from 1980 to 1994. Cropping system Soil management 1* Year(1980) Check Fertilize!•" Soil Fertilizer + Soil amendment2' amendment3' Average Cassava 30.13 27.88 18.81 32.38 26.81 27.00 20.38 18.63 27.31 26.63 22.88 28.81 27.38 24.05 25.48 Cassava/Peanut-Pigeon Cassava+ Peanut5) pea4) Average of 1st year 25.61 28.73 22.11 26.11 25.64 15th Year(1994) Cassava 8.81 21.68 24.50 13.81 19.81 20.25 12.56 16.87 23.12 16.12 16.79 20.88 12.91 Cassava/Peanut-Pigeon Cassava+ Peanut5* pea4) 15.68 9.18 Average of 15th year Relative to l"year(%) 11.22 44 19.99 17.54 70 79 18.70 72 16.86 66 " Applied 50-50-50 kg/ha of N-P205-K20 to cassava or 18.75-56.25-37.50 kg/ha of N-P205-K20 to peanut in crop rotation treatment4' % Applied 1250 kg/ha of lime and rock phosphate(3%P205)with 18.75 t/ha of munucipal compostin the 1st, 5th, 9th, and 13th year(1980, 1984, 1988 and 1992) 3' l)+2). 4) Cassava and peanut-pigeon pea were planted in alternate years After the harvest of sequentially planted legumes, the residues were incorporated into the soil before the following years' planting of cassava. 51 2 rows of peanut were intercropped between cassava rows. After harvest of peanut the residue was used as mulch. Source: Khon Kaen Field Crops Research Center, Annual Reports 1992 and 1995. Weed Control Weeds are known as a major problem, causing serious cassava yield reductions. Nowadays, labor for weed control is hard to find and the cost of manual weeding is 228 getting higher and higher each year. To solve these problems several studies were conducted at Khon Kaen and Rayong Research Centers during two years from 1993 to 1995 to determine the best methods and times of weed control in cassava. 1. Weed control for three cassava cultivars Experiments on weed control for cassava were conducted in 1993/94 at Khon Kaen Field Crops Research Center in the northeast of Thailand. Rayong 1, Rayong 60 and Rayong 90 cultivars were planted in both the early and late rainy seasons. Plots were maintained weed free for either 0 (non-weeded check), 2, 3 or 4 MAP by manual weeding, compared to a typical "farmer" practice of manual weeding only at 2 MAP and with no fertilizer application. The objective was to determine the optimum period for controlling weeds, which would produce high yields when cassava was planted in either the early or late rainy season. Table 11 shows that yields of Rayong 1, Rayong 60 and Rayong 90 cultivars, averaged over two years and two seasons, were not significantly different, ranging from 17.6 to 20.2 t/ha. There were no significant interactions between cultivars and weed-free periods when cassava was planted in either the early or late rainy seasons. However, weed control treatments produced highly significant differences in yield (averaged over cultivars and seasons/years), ranging from 20.1 to 23.4 t/ha, compared to only 7.7 t/ha in the non-weeded check. The results indicate that weed control is extremely important during the first two months after planting, but that weed control beyond 2 MAP did not significantly increase yields any further. The highest cassava yields were obtained when plots were maintained weed-free for 3 MAP. Thus, when cassava was planted in either early or late rainy season these three cassava cultivars needed to be free of weeds about 2-3 months after planting to produce highest yields. 2. Weed controlfor cassava intercropped with mungbean and peanut Experiments on weed control for cassava intercropped with mungbean or peanut were conducted during two years (1994-1996) at Rayong Field Crops Research Center in the East of Thailand. The objective of the experiment was to determine the most appropriate methods of weed control in cassava-legume intercropping systems, which would produce high yields of intercrops and cassava, resulting in a high income. Cassava, cv. Rayong 90, was planted in three cropping systems, i.e. sole cassava, intercropped with mungbean or with peanut. Four different methods of weed control involved various combinations of pre-emergence herbicide Metolachlor (1.5 kg ai/ha, a post-emergence herbicide Paraquat (0.5 kg ai/ha), and manual weeding. Weeds were controlled by either Metolachlor (alone), Metolachlor followed by hand weeding or Metolachlor followed by application of Paraquat, as compared to hand weeding alone. In sole crops of peanut and mungbean, weeds were controlled by either Metolachlor (1.50 kg ai/ha) or by hand weeding; these functioned as check plots of the intercrops. 229 Table 11. Effect of weed control on yield (t/ha) of Rayong 1, Rayong 60 and Rayong 90 planted in the early and late rainy seasons at Khon Kaen, Thailand in 1993/94 and 1994/95. 1993/94 1994/95 Average 2 years Average Cultivars(C) ER LR ER LR ER LR 2 seasons Rayong 1 28.33 19.53 10.86 17.23 20.97 18.38 19.67 Rayong 60 23.33 27.68 15.11 14.59 19.22 21.13 20.18 Rayong 90 25.03 21.88 11.33 12.25 18.18 17.06 17.62 F-test(C) NS * * NS NS * NS Weed-free period(W) 0 month (check) 2.61 13.48 4.49 5.63 5.83 9.56 7.69 2 months 31.98 26.43 16.71 15.52 24.34 20.98 22.66 3 months 34.71 26.03 13.84 19.20 24.28 22.61 23.44 4 months 31.47 24.96 13.73 17.54 22.59 21.25 21.93 "Farmer" practice" 27.07 24.25 13.39 15.54 20.23 19.89 20.06 LSD(0.05)for W 6.73 7.38 4.97 5.82 5.51 4.70 3.56 F-test(W) ** ** ** ** ** ** ** F-test(CxW) NS NS ** NS NS NS NS CV(%) 27.00 29.60 41.40 40.70 42.60 37.40 39.90 ""Farmer" practice = manual weed control at 2 months with no fertilizers applied Source: Khon Kaen Field Crops Research Center, Annual Report 1995. Table 12 shows that the yields of mungbean, both when grown as intercrops and as sole crops in 1994/95 and 1995/96 were not significantly affected by different weeding methods; yields ranged from 319 to 450 kg/ha in 1994 and from 175 to 231 kg/ha in 1995. The same results were obtained with peanut, the yield of which ranged from 419 to 556 kg/ha in 1994, and from 300 to 356 kg/ha in 1995. The results indicate that neither cassava competition nor weeding method were significantly affecting the yields of the two legumes. Table 13 shows the cassava yields in three different cropping systems as affected by different weeding methods. In 1994/95, there was no significant effect of cropping system on cassava yield, but the highest yield of 16.9 t/ha was obtained when cassava was intercropped with peanut. However, this same treatment resulted in a significantly lower cassava yield in 1995/96. Cropping systems had no significant effect on cassava 230 root starch content in either year. There were also no interaction effects between cropping systems and weeding methods, both in terms of cassava yield and starch content. However, the effects of different weeding methods on cassava yields were highly significant in both years, with the lowest yields of 7. 14 and 7.46 t/ha obtained in 1994/95 and 1995/96, respectively, when weeds were controlled by application of Metolachlor only at planting. Thus, in that treatment cassava yields were significantly depressed by weed competition. The other three weeding methods resulted in yields that were not significantly different, ranging from 14.4 to 18.4 t/ha in 1994/95 and from 1 1 .5 to 14.2 t/ha in 1995/96. The highest cassava yields of 18.4 t/ha in 1994/95 and 13.8 t/ha in 1995/96 were obtained with application of Metolachlor at 1.50 kg ai/ha, followed by two manual weedings. The weeding methods caused no significant differences in cassava root starch contents. The net income obtained with the various cropping systems and methods of weed control are shown in Table 14. With respect to weeding methods in each cropping system the highest net income in the 1994/95 experiment was obtained by intercropping cassava with peanut (1 153 US $/ha) or when grown as a monocrop (832 US $/ha), when weeds in both cropping systems were controlled by application of Metolachlor followed by two manual weedings. In contrast, the highest net income of 938 US $/ha in the cassava-mungbean intercropping system was obtained when weeds were controlled by application of Metolachlor followed by two spot applications of Paraquat. In the 1995/96 experiment (Table 13), there was a highly significant competitive effect when cassava was intercropped with peanut, resulting in the lowest yield at 10.2 t/ha. The yield of cassava intercropped with mungbean was not significantly different from that of sole cassava, since mungbean has a shorter crop duration and thus causes less competition than peanut. Root starch contents of cassava were not significantly different in all cropping systems. There were again no interaction effects between cropping systems and weeding methods, both in terms of yield and starch content. However, the different weeding methods caused highly significant differences in cassava yields, with the lowest yield of 7.5 t/ha obtained when weeds were controlled by only one application of Metolachlor at planting. The highest cassava yield of 14.2 t/ha was obtained with three manual weedings. The application of Metolachlor followed by hand weeding was not significantly different from hand weeding alone. Weeding methods had no significant effect on root starch content except that the application of Metolachlor alone resulted in a higher starch content than hand weeding alone. Since the last hand weeding was done at 2.5 months before the cassava harvest, the mulch of dead weeds might have increased soil moisture causing a reduction in starch content. 231 Table 12. Effect of different weed control methods on the yields of mungbean and peanut, grown either intercropped in cassava or as a monocrop in Rayong Field Crops Research Center, Rayong, Thailand in 1994/95 and 1995/96. Mungbean" Peanut2) Treatment 1994/95 1995/96 1994/95 1995/96 Intercropped with cassava (kg/ha) (kg/ha) (kg/ha) (kg/ha) 1. Metholachlor (1.50 kg ai/ha) 325 175 419 300 2. Metholachlor (1.50 kg ai/ha) 344 181 425 306 +hand weeding 3. Metholachlor (1.50 kg ai/ha) 344 187 444 350 + Paraquat (0.50 kg ai/ha) 4. Hand weeding 319 206 506 356 Sole crops 5. Hand weeding 356 206 494 331 6. Metholachlor (1.50 kg ai/ha) 450 231 556 319 F-test NS NS NS NS CV (%) 29.72 10.25 20.29 17.72 "Yield of dry grain 2)Yield of dry pods Source: Rayong Field Crops Research Center, Annual Reports 1995 and 1996. Table 14 shows that the highest net incomes of 314 and 269 US $/ha in the intercropping systems with peanut and mungbean were obtained when weeds were controlled by application of Metolachlor followed by three spot applications of Paraquat. The highest net income of 273 US$/ha in sole cassava was obtained when weeds were controlled by application of Metolachlor only at planting. The results from both years show some contrasting effects of cropping systems and weed control methods on cassava yields and net income, depending on the season of planting in each year. With the variation in the prices of the various crops, in rainfall and thus in weed condition and weeding costs, the net incomes varied from year to year. Nevertheless, in both years the best method for controlling weeds when cassava was intercropped with mungbean was the application of pre-emergence herbicide Metolachlor at planting, followed by spot treatments with the post-emergence herbicide Paraquat. The experiment will be repeated in 1996/97 to refine the results. 232 Table 13. The effect of three cropping systems and four methods of weed control on the yield and starch content of Rayong 90 planted at Rayong Field Crops Research Center, Rayong, Thailand in 1994/95 and 1995/96. 1994/95 1995/96 Root yield Starch Root yield Starch Cropping system(C) (t/ha) (%) (t/ha) (%) Intercropped with mungbean 11.46 22.07 12.18ab 24.74 Intercropped with peanut 16.90 23.77 10.18b 24.69 Sole cassava 13.74 23.28 12.96a 24.69 F-test(C) NS NS ** NS CV (%) a: 35.61 7.25 9.63 8.63 Weeding methods(W) 1. Metholachlor (1.50 kg ai/ha) 7.14b 23.38 7.46c 26.21a 2. Metholachlor (1.50 kg ai/ha) 18.39a 23.24 13.83ab 24.59ab + hand weeding 3. Metholachlor (1.50 kg ai/ha) 16.17a 23.00 1 1 .54b 24.50ab 4. Hand weeding 14.43ab 22.53 14.24a 23.53b F-test(W) ** NS ** ** F-test(CxW) NS NS NS NS CV (%)b: 39.61 7.05 14.90 5.44 Source: Rayong Field Crops Research Center, Annual Reports 1995 and 1996. Recommended Cultural Practices Table 15 shows the currently recommended varieties and their major characteristics, as well as the recommended cultural practices, based on research results obtained by various research institutes in Thailand during the past 30 years. FUTURE RESEARCH PRIORITIES Studies on the effect of cassava production on soil fertility, and studies on soil conservation using combinations of cropping systems and soil management practices for erosion control and fertility maintenance, so as to sustain high cassava yields in the major cassava producing areas, will remain of high priority. Studies on optimum cultural 233 practices, ie. rate, time and method of fertilizer application, weed control for various cassava-based cropping systems and for the newly recommended cultivars, will also continue. Table 14. The effect of three cassava-based cropping systems and four methods of weed control on the total net income obtained at Rayong Field Crops Research Center, Rayong, Thailand in 1994/95 and 1995/%. Net income (US$/ha)' 1994/95 1995/96 C + M21 C + P2) C2) C +M C +P C 315 716 310 235 238 273 713 1153 832 230 189 153 938 1116 663 269 314 260 499 800 684 228 247 167 Weeding Methods 1. Metholachlor (1.50 kg ai/ha) 2. Metholachlor (1.50 kg ai/ha) + hand weeding3' 3. Metholachlor (1.50 kg ai/ha) + Paraquat (0.50 kg ai/ha)3) 4. Hand weeding" " Net income is total gross income minus weeding costs. 2) C +M = Cassava intercropped with mungbean. C + P= Cassava intercropped with peanut. C = Cassava sole crop. 3) in 1994/95 hand weeding and application of Paraquat were done twice, while in 1995/96 this was one three times. Source: Rayong Field Crops Research Center, Annual Reports 1995 and 1996. 234 Table 15. Recommended cultural practices for cassava production in Thailand in 19%. Varieties: Rayong 1: high yield stability, bitter Rayong 2: good eating quality Rayong 3: high starch content, bitter Rayong 60: high yield even when harvested at 8-10 months, relatively high starch content, bitter Rayong 90: high starch content, higher yield and longer stems than Rayong 3, bitter Rayong 5: high yield, relatively high starch content, bitter Kasetsart 50: high yield, relatively high starch content, bitter Planting time: Land preparation: Planting material: Planting method: Plant spacing: Fertilization: Weeding: Harvest: Intercropping: May-June or Oct-Nov. Plowing once or twice at 15-20 cm depth with tractor; followed by once or twice disc-harrowing; ridging by oxen or tractor. Select healthy 10-12 month old plants, cut mature stems, store in shade (less than 30 days), cut stakes of 15-25 cm length Planting on ridges is needed when soil is wet and contour ridges are also very effective in reducing erosion on sloping land; flat planting when soil moisture is low; plant vertically at 5-10 cm depth; 80-100cm between rows, 80-100 cm between plants. 100kg N, 50 P2Os, and 100 K2O/ha applied as compound fertilizers (such as 15-15-15) supplimented with urea and KC1; side-dressed after first weeding at 1-2 months. On sloping land; application should be split as follows: 50 kg N, 50 P2O5 and 50 K2O/ha applied as compound fertilizer, side- dressed after first weeding at 1-2 months; followed by 50 kg N/ha as urea and 50 kg K2O/ha as KC1, side-dressed at 2-3 months. 2-3 times manually, at 30-45 days and 2-3 months later; or application of Metholachlor (1.5 kg ai/ha) directly after planting, followed by 1-2 hand weedings or spot application of Paraquat (0.5 kg/ha) At 10-12 months, during dry season Cassava may be intercropped with two rows of peanut or mungbean; 20 cm between rows, 10 cm between plants. REFERENCES Khon Kaen Field Crops Research Center (KFCRC). 1992-1995. Annual Reports for 1991-1994. Field Crops Research Institute, Department of Agriculture, Thailand, (in Thai) Rayong Field Crops Research Center (RFCRC). 1992-1996. Annual Reports for 199 1-1995. Field Crops Research Institute, Department of Agriculture, Thailand, (in Thai) Sittibusaya, C, C. Tiraporn, A. Tongglum, U. Cenpukdee, V. Vichukit, S. Jantawat and R.H. Howeler. 1995. Recent progress in cassava agronomy research in Thailand. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum,Kerala, India. Nov 2-6, 1993. pp. 110-123. Tongglum, A., V. Vichukit, S. Jantawat, C. Sittibusaya, C. Tiraporn, S. Sinthuprama and R.H. Howeler. 1992. Recent progress in cassava agronomy research in Thailand. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Utilization Research in Asia. Proc. 3d Regional Workshop, held in Malang, Indonesia. Nov 22-27, 1990. pp. 197-223 235 RECENT PROGRESS IN CASSAVA AGRONOMY RESEARCH IN VIETNAM Nguyen Huu Hy\ Pham Van Men2, Nguyen The Dang3 and Thai Phien* ABSTRACT In recent years the National Root and Tuber Crops Program of Vietnam, with the cooperation and assistance of the Centro International de Agricultura Tropical (CIAT), has drawn up a plan for strengthening the research and development capacity, with the objective of improving cassava production in Vietnam. In the area of cassava agronomy, the program has obtained the following results: - Among various intercropping systems, the interplanting of cassava with peanut, mungbean or maize were the most promising on high fertility soils, while intercropping with peanut was promising on the poorer soils. - Intercropping cassava with black bean or peanut and planting contour hedgerows of Tephrosia Candida was the best way to control soil erosion. Cassava intercropped or alley cropped with hedgerow species did not show a significant benificial effect on cassava yield at Hung Loc Center, but the effect may be observed in the long term. - The cassava population of 10,000-14,000 plants/ha and 12,000-16,000 plants/ha should be recommended for the red Latosol and the grey Podzolic soils, respectively. - Long term N, P and K trials have shown that the response of cassava to fertilizers is very different for the various types of soils; on the more fertile Latosols the response was not significant even in the fourth year of cropping, but on the grey Podzolic soil in the South and on red-yellow Ferralsols in the north the response of cassava was highly significant already in the first year, with the main response to N and K, respectively. - Short term N, P and K trials showed a high response of cassava to K in Vinh Phu province, as well as a higher income due to fertilizer application in Dong Nai province. INTRODUCTION In Vietnam, the statistical data for 1992 show that agriculture accounted for 43% of the national gross production and 51.3% of the national income. Crop production accounted for 76% of the total agricultural production, in which food crop production plays an important role, occupying 98% of the total area cultivated to annual crops. Rice production is the most important, followed by maize and cassava. The annual cassava cultivated areas is about 280,000 ha, accounting for 25% of the food crop area; the cassava production areas are distributed in all the different regions of the country (Figure 1). 1 Hung Loc Agric. Research Center (HARC) of IAS, Thong Nhat, Dong Nai, Vietnam. 2 Institute of Agricultural Sciences of South Vietnam (IAS), Ho Chi Minh city, Vietnam. 3 Agro-forestry College of Thai Nguyen University, Thai Nguyen, Bac Thai, Vietnam. 4 Institute for Soils and Fertilizers, Chem, Tu Liem, Hanoi, Vietnam. 236 Province* Caaaava area I'OOO hai 1. Ha Noi 0 7 2. Ho Chi Minn 0 / 3. Hai Phong 0 4 4. Ha Giang 6 b 5. Tuyen Quang 8 1 6. Cao Bang 2.1 7. Lang Son 3 b 8. Lai Chau 8 8 9. Lao Cai S.8 10. Yen Bai 10.1 11. Bac Thai 4.8 12. Son La 13.7 13. Vinh Phu 13 2 14. Ha Bac 9 5 15. Quang Ninh 3.3 16. Ha Tay 4 3 17 Hoa Binh 14 8 18. Hai Hung 0 3 19. Thai Binh 0.0 20. Nam Ha 0 7 21. Ninh Binh 1.9 22. Thanh Hoa 16.2 23. Nghe An 14.1 24. Ha Tinh 2.3 26. Quang Binh 4 4 26. Quang Tri 3.8 27. Thua Thien-Hue 6.2 28. Quang Nam-Da N ang 17.0 29. Quang Ngai 11.3 30. Binh Dinh 12.2 31 Phu Yan 4.8 32. Khanh Hoa ; b 33. Ninh Thuan i i 34. Binh Thuan 6.8 35. Kon Tum 7.2 36. Gia Lai 9.7 37. Dae Lac 4.8 38. Lam Dong 2.6 39. Song Be 40. Tay Ninh 41 . Dong Nai 42. Long An 43. Dong Thap 44. An Giang 45. Tien Giang 46. Ben Tre 47. Vinh Long 48. Tra Vinh 49. Can Tho 50. Soc Trang 51 . Kien Giang 52. Minh Hai Figure L Cassava planted area in the various provinces of Vietnam. 237 Ranked in third position among food crops, cassava is of special importance because it is not only a human food but is also used as a source of feed for animals and as a raw material for the processing industry. Cassava is also an important crop for poor farmers in the nutrient poor soil areas, where other crops can not be planted. The soils cultivated to cassava are generally slopy or hilly, degraded and often eroded. And these soils are often of low pH, have low levels of organic matter, are deficient in K and may lack some micronutrient elements (data of soil analysis by CIAT; Howeler, 1995). Grown mainly on nutrient poor soils, farmers generally plant cassava at a population from 10,000 to 15,000 plants/ha, dependent on the type of soil, the variety and the time of cassava planting, as well as the tradition of the farmers. In some small areas, cassava is usually intercropped with legume crops, such as peanut, mungbean, and cowpea, or with contour hedgerows of leguminous trees to reduce erosion and improve soil fertility. The importance of intercropping varies between regions; it occupies about 10% in the north and 30% in the south of Vietnam (Pham Van Bien and Hoang Kim, 1996). Until now, the inputs used for cassava production are still very low and the total production costs varies between regions, ranging from US$1 16.69 to 224.83 (Figure 2). Depending on the local conditions and the likely benifits obtained, farmers plant cassava with or without fertilizers, the cost of fertilizer application being about 19% of the total cost in the north and 12% in the south of Vietnam (Figure 3). RESEARCH RESULTS Intercropping In Vietnam many people believe that intercropping of cassava can increase soil fertility and reduce erosion, but some experimental results indicate that the net return is lower than for cassava monocropping. Research by Hoang Kim and Buresova(1987) on cassava intercropping with peanut, mungbean and winged bean showed that most of the intercropping systems gave higher economic returns. Other intercropping trials have been conducted on the red Latosol soil at Hung Loc Research Center in South Vietnam for four years, using various grain legumes and maize, with the objective of controlling weeds, preventing soil erosion, improving soil fertility and increasing income. The results showed that on the red Latosol soil cassava intercropped with peanut, mungbean or maize were the most promising systems. But cassava monocropping still gave higher economic returns than any other intercropping system, while cassava planted in single rows gave higher profits than planting in double rows (Nguyen Huu Hy et al., 1995). Two intercropping trials were also conducted on red-yellow Ferralsols and on grey Podzolic soils at the Agro-forestry College in Thai Nguyen of Bac Thai province and in Thuan An of Song Be province, respectively. The results of both trials (Table 1) indicate that on these nutrient poor soils, cassava intercropped with peanut produced by far the highest total gross income in both locations, followed by mungbean. Cassava intercropped with maize and soybean produced the lowest gross income. 238 Table 1. Effect of various intercropping systems on the yield of cassava and intercrops as well as total gross income in Agro-forestry College # 3, in Bat- Thai, north Vietnam (A), and in Thuan An, Song Be, south Vietnam (B), in 1993. Cropping systems1' Cassava Intercrop Gross root yield yield income (t/ha) (t/ha) COOOd/ha) A. Agric. College, BacThai (Av. 4 years)" Cassava monoculture 21.9 - 4,927 Cassava+ peanut 20.2 0.815 7,071 Cassava+ black bean 17.0 0.376 5,141 Cassava+ mungbean 21.2 0.135 4,242 Cassava + soybean 20.4 0.170 5,270 Cassava+ maize 17.5 1.257 5,320 B.Thuan An, Song Be (1 year)2' Cassava monoculture 41.5 - 11,620 Cassava + peanut 38.1 0.70 13,132 Cassava+ cowpea 40.2 - 11,256 Cassava+ cucumber 39.5 - 11,060 Cassava+ mungbean 38.2 - 10,696 Cassava+ maize 23.8 - 7,364 11 variety: prices : Vinh Phu cassava: peanut: blackbean: mungbean: soybean: maize: 225 d/kg fresh roots 3,100 d/kg dry pods 3,500 d/kg dry grain 3,500 d/kg dry grain 4,000 d/kg dry grain 1,000 d/kg dry grain variety: prices: KM 60 cassava: peanut: maize: 280 d/kg fresh roots 3000 d/kg dry pods 2000 d/kg dry grain o o^ O a £ O |o o V) Ch oi i/> Og f° o 2 'Q i 5 ■a > v> o o c o o 3■a o o in •> OQ yield (kg/ha s -> JT3 IT, yield (t/ha) 3u Q •—Vvi is. CT3 >> in Tt rn O 00 fl\ H CO © © T* T* O U"> Tf-" —i CI -* -i —I a .a, Cu o c .5. ex o V3 o u D ■8 <-3 fit CX a. o CX c 'J D D XI X o -<" — © >< >< K X X © © © © 00 l-H ^^ ^H 1-H ^^ in c E o 243 no ridge no-tillage contour ridge black bean intercrop a black bean intercrop + Tephrosia hedgerows 4 6 8 Months after planting 600 E §400 m c 200 n K M M N Figure 4. Effect of various agronomic practices on the accumulative dry soil loss due to erosion in cassava planted on about 10% slope in Agro-forestry College ofBac Thai, North Vietnam, in 1993. 244 30 it 20 o o a. 10 10 12 14 16 18 x -D C > <0 X 60 "^--A KM 94 50 ^~~~~OKM 60 40 30 - ; - i i _i— i i i 10 12 14 16 18 c CD c o o JZ o to □ V - CO CM 1 I 1 \II 1 > \ i ! z * - > CO CO - i i 1 f \ c 1 - 1 - ) 0. * 1 O O(?■ «* CO - V - > 1 o CO o CO o O I ID O M CL I z 6 o CO o o CM - o o CO o CO o - o CO 3 JC c o ! o. -S °> S ro O) O CO o CM (eij/i) p|3|A jooj babssbq N n 01 c/i ii < O N Z Ql o o CO «fr oo o ID O Z * o o 00 00 O O 0. * o o • 0) SI § o to E a o 100 80 60 40 20 0 0.20 0.10 10 Critical K-level Critical P-level Crop cycle Figure 8. Effect of annual application of N, P and K on cassava root yield, relative yield (yield without the nutrient over the highest yield with the nutrient) and the exchangeable K and availabel P (Bray 2) content of the soil during seven years of continuous cropping in Hung Loc Research Center, Dong Nai, South Vietnam. 250 Short-term NPK Requirements In 1993 and during 1993/94 two short-term NPK trials were conducted in Tang Hoa, Vinh Phu province, and in Xuan Thanh, Thong Nhat, Dong Nai province. Table 5 shows that in Tang Hoa there was a marked response of cassava to K application. The application of 160 kg K20/ha gave the highest yields, while without K yields were lowest, both at intermediate and high rates of application of N and P. Analyses of leaf samples indicate that K concentrations in YFEL-blades were lowest in those treatments without K application. Table 6 shows that on the red Latosol soil at Xuan Thanh in Dong Nai province application of 60 kg N, 60 P2O5 and 120 K20/ha, with or without application of 500 kg lime and 40 kg MgS04/ha, produced the highest cassava yields and economic returns. Future Direction The following research topics are considered of highest priority in the future: - Intensification of research on various cropping systems involving cassava. - Development of management practices that reduce soil erosion when cassava is grown on slopes. - Release of new agronomic technologies to increase yields and reduce production costs. - Establishment of a fertilizer program to improve the efficiency of fertilizer use. - Research on more effective weed control methods. Table 5. Effect of application of various levels of N, P and K on cassava root yield, the nutrient concentration in YFEL blades at 6 months after planting in Tang Hoa, Vinh Phu province in 1993. Fertilizers Root yield Nutrient concentration%) in YFEL blades applied (t/ha) (kg N-P205-K20/ha) N P K N40 P40 K0 10.2 3.87 0.31 0.83 N40 P40 K80 13.2 3.47 0.27 1.42 N40 P40 K160 14.9 3.43 0.20 1.44 N80 P80 K0 10.3 3.50 0.32 0.92 N80 P80 K80 14.5 3.72 0.30 1.52 N80 P80 K160 15.3 3.64 0.25 1.62 Source: Nguyen Huu Hy et al. , 1995a. 251 Agric. College Bac Thai 1 a> '>- s o o 100r o o £ .5 o a) 0.20r 0.10 - Figure 9. Effect ofannual applications ofN, P and K on cassava root yield, relative yield (yield without the nutrient over the highest yield with the nutrient) and the exchangeable K and available P (Bray 2) content ofthe soil during seven years ofcontinuous cropping in Agro-forestry College ofBac Thai, North Vietnam. > *: O D (N IT> o CO o to to .c o CM o * CO tfi o os CN o a CO i z Ol -* to .C o (eq/j) p|BjA iooi babssbq so (eq/i) p|a|A 1001 BAesseQ 00 CT> 11 o -t- O O Z a. o o oo »* o ID O CO O z *: o o CO CO o to o CO o (0 o CN ■ o CN 0- (0 o CN Ol o CM - o o CO Z *: o o - CO CO -^T- ^ O o CN o to o CO o - o CD o CN Q. CO x: 3 3c 3 1 1 6 .S ft,■ — 3 ! ■s C 5 -2 IS Vs t 3 3 ci (eq/i) p|3|A iooj babssbq (%) luaiuoo ipjeis ^ m m cd o .s cx2 2 8 ■S■O aj Q< N V~> O Os m ON NO m Tt o oo »o Os VO Tj" —* CN en Tf* 0*Oh m en os o t*^ O^ © —* -H CS CN >n o O m >n s EH -C rn r- 00 r- ci en 1/1 SO • § * ^ 9 9 9 i H ITl T}- 00c 1 o£ o s o9 SI § a. t£ 9 2 9 9 .g H w © O 8 O M © d d d > 03 M o o m i—r cS (N N o iz* 9 9 9 9Ml H w 00C o o i © d d d 1 oS ID 1 i 00 2._ u9used D . *-N © ! O 00 8 ii o o o Id o o 00 • MM u8 9 1 SO S5 i 00 in in o .j? f oE_ J2 1 (N i E 0 tN i IT> Tj- (M ~M 00 (N •-^ e 8 -—V in in ITl in •pa DC U :s 00 *£ 1 i i M ITl i/l q q i ■*. •<* in r-' K s » 2 c vo _H ^H rn ~-< 1 S .2 !£ cm. od ri d |a > SO 4> If ^E ° 005 <« £ 5 oo m CO 00 ON ON ■o c s (*i Ia (N 8 (N vo' 5 a. .2 ^ o «* a. 3 O c« tn en • • 4) s So 35 .2 o g g ? .c > s a C/3 -• 5 20 15 Forest Cashew Rubber Cassava Figure 1. Clay content of Haplic Acrisols under different cropping systems. Values are the average of 7 profiles per cropping system. Bars with common letter are not significantly different at 5% by Duncan's Multiple Range Test. 3.0- o 2.0- M 1.0 b be - c d Forest Sugarcane Cashew Rubber Cassava Figure 2. Aggregate stability of Haplic Acrisols under different cropping systems. Values are the average of at least 7 profiles per cropping system and 3 horizons per profile. Bars with common letter are not significantly different at 5% by Tukey 's Studentized Range Test. Note: Aggregate stability class: 1= highly unstable, 2 — unstable, 3= relatively stable, 4=stable 262 .60 E c LI■a 3 1.50 1.40 1.30 - ab cd Forest Sugarcane Cassava Cashew Rubber Figure 3. Bulk density of Haplic Acrisols under different cropping systems. Values are average of at least 7 profiles per cropping system and 3 replicates per horizon. Bars with common letter are not significantly different at 5% by Tukey 's Studentized Range Test. r3 C C 40 r 30- 20 - E 10- Forest Sugarcane Cashew Cassava Rubber Figure 4. Infiltraton rate ofHaplic Acrisols under different cropping systems. Values are average of at least 7 profiles per cropping system and 3 replicates per profile. Bars with common letter are not significantly different at 5% by Tukey 's Studentized Range Test. 263 50 r 40 * 30 20 - 10- 0 a b ^ Forest Sx Rubber Sugarcane gg Cashew jfl Cassava lb b :•:•:•: 1 .".*.•• - ;Xv a !•!•!•! a ':■:':::.:.. •yy.' .•-;.:f--:.:.: ::::::: b b a ■*•*•*■ 1 .■.*.". '.■'.■'. an *•*•*■. :?; :•..■'..•;.■'• ■W V ¥ 1 b b _ i%*aV '•*•*•" ■ .:;-*.-/•:_■".■ 1 •,",", •gS: (; a a a a Wi 11 .Xv- 1 ::::• Soils derived from shale > Soils derived from liparit rock > Soils 270 derived from sand stone > Old alluvial soils. This research indicate that almost all cassava cultivated soils on sloping lands are poor to very poor in terms of nutrient content and availability. Therefore, for profitable and sustainable cassava cultivation on these soils, reasonable inputs are needed, together with measures to minimize soil erosion. SOIL CHANGES UNDER CASSAVA CULTIVATION 1. Soil Degradation as a Result of Cassava Cultivation The amount of soil eroded by rainfall in sloping areas depends on many factors, but among these the percent crop cover in the rainy season is very important. Table 2 presents data of soil erosion as affected by crop cover. The secondary forest had a vegetation cover of 80-90%, which resulted in the lowest amount of soil erosion. The highest amount of soil erosion occurred in the soil under cassava. Soil and nutrient losses by erosion are presented in Table 3. Under bush fallow, the amount of soil and nutrient loss was the lowest in comparison with that under tea and cassava. Contour cultivation in tea resulted in soil erosion washing away a depth of top soil of about 2.6 mm/year and a nutrient loss of about 23 kg N, 7 P and 6 K/ha, while the cultivation of cassava resulted in soil and nutrient losses that were 4-10 times higher compared to tea cultivation and bush fallow. So, it is clear that cassava monocropping on sloping lands may cause serious soil degradation by soil erosion. It is therefore necessary to use reasonable fertilizer inputs and erosion control measures. The changes in soil organic matter content under cassava monocropping are presented in Table 4. After continuous cassava monocropping the soil organic matter content decreased due to soil erosion and organic matter decomposition. After six years the soil organic matter content decreased to about half, and after ten years to only one third of that present in the first year. Data in Table 5 show that after 1-2 years of cassava cultivation the soil nutrient content decreased if no fertilizers were applied and no erosion control measures were used. The pH decreased 0.1-0.4 units in comparison with the same soil under forest. Organic matter was about 59-72% compared to the soil under forest. The contents of all nutrients also decreased after cassava cultivation. 2. Improved Management of Soil Nutrients in Cassava-based Cropping Systems To minimize soil degradation resulting from cassava cultivation, soil conservation measures should be applied. Some research results on the effects of fertilizer application to conserve soil fertility and increase crop yields are presented and discussed. Effect of application of mineral fertilizers on cassava yield The experiments were conducted for three years on soil derived from shale. 271 Figure 1 shows that during three years of continuous cassava cultivation with no fertilization, the yield of cassava decreased from 12.3 to 8.6 t/ha; thus, the productivity of soil decreased to about 70% of that in the first year. Table 2. Vegetation cover and amount of soil erosion under different crops in soil derived on shale in Thai Nguyen province. Vegetation/crop Crop cover (%) Soil erosion (t/ha) Secondary forest 80-90 12.4 Maize 30-50 14.7 Upland rice 10-15 95.1 Cassava 10-15 98.6 Source: Nguyen Tu Siem and Thai Phien, 1993. Table 3. Soil erosion and nutrient losses in eroded soil due to the cultivation of cassava and tea in comparison with bush fallow in Than Hoa district of Vinh Phu province. Crops Depth of soil loss (mm/year) Dry weight soil loss (t/ha/year) Nutrient loss (kg/ha) N P K Cassava 12.3 2.6 1.0 145 33 145 23 10 48 26 7 6 2 3 Tea Bush fallow 12 Source: Thai Phien and Nguyen Tu Siem, 1993. 272 Table 4. Change in organic matter (OM) content and composition due to continuous cassava cultivation for various years at Phu Quy in 1994. Soil Total OM (%) Humic acid (%) Fulvic acid Humic/Fulvic acid ratio Cassava 1 year 1.72 0.48 0.64 0.75 Cassava 6 years 0.80 0.21 0.33 0.64 Cassava 10 years 0.55 0.09 0.36 0.25 Source: Nguyen Tu Stem and Thai Phien, 1993. Table 5. Chemical characteristics of basaltic rock derived soil under forest and cassava cultivation at Phu Quy in 1994. Soil depth (cm) Total (%) Exchangeable (me/ 100 g soil) pHKa OM N P Ca Mg Forest 0-10 20-40 40-60 4.2 4.1 4.4 5.80 3.30 3.01 0.26 0.11 0.10 0.11 0.04 0.04 2.00 1.18 1.60 0.80 0.40 0.76 Cassava (1 year) 0-20 20-40 40-60 3.8 4.1 4.3 4.18 2.19 1.08 0.08 0.11 0.10 0.05 0.09 0.05 1.46 1.40 1.20 1.20 0.40 1.60 Cassava 0-20 20-40 3.8 3.8 3.40 2.08 0.14 0.06 0.10 0.09 0.12 0.04 0.04 0.04(2 years) Source: Nguyen Tu Siem and Thai Phien, 1993. 273 o o > (/l U 25- 20 - O 2 15 10 - - - - -° N-P205-K20(kg/ha) D = 0-0-0 X = 30-30-60 O = 60-60-120 1992 ~l— 1993 1994 Figure 1. Effect offertilizer application on cassava yield during three years ofcontinuous cropping on a soil derivedfrom shale in north Vietnam. Application of NPK fertilizers increased the yield 71-1 12% in comparison with the control. The effects of N, P and K are presented in Table 6. Application ef 30 and 60 kg N/ha increased cassava yields from 16.9 to 17.1 and 17.7 t/ha, respectively. Application of 30 and 60 kg P2O5/ha increased cassava yields from 15.6 to 17.1 and 19.6 t/ha, respectively, 10-24% higher than the control. Application of 60 and 120 kg I^O/ha increased cassava yields 31 and 41%, respectively, compared to the control. The response of cassava to nutrient application in this soil can be ranked as K> P> N. 274 Table 6. Effect of annual application of N, P and K fertilizers on cassava yields. Data are average values for three years. Fertilizer Cassava fresh root yield Relative to applied Treatments" (t/ha) control (%) NPK NoPoKo 10.0 100 N,P,K, 17.1 171 N2P2K2 21.2 212 N N0P,K, 16.9 169 N,P,K, 17.1 171 N2P,K, 17.7 177 P N,P0K, 15.6 156 N,P,K, 17.1 171 N,P2K, 19.6 196 K N.P.Ko 13.1 131 N.P.K, 17.1 171 N.P.K, 19.5 195 " No=0 Po=0 Ko=0 N, = 30 kg N/ha P, = 30 kg P205/ha K, = 60 kg K20/ha N2=60 kg N/ha P2 = 60 kg P20,/ha K2= 120 kg K20/ha Effect of farm-yard manure, inorganic fertilizers and lime in a cassava-grain legume intercropping system Soil fertility improvement can also be achieved by intercropping cassava with grain legumes in addition to fertilization. Table 7 shows that application of NPK fertilizers combined with lime (T2, T4) increased the yield of crops, and that these yields were higher than in the treatment with only farm yard manure (FYM). The combination of NPK and FYM (T3, T5) produced higher yields than that of NPK and lime. This indicates that in these hilly soils application of FYM was more effective than that of lime. The combination of NPK with lime and FYM resulted in the highest yield. It shows that the high level of NPK together with FYM and lime promoted the growth and yield of crops. Cassava-legume intercropping systems can achieve yields of 1 1-19 t/ha of cassava and 78-160 kg/ha of black bean or 550-800 kg/ha of peanut; they tend to be more economical than monocropping of cassava. 275 Table 7. Effect of three years of annual fertilizer application on crop yields in a cassava-grain legume intercropping system at Hoa Son in 1995. Cassava yield (t/ha) Intercrop yields (kg/ha) Black bean Peanut Treatment1 Mean (1992-1994) (1992) (1993-1994) 1. FYM 11.5 78 546 2. low NPK+ lime 14.4 96 730 3. low NPK +FYM 16.8 129 836 4. high NPK + lime 14.8 104 731 5. high NPK +FYM 16.5 143 788 6. high NPK + lime+FYM 18.8 162 870 " FYM = 3 t pig manure/ha. low NPK = 25 kg N, 50 P205, 50 K20/ha. high NPK = 50 kg N, 100 P205, 100 K20/ha. Source: Thai Phien, Nguyen Tu Siem and Nguyen Cong Vinh, 1995. Soil nutrient balance as affected by fertilization Every year a considerable amount of nutrients are removed by run-off and soil erosion, as well as by crop removal (Table 8). In the case of monocropping of cassava, the annual amount of nutrients removed by cassava was 62-153 kg N, 83-181 kg P205 and 67-147 kg K20/ha. Soil fertility will decline unless nutrients are returned to the soil in the form of fertilizers or manures, or the crop is combined with leguminous crops like black bean, peanut etc. between cassava rows, and the residues of these intercrops are returned to the soil. Even with fertilization of 30-60 kg N, 30-60 P205 and 60-120 K2O/ha, the nutrient imbalance that occured varied from 62-120 kg N, 25-59 kg P and up to 72 kg K/ha (Table 8). On the other hand, nutrients can be returned by intercropping and mulching with crop residues. Crop residues can provide organic matter and nutrients to the soil (Table 9) and improve the soil's physical properties. The nutrient loss in the crop harvest of cassava and beans in a cassava-bean intercropping system was found to be 99-153 kg N, 25-49 P, 31-40 K, 36-58 Ca and 11-19 Mg/ha. Valuable nutrient resources would be taken out of the soil if no crop residues were returned to the soil. The return of crop residues, especially intercropped leguminous residues, can return to the soil about 49-80 kg N, 15-25 P, 10-15 K, 17-28 Ca, and 5-9 Mg/ha. Those are equal to about 110-180 kg urea, 300-500 kg fused magnesium phosphate and 20-30 kg KCl/ha. The effect of hedgerow farming and intercropping on soil fertility In Vietnam, cassava is mostly planted on steep lands without or with very low 276 inputs, and without soil erosion control measures. One effective way to minimize soil erosion is the planting of hedgerows of certain plant species along contour lines. The effect of crop management on soil erosion is presented in Table 10. Intercropping with grain legumes and contour hedgerows in combination with fertilizer application reduced soil erosion. On bare land soil loss and run-off were highest. Run off under monocropped cassava (T2) was reduced by 13% compared to bare land. In cassava intercropped with beans (T3), run-off was only 84% of that of bare land (T3), while intercropped black bean/peanut and contour hedgerows further reduced run-off and erosion. Fertilizer application also had a marked effect on soil erosion: in the treatments of hedgerow farming with high inputs of fertilizers, soil loss was only 2.8-3.7 t/ha, while with the low input treatment soil loss was 4.8 t/ha. A large amount of nutrients can be washed out in the eroded soil. It depends on the cultivation method and fertilizer application. These are presented in Table 11. In an experiment conducted on an exhausted soil derived from Liparit in Tarn Dao, Phu Tho province, 7.8 kg N, 3.9 kg P and 7.2 kg K/ha were washed out as part of the eroded soil from bare land. In monocropped cassava (T2) nutrients are not only lost by soil erosion, but also by removal of the crop products and residues. By intercropping cassava with peanut and planting contour hedgerows, some nutrients are returned by mulching with crop residues (T3, T4, T5, T6). Nutrient losses were lower in treatments with high inputs than in those with low inputs. With high inputs and intercropping, more nutrients were returned than lost. When peanut/black bean were intercropped with cassava, the positive nutrient balance varied from 47 to 63 kg N, 0.3 to 2.0 kg P and 19 to 27 kg K/ha/year. Table 8. Effect of fertilization on the nutrient balance in soil after cassava monocropping at Hoa Son, 1995. kg N/ha kg P205/ha kg K2O/ha Treatment FD R" F-R F R F-R F R F-R Tl 0 62 -62 0 36 36 0 56 56 T2 30 108 -78 0 54 54 42 85 43 T3 30 76 -46 13 49 36 0 72 72 T4 30 135 -105 13 38 25 84 73 11 T5 30 125 -95 13 63 50 42 97 55 T6 0 120 -120 13 61 48 42 94 52 T7 60 134 -74 13 72 59 42 111 69 T8 30 138 -108 26 69 43 42 108 66 T9 60 153 -93 26 79 53 84 122 63 " p =F = applied in fertilizers; R = removed in crop 277 Table 9. Biomass and nutrients removed or returned (kg/ha) to the soil in a cassava-bean intercropping system at Hoa Son, 1995. Biomass Treatments returned N P K Ca Mg Tl L" 99 24.9 31.2 35.7 11.4 R" 3910 49 14.7 10.1 17.1 5.4 T2 L 118 36.9 36.3 45.7 14.5 R 4050 60 18.5 12.1 20.7 6.6 T3 L 121 39.3 39.3 45.0 15.7 R 4220 64 19.7 13.1 22.9 7.2 T4 L 135 43.1 41.3 50.7 16.9 R 4450 76 22.2 14.1 25.7 8.4 T5 L 136 43.1 45.4 54.3 17.5 R 4910 79 24.6 15.2 27.1 9.0 T6 L 153 49.2 46.4 57.8 19.3 R 507 80 24.8 15.2 27.9 9.0 " L = Lost in crop harvest; R = Returned in crop residues Source: Thai Phien, Nguyen Tu Siem and Nguyen Cong Vinh, 1995. Table 10. Soil loss and run-off from degraded soil derived from Li pari t (slope: 9-12%) at Tarn Dao, Vinh Phu in 1994. Run-off Dry sofl loss Treatment Cultivation measures Tl Bare land (control) T2 Cassava monocrop T3 Cassava + black bean or peanut with low input T4 Cassava + bean or peanut + hedgerow + low input T5 Cassava + bean or peanut + hedgerow + high input T6 Cassava + bean or peanut + mixed hedgerow + high input m3/ha/year t/ha/year 14,539 6.9 12,678 6.9 12,233 6.1 12,031 4.8 11,473 2.8 10,674 3.7 Source: Huynh Due Nhan et al. , 1995. 278 Table 11. Nutrient loss in eroded soil and return in crop residues (kg/ha/year) at Tarn Dao, Vinh Phu in 1994. N P K Treatment" L: R2> D L R D L R D Tl 7.8 0 -7.8 3.9 0 -3.9 7.2 0 -7.2 T2 7.6 0 -7.6 5.7 0 -5.7 6.2 0 -6.2 T3 6.7 66.2 + 59.5 5.6 5.9 + 0.3 7.3 30.4 +23.1 T4 8.1 55.3 + 47.2 4.8 5.3 + 0.5 7.6 26.6 + 19.0 T5 4.5 67.2 +62.7 4.4 6.3 + 1.9 5.9 33.4 + 27.5 T6 6.7 60.3 +53.6 4.3 6.3 + 2.0 6.7 31.8 +25.1 " Treatments as in Table 10 21 L = Loss; R = return with crop residues, D = R-L. Source: Huynh Due Nhan et al., 1995. The effect of fertilization and biological technology on crop yield The effect of fertilization and intercropping with grain legumes on crop yields is shown in Table 12. Fertilization and contour hedgerows increased the total gross income obtained in a cassava-peanut intercropping system. Monocropping of cassava produced a yield of 10.8 t/ha, while the intercropping system produced 7-9 t/ha of cassava and 450-500 kg/ha of peanut per year, resulting in a gross income of 5.5-6.6 million dong/ha, compared with 4.9 million dong/ha for monocropping. Table 12. Effect of soil and crop management practices on the yields of cassava and intercrops, as well as on total gross income in Tan Dao, Vinh Phu in 1994. Yield of cassava Yield of peanut Gross income2) Treatments" (t/ha) (kg/ha) C000d/ha) T2 10.8 0 4860 T3 9.1 498 6585 T4 7.6 450 5670 T5 7.9 466 5885 T6 6.9 479 5500 " Treatments as in Table 10 2) Prices: cassava d 450 /kg fresh roots, peanut 5000 /kg dry pods. Source: Huynh Due Nhan et al. , 1995. 279 CONCLUSIONS 1 . In nearly all soils used for cassava cultivation adequate soil erosion control measures have not been used; as a result, they have become degraded in terms of soil physical properties and nutrient contents. 2. Application of mineral fertilizers to cassava markedly increased cassava root yields, the increase ranging from 20 to 112% compared with the control treatment without fertilization. 3. The combined application of organic and mineral fertilizers increased cassava yields 43-46% compared to the application of organic manures alone. Highest cassava yields were obtained when combining organic and mineral fertilizers with lime. 4. Intercropping grain legumes (bean, peanut) between cassava rows reduced tbe amount of soil loss by 22-59% compared to cassava monocropping, and contributed to a positive nutrient balance, improved soil fertility and increased income. REFERENCES Huynh Duc Nhan, Nguyen Duong Tai, Tran Duc Toan, Thai Phien and Nguyen Tu Siem. 1995. The management of acid upland soils for sustainable agricultural production in Vietnam: I. Tarn Dao site. IBSRAM, Bangkok, Thailand. Thai Phien, Luong Duc Loan, Nguyen Van Truong, Dau Cao Loc and Nguyen Thi Dan. 1994. The management of sloping land for sustainable agriculture in Vietnam. IBSRAM, Bangkok, Thailand. Thai Phien, Nguyen Tu Siem and Nguyen Cong Vinh. 1995. The management of acid upland soil for sustainable agricultural production in Vietnam: II. Hoa Son site. IBSRAM, Bangkok, Thailand. 280 PROGRESS IN AGRONOMY RESEARCH IN INDIA C.R. Mohankumar1, V.P. Potty1, C.S. Ravindran\ S. Kabeerathummd and C.R. Sudharmai Devi1 ABSTRACT Nutritional studies of cassava have shown that in high-P acid laterite soils, there was no response to P application during four years of consecutive cropping. In low-P soil, even though cassava initially responded to 100 kg P2O5/ha, the response gradually declined. The optimum economic dose of P for cassava in acid laterite soil is about 45 kg P2O5/ha. Studies on the NPK requirement of a short duration variety of cassava in a rice-based cropping system, showed that there was a response to the application of 100 kg each of N and K2O/ha, but no significant response to P. In upland rice fields the sequential cropping of vegetable cowpea and cassava can eliminate the need for FYM application by incorporating the crop residue of vegetable cowpea before planting cassava. Application of N and K had a significant effect on root yield of short-duration (7 months) lines of cassava, CI-649 and CI-731, up to 75 kg N/ha and 100 kg K2O/ha. Studies conducted by Kerala Agric. Univ. have shown that up to 50% of the KC1 requirement of cassava can be substituted by sodium chloride (NaCl) without any deleterious effect on crop yield. Another study revealed that under partial shade of adult coconut palms (76% PAR), rainfed cassava (cv Sree Visakham) yielded 77 % of that grown in the open. Transplanting of rooted cassava cuttings, inoculated with the VA-mycorrhizal fungus (VAMF) Glomus microcarpum var. microcarpum, was found to enhance the total dry matter and root yield, besides increasing the concentration of micronutrients like Cu and Zn in the leaves. Mass inoculation with VAMF in farmers' fields by planting rooted infected cuttings showed that 70% of the farmers were convinced that they could obtain a higher yield of roots using this practice. The present practice of planting cassava is to plant stakes directly in the field with the onset of pre-monsoon rains. In this practice, the uncertainty of rainfall may cause poor establishment. In order to overcome this, cassava stakes are first planted in a nursery at a very close spacing of 4.5 x 4.5 cm so as to accommodate about 500 stakes/m2. Uprooting is easiest when a saw dust media is used in the nursery. Root yield was not affected by the time of uprooting and transplanting in the field between 15 and 25 days, but beyond that age there was a significant reduction in root yield. This technique can be effectively used when short duration varieties are grown under rainfed conditions in areas where rainfall is limited to 4 to 5 months per year. The plants are first raised in a nursery for 25 days and then transplanted in the field with the onset of rains. INTRODUCTION Cassava (Manihot esculenta Crantz), locally known as tapioca, was introduced in Kerala state of India by the Portuguese in the 17th century. Of the different tropical root and tuber crops grown in India, cassava is of significant importance since it can produce more calories per unit of land area. Its importance in tropical agriculture is due to its 1 Central Tuber Crops Research Institute (CTCRI) Triruvananthapuram, Kerala, 695017, India. 2 Kerala Agricultural University, Thiruvananthapuram, Kerala, 695017, India. 281 drought tolerance/drought avoidance, wide adaptability to diverse soil, nutrient and management conditions, including the time of harvest. The drought tolerance is mainly due to the in-built mechanism to shed/drop the leaves under low soil moisture conditions to facilitate a slow down of all the vital activities of the plant. When soil moisture increases it again puts forth fresh growth and starts accumulating starch in the storage roots. Being a photo-insensitive crop, cassava can be profitably cultivated throughout the year with irrigation. About 70% of the cassava production in India is used as food, either directly or in processed forms. The most popular and traditional mode of consumption is in the form of cooked and mashed roots. After removing the outer rind and inner thread-like fibrous center core, fresh roots are cut into pieces, cooked in boiling water, decanted and eaten with fish or coconut gratings. Other methods of consumption are as chips fried in oil, or sun-dried chips, which are made into flour and used for preparations similar to those of rice or wheat flours. "Sago" is an important food product derived from cassava starch. It is consumed as a convalescence food in many parts of India. In India the cultivation of cassava is mainly confined to the southern states, i.e. Kerala, Tamil Nadu, Andhra Pradesh and Karnataka, and parts of the north-eastern regions (Table 1). In Kerala where the annual mean rainfall is about 3000 mm, distributed over a period of 7-8 months from April-November, cassava is grown under rainfed conditions. Meteorological data of the major cassava growing states in India are presented in Figures 1 and 2. In Tamil Nadu where the annual rainfall is only about 1000 mm, distributed over a period of 4-5 months, the crop is grown mainly under irrigation and with high-input management practices. Cassava Soils and their Characteristics In Kerala, cassava is mainly grown on laterite soils (Ultisols), followed by forest soils (Mollisols) and red soils (Alfisols). In Tamil Nadu, the major soil groups under cassava are red soils (Alfisols) and black soils (Vertisols). In Andhra Pradesh the cultivation is mainly confined to the alluvial (Entisols) and red soils (Alfisols). The main characteristics of these soils are shown in Table 2. Micronutrient status of cassava soils A critical appraisal of the availability of micronutrients in the major cassava growing states revealed that in most cases the soil have micronutrient contents that are well above the critical limits (Table 3). Inspite of that, there are locations where there is a response to the application of zinc (Zn). It was noted that Zn availability in laterite soils did not follow any relationship with the physio-chemical properties of the soil. Ttgs Si os _H o■ ON 1 i*-■^^ 213 00 £ o 00 OO 03••a IB £ Q. J* C ed c tj■33 '£• ■o c ■1 i 1 Q. ed H § 1 O Os ON o Os O OO O ON■~ ON ON o 00 o ON 00 O Tf- r~- m r~ oo Os Os Os -h n Os ON - OO ci Os ON 00 O i ON 00 SOsr- 1H m c> q Os 00 — Ov 8 Os' C^ IT) « m N© r^ i m rJ Os' NO3 W 3 N NO o - W N n Os t- O vi ci fi O c■> n so oo n n m -- -h od o ;* m - - «s NO NO O Ml OO NO 1-h c> OO ft O O o r•~ Os n ■-> e»> O Z a Months Figure 7. Rainfall (top) and temperature (bottom) at CTCRI in southern Kerala, India, during 1993 to 1995. 285 ANDHRA PRADESH + = Max temperature ■ = Min temperature m = 1993 3 = 1994 B = 1995 4c I . I TAMIL NADU c U Months Figure 2. Rainfall and maximum and minimum temperature in Andhra Pradesh (top) and Tamil Nadu (bottom), during 1993 to 1995. 286 In about 47% of the surveyed area of black soils of Tamil Nadu, the available Zn content was below the critical limit and foliar yellowing and low productivity were noticed. Both red and black soils of Tamil Nadu have a high pH (7.4 and 8.5, respectively) and Zn availability in these soils was negatively correlated with pH. The red soils of Andhra Pradesh had comparatively higher levels of available Zn. Regarding the availability of iron (Fe) all soils had available Fe contents above the critical limit. In Tamil Nadu red soils had a high level of Fe. The black soils recorded the lowest content of Fe. Iron deficiency in some parts of these black soils are due to the calcarious parent material of these soils and, consecuently, a high soil pH (8.4). Lime-induced Fe chlorosis was observed in black soils of Tamil Nadu. In Andhra Pradesh red soils had high Fe availability. With respect to the availability of manganese (Mn), all soils are considered sufficient in Mn availability. Manganese availability was negatively correlated with pH. The availability of copper (Cu) in the major cassava growing soils of Kerala, Tamil Nadu and Andhra Pradesh is also considered adequate as all soils had available Cu contents above the critical limit. Nearly all cassava soils in Kerala are low in available boron (B), while red soils in Tamil Nadu and Andhra Pradesh also have very low levels of B. Nevertheless, significant responses of cassava to B application have seldom been reported. RESEARCH RESULTS A. Cultural Practices for Cassava 1 . Standardization of nursery techniques for cassava The present practice of cassava planting in traditional areas is to plant the stakes directly in the field after the receipt of pre-monsoon showers. But in non-traditional areas, where the rainfall is limited to 4 or 5 months, it is recommended to plant stakes first in nursery beds at a very close spacing of 4.5x4.5 cm under irrigation during three weeks before the onset of regular rains. The settlings thus raised in the nursery are transplanted in the main field with the onset of monsoon rains. 1.1 Effect of nursery medium and time of transplanting Research results (Table 4) indicate that time of transplanting had significant effects on root yield. Maximum root yields were obtained when 20 day old settlings were transplanted in the main field. The type of rooting media had no significant effect on root yields. The main advantages of the nursery technique are: 1 . Settlings expressing symptoms of cassava mosaic disease can be discarded in the nursery itself, thereby greatly reducing the incidence of mosaic in the main field. 2. As the settlings are raised prior to the rainy season and transplanted to the main field 287 with the onset of the monsoon rains, the crop can more effectively utilize the short rainy season for growth and yield. This is especially advantageous in areas having a very short rainfall period. 3. In areas where the crop is raised under irrigation, this technique can reduce the number and quantity of irrigation to the crop. 4. When stakes are planted directly to the main field about 10% of the stakes may fail to germinate. This technique will reduce the percentage of failure in the main field. 5. Transplanting of healthy and uniform-sized settlings generally results in a more uniform crop stand in the main field. 1.2 Effect of stem portion and stake treatment A study was conducted to determine the effect of stem portion and treatment of nursery-grown stakes on the root yield of cassava. Stakes collected from the bottom 1/3, middle 1/3 and top 1/3 portion of the stem were treated with either VA-mycorrhizal fungi (VAMF), plant protection chemicals, 2% zincsulfate solution or 50 ppm IBA. Water- treated stakes were used as control. The stakes thus treated were planted in the nursery and 25 day old settlings were transplanted in the main field. The results (Table 5) indicate that there was no significant difference in yield due to stake treatment. Regarding the stem portion, cuttings derived from the bottom and middle portion of the stem were significantly superior to those obtained from the top portion. From this study, it is concluded that while selecting planting material, stakes derived from the bottom and middle portion (lower 2/3 length of the stem) are superior to those from the top part. 1.3 Field trial of rooted infected cuttings The production of rooted infected cuttings using different media like sawdust, sand and polypack with garden mixture were assessed. The sawdust was found to be the best medium. Among dates of transplanting, the 25th day seemed to be good compared to the 20th day of transplanting. 2. Sequential cropping with vegetable cowpea The recommended dose of farm-yard manure (FYM) to be applied to cassava is 12 t/ha. In order to substitute for this high dosage of FYM a short-duration leguminous crop of vegetable cowpea was grown preceding cassava, both under lowland and upland conditions. Under lowland conditions, vegetable cowpea followed by cassava was found-to be a feasible alternative to the use of FYM. The yield reduction under such conditions was only 12% when compared to the control, where there was no preceding crop of vegetable cowpea (Table 6). The vegetative matter produced by the seasonal crop was sufficient to provide enough organic matter to cassava; however, under upland 288 conditions, the cassava crop that followed vegetable cowpea suffered from a root yield reduction of 30%. The significant reduction in yield was due to moisture deficiency as a result of late planting and subsequent drought affecting the crop at the root bulking stage. When cassava was planted in May, up to the harvest of the crop in December, there was no serious moisture stress, as the crop was receiving the monsoon rains both for growth and yield. 3. Influence of photosynthetically active radiation (PAR) in intercropped cassava A trial was conducted at the Kerala Agricultural University to study the influence of PAR on the root yield of cassava grown as an intercrop under coconut trees during 1992/93 and 1993/94. Cassava was grown as an intercrop under the partial shade of about 40 year old coconut palms planted at a spacing of 7.5x7.5 m. The planting patterns of the cropping systems used are shown in Figure 3. PAR incidence on the cassava canopy was measured at monthly intervals with the help of a line quantum sensor. PAR received on the canopy of the intercropped cassava was less (76%) compared to cassava grown under full sunlight (Figure 4). During the rainy months not much variation was noticed between the quantum of PAR under shaded and open conditions, whereas in the dry months the PAR received in the open was considerably higher than that in the shaded area. The root yield of cassava grown under the partial shade of coconut palms was less (68%) compared to cassava grown in the open (Table 7). As the shade was more or less uniform in the coconut garden, no significant differences in cassava root yield were observed due to the various crop combinations. B. Soil Fertility Management 1. Long-term effect of manures and fertilizers in acid cassava growing soils Table 8 shows that increased cassava yields were obtained with an increase in the level of fertilizer application. Among different nutrient levels, the highest yield was obtained with 125:50:125 kg/ha of N, P2O3 and K20, in addition to 12 t/ha of FYM. This treatment was not significantly different from 100:50:100 + FYM. Application of FYM had a significant effect on yield. Application of the recommended dose of N, P205 and K2O, i.e 100:50:100 kg/ha but without the recommended dose of 12 t FYM/ha, resulted in significantly lower yields. Sources of P, i.e. triple superphosphate or Mussorie phosphate, had no significant effect on yield. Application of Zn with the recommended rate of NPK gave a significantly higher yield. 289 Table 4. Effect of nursery media and time of transplanting of settlings in the main field on cassava root yield (t/ha) at CTCRI, Trivandrum, Kerala, India. Rooting Time of transplanting of settlings in the main field (days) media 10 15 20 25 30 Mean Sawdust 25.5 26.7 26.1 24.0 22.6 25.0 Sand 27.8 26.6 27.1 24.5 21.9 25.6 Pot mix 25.2 26.5 28.3 24.5 20.3 25.0 Direct planting 25.0 25.5 26.7 23.2 18.6 23.8 Mean 25.9 26.3 27.1 24.0 20.8 - CD(5%): media NS; time 2.588 Source: Mohankumar, 1993. Table 5. Effect of stem portion and various treatments of nursery-planted stakes on cassava root yield (t/ha) in the field at CTCRI, Trivandrum, Kerala, India. Stakc treatments Top Stem portion Middle Bottom Mean VA-mycorrhizal fungi PPC (Plant protection chemicals) Zinc sulfate (2% solution) IBA (50 ppm) Control (water) Mean 27.1 30.2 30.1 29.1 27.4 29.3 30.0 28.9 28.3 30.1 30.5 29.6 27.9 29.4 30.3 29.2 28.1 30.3 30.1 29.5 27.8 29.9 30.2 . CD(5%): stake treatment NS; stem portion 0.466 Source: Mohankumar and Potty, 1994. 290 Table 6. Biomass production of vegetable cowpea and root yield (t/ha) of subsequently grown cassava as compared to the root yield of cassava grown without cowpea under upland and lowland conditions at CTCRI, Trivandrum, Kerala, India. Crop (date of planting-harvest) Lowland Upland Vegetable cowpea-cassava Biomass of vegetable cowpea (May-July) Root yield of cassava (July-Febr) Root yield of cassava alone (May-Dec) 20.5 9.1 51.1 23.5 58.0 33.1 Source: Mohankumar and Nair, 1996. Table 7. Root yields of cassava grown in various intercropping systems in a coconut garden, as compared with that of monoculture cassava grown in full sunlight at Kerala Agric. Univ., Trivandrum, Kerala, India Cropping systems" Cassava root yield(t/ba) 1992/93 1993/94 Average 1. Co+ Ca 2. Co + Ca+Vcp 3. Co + Ca+EFY 4. Co+Ca+Ba 5. Co +Ca + Vcp +EFY + Ba Mean CD(5%) SEM + 25.01 19.43 22.22 19.75 20.11 19.93 29.79 31.56 30.67 27.32 22.26 24.79 30.20 29.21 29.70 26.41 24.51 25.46 NS NS 2.642 3.247 6. Cassava monoculture (full sunlight) 35.38 30.23 32.80 hCo = Coconut, Ca = Cassava, Vcp = Vegetable cowpea, EFY = elephant foot yam; Ba = banana Source: Ravindran, 1996. 291 Table 8. Effect of annual applications of manures and fertilizers on the root yield of the sixth successive crop of cassava grown in an acid Ultisol at CTCRI, Trivandrum, Kerala, India. Treatments" Root yield (t/ha) 50:25(S):50 + FYM 21.7 75:37.5(S):75 + FYM 24.4 100:50(S):100 + FYM 29.4 125:50(S):125 + FYM 30.4 50:25(M):50 + FYM 19.1 75:37.5(M):75 + FYM 20.4 100:50(M):100+FYM 25.1 125:50(M):125 + FYM 31.8 100:50: 100 (noFYM) 18.3 100:50: 100+ 1/2FYM 23.3 100:50: 100+crop residue incorporated (no FYM) 25.9 100:50: 100 + Zn (12.5kg Zn2 SO4) (no FYM) 31.1 Ash+FYM 15.5 Ash+crop residue incorporated (no FYM) 10.2 CD(5%):5.65 "(S)=Triple superphosphate; (M)= Mussorie rock phosphate; FYM= 12 t/ha of farm yard manure Source: Kabeerathumma and Ravindran, 1996. 292 A. COCONUT B. COCONUT+ CASSAVA C. COCONUT+CASSAVA+ VEGETABLECOWPEA D. COCONUT+CASSAVA + ELEPHANTFOOT YAM E. COCONUT+ CASSAVA + BANANA [90cm • * ****** coconut cassava F. COCONUT+CASSAVA + BANANA + ELEPHANT FOOT YAM + VEGETABLE COWPEA O elephant foot yam 0 banana vegetable cowpea Figure 3. Planting patterns of component crops in various coconut-based intercropping systems. 8 8 "T- 8 (nas/jUi/gr/) uouEipEJ aAipE-XuEansqiuXsoioqd u. - 3 Q Z o < 2 Q Z o < c —° ■5 E «3 u u. + O Z « I < 5 S3 u + o I I c I CO c cI c 3 I C 5 -a 8 * a a 2 S s - 2■*■* I ! 294 2. Phosphorus requirement under upland conditions Under upland conditions of acid lateritic soils of high-P status, skipping of P application for the first four years had no significant effect on cassava yield. In low-P soils, even though cassava responded initially to 100 kg P205/ha, the response gradually declined. The optimum economic dose of P for cassava in acid lateritic soils was determined to be about 45 kg P205/ha (Nair et al., 1988). 3. Fertilization of short-duration varieties of cassava in a rice-based cropping system 3. 1 NPK requirements The common cropping system in a single crop rice field under upland conditions in Kerala is to grow a single crop of dry-sown paddy during the wet season (June-Sept); afterwards the land is left fallow or rice is followed by a pulse crop. Farmers are forced to adopt this system due to a lack of irrigation facilities to raise a second crop of rice. Cropping system studies conducted at CTCRI have shown that a short-duration (early-maturing) variety of cassava, "Sree Prakash", with seven month duration can be grown successfully in a rice-based cropping system. In order to standardize the NPK requirement of Sree Prakash grown under this system, a study was carried out with three levels of nitrogen (50, 100, 150 kg/ha), three levels of P205 (25, 50, 75 kg/ha) and three levels of K20 (50, 100, 150 kg/ha) after the harvest of wet season rice. Table 9 shows that application of 100 kg N/ha increased the yield significantly over 50 kg/ha. A further increase in N did not show any further positive response. There was no significant response of cassava to P application in this rice-based cropping system. Application of K at 100 kg K,0/ha showed a significant improvement in yield over 50 kg/ha, but this yield was not significantly different from that at 150 kg/ha. The combined effect of N and K was also significant. However, additional increases in the level of N and K beyond 100 kg/ha did not produce any significant effect on yield. From the study it could be concluded that the variety "Sree Prakash" responded to a fertilizer dose of 100 kg N, 25 P205 and 100 K20/ha in this rice-based cropping system. 3.2 N and K requirements Two promising pre-released accessions of cassava, i.e. CI-649 and CI-731 of seven months duration were tested for standardizing the N and K requirements in a rice- based cropping system. The first rainy season crop was rice followed by short-duration cassava. A short-duration rice was grown during the wet season. "Sree Prakash", a released short- duration cassava variety was used as a control with a fertilizer dose of 75:50:75 kg of N, P205 and K20/ha. Two short duration accessions of cassava, CI-649 and CI-731, were tested under three levels of N and K. 295 Table 9. Effect of levels of N, P and K application on the root yield (t/ha) of cassava, cv Sree Prakash, grown after rice under lowland conditions at CTCRI, Kerala, India. "25 "so '75 K50 Kioo K-150 Mean Njo 34.0 32.8 33.4 31.1 33.8 35.2 33.4 N100 36.8 40.2 38.3 34.5 40.7 40.1 38.4 N150 37.5 39.0 38.2 36.2 38.5 40.2 38.3 Mean 36.1 37.3 36.6 33.9 37.7 38.5 CD (5%) N = 1.385; K = 1.385; P not significant Source: Mohankumar et al. , 1996. The cassava variety CI-649 recorded a mean yield of 30.53 t/ha, which was significantly superior to that of CI-731 or the released variety Sree Prakash. Tables 10 and 11 show that the application of N had a significant effect on root production. The highest yield was recorded at 75 kg N/ha, which was significantly superior to that at 50 kg N/ha, but not significantly different from that at 100 kg N/ha. Tables 10 and 12 show that application of K had a significant effect on root yield. The highest yield was obtained with the application of 100 kg K20/ha, which was significantly superior to that at 50 or 75 kg K20/ha. Highest root yields of both varieties were obtained with 75 kg N and 100 K20/ha. 4. Effect of fertilization on the yield of triploid cassava 4.1 Effect of NPK Some of the triploid cassava varieties developed have given higher yields and had higher dry matter contents than diploids. In order to suggest a suitable fertilizer schedule, the triploids were grown under three levels of fertilization. The results are presented in Table 13. The triploid variety 2/14 recorded the highest yield of 25.58 t/ha, which was significantly superior to that of 9/76, but not significantly different from that of 237/84. Regarding levels of fertilizer, the dose of 75:50:75 resulted in a significantly higher yield than that of 50:25:50, but was not significantly different from 100:50:100 kg/ha of N, P2O5 and K20. 296 Table 10. Effect of various rates of N and K application on the root yield (t/ha) of two short-duration lines and a released variety of cassava grown in lowland soils at CTCRI, Trivandrum, Kerala, India. Fertilizer levels(kg/ha) NiKjO." Variety CI-649 CI-731 Sree Prakash (control) 27.8 27.7 . 29.3 27.0 - 31.7 27.6 - 30.7 28.8 - 29.7 28.2 27.2 33.5 31.0 - 30.5 28.8 - 29.8 29.1 - 31.7 29.4 - 50:50 50:75 50:100 75:50 75:75 75:100 100:50 100:75 100:100 Mean 30.5 28.6 CD (5%):variety 0.77; CD: Control vs treatment 2.33 "P was constant at 50 kg P2O5/ha Source: Mohankumar, 1996. Table 11. Average effect of various rates of N on the root yield (t/ha) of two short-duration cassava lines grown in lowland soils at CTCRI, Trivandrum, Kerala, India. Levels of N (kg/ha) Cassava lines 50 75 100 Mean CI-649 CI-731 Mean 29.6" 27.4 28.5 31.3 29.3 30.3 30.7 29.1 29.9 30.5 28.6 CD(5%):0.95 °Date are average values for three levels of K; P was constant at 50 kg P2Os/ha Source: Mohankumar, 1996. 297 Table 12. Average effect of various rates of K on the root yield (t/ha) of two short-duration cassava lines grown in lowland soils at CTCRI, Trivandrum, Kerala, India. Levels of K20 (kg/ha) Cassava lines CI-649 CI-731 Mean CD(5%):0.95 "Data are average values for three levels of N; P was constant at 50 kg P202/ha Source: Mohankumnar, 1996. 50 75 100 Meat 29.6" 29.6 32.3 30.5 28.4 28.1 29.3 28.6 29.0 28.9 30.8 _ Table 13. Effect of fertilizer rates on the yield (t/ha) of three triploid cassava varieties grown at CTCRI, Trivandrum, Kerala, India. Triploid Fertilizer rate (N:P2O5:K2O in kg/ha) cassava varieties 50:25:50 75:50:75 100:50:100 Mean 9/76 22.78 23.25 24.64 23.56 2/14 23.61 25.95 27.18 25.58 237/84 24.50 25.61 25.48 25.20 Mean 23.63 24.94 25.77 - CD(5%):0.953 Source: Mohankumar and Nair, 1993. 298 4.2 Effect of lime application Since some triploids showed foliar symptoms of Ca deficiency, an experiment was conducted on the potential need for lime application. The results showed that application of lime had a significant effect on the yield of all varieties (Table 14). No symptoms of calcium deficiency were noticed, however, in any of the varieties. The round-shaped leaf lobe of 9/76 was a varietal character and was observed both in limed and unlimed plots. Table 14. Effect of lime application on the yield (t/ha) of three triploid cassava varieties as compared to that of a standand diploid variety grown at CTCRI, Trivandrum, Kerala, India. Levels of lime applied (kg/ha) Triploid cassava varieties 0 600 Mean 22.53 24.61 23.57 25.08 26.08 25.58 24.00 26.38 25.19 23.87 25.69 - 9/76 2/14 237/84 Mean CD(5%): 0.78 H- 1687 (control) 19.62 21.87 20.75 Source: Mohankumar and Nair, 1993. 5. Substitution of KCI with sodium chloride (NaCI) 5. / Effect on cassava yield Studies conducted at Kerala Agricultural University have shown that up to 50% of the K requirement of cassava can be substituted by that of Na by the application of sodium chloride (NaCI) without any deleterious effect on crop yield (Table 15). 5.2 Effect on nutrient uptake The uptake of N at different growth stages and with different levels of substitution of K by Na is shown in Table 16. Up to six months after planting (MAP), in all the treatments the uptake of N showed an increasing trend and thereafter a decline. The decline in N-uptake was more conspicuous at 50% substitution of K by Na than in other treatments. At 2 MAP and 6 MAP, this treatment with 50% substitution recorded the highest uptake of N. But at harvest time K applied at full dose resulted in the highest uptake of N. 299 Table 15. Effect of partial substitution of K by Na application on the root yield of cassava grown at Kerala Agric. Univ., Trivandrum, Kerala India. Treatments Root yield (t/ha) 1992/93 1993/94 Mean 21.91 16.70 19.30 19.05 20.30 19.70 26.04 24.50 25.30 18.42 17.90 18.20 11.43 15.30 13.40 13.81 18.30 16.10 16.19 17.80 17.00 100% KC1 75% KCl +25% NaCl 50% KCl +50% NaCl 25% KCl + 75% NaCl 100% NaCl 50% wood ash +50% NaCl 50% KHCOj + 50% NaHCO3 CD(5%) 7.87 6.55 3.44 Source: Sudharmai Devi, 1995 Table 16. Nitrogen uptake at different growth stages as affected by partial substitution of K by Na in cassava at Kerala Agric. Univ., Trivandrum, Kerala India. Nitrogen uptake (kg/ha) Treatments 2 MAP 4 MAP 6 MAP 8 MAP Harvest" 100% KC1 17.08 33.90 70.36 59.73 60.94 75% KCl + 25% NaCl 11.13 48.89 58.81 54.60 45.74 50% KCl + 50% NaCl 18.85 42.30 102.42 49.65 37.01 25% KCl +75% NaCl 14.95 37.39 69.25 50.53 34.93 100% NaCl 10.20 43.32 66.48 53.58 23.97 50% wood ash + 50% NaCl 8.86 57.10 89.36 54.04 30.50 50% KHCO3+50% NaHC03 11.20 33.18 77.25 49.54 36.93 CD(5%) 4.16 NS 22.39 NS 28.40 " At 10 months after planting (MAP). Source: Sudharmai Devi, 1995. 300 The uptake of P also increased up to 6 MAP and thereafter declined (Table 17). Treatment differences were significant only at 2 MAP. During most of the growth stages the highest uptake of P was observed when 50% of the K requirement was substituted by NaCl. Table 17. Phosphorus uptake at different growth stages as affected by partial substitution of K by Na in cassava grown at Kerala Agric. Univ., Trivandrum, Kerala India. Phosphorus uptake (kg/ha) Treatments 2 MAP 4 MAP 6 MAP 8 MAP Harvest 100% KC1 1.92 4.36 8.87 8.65 6.27 75% KCl + 25% NaCl 0.61 5.47 7.29 6.32 6.61 50% KCl + 50% NaCl 1.82 5.19 11.14 9.55 7.02 25% KCl + 75% NaCl 1.11 4.65 8.70 7.77 5.00 100% NaCl 0.65 4.84 6.60 7.53 3.09 50% wood ash +50% NaCl 0.71 5.80 10.40 6.62 4.03 50% KHCO3+50% NaHCOj 0.92 3.75 8.14 7.15 4.36 CD(5%) 0.36 NS NS NS NS Source: Sudharmai Devi, 1995. At all stages of growth the highest uptake of K was observed in the 50% substitution treatment (Table 18). At later stages of growth, Na substitution above 50% or substitution with other alternate sources decreased K uptake. The Na uptake followed the same pattern as that of K (Table 19). The uptake increased up to 6 MAP, and thereafter there was a slight decline. 6. Response of cassava to mycorrhizal inoculation Studies on the effect of inoculation with VA mycorrizal fungi (VAMF) to nursery-grown stakes clearly showed that plants inoculated with Glomus microcarpum var. microcarpum and Glomus fasciculatum enhanced total dry matter and root yields, besides increasing the concentration of micronutrients like Zn and Cu in the leaves (Figure 5 and Tables 20 and 21). About 70% of the farmers who participated in the program on field inoculation of VAMF infected cuttings were convinced of the benefit of mycorrhizal inoculation in improving the growth and yield of cassava. C. Present Recommendations for Cassava Production Table 22 shows a summary of the present varieties and recommended practices for cassava production in Kerala. 301 Table 18. Potassium uptake at different stages of growth as affected by partial substitution of K by Na in cassava grown at Kerala Agric. Univ., Trivandrum, Kerala India. Potassium uptake (kg/ha) Treatments 2 MAP 4 MAP 6 MAP 8 MAP Harvest 100% KC1 12.44 11.03 46.66 35.35 28.45 75% KCl + 25% NaCl 4.40 18.25 38.00 35.77 38.53 50% KCl + 50% NaCl 12.14 16.49 48.36 37.54 38.76 25% KCl + 75% NaCl 6.44 10.05 29.76 24.33 14.13 100% NaCl 4.06 11.88 18.58 21.96 14.57 50% wood ash + 50% NaCl 3.84 23.34 46.56 31.49 15.95 50% KHCO3 +50% NaHCOj 2.79 10.54 30.56 29.82 14.37 CD(5%) 2.43 NS NS 10.19 17.50 Source: Sudharmai Devi, 1995. Table 19. Sodium uptake at different stages of growth as affected by partial substitution of K by Na in cassava grown at Kerala Agric. Univ., Trivandrum, Kerala India. Sodium uptake (kg/ha) Treatments 2 MAP 4 MAP 6 MAP 8 MAP Harvest 100% KC1 0.48 2.17 2.76 2.70 1.45 75% KCl +25% NaCl 0.24 2.47 1.96 1.81 1.42 50% KCl + 50% NaCl 1.00 1.68 2.75 2.38 1.65 25% KCl + 75% NaCl 0.63 1.49 3.87 2.92 1.14 100% NaCl 0.62 2.11 2.56 2.40 1.00 50% wood ash + 50% NaCl 0.45 1.94 3.69 2.03 0.98 50% KHCO3 +50% NaHCOj 0.34 1.75 2.82 1.99 1.04 CD(5%) 0.29 NS NS NS NS Source: Sudharmai Devi, 1995. 302 Table 20. Effect of VA-mycorrhizal inoculation and P application on the P concentration of various cassava tissues. Concentration of P in various cassava tissues (%) Treatments" Root Stem Leaf 0.13 0.10 0.20 0.19 0.15 0.50 0.21 0.25 0.65 0.19 0.30 0.75 MoP0 M,P0 MoP, M,P, Mo=no mycorrhizal inoculation; M,= mycorrhizal inoculation P0=no P added; P,= half the recommended dose of P Source: Potty, 1993. Table 21. Effect of VA-mycorrhizal inoculation and P application on the Cu and Zn concentration of upper leaves of cassava. Treatments' Months after inoculation 10>h A. Zn concentration (mg/kg) MoP0 M,P, MoP, M,P, 145 150 125 100 75 120 140 100 110 70 150 125 100 90 80 150 145 100 90 80 B. Cu concentration (mg/kg) MoP0 M,P, MoP, M,P, 11 12 13 13 14 11 12 13 14 15 9 10 12 13 16 9 11 13 13 17 " Mo = no mycorrhizal inoculation; M, = mycorrhizal inoculation Po = no P added; P, =half the recommended dose of P Source: Potty, 1993. rn o cu o 2 2 c u E o Cu 0) Nl L- 1) o t/> en >• m eq 3 ■ rt E4 55 Z 0- O o 55 o in Q O CO O o o Q- O (Bq/J) p|3!A S 5, c ^ C 0 I 1 I 1 2 1 ! sl e 304 Table 22. Recommended cultural practices for cassava production in Kerala, India in 1996. Varieties: duration Sree Visakham Sree Sahya Sree Prakash Planting time: Land preparation: Planting material: Planting method: Plant spacing: Fertilization: M-4 : excellent eating quality H-97 : high starch, good cooking quality, 10 months H- 165 : early(7 months) variety, desease and pest resistent H-226 : good cooking quality (H-1687) : good cooking quality, high yield (H-2304) : high yield, drought resistent, high starch content (S-856) : early harvestable (6-7months) April-May or Sept-Oct mounds of 25-30cm high, or contour ridges of 25-30cm high select plants free of CMV, cut mature stems, 15-20cm stakes vertical, 5 cm deep; replant missing hills at soon as possible 90x90 cm for branching varieties; 75x75 cm for non-branching types 1 0-15t/ha of FYM, incorporated; 1 00 kg N , 45 P2Oy 100K2O/ha; P all at planting, N and K split at 0 and 45-60 days; only 25 P2O5 if soil is high in P under lowland conditions FYM can be eliminated if vegetable cowpea is grown before cassava and its residues are incorporated into the soil Sprout removal: Weeding and earthing up: Intercrop: Harvest: retain only two shoots, removing excess sprouts at 30-45 days. at 45-60 days and 1-2 months later peanut, french bean, cowpea at 7-10 months, depending on variety (see above) FUTURE DIRECTIONS AND THRUST AREAS 1. Standardization of agro-techniques for cassava grown in rice-based cropping systems It has been observed that in rice fields where no water logging is experienced, cassava would be a better crop compared to rice. In the present condition of high labor costs, farmers are eagerly looking for a less labor-intensive crop like cassava. It is therefore necessary to standardize the agro-techniques for cassava in a rice-based cropping system. 2. Cassava as an intercrop with perennials trees Early bulking, short-duration varieties can be tested under different plant density/geometry, so as to determine the best planting arrangement when intercropped 305 under perennial trees. 3. Cassava leaves as a source of animal feed The high-yielding varieties with high rates of leaf production may be tested under varying levels of inputs to identify varieties suitable for the animal feed industry. Prunning of tops may be intensified to monthly intervals starting from the 7th month to the 15th month to determine whether a regular supply of biomass for cattle feed can be obtained. 4. Rapid multiplication of disease-free planting material It was found that a rapid propagation method of growing mini-setts in the nursery, and transplanting to the main field, was very effective as compared to the cumbersome method of tissue culture. The technique needs to be popularized among farmers also for the elimination of virus diseas 5. Identification of short-duration high yielding varieties of cassava. Short-duration varieties of cassava are in high demand among farmers to fit in a rice-based cropping system, especially in areas where the 2nd crop of paddy is not remunerative due to lack of an assured source of water for irrigation. 6. Studies on the effect of cassava intercropping in reforested areas on soil nutrients, surface runoff and soil loss Wherever reforestation is initiated, cassava may be introduced as a companion crop in the first 2 or 3 years, which will give additional income to the government. No detailed studies have been carried out on the impact of interplanting of cassava with forest species. A detailed study is required to find out the changes that are taking place in the flora and fauna of the ecosystem, the soil nutrients, surface runoff and soil erosion loss as the land is of varying topography. 7. Integrated nutrient management for sustainable crop production by complementing chemical fertilizers with green manures and biofertilizers 8. Identification of varieties for non-traditional areas At present, the area under cassava in non-traditional areas is increasing as the cassava roots constitute the raw material for various industrial products, such as starch, sago etc. So, suitable agro-techniques have to be developed for popularizing cassava cultivation in non-traditional areas. Wherever irrigation facilities are available, irrigation experiments have to be initiated to determine the water requirement. 9. Regional trials Location-specific trials have to be undertaken in different agro-climatic zones, so as to identify varieties better suited to various agro-ecological conditions. REFERENCES Kabeerathumma, S. and C.S. Ravindran. 1996. Long-term effect of manures and fertilizers in acid Ultisol growing cassava. Annual Report, Central Tuber Crops Research Institute, Trivandrum, India. 306 Mohankumar, C.R. 1993. Standardisation of nursery techniques for cassava and sweet potato. Annual Report, Central Tuber Crops Research Institute, Trivandrum, India. Mohankumar, C.R. 1996. Standardisation of agronomic requirements for advanced breeding lines of cassava. Annual Report, Central Tuber Crops Research Institute, Trivandrum, India. Mohankumar, C.R. and P.G. Nair. 1993. Standardisation of agronomic requirements for hybrid selections of cassava. Annual Report, Central Tuber Crops Research Institute, Trivandrum, India. Mohankumar, C.R. and V.P. Potty. 1994. Effect of set treatment on yield of cassava. Annual Report, Central Tuber Crops Research Institute, Trivandrum, India. Mohankumar, C.R. and P.G. Nair. 1996. Dry matter production and nutrient uptake (NPK) in a tuber crop-based cropping system under upland rice situations. In: Tropical Tuber Crops. Problems, Prospects and Future Strategies. Oxford and IBH publishing Co. Pvt. Ltd. New Delhi 110015, India. pp. 100-206. Mohankumar, C.R. , P.G. Nair and P. Saraswathy. 1996. NPK requirement of a short-duration variety of cassava in a rice-based cropping system. In: Tropical Tuber Crops. Problems, Prospects and Future Strategies. Oxford and IBH publishing Co. Pvt. Ltd. New Delhi 1 10015, India. pp. 233-237. Nair, P.G., B. Mohankumar, M . Prabhakar and S.Kabeerathumma. 1988. Response of cassava to graded doses of phosphorus in acid laterite soils of high and low P status. J. Root Crops 14(2): 1-9. Ravindran, C.S. 1996. Nutrient-Moisture-Light Interactions in a Coconut-based Homestead Cropping System. Ph.D thesis, University of Kerala, Trivandrum, Kerala, India. p. Potty, V.P. 1993. VA mycorrhizal association in tuber crops and their role in crop nutrition. Annual Report, Central Tuber Crops Research Institute, Trivandrum, India. pp. Sheeja, S. 1994. Micronutrient Status and their Distribution in Major Tuber Crop Growing Soils of India in Relation to Soil Properties. Ph.D thesis, Kerala Agricultural University, Trivandrum, Kerala, India. p. Sudharmai Devi, C.R. 1995. Partial Substitution of Muriate of Potash by Common Salt for Cassava (Manihot esculenta Crantz) in on Oxisol of Kerala. Ph.D thesis, Kerala Agricultural University, Trivandrum, Kerala, India. p. 307 RECENT PROGRESS IN CASSAVA AGRONOMY RESEARCH IN INDONESIA J. Wargiono1, Kushartoyo2, Suyamto H.2, and B. Guritnd' ABSTRACT Annual growth of cassava production in the main cassava production areas during the last five years seems to have kept pace with the increase in population, while the harvested area of cassava increased only 0.6% annually. Of the total cassava produced, about 54% is used for human food, 28% for industrial purposes, 2% for feed and 16% for export. Therefore, the development of this crop should be in line with the development of other food crops and agro- industry, as well as with that of dried casssava for export. Of the total cassava area planted in the country, more than 60% is harvested during June to October. Delaying the planting of both cassava and the intercrops tends to give a similar gross return compared to that of planting in the early rainy season. Therefore, most farmers in Sumatra are now planting cassava from the early rainy season to the early dry season. Thus, the cassava harvest is more spread out and cassava becomes a more dependable crop for both the starch and pellet industries, as well as a food security crop in rural areas. Most farmers do not fertilize cassava due to lack of capital; therefore, a closer cooperation between the farmers and starch and pellet factories should be considered to solve this problem. Cassava agronomy research is mainly conducted by the Bogor and Malang Research Institutes for Food Crops, by Brawijaya University in Malang and by the Umas Jaya cassava plantation in Lampung. The yield of both cassava and intercrops grown on an ultisol in Lampung decreased more than 60% during the third consecutive cropping without any fertilizer application. Interplanted crops like maize, rice and peanut grow faster than cassava and the amounts of nutrients absorbed by these crops during the first two months are higher than for cassava under these intercropping systems. But, there is an indication that cassava absorbs residual fertilizer applied to these interplanted crops. Crop productivity could be maintained by the application of adequate amounts of fertilizer and by the incorporation of cassava stems and leaves into the soil. A balanced fertilizer rate of 75-100 kg N, 25-100 kg P2O5 and 60-100 kg K2O/ha for monoculture cassava and 100-150 kg N, 50-100 kg P2O5 and 100-150 kg K2O/ha increased net income by 70 and 370% and reduced erosion by 11 and 35%, respectively. Soil loss due to erosion in monoculture cassava during 10 months in Malang, E. Java, was more than 50 t/ha of dry soil. Fertilizer application, ridding, elephant grass strips and intercropping decreased the amount of eroded soil significantly, compared to monoculture cassava without fertilization. Cassava clones with wide leaves and no branching are considered suitable for intercropping systems. INTRODUCTION Indonesia has the fifth highest cassava production in the world. Annual production is quite variable, fluctuating from 15.8 million tons in 1990, 17.3 million tons in 1993 and 15.7 million tons in 1994 (Table 1). This fluctiation is caused by fluctiations in prices for cassava products both in Indonesia and in other cassava 1 Research Institute for Grain Legumes and Tuber Crops (stationed in Bogor, W. Java). 2 Research Institute for Grain Legumes and Tuber Crops, Malang, E. Java. Brawijaya University, Malang, E-Java. 308 producing countries. Higher prices of pellets and chips, both inside and outside the European Union countries, as well as in the Indonesian market, stimulate cassava production in the following season. Then, the increase in cassava production always brings about a decrease in price. This low price discourages farmers from growing cassava in the next season. These price fluctuations should be minimized through the development of more rural agro-industry and food processing as well as through greater diversification of end markets. Table 1. Cassava area, yield, production and export in Indonesia, as well as prices of pellets and chips in the EEC and outside the EEC Market from 1990 to 1994. Year Harvested area(.OOOha) Yield (t/ha) Production COOOt) Total exported Prices($/ton) COOOt) In EEC Outside EEC Chips Pellets Chips Pellets 1990 1,311 12.07 15,830 1,009 118 124 61 98 1991 1,319 12.10 15,955 956 132 138 58 144 1992 1,351 12.23 16,516 932 127 132 72 143 1993 1,402 12.33 17,285 1,150 109 115 52 51 1994 1,357 11.59 15,729 776 93 100 72 50 Source: BINUS, 1995 Of the total cassava produced in Indonesia about 54% is used for human food (0.83 kg fresh root/day/person), 28% for industrial purposes, 2% for feed and 16% for export (BINUS, 1995). Crop and land productivity are the determining factors in the farmers' income. In Indonesia cassava is mostly grown on marginal upland soils and the average cassava yield of about 12.2 t/ha is much below the potential yield obtained in experiments, which ranges from 20-40 t/ha. Therefore, there is still a potential to improve production technologies in order to increase yields and farmers' income, while at the same time conserving the soil and water resources. The main cassava producing areas in Indonesia are located in relatively dry regions and on marginal soils of Java and Nussa Tenggara, where the crop is grown mainly by small-scale farmers which have limitations of land, capital and labor. Consequently, the cassava planting time tends to coincide with the beginning of the rainy season and cassava is usually intercropped with other food crops. In transmigration areas of Sumatra and Kalimantan cassava yields and harvested area vary due to limited availability of labor and capital to grow cassava as well as other food crops. For cassava to be a dependable crop, both as a staple food and as a cash crop, farmers will grow cassava depending on the availability of capital and labor in the family, usually from the early to the end of the rainy season. Cassava plants absorb large amounts of nutrients from the soil. Therefore, 309 fertilization is often the only way to maintain soil fertility and land productivity. As the goverment plans to stop subsidizing fertilizers, it means that the price of fertilizers will increase and farmers will reduce fertilizer applications to cassava. In Indonesia most cassava plant parts are utilized and removed from the field. Thus, organic matter additions are also needed to maintain soil fertility in the long-term. Intercropping cassava with upland food crops not only improves the land use efficiency, farmers' income and the distribution of this income through the year, but it also adds more organic matter to the soil than planting cassava in monoculture. Cassava growth during the first three months is relatively slow. Erosion during this period is relatively high as the soil surface is not adequately covered by the cassava plant canopy. Cropping systems and cultural practices that cover the soil more quickly must be identified to reduce erosion. During the past ten years, the Bogor and Malang Research Institute for Food Crops, the Agricultural Faculty of Brawijaya University in Malang, and the Umas Jaya cassava plantation in Lampung have been conducting cassava agronomy research with emphasis on solving the above-mentioned problems. Production Areas and their Agro-ecological Characters Recent trends in cassava area, production, yield and cassava product prices, as well as the total amount of cassava exported are presented in Table 1 . Annual growth rates of cassava production and harvested area from 1990 to 1993 were 3.0 and 2.2%, respectively. The decrease in harvested area (7.5%) and production (9.0%) in 1994 was mainly due to drought. Cassava yields were also affected by drought conditions. Plant productivity is determined by NAR (net assimilation rate), where NAR depends on soil conditions and water availability (Hozyo et al., 1984). Cassava root yields decreased significantly when 2-4 months before the harvest the top soil became hard due to drought (Wargiono, 1993). During the 1993/94 cropping season, rainfall during the 3rd to 5th month after planting was substantially lower than during the same period in the 1992/93 season. This resulted in considerably lower yields in 1994 in all production zones except in Kalimantan (Figure 1). Limited availability of capital is one of the constraints for the farmer to fertilize his cassava crop and to maintain land productivity. Since cassava absorbs large amounts of nutrients from the soil, the application of farm yard manure (FYM) or fertilizers have to be considered. Limited labor in the family is another production constraint and farmers may need to rent a tractor for land preparation and use herbicides to control weeds. More than 60% of cassava production is located in Java and the remaining 40% mainly in Sumatra, Nussa Tenggara and in the Sulawesi islands. Most cassava in Java is planted in the early rainy season due to limited arable land, whereas in Sumatra land holdings are larger, generally more than one ha, so cassava is planted from the beginning 310 to the end of the rainy season. The wider spread of cassava planting time is one alternative to reduce temporary over-production (Wargiono, 1993). Since the delay in cassava planting time is possible in Sumatre due to a longer wet season, cassava production in Lampung is better distributed throughout the year compared to that in Java (Wargiono et al., 1995). Cassava in Indonesia is grown on a variety of soils, such as Alfisols, Ultisols, Entisols, Inceptisols, Vertisols and various soil complexes (Wargiono, 1988). These soils are usually infertile and are often susceptible to erosion. Therefore, better soil conservation technologies, which are able to maintain or improve land productivity and are acceptable to farmers, are urgently needed. RECENT CASSAVA AGRONOMY RESEARCH Cassava agronomy research in Indonesia has as its main objective to increase cassava yields and farmer income while maintaining or improving soil fertility and land productivity. The research, therefore, includes the identification of cassava production constraints, the evaluation of new cassava clones under intercropping systems, the control of erosion, and the improvement of cultural practices and fertilizer use efficiency. 1. Case Study to Determine Cassava Production Constraints A study was conducted in Nov 1995 in transmigration areas of northern Lampung. The sub-districts studied were Manggala, Tulangbawang Udik, Tulangbawang Tengah and Mesuji. Cassava in this area is a staple food, consumed as steamed cassava/rice pellets (tiwul) during 3 to 9 months of the year. Tiwul is made from cassava flour mixed with rice in a ratio of 2-4: 1. Based on an average cassava flour consumption of 0.33 kg/day/person, the cassava consumed as tiwul is about 25 kg fresh roots/month/person. Table 2 shows supply and demand of cassava in the sub-districts studied, where utilization for human food and industrial purposes were only 9.3% and 32.3%, respectively. Table 3 shows cassava yield and production during 1990-1994 in the sub- districts studied. In some cases production increased in spite of a decrease in yield; this is due to a marked extension of planted area. This production was affected by demand by the industries and by availability of infrastructure. Good infrastructure reduces transportation costs and facilitates the sale of cassava. Therefore, the farmgate cassava price of these areas is higher compared to those in areas with poor infrastructure. The cropping system used in this area is affected by the availability of labor and capital. Average available labor is two persons per family. The capability of these two laborers to prepare land and grow food crops was 0.25-0.5 ha/month. Therefore, farmers have to delay land preparation and the planting of food crops when they want to grow crops on more than 0.5 ha. 311 15 r 2 10 y .>< c o 12.7 12.8 11.6 11.4 D1993 Q1994 ^MIW 10.2 9.5 11-2 11 10.8 10.1 Java Sumatra Southeast Sulawesi Kalimantan Indonesia Figure 1. Cassava yields by region in 1993 and 1994. Source: CBS, 1995. The Ultisol soils of the areas studied are mostly infertile; thus fertilization and good crop management (controlling weeds and pests/deseases) are essential for obtaining good yields. Since cassava grows better than other food crops on infertile land and with low inputs, more and more cassava is grown by those farmers that have limited capital. Table 3 shows that cassava yields decreased dramatically over time in all four sub- districts. In 1990 the land was newly opened from the bush and high yields could be obtained without any fertilization. However, cassava extracts large amounts of nutrients from the soil and soil fertility decreased, resulting in a decrease in cassava yield as no fertilizers were applied by the farmers in each growing season. With adequate fertilization high yields of cassava could probably have been maintained (see below). But most cassava farmers have no money to buy fertilizers; thus, the subsidizing of fertilizers may still be needed. As fertilizer subsidies from the government will be stopped, fertilizer subsidies from private companies through a cooperative system with farmers' groups may have to be considered. A Cooperative Model (Figure 2) with clearly defined relations between farmers' groups and a tapioca factory, might be recommended. 2. Cropping Systems Cassava is planted in monoculture only around urban areas, in starch factory plantations and on non-productive land, which cannot be planted to other food crops. Most farmers, however, plant cassava intercropped with other food crops, since this will enable them to increase their land use efficiency and income, improve the soil's physical 312 Table 2. Supply and demand of cassava in various subdistricts studied in northern Lampung in 1995. Sub-district Demand(%) Production ('OOOt) Surplus Food Industrial purpose ('OOOt) Manggala 483.6 5.9 32.7 297.2 T.B. Tengah 137.2 16.4 49.6 46.7 T.B. Udik 111.4 8.5 0 102.0 Mesuji 111.9 16.0 41.8 47.3 Source: Ditpertan, North Lampung 1995 Table 3. Cassava yield and production in four subdistricts of northern Lampung from 1990 to 1994. Year Manggala T.B.Tengah T.B.Udik Mesuji Yield Production Yield Production Yield Production Yield Production (t/ha) (.OOOt) (t/ha) (.OOOt) (t/ha) (.OOOt) (t/ha) (•OOOt) 1990 27.0 326.0 26.3 138.8 23.9 72.9 22.4 222.9 1991 24.7 436.1 23.1 128.2 19.7 162.9 20.2 133.3 1992 21.3 269.7 20.0 80.2 19.5 153.7 20.0 83.1 1993 13.9 503.7 14.2 271.9 13.9 84.5 13.8 89.8 1994 11.9 882.3 12.4 66.8 11.1 83.2 13.4 30.6 Source: Ditpertan, North Lampung, 1995. condition and reduce erosion, as compared with planting in monoculture (Guritno, 1989; Wargiono et al., 1992). Table 4 and 5 show that growing food crops, either in monoculture or in an intercropping system, without any fertilization is generally not recommended because of the low yields obtained. Intercropping cassava + maize+ upland rice-peanut may be recommended because higher total crop values can be obtained compared with monoculture upland rice, maize and peanut, followed by cassava. In the latter cropping 313 systems, the low yield of cassava (Table 3) is due to drought during the last four months and the harvesting of cassava at only 6 months. The higher crop value of the peanut- cassava system compared to that of upland rice-cassava or maize-cassava is due to the higher crop value of dry pods of peanut than dry grain of either maize or rice. The cropping system of peanut-cassava produced a high crop value as well as an effective control of erosion (Figures 3 and 4). More research of this system is needed before it can be recommended. Tapioca Factory Agric Ext. Field Staff Processing Staff Field Assistance Field Specialist Administration Staff Production input (Fertilizer,etc) Farmers Group Farmers Figure 2. Diagram of Cooperative Model offarmers' group with tapioca factory at Lampung, Indonesia. Table 5 shows that the intercropping of cassava + maize+ upland rice-soybean increased net income 17% compared to that of monoculture cassava, while the LER of this intercropping system was 1.44. Monoculture plantings of upland rice-soybean, maize-maize, peanut-peanut or soybean-soybean usually yielded a higher net income than the intercropping system of cassava + maize+ upland rice-soybean. For farmers having 314 land that is very susceptible to erosion, intercropping systems are still recommended since these systems tend to control erosion more effectively. Table 6 shows the performance of various cassava clones and the corresponding interplanted crops. Some characteristics of Adira 1 are the lack of branching short maturity and narrow canopy diameter, which favor high yield of interplanted crops due to the low competition from cassava. The average land holding of most farmers in Java is relatively small; thus, high yields of both cassava and interplanted crops are expected to enhance food diversity and increase farmers' income. UB1-2 is another cassava clone well suited for intercropping systems. 3. Erosion Control by Cultural Practices In Indonesia cassava production areas are mainly located in mountainous areas of Java and in regions of undulating topography outside of Java. In Java land holdings are relatively small and erosion could be reduced substantially by terracing; however, this is still very expensive. Land holdings outside of Java tend to be larger, so reducing erosion by terracing is too expensive and this method is seldom practiced by the farmers. Soil erosion, which is the main cause of soil degradation (Suwardjo and Sinukaban, 1986), can also be reduced effectively by cultural practices, such as contour ridging, contour barriers/hedgerows, mulching and minimum tillage (Evangelio et al., 1993; Sittibusayaera/., 1993; and Wargiono, 1993). Table 7 shows that soil erosion varied markedly from year to year but was consistently higher in unfertilized than in fertilized plots. Erosion losses could also be reduced by contour ridging, by intercropping with peanut or the planting of a covercrop of Mimosa envisa; however, the latter markedly reduced cassava yields, which is probably unacceptible to farmers. Most effective in reducing erosion, while also increasing cassava (and sometimes maize) yields, was the establishment of contour hedgerows of Gliricidia sepium or Flemingia congesta. When the pruned leaves and branches of these leguminous trees were mulched between cassava plants, they obviously supplied additional N to the crop, resulting in greener and more vigorous cassava plants. The beneficial effect of these hedgerows increased over time. Contour barriers of elephant grass were also quite effective in reducing erosion and increasing yields as long as the elephant grass was adequately fertilized. This system also provides a regular supply of feed for the farmer's animals. A similar trial conducted on a farmer's field with 10% slope in Tarokan village of Kediri district (Table 8) also shows that intercropping cassava with peanut and planting contour barriers of elephant grass was the most effective way to increase cassava yields and reduce erosion. This system is easier and cheaper to implement than the terracing system. Fertilizer application to cassava grown in monoculture or intercropped in an Ultisol in Lampung not only increased crop value by 56% and 23%, respectively, but also reduced erosion (Wargiono et al., 1995). a o S i ■a -" 5 -a .3 CO E I .3 se c © -o c e 3 a. s 3£ eg os U f- 4j v. 1 c a 09 « CO a> eg CO H cv .s o CO o 5, 5 2 .£. n. "o u in 1 c o 1 SC a .5 oS V — * j3 1/5 *-* Of ■a .5 c a. CO a. (A e c o 3 V3 « 3 U .2 « eg c>• T3 v- co CO (/! te £ u ec i. Oy 3 X H fr-2 a: a. oc CO U .5 S U cd CO CO s oo r- oo rl O N * o STF oo c> c> r- o « o - CIMinI^—• T* «*Mvin't-.«n *Oc>-^rl — I^ N ^ N VO VO M * >i no ci oo ■* ■* r- 2 s d d o r- 5 ? d d in w n >o - CO BO co u - N ci -t «i * r- ri B." 5 + at + u o 8"a X 8 1- D Cl a. a. S S + + 06 q£ + + U U e 5 e *j ^ -j J | J I sa | j= s * e i i 8 § 8 sss u 9 u 06 S oL S oo o 5 SS ° Q. *- i :l ■ t: u 3 o H Z SO en & £ » 1? Q ~ z E o u .s s _^ a. cox v X U-, P O | V c I OO ooh n t t; oo -H m in m in M 8 OM—^IOiNfiO * * m O m — h q ts -h" wi' -7 ci cN -H* On ©\ 00 O O 00 OOO00Ot-^TtTtg;• > J* 3 >■ -a -c E 3 s 8 M .5 u 9 ■a "S '?. u .a a E TS a cs g>• CO S/i(/; CC 0 c o ■o c S •S s 0 c *■§ SJ - 3 M■a a 1 1.■a * SI tT c .2 g Vi 'S +£ u e cc 6 a .5 i, a. 1 w SO o It E g•■= J5 laa"3 B .2 9 « > a o °Z $ Is. !--■ 2 N H ■3 '?, u .a s M u 'Si a > u ON Os n gs O os n os gs ri Os n os 7Z3 O vz ON^~ n Os rl Os ri Os 00 o 00 00 OOriri t- 00 00 so Os —i KJ n h h NO en so O —i —i ri -- —■ 00 l-l vS «ri h so 00 CNj m o " O u o C s». « e £ h h 4> Ulili i ,§ * « o o -5 -5 ssss + + + + u u u u - ri ci * so o os rl 3$ so so n N Os « oo rs| r~ oo 00 l> Tl- Os O r» so —i E C §5 « .S li U -J an t/j -o -o C C E 2* u u J .a is + + u q iri so SO so 5 Os rl ri oo ri ri oo a. o > o s 5 5 'C if u | '€ Os NO ri oo — NO a. c o ■a ■c .-3 fi I I a. p o S 5 + + u u r~ oo 322 Table 8. Effect of various cropping systems on dry soil loss due to erosion and on the fresh root yield of cassava (cv Adira 4) planted on 10% slope in Tarokan village of Kediri district in E. Java, Indonesia in 1993/94 Dry soil loss Fresh root yield Treatments (t/ha) (t/ha) 1. Cassava" monoculture 31.90 16.75 be 2.C + maize 27.54 14.38 a 3.C+peanut 16.60 20.13 d 4.C + 2 rows of elephant grass 23.81 18.25 c 5 . C + Crotalariajuncea intercrop 19.18 15.13 ab 6.C + maize+ 1 row of elephant grass 21.24 15.38 ab 7.C + peanut + 1 row of elephant grass 13.99 23.38 e 8. No crop (bare fallow) 38.09 - "cassava planted at 1.0x1.0 m, maize 1.0x0.5 m, peanut 0.25x0.25 m, Crotalaria juncea 1.0x1.25 m. Source: Soemarjo Poespodarsono and Nur Basuki, 1994. In this case, cassava was intercropped with maize and all plant tops were removed after harvest. There was a highly significant response to NPK application already in the first year. The response to N increased over time, while the response to P and K varied over the years, but was usually not significant. Although this soil has an intermediate level of OM (1.2-1.8%), the main cassava response was to N application, which was probably accentuated by the strong competition from the intercropped maize. Without N, cassava yields dropped to less than 4 t/ha in the 7th crop cycle. While the soil P level was slightly below and the exchangeable K was far above the critical level, there was no response to P but a significant response to K during the 7th consecutive cycle (Figure 7). The annual application of intermediate and high levels of NPK increased yields from 2 to 23 and 30 t/ha, respectively. Figure 8 shows the effect of NPK fertilizers on the yield of intercropped cassava, maize, upland rice and mungbean during the 4th cropping cycle in an Alfisol in Yogyakarta. A fertilizer application of 90 kg N + 50 P203 + 90 Kj0/ha tended to maintain low but stable yields of cassava and high yields of interplanted crops during four years of cropping. The growth of mungbean as a second intercrop (planted after the harvest of upland rice) was affected by light competition from cassava. Fertilization stimulated the cassava growth, resulting in an increase in cassava yield but a decrease in mungbean yield. Fertilization of 90 kg N + 50 P205 +90 K20/ha seemed to be a good fertilization rate as it maintained both high yields of crops and net income. Figure 9 shows the effect of NPK fertilization on cassava intercropped with 323 Umas Jaya a 30 .c 5 o O 20 10 voN« 100 * 80- .2 60> > 40 1 20 -ON0 ■DP. -A „ Crop cycle Figure 5. Effect ofannual application ofN, P and K on cassava root yield, relative yield (yield without the nutrient over the highest yield with the nutrient) and exchangeable K and available P (Bray 2) content of the soil during ten years of continuous cropping in Umas Jaya, Lampung, Indonesia. 324 100 I 80 ■a » 60 '>. ? 40 S 20 IT /\ /N, ^^,p»£ n ^K0 1 ■ ■ ■ L. ... J. - J . 1 en o 2 0.8 E 0.4 Critical K- level E a. Figure 6. Effect ofannual application ofN, P and K on cassava root yield, relative yield (yield without the nutrient over the highest yield with the nutrient) and exchangeable K and available P (Bray 2) content ofthe soil during eight years of continuous cropping in Jatikerto Experiment Station in Malang, E. Java, Indonesia. in o Z Q. O O O ID o CM z *: o o o o _1_ o CO o CM o o (M 6 o CD 6 ^o o ts CM O CM & £o 1 t\T— o 6 CM %> in Q. -s 6 o z ^ O) "C o ■M a■^s o k6 1 a, o ^o CM CD s £ Ka--. o Si o o CM -^ 1— 05 3 .* o o c IT) o I o 1o CO 1.c a ID O 1 So CM IT) Q. ■s & in ^ 2 "*"' CM ■3 .2 O g 1 6 8 O I Io CM CD ^ KJ o z o 3 C O IT) O 1 (eq/j) p|8;A jooj babssbq («M/ a \ 1 1 «> a. E \1 1 > o o ^il o -* o CN o Z Q. O O o in O o o a. IT) o> If) M CM o in CM (eq/j) p|3jA loot eaesseq (Bq/») ueaqBuniu 'azieiu 'aou p|8|A ihejo 1! C3 I5sfi> ■5 ■55 K b: 327 upland rice and maize in an Ultisol of Lampung during the 4th cropping cycle. Cassava in this area is an important crop, so cassava is grown in the cropping system with a higher plant population than is used in Java. As such, the yield of cassava is higher but the yield of interplanted crops is lower compared to those in Java. Most Ultisols have relatively low nutrient contents; consequently, crop productivity is low. To obtain a high net income, the rate of fertilizers applied must be relatively high. In Yogyakarta the fertilizer application of 90 kg N + 50 P205 + 90 Kj0/ha in the intercropping system of cassava + maize + upland rice-mungbean yielded a net income of Rp 1 ,900,000/ha, whereas in Lampung the higher fertilizer rate of 180 kg N + 100 P205 + 190 K2O/ha yielded a net income of only Rp 900,000/ha. Based on on-farm research in cassava production areas of southern Malang in East Java a balanced rate of fertilization for both cassava and intercropped maize (Table 9) increased net income by about 50%. Therefore, a rate of 170-67-60 of N-P2O5- K20/ha (corresponding to C1-M2 in Table 9) can be recommended when farmers have access to sufficient capital. FUTURE PLANS Some soil conservation practices, such as contour ridging, fertilizer application, intercropping with peanut and alley cropping with Gliricidia or Flemingia offer good possibilities to improve farmers' income while reducing soil losses due to erosion. Which practice is the most beneficial to the farmers still needs to be investigated. There is also a need to conduct more research about the efficient management of crop residues, improved fertilizer efficiency and cassava clone performance under specific conditions, since optimum agronomic practices are different in each agro-ecological zone. CONCLUSIONS Most cassava is grown on poor soils that are susceptible to erosion, and where farmers have limited availability of capital and labor. Intercropping is recommended due to its capability to reduce erosion, increase LER and net income, and improve food crop's diversification and support the food security program. The limited availability of labor and capital of most farmers tend to encourage the growing of low-input crops, such as cassava, and to delay cassava planting. Better cooperation with private companies to supply fertilizers and herbicides for their food crop cultivation is a possible solution. A balanced rate of fertilizer application, such as 75-100 kg N + 25-50 P2O5 + 60-100 K20/ha for monoculture cassava, and 100-150 kg N + 50-100 P205 + 100-150 K2O/ha for the intercropping system of cassava+ maize+ upland rice-peanut or soybean is able to increase net income by 70 and 370% and to reduce soil erosion by 11 and 35%, respectively, as compared to that of unfertilized crops. Moreover, adequate fertilizer application can maintain soil productivity. 328 Table 9. Effect of fertilizer application to cassava and/or maize in cassava/maize intercropping systems on yield, crop value and net income. Data are average of three farms in Sempol village, southern Malang, Indonesia in 1991/92. Treatments" Yield (t/ha) Fertilizer Totalcrop Net Cassava-Maize cost2' < value2)mOORn/ViaV income cassava maize > C2-M2 30.73 a 1.20 a 175.50 1716 1541 CI - M2 29.66 a 1.19a 136.25 1661 1525 CO- M2 20.49 c 1.19a 97.00 1203 1106 C2-M1 28.23 b 1.03 b 127.00 1566 1439 CI -Ml 22.87 d 0.99 bc 87.75 1292 1204 CO -Ml 17.56 f 0.94 c 48.50 1019 970 C2-M0 26.01 c 0.61 d 78.50 1392 1313 CI -MO 21.89 de 0.57 d 39.25 1180 1141 CO- MO 18.06 f 0.54 d 0 984 984 " C2 = full recommended dose for cassava = 150 kg urea, 100 kg TSP and 100 kg KCl/ha CI and Co are half recommended dose and no fertilizer, respectively, for cassava M2 = full recommended dose for maize = 300 kg urea, 100 kg TSP and 50 kg KCl/ha Ml and Mo are half recommended dose and no fertilizer, respectively, for maize :' Prices: cassava fresh roots Rp 50 /kg maize dry grain 150 /kg urea 210 /kg TSP 210 /kg KC1 260 /kg <0M/dy 000.) auioouj jau jo »n|8A dojo |«;oi I § CO -T" o o -r- CN CO Z tt o o 0) 0) o W)lo eg c> o o a * o o in tn o CO i— 6 o oo i; - O o J- °? . in O z CO .c O o o •- CO ° € S O o 00 O * o o T- CO .c Ifi O o in a. O) in .* CM - o - o o - 00 CO .c oH 05 - o O) CO .C 13 oo «N in °- O) in * CM O CO in Z - o o cc o o CM (bl|/j) p|9iA jooj BAesseo o o o (ei|/i) azieua pue aoii p|3;A uiejg 5 « a. a >T-S •= 5 330 REFERENCES Central Bureau of Statistics (CBS). 1995. Statistik Indonesia. Produksi dan luas panen ubikayu. (Production and harvested area of cassava). Jakarta, Indonesia. Direktorat Bina Usaha Tani dan Pengolahan Hasil (BINUS). 1995. Laporan Temu Asosiasi Direktorat Bina Usaha Tani dan Pengolahan hasil - ASPEMTI - ATTI (Proc. Meeting of Association of ASPEMTI and ATTI). Jakarta, Indonesia. Evangelio, F.A., F.G, Villamayor Jr., A.G. Dingal, J.C. Ladera, A.C. Medellin, J. Miranda and G.E. Sajise Jr. 1995. Recent progress in cassava agronomy research in the Philippines. In: R.H. Howeler (Ed). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp.290-305. Guritno, B. 1989. Tumpangsari pada tanaman ubikayu sebagai usaha peningkatan penyediaan pangan dan pendapatan petani (Cassava intercropping, a way to increase farmer income and food need). PT. Jawa Post, Surabaya. 44p. Howeler, R.H. 1981. Mineral Nutrition and Fertilization of Cassava. Centro Internacional de Agricultura Tropical. Cali, Colombia. 52p. Hozyo, Y., M. Megawati and J. Wargiono. 1984. Plant production and the potential productivity of cassava. Contribution Central Res. Inst. Food Crops. Bogor. 73:1-20. Moreno, R.A. and R.D. Hart. 1978. Intercropping with cassava in Central America. In: Proc. Intern. Workshop on Intercropping with Cassava, held in Trivandrum, Kerala, India. pp. 17-24. Nayar T.V.R., S. Kabeerathumma, V.P. Potty and C.R. Mohankumar. 1995. Recent progress in agronomy research in India. In: R.H. Howeler (Ed). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp.61-83. Sittibusaya C, C. Thiraporn, A. Tongglum, U. Cenpukdee, V. Vichukit, S. Jantawat and R.H. Howeler. 1995. Recent progress in cassava agronomy research in Thailand. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp. 1 10-123. Suwardjo, H. and N. Sinukaban. 1986. Masalah erosi dan kesuburan tanah di lahan kering Podsolik merah kuning di Indonesia. Proc. Lokakarya Usahatani Konservasi. (Erosion and soil fertility problems of Red Yellow Podzolic soils in Indonesia) Litbangtan-Ditjenbun-PT. Mosanto. Palembang. pp. 1-20. Wargiono, J. 1988. Agronomic practices in major cassava growing areas of Indonesia. In: R.H. Howeler and K. Kawano (Eds.). Cassava Breeding and Agronomy Research in Asia. Proc. 2nd Regional Workshop, held in Rayong, Thailand. Oct 26-28, 1987. pp. 185-205. Wargiono, J. 1993. Penampilan clon ubikayu pada pola tanam berbeda (Performance of cassava clones in different intercropping systems). Sem. BORIF, Bogor. Wargiono, J., B. Guritno and K. Hendroatmodjo. 1992. Recent progress in cassava agronomy research in Indonesia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Utilization Research in Asia. Proc. 3rd Regional Workshop, held in Malang, Indonesia. Oct 22-27, 1990. pp. 185-198. Wargiono, J., B. Guritno, Y. Sugito and Y. Widodo. 1995. Recent progress in cassava agronomy research in Indonesia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala. India. Nov 2-6, 1993. pp. 147-174. Widjaya, I.P.G., I.P.G. Wigena, W. Hartatik and A. Sofyan. 1990. Efisiensi penggunaan pupuk di lahan kering (Efficiency of fertilizers on upland soils). Lokakarya Nasional Efisiensi Penggunaan Pupuk V. Bogor, Indonesia. 28p. 331 RECENT PROGRESS IN CASSAVA AGRONOMY RESEARCH IN THE PHILIPPINES Fernando A. Evangeliol and Julieta C. Ladera2 ABSTRACT During the last three years cassava production in the Philippines did not progress markedly, neither in area nor in volume. In terms of area, there was only an increase of 14,424 ha or 6.4%, while in terms of volume the increase was 112,177 metric tons or 5.7%. One of the reasons for this was the unstable price for chips, especially in 1994/95. Up to the present, cassava cultivation is still concentrated in Mindanao where there are greater opportunities for marketing, particularly trading of dry chips by the San Miguel Corporation and animal feed millers, as well as fresh roots for production of starch. In terms of research, not much data has been generated, since only very few experiments were conducted. Some experiments were terminated, like the one on cropping systems trials in Negros Occidental. In Leyte the fifth cropping cycle of the long-term fertility trial under coconut showed no significant responses to fertilizers, but there were some responses in the sixth year of consecutive cropping. In the erosion control trial, highest soil losses were observed in plots with application of inorganic fertilizers (14-14-14), while the lowest soil losses were observed in plots with vetiver grass barriers or with grass mulch. Root yields were highest with the application of grass mulch or when Crotalaria juncea was intercropped and mulched at 2 months; yields were lowest in plots with lemon grass barriers. The variety x fertilizer trial in Bontoc, southern Leyte, showed a clear varietal response to fertilizers. The variety with a heavier canopy responded more markedly to fertilizer applications. In an intercropping trial in Bohol, even after three cropping cycles, cassava yields were not significantly affected by interplanting of either soybean, mungbean, cowpea, peanut or pole sitao (yard-long bean). However, row spacing significantly affected the yields of cassava and intercrops. In another trial, marked increases in cassava yields were obtained when the plant densify was increased to 15,000-25,000 plants/ha. On the other hand, no significant differences were observed when the age of pruning cassava was varied. INTRODUCTION The cassava industry in the Philippines is now gaining momentum with the existence of various market opportunities. Cassava is grown not only for human food, but also for starch, feed and industrial uses such as alcohol. Aside from the San Miguel Corporation, which currently exports cassava chips to Europe and uses cassava meal as an ingredient in animal feed, other firms, like La Tondena, are also working with cassava as a raw material for production of alcohol. This is due to scarcity of molasses resulting from low sugarcane production. Moreover, various food products from cassava are now developed, further increasing the demand for the crop. 1 Director, Philippine Root Crops Research and Training Center (PRCRTC), Baybay, Leyte, Philippines. 2 Central Visayas Integrated Agric. Research Center (CVIARC), Gabi, Ubay, Bohol, Philippines. 332 Cassava Area and Production During the last three years (1993-1996) cassava production in the Philippines did not increase very much, neither in area nor in volume (Table 1). In terms of area, there was only an increase of 14,430 ha from 21 1,420 to 225,850 ha, which is an increase of 6.8%. In terms of volume, the increase was 112,177 metric tons or 5.7%. One of the reasons for the slow growth of cassava production was the unstable price for chips, especially in the 1994/95 season. Up to the present, cassava cultivation is still concentrated in Mindanao because of the greater market opportunities, particularly the presence of chips traders like the San Miguel Corporation, feed millers and starch processors. Table 1. Cassava area, production and yield by region in the Philippines during the last five years (1991-1995). Region 1991 1992 1993 1994 1995 Cassava area (ha) Philippines 211,036 204,313 211,420 213,090 225,852 Ilocos 1,807 1,805 2,207 2,210 2,218 Cagayan Valley 354 411 384 350 398 C. Luzon 1,240 1,148 1,160 1,230 1207 S. Tagalog 10,264 10,048 10,418 10,460 10,617 Bicol 30,960 31,188 31,048 28,990 28,417 W. Visayas 9,568 9,548 8,993 8,800 8,661 C. Visayas 20,595 19,387 20,751 20,940 20,136 E. Visayas 25,111 24,009 24,033 24,280 26,679 W. Mindanao 22,643 22,512 22,588 22,820 22,650 N. Mindanao 10,949 12,588 15,349 15,460 13,780 S. Mindanao 3,017 3,134 3,278 3,300 2,610 C. Mindanao 1,636 1,589 1,655 1,670 1,570 CAR" 301 294 312 320 319 ARMM2) 63,913 58,430 60,742 63,740 79,140 CARAGA3' 8,678 8,222 8,502 8,520 7,450 333 Table 1. (continued) Region 1991 1992 1993 1994 1995 Cassava production (t) Philippines 1,815,700 1,784,897 1,844,377 1,891,780 1,956,574 Ilocos 10,656 10,269 11,771 12,134 12,868 Cagayan Valley 1,317 1,321 1,312 1,565 1,832 C. Luzon 7,776 7,972 7,532 7,890 7,744 S. Tagalog 62,800 64,066 63,515 63,751 65,861 Bicol 249,587 254,930 251,362 240,990 236,227 W. Visayas 50,238 47,857 46,822 45,979 45,000 C. Visayas 160,866 148,747 159,896 160,462 162,287 E. Visayas 96,722 92,849 93,975 94,856 92,678 W. Mindanao 218,422 215,652 220,528 227,391 226,634 N. Mindanao 103,041 1 17,727 156,709 157,086 139,139 S. Mindanao 24,949 25,679 26,547 26,245 23,020 C. Mindanao 10,773 11,066 11,230 12,691 12,864 CAR" 2,836 2,950 2,972 3,008 2,936 ARRM2) 783,734 753,886 758,609 806,135 894,815 CARAGA3' 31,983 29,926 31,597 31,597 32,669 Cassava yield (t/ha) Philippines Ilocos Cagayan Valley C. Luzon S. Tagalog Bicol W. Visayas C. Visayas E. Visayas W. Mindanao N. Mindanao S. Mindanao C. Mindanao CAR" ARMM2) CARAGA3' 8.60 5.90 3.72 6.27 6.12 8.06 5.25 7.81 3.85 9.65 9.41 8.27 6.58 9.42 12.26 3.69 8.74 5.69 3.21 6.94 6.38 8.17 5.01 7.67 3.87 9.58 9.35 8.19 6.96 10.03 12.90 3.64 8.72 5.33 3.42 6.49 6.10 8.10 5.21 7.70 3.91 9.76 10.21 8.10 6.79 9.53 12.49 3.72 8.88 5.49 4.47 6.41 6.09 8.31 5.22 7.66 3.91 9.96 10.16 7.95 7.60 9.40 12.65 3.71 8.66 5.80 4.60 6.42 6.20 8.31 5.20 8.06 3.47 10.01 10.10 8.82 8.19 9.20 11.31 4.39 "CAR = Cordillera Autonomous Region 2)ARMM = Autonomous Region of Muslim Mindanao 3)CARAGA = Newly created region comprising the provinces of Surigao del Norte, Surigao del Sur, Agusan del Norte and Agusan del Sur Source: Bureau of Agricultural Statistics 334 Cropping Systems The most common cropping system used by farmers is intercropping cassava with maize. This is practiced by farmers in Bukidnon , Mindanao and Camotes Island in the Visayas, where cassava is planted in the furrow right after the hilling up of maize. In this manner, the farmers can save labor and reduce costs in land preparation, cultivation and even fertilizer application. When the maize is harvested, the stover is either cut or pulled down to the ground as dried mulch, thus allowing the full exposure of cassava plants to sunlight. This practice has been proven to be successful and profitable, and is now being adopted by other cassava farmers, especially if their farms are flat and in open fields. Some farmers also grow cassava in rotation with legumes or maize, or plant casssava under coconut. In large plantations like Matling in Malabang, Lanao, cassava is rotated with sorghum, which is used for livestock feed. RESEARCH RESULTS Recent Experiments in Leyte 1. Long-term fertilizer trial under coconut. A long-term fertility trial under mature coconut trees spaced at about 8 x 8m, showed that yield differences due to fertilizer application were not significant in the fifth, but were significant in the sixth cropping cycle (Table 2). Highest yields were generally obtained from the treatment with 60 kg N, 90 P205 and 60 K20/ha (F8), while lowest yields were obtained in treatments without N and K application (F2 and F9), suggesting mainly a deficiency of these two nutrients (Figure 1). Soil analysis of the NPK check plots in 1996 indicated an intermediate level of 4.09% organic matter, but a very low level of K (0.07 me/100 g) and P (0.35 ppm Bray-H-extractable P). Thus, significant responses to N, P and K applications are in accordance with soil analysis results. 2. Variety x fertilizer trial. A variety x fertilizer trial was conducted in Bontoc, southern Leyte in an area with a slope of 20%, but with double hedgerows of Gliricidia sepium and vetiver grass. Results show that different varieties had significantly different responses to fertilizer (Table 3). The cassava variety with the heavier canopy (VC-3) responded most markedly to fertilizers. However, there was no significant difference between the levels of fertilizer application. 3. Cultural practicesfor erosion control. Table 4 indicates that during the 6th cropping cycle of the erosion control trial in Baybay, Leyte, soil loss was highest at 45.0 t/ha in plots with inorganic fertilizer (14-14-14) and lowest in plots with vetiver grass contour barriers (8.1 t/ha). On the other hand, cassava yields were highest in plots applied with dried grass mulch (14.5 t/ha) and lowest in plots with vetiver or lemon grass as contour barriers. ^ - o IT) n CT> _ O - o to SI O (N O) TO SI 3> O 0. O) o O) o TO Dry soil loss during the cropping cycle. Recent Experiments in Bohol 7. Intercropping. Table 5 shows the yields of cassava and intercrops as influenced by cassava planting arrangement during three cropping cycles at Bohol Experiment Station in Ubay, Bohol, Philippines. It was observed that even after three cropping cycles cassava yields were not significantly affected by interplanting of either soybean, mungbean, cowpea, peanut or pole sitao (yard-long bean). However, plant spacing significantly affected the yields of cassava and intercrops. Double-row planting of cassava (0.81 x 0.71 m in the double row, 2.0 m between double rows) generally resulted in the highest cassava yields, but a normal square arrangement (1.0 x 1.0 m) gave the heighest intercrop yields during all three years. 2. Prunning. Cutting of cassava plant tops (pruning) at 30 cm above ground level before or right after a typhoon damages the plantation, may result in improved root yields, depending on the time of pruning. Moreover, planting at closer spacing has also been found to mitigate the damaging effect of typhoons. For that reason, a time of pruning x plant spacing trial was initiated in Bohol in 1995. Table 6 presents the first year's result. There were no marked differences in the yield of cassava due to time of pruning but pruning did reduce yields on average 13-21%; this yield reduction was greater at wider than at closer spacing. Spacing had a significant effect on cassava yield. Highest yields were obtained in plots with planting distances of 0.80m x 0.80m or 0.60m x 0.60m. This implies that higher population densities can result in higher cassava yields, either with or without pruning. oo en CO U u st c c u u I! U x: -w St C °u 3 £ c J3 u l_ eg st c -2 Q. > 1 $ ;►• c *> -s. If e j= 4) a, 3 c —r •Si Is. E« i r » 3 a) a c > C3 o CO > < ON r~ en o ON X•*"■* en oo Os o CM + ON O CN CM ON m d X o E to US CO g o U ON ON en ON en Os Os a en os in ifl en * NO NO CO 0O en sen en en en oo en s ON ON Tt rm NO NO Tt Tr Tt en in ir> m © ci ri -h r•i r•i r~ ri CO CO rn cn en CN en r- o oo •T*• Tt so en en en en en i « -iflr~n -nio c c -^ S C Q + + + Q h + 4> oo Z E u c/5 >-, oo 60 g 'a. D. O i_ U IS uuuuun ON ON OO — O o — — — 5 t-~ ONCO oo CNNO — o o o o o oN os Tt r- o Tt Tj- on O Os Os k —•' o —•' d d —' so en — ON ON oo oo en o o o o — — — X) oo rm os rs o Tt oo en es os en m d d d d o Tj- r~ W O0 J> E Q o + + + + + o 1> co 4> > O0 Z co -g '■j a. oo E ■■—i oo 60 -c CT3 n. oo # * * on co -g'5.a.o u fl u u u o u < 339 Table 6. Effect of pruning and planting distance on the yield of cassava at Bohol Experiment Station, Ubay, Bohol, Philippines in 1995/96. Planting distance 1 Pruning treatments" Average 1 x lm 0.80 x 0.80m 0.60 x 0.60m Average 19.92 14.61 14.58 15.63 16.18 23.67 17.61 21.36 21.26 20.97 23.55 20.64 21.69 21.52 21.85 22.38 17.62 19.21 19.47 19.67 " 1 =no pruning 2- pruning at 5 months after planting (MAP) 3 = pruning at 7 MAP 4 = pruning at 9 MAP CONCLUSIONS During the past three years there has only been a slight increase in cassava production in the Philippines, both in area and in volume. Cultivation of the crop is still concentrated in Mindanao where there are better market outlets, such as chips traders, feed millers and starch processors. In Baybay, Leyte, cassava planted under coconut trees responsed more to K and N and less to P after six years of continuous cropping. Contour barriers of vetiver grass and application of dry grass mulch resulted in low levels of soil loss due to erosion, while cassava root yield was high in plots with application of grass mulch or Crotalaria juncea as intercrop. A variety with a heavier canopy responded most markedly to fertilizer application in Bontoc, Leyte. In intercropping trials in Bohol, cassava yields were not affected by any of the legume intercrops, but yields were affected by plant spacing. Pruning did not affect cassava yields, but closer spacing significantly increased yields. REFERENCES Bureau of Agricultural Statistics. 1996. Agricultural Statistics for 1995. Mimeographed sheets. Evangelio, F. A., F. G. Villamayor, Jr., A. G. Dingal, J.C Ladera, A.C. Medellin, J. Miranda and G.E Sajise Jr. 1995. Recent progress in cassava agronomy research in the Philippines. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp. 290-305. 340 VARIETAL IMPROVEMENT AND AGRONOMY RESEARCH IN MALAYSIA S.L. Tan' ABSTRACT Over the period 1994-1996, seven batches of seeds and two of meristem tissue cultures were evaluated and selected. As important as root yield is the trait starch content (or dry matter content) if a clone is to be accepted as a starch cultivar. At the time of reporting a number of clones have been identified as having yield levels similar to the high-yielding early cultivar MM 92, while showing dry matter contents equivalent to that of the commercial starch cultivar Black Twig. Two of these promising clones have also been evaluated positively for processing into oil- fried crisps, a popular local snack. Agronomic research included various studies on nutrient inputs and the effects of flooding on cassava performance when the crop is planted on drained peat. With the early cultivar MM 92, fertilizer inputs may be halved and applied once every alternate cropping without affecting root yields. While the effect of Ca application was not clear-cut, there was some indication that Ca applied as a 9 % foliar solution of Ca(NO3)2 or as 1 ,500 kg/ha of CaCO3 to the soil improved root dry matter content. Solid wastes from starch processing factories may be returned to the soil at the rate of 1.0-1.5 t/ha as a supplement to chemical fertilizer inputs. MM 92, in contrast to 12- month Black Twig (critical stage at 3 1/2 months), showed the greatest yield reduction when flooding occurred at 4-5 months after planting. Four days of continuous flooding resulted in yields declining as much as 45%. INTRODUCTION Malaysia and MARDI (the Malaysian Agricultural Research and Development Institute) have reached a major cross-road as far as cassava research is concerned. Labor costs, and indeed availability of agricultural labor, have become a major constraint to small-scale cultivation of cassava. Technology packages for mechanized production have been formulated by MARDI, but require a scale of production of at least 400 ha of cultivated area (Sukra et al., 1992). On the other hand, to keep a starch factory with a minimal daily capacity of processing 150 t fresh roots (Mat Isa, A., Food Technology Research Centre, MARDI, Serdang,pers. comm.) well-supplied with roots would require a cultivated area of at least 1,650 ha. This assumes fresh root yields of 25 t/ha, 275 working days in a year, and that the factory operates a single 8-hour shift per day. However, land of less than 14% slope (to allow for mechanization without accompanying risks of soil erosion) of the required area in one contiguous piece is no longer so easily available for cassava cultivation in Peninsular Malaysia, where the demand for starch is strong. Sustained research on cassava, particularly in the fields of breeding/selection and agronomy, has been going on for more than two decades in MARDI, and many practical and cost-effective technologies have been generated. Adoption of those technologies by the end-user, however, has been less than enthusiastic or satisfactory, largely due to the 1 Sweet potato and cassava breeder, Food and Industrial Crops Research Centre, MARDI, P.O. Box 12301, 50774 Kuala Lumpur, Malaysia. 341 constraint of land availability. For these reasons, in Malaysia.s Seventh Development Plan, covering the period 1996-2000, funding for cassava research has been discontinued. Starch, on the other hand, is anticipated to have a growing demand, especially in the widely varied processed food industries (Tan, 1996). Hence, research emphasis has shifted to sweet potato as a possible alternative starch source. MARDI is also on the treshold of embracing a corporate image, which brings with it an increasingly important role in income generation for the institute through such activities as consultancy services, contract research, patents and copyrights. Having built up strong expertise in cassava production technology, prospects are sound in attracting consultancy services. There has been encouraging interest from the private sector to invest in cassava production in the east Malaysian states of Sabah and Sarawak, as well as in other Southeast Asian countries where labor availability is still secured. Thus, while active cassava research in MARDI may wind down in the coming years, the promotion of cassava technologies will remain a significant role of the institute. The cassava breeding and selection program has CIAT to thank for keeping it alive for a few more years through the mechanism of the small contract agreement. These modest funds will ensure that the evaluation and selection of short-listed clones may be completed satisfactorily. VARIETAL IMPROVEMENT 1994-1996 Anticipating the gradual closure of the cassava research program, only two seed batches were introduced since the last Asian Cassava Research Workshop. In 1994, 1,141 seeds, labeled batch (CIAT 13 (TH), were sent from the CIAT-Thai program, while 1,040 seeds (CIAT 14) came from CIAT Headquarters in Cali. The germination rate was quite different for the two batches: 59% for CIAT 13 (TH) resulting in 673 seedlings, and only 7.8% for CIAT 14, giving 81 seedlings. Over the period 1994-1996, evaluation and selection following the usual stages of Seedling, Single-row, Preliminary, Advanced and Regional Yield Trials involved materials introduced from 1989-1994 (Table 1). Past experience shows that mere high root yield (as in cv. Perintis), even with an early harvestable characteristic (as in cv. MM 92), is insufficient to attract wide use of the clones by farmers, especially those supplying the starch processors. This is due to the existence of a price discounting system based on a minimal root starch content of 28% . In an effort to ensure greater acceptability of the new clones, intermediate (i.e. as high as the local cultivar Black Twig) to high root starch content has become a significant selection criterion. Latest harvest data from the yield trials are given in Tables 2 to 5. In all the trials, three control cultivars were used, namely Black Twig, Perintis and MM 92. 342 Table 1. Stage of evaluation and selection of CIAT germplasm over the period 1994-1996. Seedstock Stage of evaluation Year No. clones CIAT8 Regional Trials 1993-1995 7 CIAT7 +CIAT9(TH)" Advanced Yield Trials 1993-1995 16 CIAT 10 Advanced Yield Trial Regional Trial 1994-1995 1995-1997 20 7 CIAT 1 1 Preliminary Yield Trial Advanced Yield Trial 1994-1995 1995-1996 20 12 CIAT 12 Single-row Trial Preliminary Yield Trial 1994-1995 1995-1996 25 5 CIAT 13 (TH)" Seedling Evaluation Single-row Trial Preliminary Yield Trial 1994-1995 1995-1996 1996-1997 673 37 14 CIAT 14 Seedling Evaluation Single-row Trial 1994-1995 1995-1996 81 16 Meristems 1990 Advanced Yield Trial Regional Trial 1994-1995 1995-1997 10 4 Meristems 1991 Single-row Trial Preliminary Yield Trial Advanced Yield Trial 1994-1995 1995-1996 1996-1997 7 7 7 TH = introduced from the Thai-CIAT program Although in a Regional Trial shown in Table 2, SM814-18 produced root yields at 6 and 12 months equivalent to those of Perintis and MM 92, its dry matter content was unfortunately as low as these two clones, so the clone did not merit final selection. In the recently harvested Regional Trial (Table 3), MKUC28-77-3 and Rayong 60 showed promise at 6 months: their root yields were not significantly different from the controls, while dry matter contents were similar to that of Black Twig. However, at 12 months, the best yield of 41.2 t/ha was obtained with CM6149-30 (although not significantly different from the three controls). Taking root dry matter content into consideration, the highest dry root yield was also recorded for the same clone. CM6149-30 had a dry 343 matter content similar to that of Black Twig, and therefore significantly higher than those of Perintis and MM 92. Rayong 60, together with CM6149-23, CM6149-54 and CM6885-75, had dry root yields (ranging from 10.7 to 11.3 t/ha) similar to the controls because of their intermediate to high dry matter contents. CM6885-75 had a root dry matter content of 43.0%. Table 2. Mean data from a Regional Trial on "CI AT 8" seed batch over two seasons at Pontian peat station, Johore, Malaysia. Clone 6-month data 12-month data Root Harvest RDMC" Root Harvest RDMC" yield index (%) yield index (%) (t/ha) (t/ha) OMR32-06-21 20.3 0.47 34.3 33.6 0.60 34.0 OMR32-06-25 20.2 0.45 30.3 29.1 0.48 30.6 OMR32-06-46 27.8 0.60 29.9 39.8 0.63 26.5 OMR32-06-64 21.6 0.50 31.5 35.2 0.58 37.1 OMR32-30-11 20.0 0.55 32.5 33.7 0.62 37.0 OMR32-30-37 23.2 0.48 30.3 30.2 0.52 36.0 SM814-18 32.6 0.60 27.8 52.7 0.66 26.0 Black Twig 20.4 0.44 27.6 34.0 0.50 29.5 Perintis 29.7 0.58 25.8 53.5 0.65 26.0 MM92 34.7 0.65 25.1 51.4 0.63 19.9 LSD(P = 0.05) 5.4 0.06 4.5 5.6 0.03 4.4 "RDMC = Root dry matter content In the Advanced Yield Trial, testing clones selected from CIAT 7 and CIAT 9 (TH), the clone SM967-1 showed promise in terms of dry matter content (higher than that of Black Twig), while having root yields equivalent to that of MM 92 at 6 and 12 months (Table 4). Harvest data from the other Advanced Yield Trial (on CIAT 1 1 clones) showed SM1565-7, SM1562-19 and SM1565-57 to be desirable in terms of dry matter content, with reasonable root yields at six months, while the first two clones as well as SM 1794-23 and CM7719-7 were promising in terms of starch content and root yields at 12 months (Table 5). 344 Table 3. Regional Trial on clones selected from "Meristems 1990" and "CIAT 10" seed batch (6-and 12-month data) at Pontian peat station, Johore, Malaysia. Clone 6-month data 12-month data Root Harvest RDMC" Root Harvest RDMC" yield index (%) yield index (%) (t/ha) (t/ha) Rayong 3 10.6 0.74 38.3 18.2 0.77 38.1 Rayong 60 19.3 0.66 39.3 28.4 0.69 39.1 CMR28-67-76 17.3 0.66 38.8 17.9 0.64 40.3 MKUC28-77-3 18.1 0.66 43.6 27.0 0.68 35.9 CM6 149-23 14.3 0.66 40.5 31.4 0.64 35.7 CM6149-30 22.1 0.63 38.0 41.2 0.67 39.3 CM6 149-54 16.1 0.60 38.0 29.8 0.67 37.4 CM6 149-55 12.6 0.61 40.8 13.6 0.63 41.5 CM6885-75 16.6 0.64 39.8 24.9 0.68 43.0 CM7752-4 11.8 0.60 35.7 18.6 0.63 40.5 CM8061-2 11.8 0.60 40.3 19.2 0.73 37.8 Black Twig 21.7 0.60 41.9 37.1 0.65 39.8 Perintis 20.2 0.69 32.3 39.5 0.80 27.5 MM92 24.4 0.78 28.5 29.8 0.74 26.9 LSD(P=0.05) 9.7 0.10 5.0 13.1 0.08 5.7 )RDMC = Root dry matter content 345 Table 4. Advanced Yield Trial of selected clones from "CIAT 7" and "CIAT 9 (TH)" seed batches (6-and 12-month data) at Pontian peat station, Johore, Malaysia. Clone 6-month data 12-month data Root Harvest RDMC" Root Harvest RDMC" yield index (%) yield index (%) (t/ha) (t/ha) OMR33-15-7 18.3 0.71 17.1 23.6 0.65 32.2 OMR33-15-21 18.3 0.60 25.9 30.3 0.59 33.6 OMR33-06-21 12.4 0.50 30.4 18.9 0.38 40.9 OMR33-63-4 16.1 0.68 35.7 24.3 0.62 36.0 OMR33-63-30 16.5 0.65 30.8 28.1 0.58 36.3 OMR33-63-9 14.3 0.68 30.3 22.7 0.59 39.6 SM836-10 20.7 0.58 26.0 32.8 0.60 31.7 SM%7-1 22.4 0.66 29.3 31.1 0.63 36.7 SM987-13 20.7 0.62 23.7 36.5 0.60 27.4 Adira 4 15.4 0.62 25.5 31.1 0.60 32.1 Black Twig 24.2 0.68 23.8 34.3 0.64 33.4 Perintis 19.3 0.69 22.8 31.2 0.72 27.1 MM 92 23.1 0.76 21.4 33.3 0.68 19.7 LSD(P=0.05) 5.0 0.08 8.6 8.1 0.06 4.7 l)RDMC=Root dry matter content 346 Table 5. Advanced Yield Trial on selected clones from "CIAT 11" seed batch (6-and 12-month data) at Pontian peat station, Johore, Malaysia. Clone 6-month data 12-month data Root Harvest RDMC" Root Harvest RDMC" yield index (%) yield index (%) (t/ha) (t/ha) CM6658(C)-2 9.8 0.20 25.3 32.2 0.42 38.2 SM 1583-2 16.1 0.37 31.1 36.4 0.52 41.5 CM6098-45 11.8 0.21 29.8 25.6 0.35 39.5 SM 1565-7 24.3 0.45 29.6 41.7 0.57 40.3 SMI 565-57 20.1 0.45 36.0 28.3 0.51 41.1 CM8223-29 9.7 0.24 29.5 22.6 0.31 40.4 CM8223-30 16.8 0.33 27.1 35.0 0.48 40.7 CM6517-9 11.8 0.30 29.6 30.6 0.44 40.0 SM1562-19 23.6 0.41 31.6 36.8 0.55 37.5 SMI 794-23 14.9 0.39 33.7 41.0 0.56 40.3 CM7719-7 12.1 0.25 32.9 43.6 0.47 37.2 CM7719-23 11.2 0.24 31.3 35.3 0.51 38.8 Black Twig 11.4 0.16 27.1 38.1 0.42 36.2 Perintis 23.1 0.43 25.8 50.8 0.55 33.4 MM92 29.6 0.48 25.1 45.7 0.56 32.1 LSD(P= 0.05) 5.6 0.12 6.5 9.7 0.10 6.6 "RDMC = Root dry matter content Recent trends show a growing demand for edible cassava types for processing into oil-fried crisps, a popular snack locally known as kerepek. This cottage industry is gaining ground, especially in Johore with its close proximity to the Singapore market. To capitalize on this demand, root samples from the three promising clones at six months (from CIAT 1 1), which also have low root cyanide contents (< 60 ^g HCN/g fresh root), were sent to a kerepek processor to test their acceptability for this type of utilization. The feedback has been enlightening: SM1562-19 was considered to be superior to the local cultivars currently used, SM 1565-57 was acceptable as a substitute, while SM1565- 7 was rejected because of its hard texture when so processed. Such information coming from the Advanced Yield Trial will be very beneficial in making the final selections for the Regional Trials. The above results show that it is becoming increasingly difficult to raise root yields above those of the released cultivars. However, there is still room for 347 improvement of the dry matter content (i.e. the starch content) at this yield level. AGRONOMY RESEARCH 1994-1996 Agronomy research on cassava terminated in 1995 with the retirement of the agronomist, Mr. Chan Seak Khen, and the resignation of his replacement. Although the breeder shouldered the responsibility of the ongoing research trials for a short time, increasing work pressure from sweetpotato research has caused her to abandon further efforts after completion of those trials. What follows is a summary of the findings of final agronomy research on cassava conducted by MARDI. Reduction of fertilizer use with an early clone Using the six-month cultivar MM 92, this trial was conducted on drained peat over six cropping cycles, beginning in 1991. Fertilizer inputs for cassava on peat have been high (200 kg N, 30 P2O5 and 160 K2O/ha) in the past when the 12-month Black Twig cultivar was used (Tan and Chan, 1994). The nine treatments included a combination of fertilizer rates and frequency of application as follows: Application rate over 6 crops" Percent of Treatment t Control 1 1st 2nd 3rd 4th 5th 6th 1 (Control 1) 1 1 1 1 1 1 100.0 2 1 0 1 0 1 0 50.0 3 1 0 0 1 1 0 50.0 4 1 0 0 0 1 0 33.3 5 Vi Vi V4 '/2 '/2 H 50.0 6 Vi 0 '/4 0 Vl 0 25.0 7 'A 0 0 '/2 Vi 0 25.0 8 % 0 0 0 lA 0 16.7 9 (Control 2) 0 0 0 0 0 0 0.0 " each crop of 6 months duration 1: full rate of fertilizers (200:30:160) XA: half rate of fertilizers (100: 15:80) 0: no fertilizers 348 Results show that when MM 92 was grown, the fertilizer rate could be reduced by half, and this half rate (100: 15:80) could be applied to every alternate crop instead of every season without affecting either fresh or dry root yield (Figures 1A and IB). Root dry matter content was unaffected by any of the fertilizer treatments. The results of this trial have been published (Tan, 1995). Effect of waterlogging on cassava performance These series of experiments conducted in fibre-glass lysimeters filled with peat soil, started in 1989 using the cultivar Black Twig (harvested at 9 months); in 1992 the test variety was changed to cultivar MM 92 (harvested at 6 months). The objective of the experiments was to determine the critical stage of plant growth when flooding had the most detrimental effects, as well as the effect of flooding duration at this critical stage. For Black Twig, the critical stage was at Vh months after planting, at the stage of storage root initiation, when the subsequent root yield could be reduced by 80%. Figure 2 shows that flooding at this stage, from 1 to 4 days, did not bring about a significant root yield reduction (although a 25% reduction was observed after 4 days' flooding). Figure 3 shows that for MM 92 yield reductions up to 80% occurred when flooding was imposed at 4-5 months. MM 92 showed significant yield declines of 25% and 45% after flooding for 4 days at Vh months and at 5 months, respectively. 349 2 Oo =s u 200 150 100 50 0 ab bc bcd cde ab de 1 = full rate '/: = half rate 0 = zero be llllll 101010 100110 \000\0'A'A'A'A'AWAO'AO'AO'AOO'A'AO '/4 000 WO 000000 Control 1 Fertilizer rate per season Control 2 B. -a u '>. o £ E?■a > u 60 50 40 30 20 10 bc bcd bcd ab 1 = full rate Vi = half rate 0 = zero cde llllll 101010 100110 100010 A•H 40 01 20 o V. 3.0 100 „ 80 60 > ■h 40 0) 20 0.30 O -SC 205 CATAS .94 a -SC 205 CATAS .95 A-SC 124 CATAS .95 » -SC 201 Nanning .95 . -Vinh Phu AC(*3 .95 •-KM 60 AC#3 .95I 3.5 4.0 4.5 5.0 % N in YFEL-blades 5.5 6.0 0.35 0.40 0.45 0.50 % P in YFEL-bladea 0.60 0.6 1.0 1.2 1.4 1.6 % K in YFEL-blades 1.8 2.0 Figure 3. The relation between the relative yield of cassava, i.e. the yield without the nutrient as a percent of the highest yield with the nutrient, and the concentration of N, P and K in the youngest fully-expanded leaf (YFEL)-blades at 3-4 months after planting of various cultivars planted in four fertility trials in China and Vietnam. 362 Figure 4 shows the relationship between relative cassava yield and the organic matter, available P and exchangeable K contents in the soil, obtained from yield data and soil analyses results of ten long-term NPK trials conducted in Asia between 1993 and 1996. Critical levels of 3.2% OM, 7 ppm P (Bray-2) and 0.14 me K/100 g (55 ppm K) were estimated from these data. These are slightly higher than those estimated from earlier data from almost the same sites, i.e. 4.5 ppm P and 0.12 me K/100 g (Howeler, 1995A), but fall within the "medium" range reported elsewhere for these soil parameters (Howeler, 1995B; and Howeler, 1996). These "critical" levels serve only as a general guide in the interpretation of soil and plant tissue analyses to enable the diagnosis of a likely nutrient deficiency or toxicity, and to estimate the need for application of a particular nutrient. Fertility Maintenance through Green Manuring Research on soil fertility maintenance or improvement through green manuring, intercropping or alley cropping has been conducted in Vietnam, Indonesia and Thailand. In Vietnam the experiment was conducted on a highly-fertile red Latosol (probably classified as Eutrustox) in Hung Loc Research Center and not much beneficial effect of green manuring would be expected. Table 2 shows that there was no significant effect of treatments in the first year, but that some treatments started to improve yields after 3-4 years of continuous cropping. Intercropping and mulching at 2-3 months after planting (MAP) of cowpea or Canavalia ensiformis increased yields, probably by increasing the soil moisture conditions during the dry season, or by the slight increase in OM and K contents of the soil. The beneficial effect of alley cropping was less pronounced, but also increased over time, especially the alley cropping with Tephrosia candida. Alley cropping with Gliricidia sepium, Leucaena leucocephala or Flemingia congesta also significantly increased cassava yields in a low-N soil of Jatikerto Experiment Station in Malang, Indonesia, after several years of continuous cropping (Wargiono et al., 1995; Wargionoef a/., 1997) In Thailand three green manure species, i.e. Crotalaria juncea, Canavalia ensiformis and pigeon pea, were either intercropped and cut at 2 MAP, interplanted at 7 MAP and cut and mulched before the following planting, or were used as green manures and pulled up and mulched before planting cassava; in the latter case, cassava was harvested after 18 months for a two-year cropping cycle. Table 3 shows that in the sandy clay soil of Rayong Field Crops Research Center the intercropping and mulching of all green manures increased yields, with Canavalia ensiformis being most effective. Interplanting at 7 MAP had no beneficial effect on cassava yields, but may benefit the next crop; the interplanted Crotalaria juncea competed strongly with cassava for soil moisture during the dry season, resulting in low cassava yields. Green manuring with Crotalaria or Canavalia markedly increased the yield of 18-months old cassava, more than doubling that of cassava harvested at 12 MAP. This may be a promising system for 363 the soil and climatic conditions of Thailand. Table 3 also shows that mulching Crotalariajuncea generally supplied more nutrients than either Canavalia ensiformis or pigeon pea. EROSION CONTROL 1. Effect of Crops on Soil Erosion A comparative study on the effect of various crops on soil erosion and on nutrient uptake and removal, conducted since 1989 in Sri Racha, Thailand (Putthacharoen, 1993; Howeler, 1995A), was continued for a second cycle of 27 months between 1991 and 1993. Seven crops were grown in adjacent plots on 7% slope and soil losses due to erosion were measured at monthly intervals. Figure 5 shows the accumulative erosion losses during the second cycle. Similar to the first cycle, cassava grown for root production (spaced at 1.0x1.0 m) resulted in the highest level of erosion, followed by cassava grown for forage production (spaced at 0.5x0.5 m). Soil loss for cassava root production was about 130 t/ha in the first crop and about 40 t/ha in the second crop. The closer spacing of cassava plants used for forage production resulted in considerably less erosion due to a more rapid canopy closure. Sugarcane also produced high levels of erosion during the first year after planting, but almost no erosion during the following ratoon crop. The short-cycle crops of maize, sorghum, peanut and mungbean produced much lower levels of erosion, ranging from 35 to 55 t/ha in two years (three crops). Pineapple caused the least amount of erosion since it was not replanted but left as a ratoon crop during the second cycle. Table 4 shows that cassava yields were low in the first cycle, but rather normal in the second. Soil losses due to erosion, however, were very high for cassava in both cycles, about two to six times higher than those of other crops grown during the same four year period. Thus, under the soil and climatic conditions of Thailand is seems that cassava indeed causes a lot more erosion than other crops. In a similar comparative study conducted on 5% slope in Tamanbogo, Lampung, Indonesia, cassava again caused higher levels of erosion than any of the other food crops, but the differences were much smaller than in Thailand, partially because the soils in Tamanbogo are less erodible than in Sri Racha, Thailand, and partially because the longer wet season in Tamanbogo allowed the planting of two short-cycle food crops per year, versus usually only one crop in Thailand. Similar to data presented by Wargiono et al., (1997), Table 5 shows that erosion in cassava was considerably higher than in other crops when no fertilizers were applied, but only slightly higher than other crops when each crop was adequately fertilized. The consecutive planting of two peanut crops in one year produced the highest net income and lowest levels of erosion, while cassava grown in monoculture produced a low income and serious erosion. For that and other reasons, most farmers in Indonesia plant cassava intercropped with other food crops, which generally increases income and reduces erosion. 364 100 ~ 60 S 40 20 1.0 100 80 A 2 60 40 20 1.5 2.0 2.5 3.0 Organic matler(%) 3.5 4.0 4.5 ■ * X 0 A A m ~ V V A V O o V A * o - / a 1 ° 0= Nanning o = Guangzhou a = Danzhou (CATAS) » = Hung Loc • = Truco * - Umas Java * = Yogyakata ■ » Jalikerlo I * = ViSCA I i 1 I i I 10 15 20 25 Available P(ppm, Bray II) 30 35 100 80 2 60 ■a 1 u 3 ■w C 3 ;►• u 3 s 61 C "u 3 t 2 so £ c E■0 c c '5. 3 Q. c£0 u V 53 .x c cd 1* ■0 c U 0 ex 2e $'n. DCQ. C • L. U u u S ex c < iJ! If s ON ca -9■■—■ m T) Os • O O Uh CO > CO «fr a Os r ) f■) Os r•-> Os r5 ON Os C E^- X> X) ■§ Cd Cd X> 'J X> w-)inTtoNOOCMOm >/-> uS Tt r~' Os r--' so sb CM/■> — O —' Os CM Os in-«NO X> O IN ON U~i O e« x> o t*~ r•"> O noorNlmrnTtooN ? Cfor Cfor Teph cowp< Leuc Gliri J .■§ S | S •§ | f 5 * * * * * *3 ■8 ■SI'S ■8 11 G C O Q, 0. q. 0, o. a. ct E Q, o. 0 O. Q, 0 O O O C O O C E ca 1- 1- -J V- Lw -J -j O O 1 OOII > h h >• h <- >• >n rd D 4> D 4> 3 .S .S « .S -E 13 73 uuuuuuuu 00 z — CM rO Tj- 10 NO P^ 00 U_ a: ■s C o E CO I ■8 > o E 5.If -2 -D 8 -S r u C ^3 a. c cd . h E Q- E D CO X 3 ■5 C/5 O cd ■2 8.S.S 3 £ » 2 « a> eu * u -s ■= « X! 3 « S5 ■S 3 <1> 00 D. — 1-9 g-1 u « u 3 ■5 in m 5 t n 3 3CM fi •§ a § a .s S3 0) « C s * 1"8 W -w "s | ■o U •m CS .2 £ 4> » £ * 5 >■ 8 5 IS S.H ua VC legu 94s Hg L. .= J3 ►w -O ° es WD~ C S3 3 a u E 2 suringor earchCes .__ a s = OS rees o9 ro o 2 ■o *> §E .2 ©■s ►< S(2 • Q. C ■= CX- -2 ?era nted moi - s oo ■a o. . i- •* i u_ ^ ^ 03 "O o3 U E a oo c •* .22 *—'"C 3 z u a 00 — N N N * N 0CO\ Tt W N 00 1D O W »J i I/-> VO f> N—-ON° - - —i o •* rl h - w en (N (N m Tj- o u-> o r- cn r- (N■* (N (N VO W J1 no c-> en rn N (N Tfr -* (N CM r- 4-■s < *s 4N i-s SE i 8 ~ .3 D- < •8 11D- < s 03 ■8 3Nn « * ^1 c 03 < *-* o3 2 .e i =o gE E oo 22 - oo -5 60 On 2 1 a oo o3 £ « - jj r^ [fl S ^* "^00 CO a> on 13 -s 1.1 E c .5 . . «> 3 S « E c D O 03 a> 03 3 E on £ V £ 2 s 60 ^ Is 03 03 > > 03 03 OO OO U U S a o 2 c? S © o? <3 g oo c a oo je U.B.OU.E? + + 03 03 > > o3 03 OO OO ol 3 u u Q. + + 03 03 > > 03 03 OO 00 o3 3 u u 60 Q 3 O O C 2 —* £ -2 > -2 03 "a00 v. o1r£ u 0 i 860 G 60 K1 > & 03 00 ^ — ri«*in>Or-ooaO"-N OO -5 c 1 03 "8 03 > 03 oo i 1o3 C E o3 l rn 0 03 ■x S u > o 8 M •- VS ._ C - 3 g3 is E vO m 0£ c .C 3■o C « J= 2 8 ca•5 ! « » . §1 P c £ o 2 S © po &£ o *©.••• On tJ- q M (s| VI fi i? B ft ft B — on — —i N m —* r~ oo m Tj- — (N —i Tt Tt (N ON SO (N — —i Tj> Tj> —. r- oo in On On m .S 3 tj T3 -5 O o o. O —i o 2 P ,o ,o ,o 73 > 03 c/3 c/j 03 01 03 E -C u- c oo CO S3 3 C Six .J ca 60 3 O0 .J >* x: 3 o 1/1 3 a. o c/3 J^ a> > 13 c 73 ca H .is f o On >• — o .^ T3 On C On cl — - 2 o on *j 00 t/3 On .2 On O On C On (N — >N I O On ^ 0O VS On 00 O | 2 ■a "3 .>- .u- 03 -*— -4-m o o 00 en ?K ii n I! •s* s « ? 2 2 - 3 -° a o si i « ? s i>2 2 to *- 3 ? c» 8 • c o c — 3 o o e id TO a. m n ii » it i i n • o > a ■ * < o o o (0 o CM o 00 o 00 o G o> [O * a 1— — .2 c k. IN •-- ♦* n o § c I- CM z- 1 o 1 CO c < o o oo 1 O C3 (eq/i) sso| |ios Ajp uinoov (uiiu) uojiBiidpajd A|i|iuow I ! &5 — I *9 IS c c £ Sj § £ >• su i3.e C*. L. c o i) iS 9 3 e 33>•■c« c V fe V d >■ V S3 3■D U a L. o 2^ 33 '3 a a ■n■o e c 33 s '5 T3 jE .£ H 'S^ Im -w c 8 u L. _c c o E a.>o 3 EP Q. c O >-. c 33 3 c e*c V a 33 c a g E*- e s & i o c IP 33 E ce■a c B 35 it & oc >-■ at § a. c ■■3 ' o 0) rr1 s 'i H**e > W u NO 00 < NO 00 NO Tt Tt ' < m Os ON ^ o o > < 8. >N •c ca > NO ^ ON n NO 3 3 60 I a X J3 | | ■2 g x u m o3 3 8 a. "i m 60 c o M (2 61) d 0 >> ca (2 m <*1 IT) n NO in -5 •c 2 -4-t 0u D. O ON SI) C O >1 3 m oo Os IT) O ri OO ^ § 1 I o su 2 > < 60 60 g .9 o o < H H H 374 25.65 16.40 3.60 25.65 13.28 3.40 25.94 15.47 7.92 27.48 18.45 7.75 23.99 14.83 7.61 21.17 12.27 6.04 Table 7. Effect of various soil/crop management treatments on cassava yield and erosion in cassava grown on 5% slope in Khaw Hin Sorn, Prachin Buri, Thailand, in 1995/96. Cassava Cassava Dry root top soil yield yield loss Treatments (t/ha) (t/ha) (t/ha) 1. Cassava monoculture, no barriers 32.53 22.67 10.01 2. Cassava monoculture, vetiver (native) hedgerows" 3. Cassava monoculture, vetiver (Sri Lanka) hedgerows" 4. Cassava monoculture, Crotalaria juncea hedgerows" 5. Cassava monoculture, pigeon pea hedgerows" 6. Cassava monoculture, Leucaena leucocephala hedgerows" 7. Cassava monoculture, Gliricidia sepium hedgerows" 8. Cassava (Rayong 1) intercropped with cassava (Rayong 60) 27.39 26.32 10.07 " The hedgerows occupied about 20% of the total area, resulting in lower yields. REFERENCES Chan, S.K., S.L. Tan, H. Ghulam Mohammed and R.H. Howeler. 1994. Soil erosion control in cassava cultivation using tillage and cropping techniques. MARDI Res. J. 22:55-66. Evangelio, F.A., F.G. Villamajor Jr., A.G. Dingal, J.C. Ladera, A.C. Medellin, J. Miranda and G.E. Sajise Jr. 1995. Recent progress in cassava agronomy research in the Philippines. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp. 290 -305. Evangelio, F.A. and J.C. Ladera. 1997. Recent progress in cassava agronomy research in the Philippines In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Howeler, R.H. 1995A. Agronomy research in the Asian Cassava Network-Towards better production without soil degradation. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp 368-409. Howeler, R.H. 1995B. Results of soil analyses in Asia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp. 447-464. Howeler, R.H. 1996. Diagnosis of nutritional disorders and soil fertility maintenance of cassava. In: G.T. Kurup et al. (Eds.). Tropical Tuber Crops. Problems, Prospects and Future Strategies. Oxford and IBH Publishing Co. Pvt. Ltd., New Dehli, India. pp. 181-193. Nguyen Huu Hy, Pham Van Bien, Nguyen The Dang and Thai Phien. 1997. Recent progress in cassava agronomy research in Vietnam. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in 375 Danzhou, Hainan, China. Nov 3-8, 1996. Nguyen The Dang, Tran Ngoc Ngoan, Le Sy Loi, Dinh Ngoc Lan and Thai Phien. 1997. Farmer participatory research in cassava soil management and varietal dissemination in Vietnam. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Putthacharoen, S. 1993. Nutrient Removal by Crops and Nutrient Loss by Erosion in Cassava in Comparison with that of Other Crops. MSc. Thesis, Kasetsart Univ., Bangkok, Thailand. 103 p. (in Thai). Tongglum, A., V. Pornpromprathan, T. Nual-on and R.H. Howeler. 1997. Recent progress in cassava agronomy research in Thailand. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Vongkasem, W., K. Klakhaeng, S. Hemvijit, A. Tongglum, S. Katong and D. Suparhan. 1997. Farmer participatory research in cassava soil management and varietal dissemination in Thailand. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Wargiono, J., B. Guritno, Y. Sugitoand Y. Widodo. 1995. Recent progress in cassava agronomy research in Indonesia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop, held in Trivandrum, Kerala, India. Nov 2-6, 1993. pp. 147-174. Wargiono, J., Kushartoyo, Suyamto H. and B. Guritno. 1997. Recent progress in cassava agronomy research in Indonesia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Zhang Weite, Lin Xiong, Li Kaimian and Huang Jie. 1997. Farmer participatory research in cassava soil management and varietal dissemination in China. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. 376 FARMER PARTICIPATORY ADAPTATION AND ADOPTION OF CONTOUR HEDGEROWS FOR SOIL CONSERVATION Sam Fujisaka1 ABSTRACT From 1987 through 1992 a team of on-farm researchers worked at Claveria in northern Mindanao to improve the productivity and sustainability of the local upland rice and maize-based agroecosystems. Adaptive research goals were to improve crop productivity, control soil erosion, and improve nutrient cycling. A "strategic" goal was to develop methods by which research programs could work with farmers to develop locally appropriate and adoptable innovations. We used contour hedgerow systems as a general technology to control soil erosion; worked with farmers on their adaptation of the technology; and eventually our results included lessons for farmer-participatory research, technology development, and technology transfer. INTRODUCTION Soil erosion and soil nutrient depletion are major problems in cultivated upland areas in southeast Asia. Agroforestry technologies developed to address these problems include vegetative strips to reduce soil erosion on sloping lands and use of legume tree biomass to improve soil nutrient cycling (Huxley, 1986; Kang and Wilson, 1987; and Young, 1986, 1987). Innovations, however, have not been widely adopted because of technical problems and lack of fit with farmers' circumstances. With a goal of increasing the local appropriateness and, therefore, adoption rates of such technologies, recent approaches have actively involved farmers in the development of such conservation technologies (Fujisaka, 1989a; Getahun and Njenga, 1990; Pahlman, 1990; Raintree and Hoskins, 1988; Roucheleau, 1987; and Scherr, 1991). This paper describes a case in which farmers adapted a "knowledge-intensive" technology of contour hedgerows to fit their needs and local circumstances. Contour hedgerows harness erosive forces (plowing and rainfall runoff) to naturally form terraces. The technology is appropriate for areas with sloping land, use of tillage, intense rainfall, and land scarcity. Researchers from the International Rice Research Institute (IRRI) and the Philippines' Department of Agriculture facilitated farmer-to-farmer training and farmers' adaptation of contour hedgerows in Claveria, Misamis Oriental province. Claveria farmers were able to: 1 . modify hedgerow establishment methods in order to substantially reduce labor requirements (Fujisaka, 1993); and 2. select locally suitable species for use in the hedgerows (Fujisaka et al., 1994). LOCATION AND METHODS Claveria is an on-farm research site located at 390 to 550 m above sea level. A 1 Centro Internacional de Agricultura Tropical (CIAT), Apartado Aereo 6713, Cali, Colombia. 377 mean of about 200 t/ha/year of soil are lost from slopes given that rainfall is moderate to high and about 59% of the cropping occurs on land with > 15% slope. Average rainfall is 2200 mm/year, with 5 to 6 wet months (July-Dec > 200 mm/month) and 2 or 3 drier months (< 100 mm/month). Soils are classified as oxic Dystropepts, ranging from clays to silty clay loams; these are acidic (pH 4.5-5.8) with low available P (1 .3-4.7 ppm), low CEC (6-12 me/lOOg), medium to high Al saturation (1 1-51 %), and moderate organic matter (3.16%) and exchangeable K (113.1 ppm). Farmers settled in the area in the 1950s following deforestation by logging and used some of the resulting grasslands for large-scale cattle ranching in the 1960s and 1970s. Agriculture is now semi-permanent with all lands de facto or de jure owned; some land fallowing is practiced; and lands are prepared by oxen-drawn plow. Main crops are traditional and improved maize, traditional and improved upland rice, cassava, and, on smaller plots, tomato for commercial sale. Farmers adopting contour hedgerows grow these same crops in the alleys formed between hedgerows. Farmer-to-farmer procedures were used to introduce contour hedgerows to Claveria farmers (Fujisaka, 1989b). In 1987, six farmers who reported yield declines due to soil erosion and two IRRI technicians learned from farmers in the non-government World Neighbors project in the neighboring island of Cebu how to use an A-frame to establish contour lines, construct contour bunds and ditches, and plant hedgerows. The "original" hedgerows were comprised of one or two rows of the leguminous tree Gliricidia sepium, and one or two rows of Napier grass, Pennisetum purpureum. These farmers and later adoptors trained more than 200 farmers from 1987 through 1991 using the farmer-to-farmer techniques applied in Cebu. Approximately 80 Claveria farmers adopted some form of contour hedgerows by late 1992. Researchers at the site recorded field areas and slope, length of hedgerows, area occupied by hedgerows, heights of embankments formed over time, labor used for establishment, farmers' technical changes and adaptations, and effectiveness of methods. Open-ended interviews were used to obtain farmers' perceived benefits, associated problems, and evaluations of technology components. Slopes of alleys (areas between hedgerows) were calculated based on measurements of the embankments, assuming that hedgerows were placed at 1 m vertical intervals. Fifty-five of the adoptors were again interviewed and their 60 hedgerow fields monitored at the end of 1992. Many later adoptors (within the 1987-92 period) relied- solely or combined with other species-upon naturally occuring "weeds" in their hedgerows. We identified these in each of 57 fields (three fallowed fields were not included). Forty-six of the fields with hedgerows utilized at least some weeds. For each of these fields and during the main cropping season, ten 10-meter lengths of hedgerow were randomly selected across each field; the five weed species with the greatest number of individuals were identified, with actual counting done if there was any doubt as to 378 which weeds were the most frequent. "Main" weeds presented include any listed as one of the five most frequent in any of the 57 fields monitored. Weed species encountered in the natural hedgerows were compared to major weeds in upland rice and maize identified by farmers in an earlier study. Farmers also evaluated the contour hedgerows. We asked, "How do you now evaluate your contour hedgerow system?" The question was intentionally open-ended to allow answers reflecting what was important to each individual, e.g. hedgerow functions, problems, and/or species choice. No specific or further prompts were used. Each response is presented in terms of its percentage of the total number of answers given. Farmers who did not adopt the technology after farmer-to-farmer training were interviewed about reasons for non-adoption. Slopes of the lands of adoptors and non- adoptors were measured and compared. Crop yields on farmers' fields with and without contour hedgerows were monitored in the wet season of 1991, a severe drought year, by taking crop samples. Other data on yields and on soil erosion rates with and without contour hedgerows were provided by other researchers working at the site. RESULTS Contour Hedgerow Adoption in Claveria By the end of 1990, each of 75 adoptors in Claveria contoured a mean of 0.7 ha and established 761 m of hedgerow per field. Mean slope of fields was 22%. On average, hedgerows occupied 9% of the field areas on which they were established. Because farmers placed one hedgerow per meter of vertical drop within a field, length of hedgerows per ha increased as initial field slope increased, with a corresponding increase in labor required, although this was influenced by evolving hedgerow establishment methods. Terracing took place rapidly after hedgerow establishment. For nine fields established in 1987, mean heights of hedgerow embankments were 36 cm after one year, 44 cm after two years, and 49 cm after three years, indicating that slopes of alleys decreased from an initial mean field slope of 16% to 9% after one year, 8% after two years, and 7% after three years. Furthermore, 75% of the embankment formed after three years had already formed in the first year, suggesting that alleys reached relatively stable slopes after one year. Height of terrace embankments in front of the hedgerows tended to increase as slope increased. Farmer Modification of Hedgerow Establishment Methods Working as a group, farmers established hedgerows on 10 parcels of land having an average size of 0.8 ha in the first year. Labor for establishment (average 29 days/ha with 55% for shoveling) and hedgerow density (673 to 1555 m/ha) depended on field slope and resulting variation in distances between strips. Farmers established 17 to 57 379 m/person/day of hedgerow; labor to establish hedgerows was least on fields already plowed after initial rains and greatest on grassy fallowed fields on which hedgerows had been established prior to the rains. In 1987 farmers spent an average 14 hours per 100 meters of hedgerow established, and 29 days of labor per ha on which hedgerows were placed. By 1990 farmers needed only 8 hours per 100 meters and 16 days per ha (Table 1). The original hedgerow establishment method consisted of using the A-frame to determine contour lines, double plowing to create a bund, shovel work to reinforce bunds, and planting of a double row of trees plus a single row of grasses. Farmers saved labor by reduced plowing, virtual elimination of shovel work, planting of either trees or grasses rather than combinations, or-by 1989 and 1990-the staking of contour lines (usually but not always bunded), which were then left to be covered by "weeds" or native grasses (discussed below). There were no differences among farmers' various hedgerow establishment methods in terms of embankment formation and terracing, but establishment labor per ha also decreased with increasing field size, suggesting some "economy of scale". Table 1. Farmer contour hedgerow adoption and labor used for hedgerow establishment, Claveria, Misamis Oriental, Philippines, 1987-1990. 1987 1988 1989 1990 Number of adoptors 14 13 29 6 Mean labor (hrs) per 100 m of hedgerow 14 14 10 8 Farmers' Evolving Hedgerow Species Choice Following what they learned from farmers in Cebu, initial adoptors in 1987 and 1988 planted hedgerows largely of Glyricidia sepium and Pennisetum purpureum. Farmers gradually shifted species in the hedgerows. Those adopting in 1989 planted mulberry and Flemingia congesta in addition to G. sepium; and planted Setaria spp and Panicum maximum (Guinea grass) in addition to P. purpureum. Farmers also started to develop contour hedgerows of purely natural vegetation in 1989. In 1990 and 1991, new adoptors planted mainly Setaria or Guinea grass, or developed hedgerows of weeds. Other less used species included Chloris sp (signal grass), Stylosanthes spp, Helianthus annuus (wild sunflower), "ginseng" (an unidentified medicinal plant), cassava, taro, coffee, and fruit trees. Table 2 shows the species planted (or utilized in the case of natural weed contour hedgerows) in the hedgerows by year of establishment (1987-1992). Entries per column 380 total more than the number of fields as more than one species were often planted or utilized. By the end of 1991 and for 60 fields belonging to 55 hedgerow adoptors: 47% included G.sepium, 38% included P.purpureum, 32% included Setariaspp, 20% included P. maximum, 17% included mulberry, and 18% were of weeds alone. Adoption of G.sepium and of P.purpureum decreased; and adoption of Setaria, P. maximum, and weed strips increased over the study period. Mulberry adoption was high only in 1989. For the 15 of 60 fields established in 1990 and 1991, only one included trees (mulberry), 8 fields had planted grasses, and 7 had weeds alone. Farmers evaluations of their different hedgerow species are shown in Tables 3 and 4. Farmers had several sources of seed or planting material. They used cuttings of locally indigenous G.sepium, P.purpureum and H.annuus. Seed of F.congesta came from a stand of trees planted by a farmer several years prior to our work in the area. IRRI conducted trials on forage species and small quantities of seed of Chloris sp, Stylosanthes spp, and P. maximum were given away upon request. Others later obtained seeds of these forages from the first farmers planting them in hedgerows. Mulberry cuttings came from a 10 ha area where trees had been planted in the mid-1980s for a government silkworm project. Farmers had tried to eradicate the mulberry; and their lack of success meant that there was ample planting material when farmers were again induced to rear silkworms. Although farmers had free access to researchers' forage trials, their selection and testing reflected their own search for useful and suitable hedgerow species. The main species found in the contour hedgerows utilizing natural, in situ weeds (and in hedgerows combining both planted species and weeds) were Pennisetum polystachion (in 67% of 57 fields with hedgerows), Paspalum conjugatum and Borreia laevis (58% each), Ageratum conyzoides (44%), Chromolaena odorata (42%), Digitaria longiflora (37%), Mimosa invisa (32%), Rottboellia cochinchinensis (25%), Hyptis suaveolens (25%), and Imperata cylindrica (23%). Twenty-two other weeds occured as a main weed in farmers' hedgerows, but each of these in less than 10% of the hedgerows. Only 1 1 % of fields had hedgerows with no natural grasses or weeds (Table 5). A previous study (Elliot and Moody, 1986) elicited farmers' worst weeds in terms of effort needed to control each one. These were, in order, R. cochinchinensis, D. longiflora, P.polystachion, B. laevis, M. invisa, I. cylindrica, P.conjugatum and Brachiaria mutica. These same problem weeds—and especially creeping weeds such as D. longiflora and P.conjugatum-hdve provided a good alternative for farmers interested in soil erosion control and in reducing hedgerow establishment costs, but not interested in other supposed "multipurpose" uses (i.e., fodders, green manures, or cash crops such as coffee or fruit trees) of hedgerow species. Farmers knew that they needed to slash the weedy hedgerows before seeding in order to avoid creating additional weed problems in their crops. oo CO 2 ca "3 > c/5 i= ~ s T3 •— -J r- r- o r~ c rt — — 5 6 1ablish ines. Os || o o o o ts »• 8 3 /—s •Si • — CO Z =F S T© K Os II o o o — u 0s c V — ■8hedg iftta x: X01 ^ = 6 S <* *? 5 w ct 00 II oo - pi r- ^ I 4> - £ 9 rn •s:§ > oo '"■'oo || rN _lant 991 Q.-H <3 £ £ ||2 Sped in191 * % « «s ■2 B.-2 >• G .2 -5 g— :c 1 s£ H 1 s- g >o Tt O Tt -H VO C o ■8 5 60 c 60 o c 3 & 6 o a. Q. Ou- -u 60 13 SZ O CA a -g 95 %II Si a> c II o - o ■I T3 a■8 o. s^i o 73 ■3 ss c 1> o u, a> 60S ■8 C JS 3 O Z c 382 Table 3. Farmer evaluations of napier grass (Pennisetum purpureum) and Gliricidia sepium planted in hedgerows, in Claveria, Philippines in 1990. Number of farmers Pennisetum purpureum Positive evaluations Controls soil erosion 15 Provides animal feed 11 Grows easily 4 Competition not a problem if maintained 1 Negative evaluations Crops near hedgerows were stunted and yellowed 11 Competitive and too vigorous 6 Roots spread into alleys and make plowing difficult 3 Deteriorates with constant pruning 2 Shelters rats 1 Total 54 Gliricidia sepium Positive evaluations Source of organic fertilizer 16 Improves the soil and gives a higher yield 1 Provides feed for animals 1 Reduces soil erosion 1 Negative evaluations Caused shading if not maintained 1 Difficult to plow when roots spread to the alley 1 Total 21 383 Table 4. Farmer evaluations of other contour hedgerow species in Claveria, Philippines. Number of farmers Mulberry prevents soil erosion and provides income 13 Will extend mulberry hedgerows to other parcels 2 Flemingia congesta does not compete with the alley crop and is easy to maintain 1 Grasses control soil erosion and provide fodder 5 Grasses hold the soil better than trees 2 S.guyanensis is good, but competes with crops 1 Andropogon sp is good, but spreads into the alley 1 Roots of Desmanthus virgatus do not hold the soil 1 Pineapple holds the soil and provides cash and food 2 Taro holds the soil and provides cash and food 1 Sunflower holds the soil, provides green manure, but can be a weed problem 1 Total 30 Adoptors' Evaluations of the Technology Ninty-six percent of the 55 adoptors interviewed in 1992 viewed hedgerows as a way to control soil erosion (Table 6). Almost half (45%) mentioned production of fodder-especially in the dry season-for their draft animals as a desirable function of the hedgerows. Fifteen percent reported that biomass from G.sepium could be used as a green manure for the alley crop, although only three cases during one rice growing season (1991) of biomass use for green manure were observed. The same percentage of farmers thought that inorganic fertilizers could be applied to their alley crops without being washed downslope by rains. A few farmers reported that natural grass strips were easy to maintain (and survived drought periods better than introduced grasses); while others thought that hedgerow pruning and deep plowing in the alley solved the problem of crop-hedgerow competition. On the other hand, 35% of the farmers reported that hedgerows—especially those with grasses such as P.purpureum-compeled with the crop grown in the alley. Our observations made it clear that rice and, to a lesser degree, maize were very affected by such competition for moisture and soil nutrients (and light in the case of trees in the hedgerows). Competition between hedgerow and crop led to farmer testing and selection of species (including weeds) other than the orignal P.purpureum. 384 Almost a third (31 %) of the farmers found it necessary to fallow land on which they had established contour hedgerows due to combinations of soil nutrient depletion and competing demands and higher opportunities from their flatter lands and from off-and non-farm income sources. Continued fallowing of adopters ' fields was noteworthy because researchers had regarded contour hedgerows as a possible way to sustain permanent cropping. Some landowners fallowed parcels on which they had established contour hedgerows, and then selected and became tenant farmers on other parcels-in a unique form of shifting cultivation. Fallowed fields have traditionally been an open-access grazing resource in Claveria. Neighbors' staked or freely grazing animals were mentioned as a problem by 18% of the hedgerow adopters. We observed that grazing was especially a problem in cases in which forage species preferred by animals were planted in the hedgerow. Destruction of hedgerows by neighbors' cattle is a "social" problem in that the solution would require community or group cooperation and action. About a fifth of the farmers planted mulberry trees in their hedgerows after a silkworm project was started in the area. By 1992, however, the project was no longer in business and 16% of the adopters thought that mulberry was useless without the project. Non-Adoptors Farmers who did not adopt hedgerows after attending farmer-to-farmer training sessions gave their reasons. These were lack of labor or draft animals or competing demands for labor on non-sloping areas for farmers with large proportions of such lands or for off-farm and non-farm activities (Table 7). Although a few non-adoptors mentioned that their tenant status barred adoption, about 16% of adoptors were share tenants. In light of non-adoptors' responses, adopters' and non-adoptors' land sizes and slopes were measured. Differences between total land holdings and between percent slopes on sloping lands did not differ significantly between the groups, but a significantly higher proportion of adopters' lands was sloping (> 7% slope) compared to that of non- adoptors' (Table 8). Crop Performance and Soil Erosion Yields of maize and rice on farmers' fields with or without contour hedgerows were similar in the wet season of 1991, a year of severe drought (Table 9). The lack of response to hedgerow adoption was not discouraging because hedgerow species are susbtiantally more competitive for moisture than rice or maize in drought periods. Data from researcher designed and managed trials in Claveria also showed that cereal yields were similar with and without hedgerows when leaf prunings from the hedgerows were 385 not applied to the crop, although there was a substantial maize yield increase following incorporation of Cassia spectabilis biomass (Mercado et al., 1991). Soil erosion was monitored by researchers in controlled experiments. In normal rainfall years, approximately 200 t/ha were lost on open slopes and 20 t/ha from fields with contour hedgerows. In the 1991 drought year less soil was lost, but some 40 t/ha were still lost on open slopes and 4 t/ha on fields with contour hedgerows (Garrity et al., 1993). Table 5. Main "weeds" in 57 contour hedgerows", Claveria, 1992. Weed species % of Hedgerows Worst weeds2 No "weeds" used in hedgerows Pennisetum polystachionv Paspalum conjugatum Borreria laevis Ageratum conyzoides Chromolaena odorata Digitaria longiflora Mimosa invisa Rottboellia cochinchinensis Hyptis suaveolens Calopogonium mucunoides Imperata cylindrica Bidens pilosa Digitaria setigera Mimosa pudica Sida rhombifolia Elephantopus tomentosus {Brachiaria mutica) 11 67 58 58 44 42 37 32 25 25 23 23 21 16 14 12 12 (2) 3 7 4 (7) " Three fallowed or abandoned contour hedgerow fields not included. 21 Eight most difficult to control weeds in farmers' order. 31 "Main" weeds were among the five most frequently occurring in the given % of hedgerows; 22 other species were each one of the five highest occurring, but each in less than 10% of the hedgerows. 386 Table 6. Farmers" evaluations of contour hedgerows established in 1987-1991, in Claveria, Misamis Oriental, Philippines. Evaluation % of farmers Postitive27 Hedgerows control soil erosion 96 Hedgerows provide fodder 45 G.sepium provides green manure 15 Inorganic fertilizer can be applied and not lost downslope 15 Natural grasses are easy to maintain 9 Negative37 Hedgerows (especially grasses) compete with crop 35 Fallowed due to poor soil, flatter lands, off/non-farm work 31 Neighbors' animals graze and destroy hedgrows in dry season 18 Mulberry serves no purpose without silkworm project 16 " n = 55 27 Other positive evaluations: Setaria is good for erosion control and fodder; hedges are not crop-competitive if pruned and plowing in the alley is deep; and G.sepium can be used for fence stakes ?/ Other negative evaluations: crop stunted in upper rows due to soil scouring; pruning hedgerows required too much labor; contours on uneven slopes mean uneven alley widths, which are difficult to plow Table 7. Trained farmers' reasons for non-adoption of contour hedgerows (140 observations). No. of observations Work demands on non-sloping or lowland parcels 40 High labor for contour hedgerow establishment 27 Off-farm and non-farm work opportunities 18 Lack of draft animal 16 Lack of capital for labor and inputs 12 Left the area 10 Does not own the land 7 Miscellaneous 10 Total 140 387 Table 8. Areas and slopes of lands of adoptors and non-adoptors of contour hedgerows in Claveria, Philippines. Adoptor Non-adoptor Total land area (ha) area of flat land (ha) area of sloping land (ha) % slope of sloping land 1.70 1.68 0.43 0.79 1.27 0.89 21 25 Table 9. Farmers' yields of rice and maize (t/ha) of fields with and without contour hedgerows in Claveria, Philippines, wet season 1991 (number of cases in parenthesis). With hedgerows Without hedgerows Rice 0.68 (6) 0.77 (16) Maize 1.22 (12) 1.23 (10) CONCLUSIONS Findings suggest that a permanent, sustainable contour hedgerow system would have several features. 1. Sufficient soil tillage and rainfall is needed to cause both soil erosion as a problem and to enable natural terracing to take place as a solution. 2. Farmers should have mostly sloping land and a lack of off- and non-farm labor opportunities if they are to invest in their land. 3. The land frontier needs to be closed if farmers are not to shift more profitably to other parcels. 4. Hedgerow species are needed that survive the dry season, do not overly compete with the crop in the alley, and are fairly easy to maintain. 5. Farmers should be able to apply inorganic fertilizers to the alley crops or should have sufficient labor to apply the biomass from perennial legumes planted in the hedgerows to maintain soil fertility. 388 6. A social system is needed in which community members cooperate in the control of grazing animals in a way that animals are not allowed to graze and destroy hedgerows. Perhaps more importantly, both researchers and farmers contributed to developing appropriate and adoptable technologies. Researchers facilitated initial farmer-to-farmer technology transfer and monitored farmers' progress in solving problems. Farmers modified the basic technologies-use of the A-frame and vegetative contour strips—in order to save labor and reduce crop-hedgerow competition. Adoptors greatly reduced soil erosion on their sloping lands, but now need to solve the problem of soil nutrient depletion. REFERENCES Elliot, P. and K. Moody. 1986. Weed control studies in upland rice-based cropping systems. Unpublished paper presented at the Claveria Cropping Systems Review, held in Cagayan de Oro City, Philippines. March 6-7, 1986. Fujisaka, S. 1989a. The need to incorporate farmer perspectives: lessons from a comparison of selected upland projects and policies. Agroforestry Systems 9:141-153. Fujisaka, S. 1989b. A method for farmer-participatory research and technology transfer: upland soil conservation in the Philippines. Experimental Agriculture 25:423-433. Fujisaka, S. 1993. A case of farmer adaptation and adoption of contour hedgerows for soil conservation. Experimental Agriculture 29:97-105. Fujisaka, S., E. Jayson and A. Dapusala. 1994. Trees, grasses, and weeds: species choices in farmer-developed contour hedgerows. Agroforestry Systems 25:13-22. Garrity, D.P., D.M. Kummer and E.S. Guiang. 1993. The upland ecosystem in the Philippines: alternatives for sustainable farming and forestry. In: Agricultural Sustainability and the Environment in the Humid Tropics. National Academy of Sciences, Washington DC, USA. Getahun, A. and A. Njenga. 1990. Living stakes: Kenyan farmers introduce an agro-forestry technology. Forest, Trees and People Newsletter 9, 10: 38-39. Huxley, P. A. 1986. Rationalising research on hedgerow intercropping: and overview. International Council for Research on Agroforestry (ICRAF) Working Paper 40, Nairobi, Kenia. Kang, B.T. and G.F. Wilson. 1987. The development of alley cropping as a promising agroforestry technology. In: H.A. Steppler and P.K.R. Nair (Eds.). Agroforestry: A Decade of Development. ICRAF, Nairobi, Kenia. Mercado, A., A. Montecalvo, D.P. Garrity and I. Bashri. 1991 . Contour hedgerow systems using Cassia spectabilis and their effect on upland rice and maize crops on sloping acid upland soils. Agroecology Unit Paper. IRRI, Los Banos, Philippines. Phalman, C. 1990. Farmers. perception of the sustainability of upland farming systems in northern Thailand. The Sustainable Agriculture Newsletter 2:2:25-26. Raintree, J.B. and M.W. Hoskins. 1988. Appropriate R&D support for forestry extension. Paper for the FAO Consultation on the Organization of Forestry Extension, held in Bangkok, Thailand March 7-11, 1988. Roucheleau, D.E. (n.d.) The user perspective and the agroforestry research and action agenda. ICRAF, Nairobi, Kenia. Scherr, S, (Ed.). 1991. On Farm Agroforestry Research. A special issue of Agroforestry Systems 15:2, 3:91-308. Young, A. 1986. The potential of agroforestry for soil conservation. Part I. Erosion control. Working Paper 42. ICRAF, Nairobi, Kenia. Young, A. 1987. The potential of agroforestry for soil conservation. Part II. Maintenance of fertility. Working Paper 43. ICRAF, Nairobi, Kenia. 389 FARMER PARTICIPATORY RESEARCH IN CASSAVA SOIL MANAGEMENT AND VARIETAL DISSEMINATION IN CHINA Zhang Weite, Lin Xiong, Li Kaimian and Huang Jie1 ABSTRACT This paper mainly describes the objectives of the FPR project, the results of the RRA that was conducted in Hainan and the demonstration plots at CATAS, as well as the demonstration and FPR trials at the pilot sites (especially in Kongba village of Baisha county). The major existing problems and suggestions to improve the FPR trials in the future are also presented. INTRODUCTION During the 4th Asian Cassava Workshop in India in 1993, a new project on "Improving the Sustainability of Cassava Production in Asia" was first discussed and the possibility explored of having a pilot site in Hainan, not only because of serious soil fertility and erosion problems in cassava producing areas in the mountains of central Hainan, but also because of the active collaboration between cassava researchers at CATAS with agricultural officers and farmers in the area. This project proposed to use farmer participatory research (FPR) methodologies to improve the development and adoption by farmers of more sustainable management practices. Objectives of the FPR Project The objectives of this project are: a) to reconcile the short-term needs of farmers to increase crop yields and income with the long-term objective of preserving the soil's productivity, i.e., to provide benefits to both farmers and society; and b) the essence of this approach is that farmers own the process: they develop the most suitable practices for their own conditions by testing a range of options on their own fields. Workshop on FPR Methodologies The Nippon Foundation project organized an FPR training workshop in Rayong, Thailand, in July of 1994, to explain about the importance of the FPR approach, and to develop a work plan of activities for the project in each participating country. Also included were discussions on Rapid Rural Appraisal (RRA) methods as well as practical training in interviewing farmers. During the workshop, the Chinese participants made a work plan for the FPR project in Hainan, as shown in Figure 1. FPR PROJECT IN HAINAN 1. Rapid Rural Appraisal (RRA) An RRA was conducted in Hainan from August 14 to 20, 1994. The major ' Chinese Academy of Tropical & Agricultural Sciences (CATAS), Danzhou, Hainan, China. 390 Establish demonstration plots CATAS, March, 1994 Conduct an RRA Select two pilot sites Conduct formal survey Farmers visit demonstration plots CATAS, Jan, 1995 Establish plots in pilot sites Conduct FPR trials in pilot sites Collect data of FPR trials Monitore and evaluate FPR trials Figure 1. Design ofFPR project in Hainan. 391 objectives of this RRA were: a) To gain a better understanding of cassava production and utilization in Hainan and to analyze the major constraints and opportunities. b) To select two pilot sites for the FPR project. RRA Methodololy Before conducting the RRA, we discussed the methodology to be used in the group and individual interviews with farmers and the type of information to be collected. The selection of locations to be visited was based on: a) the importance of cassava in the area; and b) the representativeness of the production and/or processing system. As such, based on secondary information, a pre-stratification was attempted as follows: 1) lowland vs. highland areas; 2) near vs. far away from processing factories; 3) collective/state farm vs. "private" farm production. In most districts or counties we first contacted the local government officials, mainly from the Agricultural Bureau or the Agricultural Committee. They usually provided us with secondary data about climate, soils, landscape etc, and the production of the major crops. We also discussed the problems of cassava production, processing and marketing as well as future opportunities. After talking with the government officials, we were usually taken to a cassava growing village by local extension agents. In each village, we separated our team into two groups. One group would usually discuss socio-economic issues and processing with one group of farmers, while the second group would discuss varieties and cultural practices with another group of farmers; usually, about ten to fifteen farmers were interviewed in each village. Most discussions were held in a farmer's house or in the courtyard in front of the house. Each session lasted about one and a half to two hours. After the discussion we were often taken to a farmer's field, where we could see different cropping systems and evaluate the crops. Sometimes we were taken to steep mountains slopes where cassava was planted, providing an opportunity to ask farmers questions about erosion control and maintenance of soil fertility. Results of RRA In recent years, the cassava growing area in Hainan has been about 24,000 ha with a total annual production of 320,000 ton of fresh roots; this corresponds to about 6% of total cassava production in China. In Hainan, cassava production is concentrated in the foothills of the central mountain range, with greatest production in Baisha and Qiongzhong counties (Figure 2), In the past decade, cassava production area and yield have fluctuated for several reasons. The decline in cassava area and production in Hainan is mainly due to the aggressive promotion of other more valuable crops, like rubber and fruit trees, as well as the absence of a significant expansion of cassava processing facilities. While in Guangdong and Guangxi cassava further-processing into 392 glucose, MSG, maltose, alcohol and sorbitol is being promoted, in Hainan the processing is still confined mainly to production of raw starch, using rather antiquated processing facilities. However, farmers reported that during the past couple of years the cassava area may have expanded again due to increasing prices as well as problems in the transport of competing sugarcane. In the counties visited, cassava was generally the third or fourth most important crop after rubber, rice and/or sugarcane (Table 1). Hainan has a tropical and sub-tropical climate and is influenced by tropical monsoons and typhoons, but with sunshine all year round. Its annual mean rainfall is 1500-2000 mm, 75-90% of which is concentrated during the rainy season of May to October. Sometimes the precipitation per hour reaches 80-100 mm, which is a major factor causing high levels of soil erosion. The annual mean temperature is 23-25°C. In Hainan, being located between 18 and 20°N, cassava can be planted year-round without danger of frost but with occasional short-term cool temperatures of 4-7°C during the winter months, especially in the northwestern region and the northern part of the central region; however, this does not affect cassava yields. Cassava is planted in Jan-March in most of Hainan, but planting is delayed to March-April in the western part of the island because of drought. Meanwhile, Hainan is the only place in China where cassava pollination and hybridization can be successful. In Hainan the cropping area per family is similar or slightly larger than in Guangdong and Guangxi, except in Kongba village of Qifeng town where farmers cultivate rather large areas by planting crops also on illegal land with steep slopes. Moreover, in that area most farmers also manage 0.5-1 ha of rubber trees. Thus, time for attending the crops is limited to 4-5 hours a day, usually in the afternoon. This explains the prevalence of cassava monocropping, the less intensive management of the crop and the use of herbicides in some locations. Table 2 shows that cassava is grown mainly on gentle slopes with some production on very steep slopes (up to 100%), especially in Tongzha city. Cassava in Hainan is seldom grown on terraced fields except on State Farms; in contrast, rubber is often planted on narrow terraces in the mountains. Cassava is generally planted as a monocrop, although some intercropping with maize or peanut, or with maize plus sweet potato was found; in some places the crop is also rotated with sugarcane. In Baisha and Tongzha counties cassava is sometimes grown between young rubber or fruit trees. Where cassava is grown on steep hillsides, the land is usually prepared only by making individual planting holes of 10 x 20 cm with a hoe, after slashing and burning the weed or bush vegetation (Table 3). On more gentle slopes the land is prepared by buffalo or oxen, usually 1-2 plowings followed by 1-2 rakings. Ten to twenty cm long stakes are planted horizontally 6-10 cm deep in each planting hole, spaced at random or at 0.6 x 0.6 or up to 0.8 x 0.8 m. Plant populations are very high and farmers think that this may be useful for reducing soil erosion. Depending on the type of previous vegetation, the land clearing, soil preparation and cassava planting may take up to 15 393 man-days/mu (225/ha). On the Bayi state farm and in Kongba village, farmers spray with pre-emergent and/or contact herbicides after planting, followed by one hand weeding at 2-4 MAP. In other areas weeds are controlled with a hoe or knife, while in Longtang town fields are weeded using a buffalo and plow. In the mountainous areas of Hainan, fertilizer and/or FYM are seldom applied to cassava, or are applied only after the second crop. Farmers often reported that yields increase from the first to the second year of cropping due to the slow decomposition of the previous fallow vegetation, but yields decrease again in the third and subsequent years due to nutrient extraction and erosion. Thus, 2-3 crops of cassava are usually rotated with sugarcane or the land is returned to fallow for 1-2 years. If cassava cultivation continues beyond two years, farmers usually apply some urea, SSP or 15-15-15 fertilizers, either at planting or at 2-3 MAP. In most areas of Hainan, cassava is harvested in Nov to Dec, but in some cases this extends to Feb or even March. In the highland area of Tongzha city some farmers leave cassava for a second cycle, harvesting only after about 18-20 months due to labor shortage or because the crop did not grow well. If a starch factory is nearby, such as in Maoyang and Maoqui towns of Tongzha and in Hongdao town of Qiongzhong county, cassava is mainly sold as fresh roots (Table 4). But in the absence of a nearby factory, cassava is chipped and dried; in the Bayi state farm and in Qifeng most of this is sold to traders, while in Longtang town the chips are ground to a powder in a local mill and this is fed mainly to the farmer's own pigs or is sold at the local market. Some of the chips are taken to the animal feed mill in Wenchang county. In Hainan, on the average, 80- 90% of cassava is sold off-farm. In many areas, especially near urban centers or main roads, on-farm pig feeding is highly lucrative. It was estimated that cassava-based pig- feed (mixed on-farm) is 15-25% cheaper than commercial pig feed. With the current strong economic development, especially around the larger industrialized cities, demand for pork is very strong. In China, pork is the animal protein most consumed. Economic data show that labor requirements for cassava production in Hainan are higher than on the mainland. It ranges between 12-20 man-days/mu (180-300/ha). To a large extent this can be explained by the fact that cassava fields are on steeper slopes, farther away from the village and because of climatic conditions chip drying takes longer. Cassava income (labor not accounted for) ranges from Y250-300/mu. Although sugarcane is often used as a crop alternative, and has a higher income (Y300-400/mu), in several areas there were demand problems. Also, cane, once harvested, needs to be factory processed at once. Cassava, on the other hand can be chipped and stored, and has more utilization alternatives. As such, these are typical criteria entering the farmer's decision-making process regarding crop choices. In Hainan most farmers had mixed feelings about cassava. On the one hand they like the crop because it is well-adapted to the local soil and climatic conditions, it is easy to grow with a minimum of inputs and it gives a quick return on labor; the harvest and I 1 1 •5 3 * 1 § i 1 ft, 3 I 395 Table 1. Secondary data of Hainan province (1994). Bahyi State F. Bayi State F. Baisha Tongzha Qiongzhong Dingan Tongshan Production County city county county branch Team #2 Longtang town LatitudeCN) 19° 30' 19° 30. 19° 20. 18° 45. 19° 5. 19° 25. Altitude(masl) < 150 <150 150-300 300-500 250-500 100-200 Soils Hapludult Hapludult yellow-red rocky rocky dark Landscape Hapludult Paleustult Paleustult Hapludult -general hilly hilly hilly mountain hilly-mount. 60% upland 40% irrig. -cassava on flat and on flat and on flat and on steep on gentle onflat and Climate steep slopes steep slopes gentle slopes slopes slopes gentle slopes Average temperature (°Q 23.0-23.5 23.0-23.5 23.5-24.0 22.5 22.0-22.5 22.0-23.0 Absolute minimum temp. Mean minimum temperature 5-7 5-7 4-5 4-5 5-7 Absolute maximum temp. Mean maximum temperature Rainfall (mm) 1,600-1,800 1,600-1,800 1,600-1,800 > 2,400 2,000-2,400 No. of frost-free days 365 365 365 365 365 365 Typhoon incidence medium medium low low medium high Main crops (mu)" rice - irrigated 1,416 30,500 72,773 15,600 - upland 4,032 3,087 cassava 2,500 300 36,778(#3) 15,748 42,012 4,100 maize 9,382 904 peanut 2,866 6,738 4,600 sugarcane 5,136 100 (#2) 8,182 14,000 fruit trees 410 10,300 4,200 vegetables 5,023 1 1 ,303 rubber 1,000 (#1) 4,800 sweet potato 239 62,937 8,050 4,500 bamboo 1,000 pineapple 14,167 Cassava yield (kg/mu) - fresh 1,419 988 - dry 500-750 Total farm size (mu/lf)31 - upland 5.0 - irrigated 0.5 Average income (Y/person) 3,200 " 1 ha = 15 mu 21 If = Labor force * 1 US$ = 8.4Y 3 w S'S'l.s M On rn I S .a fr a £ ! I I .- w> I s 1 = I c 1 O O ea 8 ffl ■s 00 6f> = .5 g>€ .2 — E a, « l ^ ■= 2 c « 2 c SI I -a I s 1 I 00 w (0 c ii £ 5 3 ! 8> 8 ! » 1 8 <§ 8 -8 . 2 1> el S B o O « as .. 3 h at (a •S g w « en ?o"3 ° ■S ex * 8 to U a u «5 c 22 o 1 + o 8 3 8 O u B S + C ^ a « "> — =S o a 4 t no d> oo -4 ri ri i o .A *r "5 7— in N *fi oe o tr « - n4 vi g 8 6 o n 6 ? cn —i •-< •> 00 (N O ~ *5 <§ -2 J 6 5 § St" .« _3 *° •" 2 p i = = ail all U V "T "? °f 8 as i ,8 iqv 8 « - 5 s 5 c u u 1 1 : 1 S" "S 3 | 1 •s c II o, 03 s II o 00 "E c I a c s II £ o u - z gp 3 4> •C .2 > > o .5 =a x ■> ■a a u M 5? M X u. > — 2>3 B u 72 » g « .2 2S? s s 64 « o — 15 H E ■5 « * § Nw •- ■** 3 s E U 03 Is o ap g. 5 a- CI ON O. ~ o o !c t- — r~ cj o O — ^4 °° « - n - • g, * * •ft n m M O - o o 5 C 3 V) — u U 3 o ■3 ■a « oo •*; OA C 2 o ■3 3 u 00 £2 w o § = III. > '3 £ oo ON en S 5 2 •> S .a a. I £ or* CO t3 «j .— a *> I V) on o o 00 e ■2 r> s y ato CO u. ~ d o v> 3 ri o - c to P a. U. CO o "> i/^ o 6 6 O O M ■t 3-H.eO. « &*> u. i r. .2. c z horizontal 10 10-15 hoe/knife 3 Nov-Jan C+Mwith SPinbetween in or]moocNr^~-' \o m no © * I> Ji s/1 U5 *£e 3 i Q g o o 0 0 o 611 o 0 o o o c y5 s C c c c c c c > c c c c n Styl.guia Brach.br c 60 TD .2 3 -c 9 o. a. a 9. g c e 5 oo lit § 3 Oh « C;3.2.S.2.2.2.2.2.2 o. 3 'S.'S.'E.'S.'o.'H.'S.'E. o o o o 00 00 00 00 00 00 00 00 00 00 o 00 00 00 00 00 oo o o o o o o o o o o o "x o o o o o o o X X X X X X X X 00 X X X X 00 X 00 X © o o o o o o o o o 00 00 o o o o c~"zZZZZZZZZZZZZZZ 5! -^^l(nTtvl*O^ooc>O~*NnTtV'l'<'lv s> B I —* 00 u a u <-•g ° 9. 60 00 •* 55 o § "8 u. a JS to 4: 9 60 M r- ^ * —* C E of o a.■c s Si 1- 3 O E 6 1 1 IS E a•c s »> s Z o oo a " "S s C o a1 8 ■*1 °" 2? "e z § 9, o * 1-8 8 s «S 3 £ 1 1 B o (S o r- 00 o 00 r» o ZZ O r) oo 00 o ct* O* r^"ooododr-r-ooo*cK©odr-o^or- oooo*Onr»oooooor^r^ooo*o*ooor-o*©r- a. 6 . <2 H 2 ' "" 1 s mr-r*>ooo oo o Tt r- rs r^ rs O ■* (N V> O* rs ~* (si « r^ r^ tN __ ,^-r- — ^■r^fifti *© 6 c £ c a H c 3 6 o>• * 3 .s £ H H H •> 2 n d 2 o <& V II u J3 .» OS aa 2 f•a 5 M) > 2 ft © .—i T3 '+ C X I" ■o c o U Z 2 8 ^ U. tu u 3c u e ■0 n oo m O -^ O - 00 00 -h 00 f> 00 00 00 os ■n o n m 0* CO o 00 n i ■ oo C— CO COm ■* m m m m 8 moo mm m p00 o 00 -O Os CI i m o n CO m en n CO CO m m O O o o m n c| oo m h o o n mCnmmm. OZo.uia.ZZZu. -Nn*mio'Kio'oi 0. E 9 t^ O C I a. o u. r- m CI r- sO r- n ■—i CO 00 so « -o Oi m ■* CI m o CO m — cn •—i m oo m n m i© r— m m O n m -h CI -- CI CI —i O O O O O O p o o r~ p m n Oi r^ so io n CO Tt CI CO Tt CO so oo m o — vo oo r~ m oo o — TI * Tt Tt mi m Ml c o Ml CI Ml M) Ml M) <* ■* ■* £ P Tt g o p p p 2 3 X 2 Kiuig .5 ~ c r c •C 08 3 ed CCJ eej N (- £ H H f- < OQ O Q UJ U. O o« u O II II II -C a. 8. 3 Ml in M O € s 1 1 ■a c■S 1 Ml .5 =5 O Z a. 00 00 M M r~ m m m m — II II Z BU a a on « U. 0 5 Ml 8 * £ 411 Table 10. Combined results of 15 FPR cassava variety trials conducted in Kongba village, Qifeng town of Baisha county in Hainan, China in 1995/1996. Varieties/Clones Farmer's name Improved variety Check (SC205) SC8013 Fu Yong Chuan 40.6 29.7 Tan Yin Chai 36.7 37.5 Tan Ya Zhui 29.7 37.7 Tan Huan Chong 35.7 37.7 Tan Ya-e 36.1 30.5 Zhou Shao Xiong" 35.9 42.9 Average 35.8 36.0 SC8002 Ma Guo Lei 28.4 32.7 Tan Jing Zhou 28.5 38.4 Tan Ya Zhui 24.9 37.7 Zhou Shao Xiong" 50.7 42.9 Average 33.1 37.9 ZM9038 Tan Jing Zhou 26.7 38.4 Tan Ya Lao 42.8 31.2 Average 34.7 34.8 ZM9036 Liu Ya Chun 44.4 33.3 ZM9057 Fu Jin Yu 28.5 32.6 Tang Ya Han 46.0 24.1 Zhou Shao Xiong" 53.1 42.9 Average 42.5 33.2 ZM9066 Tan Yin Chai 32.1 37.5 ZM8639 Tan Ya Qing 34.9 28.8 Zhou Shao Xiong" 34.4 42.9 Average 34.6 35.8 ZM8641 Tan Jia Chai 24.9 28.8 Zhou Shao Xiong" 42.1 42.9 Average 33.5 35.8 OMR33-10-4 Tan Wen Fu 36.3 23.4 Zhou Shao Xiong" 53.1 42.9 Average 44.7 33.1 SCI 24 Tan Yin Chai 50.4 27.7 Zhou Shao Xiong" 28.1 42.9 Average 39.2 35.3 ZM9045 Zhou Shao Xiong" 34.3 42.9 ZM9076 Zhou Shao Xiong" 48.8 42.9 ZM8803 Zhou Shao Xiong" 22.2 42.9 SM 1592-3 Zhou Shao Xiong" 32.0 42.9 ZM8316 Zhou Shao Xiong" 26.5 42.9 "These data maybe overestimated due to unclear spacing. 412 FARMER PARTICIPATORY RESEARCH IN SOIL MANAGEMENT AND VARIETAL SELECTION IN THAILAND V. Vongkasem1, K. Klakhaeng1, S. Hemvijit1, A. Tongglum2, S. Katong2 and D. Suparhan2 ABSTRACT A pilot project on the use of Farmer Participatory Research (FPR) methodologies with the objective of enhancing farmer adoption of practices that minimize soil erosion in cassava-based cropping systems, was conducted jointly by DOA and DOAE with technical and financial support from CIAT in two sites in Nakhon Ratchasima and Sra Kaew provinces. The activities involved a preliminary survey using RRA methodologies, the setting out of demonstration plots, as well as farmers meetings and farmers field trips to observe the demonstration plots. The activities also included conducting various types of FPR trials with farmers on their own land and organizing a farmer's field day to harvest these trials, to discuss the results, and to select the best treatments for next-year's trials. The results revealed that the highest yield of cassava in Soeng Saang district of Nakhorn Ratchasima province was obtained from planting contour barriers of vetiver grass within cassava plots; this practice also resulted in the lowest level of soil erosion. On the contrary, in Wang Nam Yen district of Sra Kaew province, the highest cassava yield was obtained in plots with up- and-down ridging, while the lowest level of soil erosion was observed with contour ridging. However, from the discussion between participating farmers and FPR team members it was concluded that the planting of contour barriers of vetiver grass was the most effective method to reduce erosion in the long-term, even though this treatment neither had produced the highest yield of cassava nor the highest income. Within the same pilot project, FPR trials on cassava varieties and on rates of fertilizer application were conducted in order to motivate farmers to participate m the project. In Wang Nam Yen district, Rayong 90 had the highest yield and ranked second in starch content, while Rayong S had the second highest yield but the highest starch content. In Soeng Saang district, Rayong 90 ranked first in yield and starch content, while Kasetsart 50 ranked second in both yield and starch content. With respect to the fertilizer trials, it was found that in Soeng Saang district the application of 156 kg/ha of 15-15-15 resulted in both the highest yield of cassava and highest net income. INTRODUCTION Cassava is an important cash crop in Thailand. Due to its favorable characteristics, such as relative ease of cultivation, drought tolerance and adaptation to poor soils, casava has become very popular, especially for poor farmers. During the past five years (1990-1994) the total planted area of cassava in Thailand ranged from 1 .40 to 1.52 million hectares. The annual production of fresh roots was 19.0-20.7 million tons, while the value of exports of dry cassava products was more than 18 billion baht (US $ 2 Dept. of Agricultural Extension (DOAE), Rice and Field Crops Promotion Division, Chatuchak, Bangkok, Thailand. Dept. of Agriculture (DOA), Rayong Field Crops Research Center, Huay Pong, Rayong, Thailand. 413 720 million) per year (Office of Agricultural Economics, 1995). The major cassava growing areas have spread mainly to the poor-soil and drought-prone areas in the northeastern and eastern parts of Thailand. Despite the poor soil and droughty conditions in these areas, cassava grows fairly well. However, when cassava is grown on slopy land soil erosion may be serious even in areas with gentle slopes of less than 10%. Moreover, in an intensive four-year study it was shown that soil erosion caused by cassava cultivation was about twice as high as that caused by cultivation of mungbean, and three times as high as that caused by maize, sorghum or groundnut (Puttacharoen, 1992). Due to the wide spacing used in planting cassava and its rather slow early growth during the first three months after planting, a lot of the soil surface remains exposed to the direct impact of fallen rain, causing severe soil erosion. Therefore, the Dept. of Agriculture (DOA), Kasetsart University and the Centro Internacional de Agricultura Tropical (CIAT) have conducted collaborative research into practical ways to decrease erosion in cassava production areas. The research showed that there were many ways to manage or improve cassava cropping systems that will enhance nutrient conservation and reduce erosion. Each management practice has its advantages and disadvantages: for example, some practices that control erosion require more money or more management, while the yield or income does not necessarily increase. The researchers did not know whether farmers would adopt these practices or not. Therefore, CIAT initiated collaboration with the Dept. of Agricultural Extension (DOAE) and DOA to establish a cassava-based cropping system management project and work with cassava farmers in the provinces, using a farmer participatory research approach. The objective of this project is to enhance farmers' awareness of the importance of soil conservation, to demonstrate a wide range of soil erosion control practices, to let farmers select the most appropriate ones and test these methods on their own fields, so they will develop the most useful practices for their own conditions. This in turn is likely to enhance adoption and the continued use of these practices even after the project is finished. FARMER PARTICIPATORY RESEARCH (FPR) ON CASSAVA SOIL MANAGEMENT Objective To enhance the development and adoption by farmers of improved cassava cropping systems and cultural practices that will maintain soil productivity and reduce erosion while sustaining a reasonable farm income. Location of the study 1. Soeng Saang district of Nakhon Ratchasima province in the lower Northeast. 2. Wang Nam Yen district of Sra Kaew province in the eastern part of Thailand. 414 Responsible organizations 1 . Rayong Field Crops Research Center of the Field Crops Research Institute of the Dept. of Agriculture (DOA). 2. Field Crops Sub-Division, Rice and Field Crops Promotion Division, Dept. of Agricultural Extension (DOAE). 3. Centro Internacional de Agricultura Tropical (CIAT). 4. Thai Tapioca Development Institute (TTDI) Research and Training Center in Nakhon Ratchasima province. Plan of Implementation 1. To select appropriate areas and within those select the most suitable pilot sites. 2. To collect information on the agro- and socio-economic conditions of cassava farmers in the pilot sites. 3. To develop a plan of action together with research/extension staff and farmers in group meetings in the selected pilot sites. 4. To establish demonstration plots in Pluak Daeng, Rayong province (1st year), and at the TTDI Research and Training Center in Nakhon Ratchasima (2nd year). 5. To provide training for participating farmers and let farmers select the most useful treatments in the demonstration plots: 6. To let farmers select treatments and implement farmer participatory research trials on soil erosion control, fertilization and varieties on their own fields in the two pilot sites. 7. To organize a study tour for participating farmers to the TTDI Research and Training Center to see and discuss alternative management practices. 8. To jointly harvest the trials and present and discuss the results during a farmers' field day in order to select best-bet treatments for next year's trials. Procedure 1. Appropriate Area Exploration and Selection The criteria for the selection of appropriate areas are: 1 . cassava is an important crop in the area, both at present and in the future; and 2. cassava is planted on slopes with serious soil erosion problems. The first step in the process was area exploration. Three provinces, namely Rayong, Nakhon Ratchasima and Sra Kaew, were explored by conducting preliminary rapid rural appraisals (RRA). The most suitable pilot sites were selected by analyzing the RRA results. It was found that in Rayong province most farmers favored the planting of rubber and fruit trees as the main crop and cassava only as an intercrop. Thus, the cassava growing area there is likely to decrease. After this initial exploration, only two locations, i.e. Soeng Saang district in Nakhon Ratchasima province and Wang Nam Yen district of Sra Kaew province were selected. 415 2. Exploration ofAgro- and Socio-economic Conditions ofFarmer in the two Pilot Sites A Rapid Rural Appraisal (RRA) technique was used to select the pilot sites, to ensure that the selected locations are indeed suitable for this project, and to collect baseline information about the socio-economic conditions and the agricultural practices used at the onset of the project. The results of these RRAs are summarized in Tables 1-13. 3. Group Meetings with Farmers Farmers' meetings were held in the two selected pilot sites, i.e. Wang Sombuun village in Wang Nam Yen district and Noon Sombuun village of Soeng Saang district, to discuss the objectives, principles and procedures of the project with the farmers, local extension staff and village leaders. The farmers analyzed and decided for themselves whether they wanted to participate in the project. 4. Demonstration Plots at the Research Center Demonstration plots were established along contour lines by cassava researchers to show farmers a wide range of management practices to help reduce soil erosion in cassava fields. These demonstration plots were established in two locations, one in 1994/95 in Pluak Daeng, Rayong province, and the other in 1995/96 at the TTDI Research and Training Center in Huay Bong of Nakhon Ratchasima province. Unfortunately, some demonstration plots in Pluak Daeng were lost due to flooding. There were 24 treatments in the demonstration plots, such as the growing of peanut, mungbean, soybean, sesame, pumpkin and sweet corn as intercrops in cassava. Also included were contour barriers or hedgerows of vetiver grass, lemon grass, citronella grass, sugarcane, ruzie grass (Brachiaria ruziziensis), mulberry and king grass; as well as practices like the application of grass mulch, planting cassava at closer spacing, on contour ridges, and the use of chemical fertilizers and/or chicken manure or various green manuring practices. Soil lost due to erosion collected in plastic-covered channels dug along the lower side of each plot. These sediments were weighed twice during the cropping cycle and a sample was taken to the laboratory and dried to calculate the amount of dry soil lost in each treatment. The amount of sediments in the channel of each plot was a clear demonstration to the farmers about the quantities of soil lost by erosion due to each treatment. Results of these FPR demonstration plots are shown in Tables 14 and 15. It is clear that each treatment had a different effect on cassava yield and on erosion. These demonstrations plot in two locations provided farmers with examples or alternatives for them to select some methods that may be useful in their own fields and under their own conditions. 416 Table 1. Secondary data of Soeng Saang district of Nakorn Ratchasima, Baan Khai district of Rayong and Wang Nam Yen district of Sra Kaew provinces in Thailand in 1994. District Soeng Saang Baan Khai Wang Nam Yen Province Nakorn Ratchasima Rayong Sra Kaew Latitude (°N) 14° 20.N 12° 40.N 13° 40. N Altitude (masl) -100 -400 -300 Rainfall (mm/yr) 948 -1400 1404 Rainy season ( > 60mm) April-Oct. April-Oct. April-Nov. Soils loamy Paleustults loamy Paleustults clayey Haplustults dark brown orwhite sandy loams white sandy loams red or black clay loams red clay soils with many rocks Landscape rolling rolling hilly - cassava on long gentle slopes on long gentle on short steep slopes (upper part) (10-20%) slopes - sugarcane lower part of slope - rubber lower part of slope Main cropsfraiV - cassava 122,200 63,259 - rubber 190 • - pineapple - - sugarcane 11,900 57,090 - maize 10,000 116,370 - rice 39,234 38,341 - soybean - 83,215 - fruit trees 15,651 - - cotton - 13,238 - sunflower - 5,911 - mungbean - 10,290 - sorghum - 3,722 Yield (t/rai) - cassava 2.7 2.5-5.0 2.65 - sugarcane 10 8.25 - maize 0.5 0.70 - rice 0.5 0.41 - soybean - 0.26 - mungbean - 0.11 Farm size (rai/fam.) Total 25-150 20-140 - cassava 15-20 10-60 - sugarcane 30 36 - rubber 20 - - maize 10-15 10-80 - rice - 12 Land ownership presently STK next year SPK 4-0 12) presently STK " 1 ha = 6.25 rai in future SPK 4-01 2) maximum 100 rai per family 417 Table 2. Cropping systems, cassava varieties and agronomic practices in various subdistricts of Nakorn Ratchasima, Rayong and Sra Kaew provinces of Thailand in 1994. Subdistrict Noon Sombuun Baang But Wang Sombuun District Soeng Saang Baan Khai Wang Nam Yen Province Nakorn Ratchasima Rayong Sra Kaew Cropping Systems ° C monocrop on C monocrop on C monocrop on gentle slopes gentle slopes gentle slopes, C between rubber rotated with M-Mu Varieties R60, Rl, R3 Rl, R3, R60, R90 R60, R3, R90 Yield (t/rai)2' 2.5-4.0 2.5-5.0 2-4 Land preparation tractor + 3 disc (lx) tractor + 7 disc (2x) tractor + 3 disc (lx) no ridge or ridge-up ridge with tractor ridge with tractor Planting method with weeding contour or up-down mainly up-down slope stake size (cm) 15-20 15-20 15-20 stake position vertical vertical vertical spacing (cm) 70x50 to 120x100 90x60 5-6 C rows between rubber Fertillizer use none to cassava - manure (t/rai) some - - 15-15-15 (kg/rai) 12-50 at 2 MAP - to maize or soybean - 13-13-21 (kg/rai) - 50 at 1-2 MAP3' Weeding - first hoe or Gramoxone hand, 1-2 MAP3' Pre-emerg. herb. - second hoe or hand tractor Gramoxone hoe - third Gramoxone or knife Gramoxone Time of planting - cassava - first crop March-May (red soil) April April - second crop Oct-Nov (white soil) June - - sugarcane March-May March-May - - maize - - May-June Time of harvest - cassava - first crop Dec-Febr - Nov-Febr - second crop Sept-Dec - - - sugarcane Dec-Febr Dec-Febr - - maize - - Aug Method of harvest by hand or tool by hand or tool tractor + blade " C = cassava, M = maize, Mu = mungbean 2) 1 ha = 6.25 rai 3> after first weeding 418 Table 3. Cassava utilization and socio-economics in various villages of Nakorn Ratchasima,Rayong and Sra Kaew provinces of Thailand in 1994. Subdistrict Noon Sombuun Baang But Wang Sombuun District Soeng Saang Baan Khai Wang Nam Yen Province Nakorn Ratchasima Rayong Sra Kaew Cassava utilization (%) - chipping floor 10-50 in dry season 100 - starch factory 50-90 in wet season 0 Prices - fresh roots (B/kg) " 1.25-1.50 0.9 1.10-1.15 - dry chips (B/kg) 2.80 - - - 15-15-15 fert. (B/50 kg) 285 280 - - 13-13-21 fert. (B/50 kg) - 290 - - urea (B/50 kg) - - - - Grammoxone (B/5 liters) - 470 - - pre-emerg. herbicide (B/rai) - - 55 - labor (B/day) 70 80 - Costs (B/rai) -land preparation- plowing 100 110 100-120 - discing - 70 80-90 - ridging - 90 80-100 -planting 70-100 90-100 80-100 -fertilization 85-315 350 0-185 -weeding 400-500 400 2602l-760 -harvest3 250-400 300-600 440-600 -transport4 150-600 325-650 210-240 Total 1,055-2,015 1,645-2,370 1,150-2,075 Gross income (B/rai) 3,125-6,000 2,250-4,500 3,600-4,200 Net income (B/rai) 1,485-4,625 505,2,300 1,525-4,200 " Root prices Dec.94 Aug .94 Jan'94 27 own labor not included 3/ Harvest cost (B/ton) 100 120 150 47 Transport cost (B/ton) - to local chipping floor 50-100 100 70-80 - to starch factory 130-150 130 160 419 Table 4. Topography, soils, crops and cropping systems in various villages of Noon Sombuun subdistrict, Soeng Saang district, Nakorn Ratchasima province of Thailand in 1994. Village: #1 and 2 U #10 Topography gentle slopes gentle slopes rolling with some with some hills steep slopes Soil tvpes black soil white soil (low fert.) red soil (100%) red soil red soil (fertile) white soil (s.c.l.)" black soil (fertile) Land use 2) some laterite no stones some big stones - red soil C, Sc, M ,Pa, Ch, F C, M, Veg, Sp, F C, F - white soil C, Sc, Wm, Pu C - black soil Veg, Pu - lowlands nce - irrigated land Pa Pa, Ta Crop choices SC mainly by rich Many crops to spread Ta or Pa farmers income over year in irrig areas Pa in irrig. areas Rice in lowlands Too dry for M cotton .soybean Livestock Few have cattle.pigs Few have cattle, water buffaloes, pigs, chickens, ducks, fish somme have cattlt Off-farm activities sericulture none none Cassava vield(t/rai) - initially or now with fert. 4-5 4-5 - now with little/no fert. 2-3 2.5-3 Cassava utilization (%) - to local chipping floors mainly poor mainly poor 10-20 - to starch factory in Korat mainly rich mainly rich 80-90 1) s.c.l. = sandy clay loam 2) C = cassava, M = maize, Sc = sugarcane, F = fruit trees (banana, tamarind, mango, papaya), Ta = tamarind, Pa = papaya, Sp = sweet potato, Ch = chili peper, Pu = pumpkin, Wm = watermelon, Veg = vegetables 420 Table 5. Cassava cropping systems, varieties and agronomic practices in various villages of Noon Sombuun subdistrict, Soeng Saang district, Nakorn Ratchasima province of Thailand in 1994. Village: #1 and 2 H #10 Cropping systems C monoculture continuous cropping C monoculture sometimes rot. M C monoculture continuous cropping Varieties R60, Rl, R3, R5 Pumpuang R60, Rl, R3, R90 R5, KU-50, CMC Pumpuang R60 (80%), R3 R90, KU-50, R5 Red Rayong, Red KU Land preparation tractor +3disc (lx) no ridging tractor +3disc (lx) no ridging tractor +3disc (lx) no ridging Planting method -stake position -spacing (cm) vertical vertical vertical 70x50 (white soil) Fertilization (kp/rai) 120x100 (red soil) 120x100 (red soil) -15-15-15 12-25 25-50 (white soil) sometimes 25 sometimes-manure Weeding -first -second -third hoe hoe hoe hoe or Gramoxone hoe or hand tractor knife hoe Gramoxone Time of planting -first crop -second crop knife Time of harvest -first crop -second crop March Nov Febr-Apr (red soil) Oct-Nov (white soil) March-Apr Dec Aug Dec-March Sept-Dec Jan-Febr 421 Table 6. Principal cassava varieties and their characteristics in several villages of Noon Sombuun subdistrict, Soeng Saang district, Nakorn Ratchasima province of Thailand in 1994. Village: U 1 and 2 #4 #10 Main varieties Secondary varieties R60, Rl, R3 R90, R5 R60, Rl, R3 R90, R5 R60, R3, Rl R90, R5 Characteristics Rayong 1 Drought tolerant Low yield Low starch Rayong 60 High yield Low starch Good plant type Convenient- weeding High yield Low starch Susc. root rots Good in white and red soil High yield Low starch Drought tolerant Rayong 3 High yield (esp. 1st yr) High starch Short stature High starch High yield High starch Low germination after storage Difficult harvest Rayong 90 Only good yield in white soil High starch High yield High starch Difficult weeding Low germination after storage Rayong 5 High yield High starch Drought tolerant 422 Table 7. Main problems and farmer solutions in several villages of Noon Sombuun subdistrict, Soeng Saang district, Nakorn Ratchasima province of Thailand in December 1994. Problems Solutions Village #1 and 2 Village #4 1) Low price (except this year) 2) Lack of labor 3) Decreasing yields (20 % in second yr) 4) Root rots (7-14% yield loss) 1) Erosion (5-10% yield loss) 2) Root rot in white soil during last 2 years (esp.in R60) 3) Spider mites (esp.in red soil, R60) 4) Decreasing yields Need cassava harvester Increasing use of fertilizer Lemon grass barriers Fill gulleys back in Plant bamboo in higher areas Need to apply fertilizers after 4-5 years Village #10 1) Drought (25% yield loss) 2) Erosion (14% yield loss) 3) Spider mites (50% yield loss in some areas) 4) Lack of labor (esp.for harvest) Plant vetiver grass Plant on contour ridges (but break with heavy rain) Tamarind + ruzie grass in contour strips Use cassava harvester 423 Table 8. Cassava processing in Soeng Saang district and in Korat city of Nakorn Ratchasima province of Thailand in December 1994. Soeng Saang district Korat city Manufacturer Ownership Products Production capacity Tirapong Chipping Factory private cassava chips 2.56 ha drying floor Actual production (t/year) about 3,000 chips Production period 5 months (Oct-Febr) Raw mat. needs (t fresh roots/year) Price fresh roots (B/kg) 1 .0 - 1.35-1.40" Price products (B/kg) Conversion ratio Starch content determ. Technology By products/waste 2.80 chips 2.0-2.4 fresh roots = 1 chips by variety and break roots chipped roots spread, turned over, dried in 3 days Raw material cost/total cost 95-100 Sa-Nguan Wongse Industries Co. private raw starch modified starch glucose (next year) 185,000 starch (largest in Thailand) 140,000 starch 6,000-12,000 mod. starch year-round, 24 hrs/day 875,000 -» 1.58 (at 30% starch) 5 fresh roots =1 starch measure starch content washed roots grated, centrifuged flash-dried 1) Waste fiber (50-58% starch) pressed, dried, used in pellets or chicken feed 2) Peel sold as compost or for mushroom production (B300/t) 3) Waste water sedimented,-* irrigate eucalyptus plantation 85-88 "plus B 0.02/kg for truck driver 424 Table 9. Topography, soils, subdistrict, Wang crops and cropping systems in two villages of Wang Sombuun Nam Yen district, Sra Kaew province of Thailand in 1995. Village #5 Village #3 Topography Soil types Land use Crop choices Livestock Off-farm activities Cassava yield Cassava utilization (%) -chipping floor (mainly for sale to starch factory) -starch factory (>100km) rolling, foothills reddish-brown clay loam many brown/purple stones high areas: cassava or cassava rotated with maize and soybean low areas: cassava, maize, soybean, mungbean, cotton, sunflower, eggplant, custard apple, banana, cucumber, mango 1. maize 2. cassava 3. soybean/mungbean/sunflower chickens, water buffalo beef cattle, fish part-time work mainly in construction Rayong 1: 2-3 t/rai Rayong 60: 3-4 t/rai Rayong 90: 4-6 t/rai 99 1 rolling, foothills brown/purple stones big rocks on 30 % of area cassava rotated with maize and soybean 1. maize 2. cassava 3. mango chicken, beef cattle part-time work Rayong 3: 2-3 t/rai Rayong 60: 3-5 t/rai Red Tip: 2-3 t/rai 100 0 425 Table 10. Cassava cropping systems, varieties and agronomic practices in two villages of Wang Sombuun subdistrict, Wang Nam Yen district, Sra Kaew province of Thailand in January 1995. Village #5 Village #3 Cropping systems Varieties Land preparation Planting method -stake position -spacing (cm) Fertilization (kg/rai) cassava monocropping rotated with maize and soybean R90>R3>R60>R1 tractor +3disc (2x) tractor + 7disc (lx) ridging (mostly up-down slope) vertical or inclined R60 at 50x100 R90 at 50x75 or 50-60x100 Rl at 100x100 no fertilizer to cassava or 6-20 of 15-15-15 at 3 MAP cassava monocropping rotated with maize and soybean R60>RedTip>R3>R90 tractor + 3disc (lx) ridging vertical or inclined R90 at 50-60x100 20 of 15-15-15, 13-13-21 or 25-7-7 Weed control 1) Alachlor, Diuron as pre-emerg. 1) Alachlor as pre-emerg. 2) Gramoxone as post emerg. 2) Gramoxone as post emerg. 3) hoe 3) hoe or knife Harvest by tractor + harvester manual lifting and loading Time of planting Apr-May Apr-May Time of harvest Dec-Jan Nov-Jan 426 Table 11. Principal cassava varieties and their characteristics as well as the main problems mentioned by farmers in two villages in Wang Sombuun subdistrict, Wang Nam Yen district, Sra Kaew province of Thailand in January 1995. Village Village #5 Village #3 Main varieties Characteristics -Rayong 1 -Rayong 3 -Rayong 60 -Rayong 90 R90, R3, R60, Rl low yield yields are decreasing high yield, low starch, good weed control, not always accepted by starch factory high yield, high starch good weed control R60, Red Tip, R3, R90 high yield, high starch, small plant high yield, good soil cover good price, decreasing yield -Red Tip " tall plant, low yield, good weed control Problems 1 . low price 1. lack planting material of good varieties2. irregular rainfall 3. lack of labor 2. lack knowledge about fertilizers4. lack of capital 5. erosion 3. erosion 6. good varieties difficult to find and expensive 4. diseases 7. decreasing yields when planted contiuously 5. lack of labor 427 Table 12. Cassava utilization and socio-economics in two villages in Wang Sombuun subdistrict, Wang Nam Yen district, Sra Kaew province of Thailand in January 1995. Cassava utilization (%) Village #5 Village #3 - chipping floor " 99 100 - starch factory v 1 0 Prices - fresh roots (B/kg) 1.19-1.20 in chipping yard 1.32 in starch factory - 15-15-15 fert. (B/50 kg) 285 - 13-13-21 fert. (B/50 kg) 300 - urea (B/50 kg) 250 - labor (B/day) 100 Production costs (B/rai) - land preparation - plowing 100-120 100 - discing 80-90 - - ridging 80-100 80 - planting 100 80 - weed control - pre-emergent - herbicide cost 160 60 - labor 100 40-50 - post emergent - Gramoxone cost 200 160 - labor 100 .3) - manual control with hoe or knife 200 _3) - fertilization - fertilizer cost 100 120 - labor 85 .3) - harvest - digging (by tractor) 150 150 - collecting 300-450 290 - transport 210-240 225 - land rent 400 1,310" - Total 1,500-2,195 Gross income (B/rai) 3,600-4,20041 3,600" Net income (B/rai) 1,800-2,700" 2,290" "most chipping floors do not chip, but transport fresh roots to starch factories 2)starch factories in Kabin Buri or Chanta Buri, both about 100 km away "own labor not included 4,high gross and net income due to unusually high cassava price c.e .J s ■5 5= C I § DS .Si — 60 C ■a •e 2eg t-1 > o v. 1 o M in — N N O at t/5 u .C c > B c 0 o 2 2 so c -c 2 o I o 1- © 00 in o 0n r 1 r 1 q— ri — 3 O 00 5 — c] ri O 00 Q -6 c c .J I5. Q. c c 1- sreak .read, riedin 3] 8 •s. 5= syvarietyand chippedroots turnedov r,d 1.00-1.s0 0 13 DO XIc 0 88 E. CI) > ■^j s ri 1 c 5 > 00 0 00 1 0 qz ri S3 H 3 5 £ 8 3 0 I 9 o J3 c 8 8-2 *2 •- £X.^ ■8 " 8.1 5 §5.2 111 I cu < 0- oS 2 "o cu e o o P XI £ <-> o 3 60 C epia ^ ° 2 OP ~ •a ° o — 60 3 M O .. o O — u eg 12 «5 •c -c CL Q. 3 H C « CU DC 429 5. Farmers Training Farmers from the two pilot sites that were interested in participating in the project were invited to join a two-day training course at the Rayong Field Crops Research Center with the objective of: 1 . increasing the farmers' knowledge and understanding of soil and water conservation in cassava production; 2. to discuss with the farmers how to conduct, with the help of researchers and the extension workers, FPR trials on their own fields, and 3. to increase their knowledge on new cassava varieties and production techniques. Furthermore, they visited the demonstration plots on management practices to reduce erosion and discussed the advantages and disadvantages of each treatment. Each farmer was asked to score the various soil erosion control treatments, considering their likely effect on yield and income, their effectiveness in reducing erosion, and whether they seem to be useful under the farmer's own conditions in the village. The results of these farmers' evaluations are also shown in Table 14. 6. FPR Trials on Soil Erosion Control, Varieties and Fertilizer Use After the training session and farmers' evaluation of the demonstration plots, researchers from DOA, together with extension workers from DOAE, discussed with collaborating farmers in the pilot sites the type of FPR trials that they wanted to do on their own land, as well as the treatments to be included, plot size, management etc. Farmers in Soeng Saang district decided to conduct trials on erosion control, on fertilization and on cassava varieties, while farmers in Wang Nam Yen district decided to conduct only trials on erosion control and varieties. 6. 1 Soil erosion control trials FPR team members and collaborating farmers selected the most appropriate areas and layed out the FPR erosion control trials for the farmer along contour lines. The farmers participating in the project discussed and selected the treatments they wanted to test in their own trials: - Nine farmers in Soeng Saang chose to do erosion control trials and selected five treatments, namely 1. their own traditional practice of up-and-down ridging; 2. vetiver grass contour barriers; 3. contour ridging; 4. sugarcane contour barriers; and 5. intercropping. However, farmers selected different intercrops: five farmers selected peanut, one selected sweet corn and one farmer selected mulberry bushes instead of vetiver grass as contour hedgerows. 6.2 Cassava varieties and fertilizer trials FPR trials were also conducted on varieties and fertilizer use by those farmers that wanted to participate but did not have slopy land to conduct erosion control trials. In the variety trials, five varieties were compared, namely Rayong 1, Rayong 3, Rayong 5, Rayong 60, Rayong 90 and Kasetsart 50. In the FPR fertilizer trials different combination of N, P and K were applied with the objective of determining which nutrients produced the greatest benefits at the lowest s•c > J3 Z c g)TJ " 5 M IS «fc c w „ o «*i g -t ■u .5 If X! M H .9 a! 1 | OO go '5' cs ao. -—• I/) II) B 3) c o I a. c u E s H O Os Os V~) — Tt—im O c> n i/->i/-)ini/">i/-)i/■>i/■>U■>i/■>i/■>«i/■>i/■ii/■>viini^ lO o o>/■> >n o IT! >/^ IT) NiOmifl«lOiflifliOiflmiflin«mmmm« o oo o o o o o o oooooooooNNNininminin/iiflinifliflininiflinifliniflin 1 o O O o o o o o o o O O o o o o o o o o o o o X xr X X X X X X X X X X X X X X X X X X X X X X O oo 00 00 oo 00 00 00 00 00 oo oo oo 00 oo oo 00 00 oo oo oo oo 00 oo — ooooooooooooooooooooooo o o o o c B a B oil B B B E T B O. O 3 O o o o o E oooooooooooo BEBEBBEBBBEE * * * * * * 0 n o 0 0 r. t t t t t t tfl n) M rcl low yield because of continuous intercrop competition as peanut was not harvested 4)low yield due to excessive shading from hedgerow species that the seed and stolens of the grass soon spread into the cassava field, thus becoming a serious weed problem. Vetiver grass contour barriers were only intermediately effective in reducing erosion (Table 17). Vetiver grass contour barriers were less effective in reducing erosion in this location because the grass established a solid hedgerow only gradually, while heavy rains after cassava planting caused rather severe soil loss due to erosion. 433 After the harvest, the FPR team presented the collected data to the participating farmers in the two pilot sites. They exchanged opinions and discussed together the advantages and disadvantages of the different methods for soil erosion control. Farmers in Soeng Saang indicated that vetiver grass contour barrier had the highest efficiency in erosion control, followed by contour ridging and sugarcane contour barriers. Later, farmers had a group meeting and selected the management practices they wanted to test in the trials in the following year. These include: 1 . the farmer's own practice of up-and-down ridging 2. sweet corn-cassava intercropping 3. squash/pumpkin-cassava intercropping 4. sugarcane contour barriers 5. vetiver grass contour barriers Similarly, farmers in Wang Nam Yen considered that vetiver grass contour barriers had the highest efficiency in erosion control. Although the research results did not show this grass to be particularly effective in erosion control, in some trials the grass had shown its great potential in reducing soil loss due to erosion. Farmers expected this method to be the most effective in the long-term. The farmers group of Wang Nam Yen selected the following treatments for the next year's trials: 1 . the farmer's own practice of up-and-down ridging 2. contour ridging 3. vetiver grass contour barriers 4. mungbean-cassava intercropping 5. dried grass mulch application between cassava rows 2. FPR Variety Trials In Soeng Saang, the results of the variety trials in which six varieties were compared, showed that Rayong 90 produced the highest yield of 37.0 t/ha and had the highest starch content of 29.5%, followed by Kasetsart 50 with a yield of 35.7 t/ha and starch contant of 28.1% (Table 18). In Wang Nam Yen, Rayong 90 produced the highest yield (32.5 t/ha). The next variety was Rayong 5 with slightly less yield (30.7 t/ha). On the other hand, Rayong 5 had the highest starch content (30.3%), while the starch content of Rayong 90 and Kasetsart 50 were equal (28.7%) (Table 18). 3. FPR Fertilizer Trials Fertilizer trials were conducted only in Soeng Saang. The results (Table 19) indicate that the application of 25 kg N, 25 P205 and 25 K:0/ha gave the highest yield of 31.6 t/ha and also the highest net income of 32,113 baht/ha. This treatment 434 corresponds with an application of 166 kg/ha of 15-15-15 fertilizers, which happens to be the present fertilizer recommendation. Constraints The following constraints were the main cause of problems encountered in the FPR project: 1 . The local extension workers, who are responsible for this project in the pilot sites, are often busy with other routine work, so they could not spend much time in this project. 2. Some participating farmers did not take good care of their trials. 3. In some erosion plots there were slopes in two directions causing problems in the collection of eroded sediments. 4. In the rainy season, the run-off is very heavy and farmers were not always able to protect their FPR erosion trials from run-off coming from fields above. This damaged some plots and resulted in excessive erosion in some plots not at all related to treatments. This reduced the reliability of the erosion data. FUTURE PLANS New FPR trials are already being implemented in the same districts to confirm the results of last year. Some participating farmers who showed little interest in the project have been replaced by new farmers. In addition, the project has selected some more suitable areas for the soil erosion control plots. REFERENCES Office of Agricultural Economics 1995. Agricultural Statistics of Thailand, Crop Year 1994/95 Ministry of Agriculture and Co-operatives, Bangkok, Thailand. Puttacharoen, S. 1992. Nutrient Removal by Crop and Nutrient Loss by Erosion in Cassava in Comparison with that of Other Crops. MSc thesis. Kasetsart Univ., Bangkok, Thailand. 103p. Io I c I ■t 1 o • £|s tn e = o •o 1 § c a 1 ■a o■e 85 o U B Q. C 58 E O 32 "3 o £ £ ! ► «< z vo ^H 0> X! « H p S 6 a 3 a u Z c _0 o 3 "8 c o S 1 0 a u oo .2o s 0 S c 1 o oo r- o o —i o i O » in o * " in oo m - o in oo r- t» on r- no r- t— —* no* in Tf o* r-* r4* c*l tN (N m m rl m Tt (01 -h On Oi Oi o\ On —* 00 — —• — — "n ou c .a,ooa? & £ 2 •Sob ill U to Cl to 2 E « 60 S8888 — oo — r-i ts 60 oo o on en c 01 o 8f 15 3 E O cj 1 c o S B a $ c o •c T3 c ca Eh c/> o> t*. u o B a n 3 u.a. u * 01 0) 01 2> 1/3•< H o S 53 ca i>fc EL 8r, o ^ "> 2 & 8 SQ ~ w "8 u A E o o .s ca - OD Gfi 5 © 55 S £ 8 a .s v O T3 ^v v- — ca c 55u V 3 a GO n ca o > —I 3—n 8 . U w. z °- c I ! Oi/IOOOO—.—.O —< — — — cN OMoor^ooooooO't tN-HNtNtNclNn- I/"l M N (S N On no — t O VO O o o O fI —. O o O 5 TTC-lOnOOOtN^t^r-in©oomcNOO Tj- t^ (N — Tt fi {S ci fi fi (N O O Tt Tt T}- o o Tt-t— o r^ On m o\ >o t^-coTj-inininnoN 00rnndnOiO>/">f>-HTfr wi iO N0 ci —Nci^ino^«io\ ca60 60 60 60 60 60 r* 4 <* £ 4 4 ^ os t— —i O CT\ «r> no no ca =5 Xl p I 1 1 s c 1 V) I u■5 60 .a I 3 z 437 Table 18. Results of three FPR variety trials conducted by farmers in Noon Sombuun village of Soeng Saang district, Nakorn Ratchasima province of Thailand in 1995/96. Cassava yield (t/ha) Starch content (%) Rayong Rayong Farmer s name 1 3 5 60 90 KU 50 1 3 5 60 90 KU 50 1. Mr. Prasit . - 32.4 36.8 - 31.2 - - 28.9 29.2 - 29.2 2. Mrs. Lek 22.0 9.0 20.5 15.5 20.0 20.0 26.1 25.3 25.9 19.8 29.8 27.9 3. Mrs Chuen 39.0 37.0 43.0 39.0 54.1 56.0 25.7 28.0 23.9 25.9 29.3 27.1 Average 30.5 23.0 32.0 30.4 37.0 35.7 25.9 26.6 26.2 25.0 29.5 28.1 Table 19. Average results of five FPR fertilizer trials conducted by farmers in Noon Sombuun village of Soeng Saang district, Nakorn Ratchasima province of Thailand in 1995/96. Fertilizer rate Root Gross Fertilizer Net N-P2O5-K2O yield income cost income (kg/ha) (t/ha) < - - -—(baht/ha) > 0-0-0 28.6 30,030 0 30,030 50-0-0 29.2 30,660 811 29,849 50-50-0 31.0 32,550 1,755 30,795 50-0-50 30.7 32,235 1,261 30,974 25-25-25 31.6 33,180 1,067 32,113 50-50-50 30.8 32,340 2,133 30,207 Prices: cassava fresh roots B 1.05/kg urea 365/50kg triple superphosphate 425/50kg KC1 270/50kg 15-15-15 320/50kg 438 FARMER PARTICIPATORY RESEARCH IN CASSAVA TECHNOLOGY TRANSFER IN INDIA S. Ramanathan and M. Anantharaman1 ABSTRACT The concept and methodology used in the transfer of technology (TOT) have undergone changes over the years corresponding to farmers' needs, and as a result, farmer participatory TOT has become more relevant, especially in complex diversified risk-prone (CDR) farming systems. The TOT in cassava, which largely belongs to the CDR farming system in India, is no exception to this. The National Demonstration Programme (NDP) on cassava, which started in 1970, was the first organized attempt to transfer the improved cassava technologies, especially the hybrid varieties. The main objective of NDP was to convince the farmers about the production potential of the new cassava hybrids by conducting demonstrations in farmers' fields by scientists. It restricted the scope of farmers' participation in technology transfer. Their role was mainly confined to be the passive spectator of the demonstrations. The scope of the farmers' participation was widened under the Operational Research Project (ORP) in the mid-seventies, which laid emphasis on the identification of constraints in the adoption of technologies. Though the farmers did not have much say in the technology transfer, they played a significant part in assisting the scientists in identifying the operational constraints to adopting the cassava technologies. The Lab-to-Land Programme (LLP), launched during 1979, witnessed greater participation of adopted farmers, by emphasizing direct linkages between scientists and farmers. The participation of the farmers was ensured right from the benchmark survey of farm families, the demonstration of cassava technologies, identification of potential spheres of development, and the dissemination of the impact of the program through fellow farmers and mass media. The impact assessment of the program revealed that there can not be a uniform package for transfer, and very often farmers try to blend the new technologies with their traditional practices. Hence, there is a need to evolve appropriate location-specific technologies suitable for various micro-environments. On the realization of this fact, CTCRI has implemented a farmer participatory research program on cassava technology assessment, refinement and integration, which indicates the differential pattern of technology preferences in various farming situations. INTRODUCTION Cassava is a crop of small and marginal farmers operating under the complex, diversified and risk prone (CDR) system of agriculture in India. The Central Tuber Crops Research Institute (CTCRI) is spearheading the research and development of cassava in the country, together with the extension work being carried out by the Department of Agriculture/Horticulture of Kerala and Tamil Nadu, the two major cassava growing states in India. Following the changing trends in the agricultural transfer of technology (TOT) programs at the national level, the TOT programs of cassava too have undergone changes at appropriate times, facilitating more and more farmers' participation in the TOT process over the years. 1 Central Tuber Crops Research Institute (CTCRI), Thiruvananthapuram, Kerala, 695 017, India. 439 Historical Perspective of TOT in Cassava During the early seventies (1971-74) National Demonstration (ND) on cassava was the pioneering attempt to transfer cassava technologies on a large scale, especially the hybrid varieties in India. This program, the oldest front-line extension project conceived by the Indian Council for Agricultural Research (ICAR), was implemented by various ICAR Institutes and Agricultural Universities, but with little direct participation of farmers, either in technology transfer or in their generation. The farmers' role was mainly confined to be the passive spectator of the demonstrations (Table 1). Farmers' participation in cassava technology transfer was first introduced under the Operational Research Project (ORP) on cassava, implemented during 1976-81 in a suburb village near Thiruvananthapuram city of Kerala State. Although the beneficiaries of ORP did not have a say in the technologies of cassava to be transferred, they were the dominating partners in the identification of socio-economic and technological constraints in the adoption and spread of introduced cassava technologies envisaged under the ORP. The scope of farmers' participation in cassava technology transfer was increased under the Lab-to-Land program (LLP) launched during 1979 to commemorate the Golden Jubilee celebration of ICAR. Right from the initial benchmark survey of the selected farm families up to the final impact survey of the program after withdrawal from a particular village, the beneficiary farmers enthusiastically took part in the TOT of improved cassava technologies. Occasions such as laying out the demonstration plots, group meetings, harvest festivals etc. were effectively utilized to enlist the participation of both the beneficiary as well as non-beneficiary farmers in the technology transfer process. The sharing of successful experiences in the introduced cassava technologies by the participating farmers with their fellow farmers through radio, television and the print media acted as a motivating factor in the further spread of cassava technologies (Anantharaman and Ramanathan, 1986). Impact Implications A study on the impact of LLP on cassava, undertaken across the implemented villages, revealed a differential pattern amongst the villages (Anantharaman and Ramanathan, 1996). It was seen that the socio-economic conditions operating at the micro-level of the farmers played a significant role in influencing the impact of the program in various villages. Moreover, a complete adoption of the entire package was not very common; rather, the farmers tended to use a blend of the introduced cassava technologies and their traditional practices, according to their socio-economic and cultural conditions. The slow rate of multiplication coupled with the time lag between the harvesting and subsequent planting continued to pose a severe constraint in the rapid spread of cassava technologies. The absence of a specialized agency for large-scale multiplication and distribution of planting material of hybrid cassava further aggravated this problem. Notwithstanding these bottlenecks, the cassava growers, if they are not 440 innovators, they are also not laggards in adoption; they show a keen interest in the adoption of improved cassava technologies once they are exposed to those technologies and backed up with an adequate supply of critical inputs, such as planting material and fertilizers. Table 1. Transfer of Technology Projects of cassava in India and the extent of farmers' participation. Name of the project Period Impact Nature and extent of farmers'participation National Demonstration 1971-74 An awareness about Very much limited. Program hybrid cassava Only a spectator of varieties created. the demonstrations Operational Research 1976-81 Impressive yield Participation widened. Project performance and Dominating partner income generation. in identification of Constraints in TOT various constraints. of cassava identified. Lab-to-Land Program 1979 to Impressive yield Participation from the today performance and start to the end of the income generation. program. Experience Spread of introduced sharing between farmers varieties. Overall for effective development of dissemination. adopted families. FPR Program in Since Location-specific Active partners in the cassava 1994 varieties identified. selection and spread of varieties. The overall impact of the TOT efforts in cassava clearly indicated the necessity of generating location-specific technologies, particularly new varieties suitable to various micro-climatic environments of cassava production zones in the country. On realization of this fact, and taking into consideration the worldwide recognition for farmers. 441 participation in the development of agricultural technologies and transfer, CTCRI has initiated during the mid-nineties a program of participatory research in cassava, especially for varietal evaluation and transfer. Under this program the farmers are given a basket of cassava varieties from which they can make a choice suitable to their own conditions and requirements. FARMER PARTICIPATORY RESEARCH AND TRANSFER IN CASSAVA Generally, agricultural technologies are developed under ideal conditions and based on various criteria they are assumed to be solving farmers' problems, as perceived by the scientists. But, quite often there exists a wide gap between what is perceived as needed by the scientists and what is actually needed by the farmers. As a result, the majority of the technologies turn out to be a mismatch and their transfer to the end-users is found difficult. In this regard, the methodology of farmer participatory research is expected not only to better identify the topics to be researched, but also to facilitate the effective transfer of improved technologies. Keeping this in mind, in 1994 a program of farmers' participation in cassava varietal evaluation and popularization was initiated at CTCRI, following the methodology developed by Anantharaman et al. (1994). Methodology of FPR in Cassava Figure 1 shows the five stages of the FPR methodology used at CTCRI. Stage I. Selection of villages and diagnosis of the cassava crop status. As an initial step in the conduct of on-farm trials, a suitable village has to be selected, i.e. a village where cassava plays a significant role in the economy of the village, and the selected village should have a major area under cassava. Using these criteria, Anacode village in Thiruvananthapuram district of Kerala was selected for implementing the FPR approach. A detailed study on the status of the cassava crop in this selected village was undertaken using various participatory rapid appraisal (PRA) tools and techniques. These included: /. Village transect The village transect of Anacode represents the cross section of the village, indicating the topography, soil type, crops grown, livestock raised, irrigation source, problems etc. (Figure 2). The selected village has a undulating topography like most Kerala villages, with low-lying paddy fields located at one end of the village and adjoining uplands. Though traditionally the lowland area is mainly used for paddy cultivation, of late cassava has emerged as a money-making crop under low-land conditions. A sizable proportion of the uplands in the village is under cassava, both as a monocrop and intercropped under either rubber trees during the initial period or, to a lesser extent, under coconut trees. It is clearly evident from the transect that cassava is an important crop in Anacode, raised under both upland and lowland conditions. 442 STAGE I Selection of village and diagnosis of crop status U STAGE II Selection of farmers and formation of evaluation groups u STAGE III Laying out and management of on-farm trials on cassava varieties and evaluation by various user groups and scientists u STAGE IV Evaluation of selected varieties for validating the performance u STAGE V Popularization of the most preferred variety. Release of the variety for large scale TOT Figure 1. Schematic representation of the methodology ofFPR used in cassava by CTCRI in Kerala, India. 443 Lowland Upland Soil type Crops Trees Irrigation Resources Livestock Drainage Problem Clayey Paddy, Cassava Banana, Vegetables Aroids Canal Fodder Medium Conversion of paddy land to other crops Red and laterite Rubber, Cassava, Coconut, Yams, Mango, Jackfruit Tamarind, Perumaram etc. Wells and rainfed Fuel, Grass, Cattle and poultry manure Cow, Goat, Buffalo, Poultry Good Lack of irrigation facilities Figure 2. Transect ofAnacode village in Trivandrum district of Kerala, India. 444 2. Cropping pattern and seasonality ofpests and diseases The informal interview with key informants together with a diagramming technique, cross-checked with the secondary data available at the village Agricultural Office, revealed that besides paddy, cassava is also grown to a large extent in the lowland. Normally two crops of cassava are cultivated under this condition, one during the period April/May to Sept/Oct and another one from Oct/Nov to March/April. The crop takes about 6-7 months until harvest under the lowland condition. Non-availability of mosaic-free stems and damage by rats are the two most serious problems experienced under this condition. In the upland, cassava is planted during April/May and is harvested in Feb/March. As in the case of the lowlands, cassava mosaic disease is an important constraint to increasing cassava production in the uplands too. Kariyila poriyan, Mankozhunthan and Karukannan are the popular cassava varieties grown by the farmers. The high-yielding cassava variety Sree Vishakom (H-1687) is also being cultivated. 3. Crops ranking Farmers were also asked to compare the relative importance of the various crops they grow based on the criteria that they feel are important in choosing a crop enterprise. Using these criteria, i.e. food security, profitability, low pests and diseases incidence, tolerance to drought, low input requirement and good marketability, rubber was the most preferred crop, closely followed by cassava, owing to its drought tolerance, low pests and disease incidence, low input requirement and food security (Table 2). Banana and vegetables, though considered as being the most profitable, were susceptible to pests and diseases and required high inputs for their cultivation. 4. Cassava varieties ranking A group of farmers was asked to rank their five most popular cassava varieties against the characters they considered most important, such as yield, taste, starch content, root compactness and season (Table 3). It is clearly evident from the ranking that none of the varieties were ranked first in all these aspects, indicating that each of the popular varieties excelled others in one character or the other. The hybrid cassava variety Sree Vishakom was ranked first in yield, taste and compactness of roots, but was preferred last with respect to starch content and season (non-bound). Similarly, the local variety Monkozhunthan rated first for starch content but was ranked last for taste. However, the overall ranking revealed that three of the varieties had quite similar scores. 5. Problem-cause relationship in cassava Problem-cause relationships were determined with a group of farmers for both lowland and upland conditions. Low productivity of cassava was perceived as the major problem in both the upland and lowland situations (Figures 3 and 4). The average productivity of cassava ranged from 15-17 t/ha under upland and 20-25 t/ha under 445 lowland conditions. The bio-physical as well as the socio-economic causes for such low yields are shown in the Figures. Of the various causes, non-availability of improved short-duration varieties was found to be the major limiting factor to cassava production under both production systems. Evaluation and identification of suitable cassava varieties with farmers' participation contemplated under the FPR program could be the possible solution to overcome this problem and increase the cassava productivity in Anacode village. Stage II. Selection of farmers and formation of evaluation groups /. Selection of cooperatorfarmers Two farmers with long experience in cassava cultivation, who were willing and cooperative in nature and who have a high proportion of their land planted to cassava, were chosen, one each for conducting an on-farm trial (OFT) under upland and lowland conditions. 2. Formation of evaluation groups It was not possible to establish a large number of trials to be used as replications in the on-farm research, due to the lack of mosaic-free planting material and other resources. Instead, by forming groups of various evaluator categories, such as farmers, traders and farm women, to evaluate a single trial, each member of the group was considered to be a replication/observation for the purpose of analyzing and interpreting the outcome of the trials. However, care was taken to involve the entire group in the management of OFT. Accordingly, at the village level a farmers group, a traders group and a farm women group were formed, consisting of persons with extensive experience in cassava cultivation or trade, and having the ability and inclination to participate in the evaluation and to give their assessment on the cassava varieties. In addition, a group of scientists comprising breeders, agronomists and plant protection specialists, coordinated by a social scientist, was also formed, in order to determine the degree of agreement in the evaluation of cassava varieties by the scientists with other user groups. Stage III. Laying out, management and evaluation of OFT In consultation between the farmers and scientists, and on the basis of character preference of cassava varieties, as indicated in the varieties' ranking (Table 3), it was decided to include 1 1 cassava varieties, comprising three released varieties, one improved variety, four pre-released varieties and three popular land races. The varieties selected for the trial were: Released varieties: 1. Sree Vishakom (H-1687) 2. Sree Sahya (H-2304) 3. Sree Prakash (S-856) 446 SOCIO-ECONOMIC BIO-PHYSICAL SMALL LAND HOLDINGS POOR KNOWLEDGE ABOUT PRODUCTION AND PROCESSING MANAGEMENT LACK OF CASSAVA BASED COTTAGE INDUSTRIES NON AVAILABILITY OF IMPROVED SHORT- DURATION VARIETIES LOW PRODUCTIVITY OF CASSAVA SEVERI RODEN DAMAC HIGH INCIDENCE OF MOSAIC LODGING DUE T< HEAVY WIND Figure 3. Problem-cause relationshipfor low productivity of cassava under lowland conditions in Anacode village, Trivandrum, Kerala, India. 447 SOCIO-ECONOMIC BIO-PHYSICAL SMALL LAND HOLDINGS POOR KNOWLEDGE ABOUT PRODUCTION AND PROCESSING MANAGEMENT NON AVAILABILITY OF IMPROVED SHORT- DURATION VARIETIES LOW PRODUCTIVITY OF CASSAVA LOW YIELD LACK OF CASSAVA BASED COTTAGE INDUSTRIES CULTIVATION MOSTLY AS INTERCROP LODGING DUE TO HEAVY WIND Figure 4. Problem-cause relationshipfor low productivity of cassava under upland conditions in Anacode village, Trivandrum, Kerala, India. 448 Table 2. Matrix ranking of crops by farmers in Anacode village, Trivandrum district of Kerala, India. Characters Paddy Cassava Rubber Banana Vegetables Food security 1 2 5 3 4 Profitability 5 4 1 2 3 Low pest and disease incidence 4 1 2 3 5 Drought tolerance 5 1 2 4 3 Low input requirement 4 1 2 3 5 Marketability 4 5 1 3 2 Source: Anantharaman et ai, 1994. Improved variety: 4. M-4 (popular among farmers in other areas) Pre-released varieties: 5. CI-649 6. CI-664 7. CI-731 (a selection from the local variety Kariyilaporiyan) 8. CI-732 Local varieties: 9. Karukannan 10. Monkozhunthan 1 1 . Kariyilaporiyan While implementing the OFT on cassava, the consultative participation of farmers, which emphasizes researcher managed and farmer-implemented trials, as outlined by Ashby (1986), was resorted to. However, researcher management was restricted to guidance provided to farmers in the management of the OFT. Considering the comprehending ability of the farmers and their unfamiliarity with experimental designs, the laying out of the OFT using a typical experimental design was found difficult under actual farm conditions. Hence, in discussion with the farmers and in consultation with a statistician, a suitable design (modified CRD with two replications) was formulated, which would facilitate the laying out of the trials and still be adequate for appropriate statistical analysis. Accordingly, two trials, one each under upland and lowland conditions, using the 1 1 selected cassava varieties in two replications under two types of management, namely recommended package of practices and farmers' management, were laid out. They were closely monitored through fortnightly visits to the village and with active cooperation of the participating farmers. Group meetings were also organized at the critical stages of the trials and the views and experiences of 449 Table 3. Matrix ranking of cassava varieties on the characters preferred by farmers in Anacode village of Tri vandrum district in Kerala, India. Character Sree M-4 Kariyila Mankoz Karu Visakham poriyan hunthan kannan Yield Taste Starch content Root compactness 1 Season 2 4 5 3 3 2 4 4 5 1 3 2 4 3 2 I Source: Anantharaman et al., 1994. the individual farmers were recorded. Group consensus was arrived at in the conduct and progress of the trials. Cassava roots were harvested at the 7th month under the lowland condition and at the 10th month under the upland condition. The harvested roots were exhibited in a row, masking their identities, but with labels as VI, V2 etc. The selected groups of scientists, farmers, farm women and traders were guided with an open-ended questionnaire to register their opinion on positive and negative aspects of the roots. They were also requested to rate the varieties on a five point continuum and to select the best five varieties from among the 1 1 exhibited. Results of FPR Cassava Varietal Selection The most salient features of the preliminary stage of evaluation of cassava varieties are as follows: /. Performance of cassava varieties (Phase I) The average root yields of the tested varieties are given in Table 4. The analysis of the data indicate that the yield performance of the varieties under the recommended as well as the farmers' management did not show any significant difference under both production systems. However, significant differences existed amongst the varieties, with CI-649 producing a significantly higher yield than the other varieties under the lowland condition. The yields of CI-731, Mankozhunthan, H-1687 and H-2304 were not 450 statistically different, but were higher than those of the remaining varieties. The yield performance of the cassava varieties, except H-1687 was found to be rather poor under the upland condition (Anantharaman et al., 1995). 2. Evaluation of cassava varieties Result of the evaluation by various groups on the basis of the ranks obtained by each variety in terms of root characteristics, taste and suitability of the cassava varieties for their end-use, are presented in Table 5. The Spearmen rank order correlation, determined between the rank order of the varieties given by the various groups, indicate that while the scientists' preferential order did not have agreement with that of other groups, there existed agreement amongst those other groups. This shows that there were differences in the preferential order of the varieties between scientists and end-users. Analysis of the criteria used by the scientists and the other user groups for evaluating the varieties, indicate that, while scientists gave importance to root weight, size, shape and compactness, farmers considered number of roots (optimum being 5-7), uniformity in size (optimum being 500-600 gm), skin and rind color, starch content, marketability, attachment of the root to the base of the plant, root shape, general appearance, absence of fibrous portion etc. as important criteria in selecting a variety. Farm women considered taste, fast cooking, bitterness/sweetness of the cooked roots important in a good cassava variety. The traders considered uniformity of roots (medium-sized), starch content and skin color as important criteria. In addition, the farmers were of the opinion that the cassava varieties suited to the lowland condition should be of medium stature and should not grow too tall. Otherwise, it might lead to lodging. Moreover, the variety should not have too much foliage, and at the same time, it should possess a higher starch content and have less rat damage. As regard to stems, these should be flexible and strong. Since there will be moisture at all times under lowland conditions, the roots should resist rotting. On the basis of the evaluation by various groups, five cassava varieties, i.e. CI- 649, CI-664, CI-731, CI-732 and H-1687 have been selected for further evaluation during the subsequent phase. Stage IV. Evaluation of selected clones (Phase II). The five cassava varieties selected under Stage III were grown under both upland and lowland conditions for validating the performance of these varieties for one more season, and to select one or two most preferred varieties for popularization in the selected village. The same process of evaluation of the varieties at time of harvest (excepting scientists group) conducted during the earlier stage, was repeated. The analysis of the performance under the lowland condition has been completed and the same under the upland condition is in progress. The yield performance indicate that the cassava variety CI-732 was the highest yielder (46.29 t/ha), followed by CI-649 (40.51 t/ha) and CI-731 451 Table 4. Yield performance of eleven cassava varieties in on-farm trials conducted under lowland and upland conditions in Anacode village of Trivandrum district in Kerala, India. Average root yield (t/ha) Variety Lowland Upland 29.25 30.15 26.15 12.83 19.81 12.83 26.72 14.40 32.13 19.50 55.00 14.18 22.13 6.66 21.81 12.15 30.91 8.78 19.69 17.55 28.50 16.65 H-1687 Karukannan CI-664 S-856 CI-731 CI-649 CI-732 Mankozhunthan Kariyilaporiyan H-2304 Source: Anantharaman et al., 1995. (39.72 t/ha). CI-732 also emerged as the most preferred variety by both the farmers and the farm women. Matrix ranking of varieties based on yield and other characters by farmers revealed that the preference for the varieties were in the order of CI-732, CI- 73 1 , CI-649 and CI-664. On the basis of overall performance and evaluation by various user groups, the cassava varieties CI-732, CI-649 and CI-731 were the most preferred ones, and these were taken to the next stage, popularization. Since CI-731 is a selection from the popular local variety Kariyilaporiyan, it was decided to popularize only CI-732 and CI-649 in the selected village. Stage V. Popularization of selected cassava varieties Under the popularization scheme, the participating farmers acted as "seed producers" for making available the disease-free planting materials of the preferred varieties, as the non-availability of planting materials of high-yielding cassava varieties was identified by Ramanathan et al (1987) as the most important constraint in the adoption of these varieties by the farmers. Using the stems supplied by the seed producing farmers, presently about ten farmers have grown the cassava varieties CI-732 and CI-649 in about one hectare of cassava area in the village. Efforts are underway to formally release these varieties by the State Variety Release Committee, so that large- scale TOT by the Kerala Department of Agriculture can be initiated. The program of FPR in cassava is being continued in other villages of 452 Thiruvananthapuram district and is also programmed to extend to other regions of the state. Problems Encountered in the FPR on Cassava 1. Availability of planting material Availability of disease-free high-quality planting material of new cassava varieties at the right time and in the right quantity continues to be a serious constraint in conducting these trials in large numbers and over various regions of the state. The absence of any agency for large-scale multiplication and distribution of planting material of improved varieties of cassava in India further aggravates this problem. 2. Laying out of OFT In spite of adopting a simplified experimental design which was discussed with the farmers, the actual conducting of the trial was found difficult, especially under lowland conditions. An area of 0.8-1 .0 ha was needed for conducting the trial. Getting such a big plot in the lowlands was difficult, since channels dug in between the plots for drainage rendered them unsuitable for laying out a varietal trial with 10-12 varieties in two replications. Moreover, as bunds have to be made around each subplot to demarcate the replications, additional labor was required, which in turn increased the expenditure in the cultivation. The undulating topography of the land also at times made the laying out the trials difficult, particularly under the upland condition. 3. Organizing and conducting evaluation Collection of various members of the evaluation groups, especially of the traders and organizing the evaluation on a particular day was found to be difficult. Often the traders go away for their business and the rest of the evaluators have to wait for hours. There were occasions, in which the evaluation had to be postponed until the subsequent day. 4. Harvesting and marketing of roots During the initial phase of evaluation, about 50% of the plant population was harvested to get accurate yield data of the varieties. This created problems in marketing as these could only be sold a day after harvest due to the detailed evaluation on the day of harvest. Resorting to a sample harvest also had its own problems. Since the majority of the farmers preferred to market the roots on a contract basis, sample harvesting, which created vacant spots here and there in the cassava plot, always resulted in a lower price as compared to a full plot of cassava. 5. Labelling and storage of planting materials Careful labelling of the varieties and the separate storing of the stems of each 453 variety is very important to avoid mixing up of varieties and for conducting subsequent trials in the next season. With the stems of 1 1 varieties looking more or less similar, the farmers expressed difficulty in careful identification, labelling and storage of stems of each varieties at the time of harvest. Table 5. Rank order of cassava varieties based on evaluation by user groups and scientists in Anacode village of Trivandrum district in Kerala, India. Ranking by user groups Variety F(irmers Farmi women Traders Scientists H-1687 2 6 6 3 Karukannan 3 5 3 11 CI-664 5 11 8 2 S-856 9 2 6 7 CI-731 4 3 3 4 CI-649 1 1 3 1 CI-732 8 7 9 8 M-4 6 4 1 9 Mankozhunthan 7 10 9 5 Kariyilaporiyan 3 3 4 6 H-2304 11 9 10 10 Degree of agreement between user groups: Scientists vs Farmers 0.39NS" Scientists vs Traders 0.1 3NS Scientists vs Farm women 0.20'-s Farmers vs Traders 0.65" Farmers vs Farm women 0.49'« Traders vs Farm women 0.80. " NS = Not significant Source: Anantharaman et al. , 1995. ** = Significant at \% level REFERENCES Anantharaman, M. and S. Ramanathan. 1986. Report of Lab-to-Land Programme Phase II, 1982-84. CTCRI, Trivandrum, Kerala, India. 27p. Anantharaman, M., S. Ramanathan, S.G. Nair and T.V.R. Naryar. 1994. Farmers participation in cassava varietal evaluation. RPF II (1994-95). CTCRI, Trivandrum, Kerala, India. Anantharaman, M., S. Ramanathan, S.G. Nair and T.V.R. Nayar. 1995. Farmers participation in cassava varietal evaluation. RPF II (1995-96). CTCRI, Trivandrum, Kerala, India. Anantharaman, M. andS. Ramanathan. 1996. Socio-economic discriminants for successful cassava technology transfer programme. In: G.T. Kurup, M.S. Palaniswami, V.P. Potty, G. Padmaja, S. Kabeerathumma and Santha V. Pillai (Eds.) Tropical Tuber Crops. Problems, Prospects and Future Strategies. Oxford & IBH Publishing Co. Pvt. Ltd., New Delhi, India. pp.537-543. Ashby, A.J. 1986. Methodology for the participation of small farmers in the design of on-farm trials. Agricultural Administration 22:1-9. Ramanathan, S., M. Anantharaman and K.R. Lakshmi. 1987. Constraints in adoption of high yielding cassava varieties. Indian. J. Extension Education 23(3&4):55-59. 454 FARMER PARTICIPATORY RESEARCH IN CASSAVA SOIL MANAGEMENT AND VARIETAL DISSEMINATION IN VIETNAM Nguyen The Dang\ Tran Ngoc Ngoan1, Le Sy Loi\ Dinh Ngoc Lan1 and Thai Phien2 ABSTRACT Farmer Participatory Research (FPR) in Vietnam has been carried out since 1994 in collaboration with CIAT, with the objective of improving the adoption of soil conservation practices in cassava fields. Two villages in Pho Yen district, Bac Thai province; one in Thanh Hoa district, Vinh Phu province; and one in Luong Son district of Hoa Binh province were selected as pilot sites for implementing the FPR methodology. In 1994: By using RRA and PRA methods in conducting diagnostic surveys, some main limiting factors in cassava production were identified, such as lack of suitable planting methods for soil erosion control and lack of knowledge about balanced fertilizer application and about high yielding varieties. Therefore, demonstration plot with 17 treatments on different ways to improve soil fertility and methods to control soil erosion, were established at the Agro-forestry College in Thai Ngyen, Bac Thai. Farmers' field days were held to show the demonstration plots to farmers and extensionists from two of the selected districts in mid Nov, 1994. Based on the results and discussion, seven treatments were identified by farmers as promising treatments for 1995. Farmers also discussed how to arrange these in simple FPR trials in their own fields. In 1995: Twenty five farmers of two villages in Pho Yen and ten farmers in Thanh Hoa districts participated in the project by conducting research on their own fields. At time of harvest, a farmers field day was held in both districts in mid Nov, 1995. Fanners and researchers joined in the harvest and in the discussion of the results. Some best treatments were identified. The treatments of cassava intercropped with peanut and contour hedgerows of vetiver grass, combined with balanced NPK application, was considered as the most promising practice at both pilot sites, as soil erosion losses were reduced by 20-40% compared to the check plot of cassava grown in monoculture and without hedgerows. In Pho Yen district, cassava yields in this treatment were about the same as the check plot, but net income increased 9-36%. In Thanh Hoa, cassava yields increased about 9% compared to the check plot, while net income increased by 23%. In this location cassava intercropped with peanut increased net income from 131 to 273% over cassava monoculture. Farmers who tested new promising clones considered KM60 and CM4995-7 as the most suitable for their conditions; these clones increased yields from 1 .7 to 4. 1 t/ha over the check variety Vinh Phu. These initial results are encouraging more and more farmers to participate in the FPR trials. The number of farmers participating in 1996 increased and some of them can conduct the trials now by themselves. INTRODUCTION In terms of area planted, cassava (Manihot esculenta Crantz) is the fifth most important food crop in Vietnam, after rice, maize, sweet potato and vegetables. Cassava fresh roots are used as human food and animal feed, but most cassava is processed into dried chips or cassava starch, which are used to make monosodium glutamate, alcohol, 1 Agro-forestry College of Thai Nguyen University, Thai Nguyen city, Bac Thai, Vietnam. 2 Institute for Soils and Fertilizers, Chem, Tu Liem, Hanoi, Vietnam. 455 candies, etc. Cassava is grown mainly on hilly land in Vietnam. According to the literature (Howeler, 1981), to produce a yield of 15-20 ton fresh roots per ha, cassava absorbs about 75-100 kg N, 35-50 kg P205 and 105-140 kg K2O/ha. Therefore, the crop may deplete the nutrients in the soil, especially if farmers remove from the field not only the roots but also the stems and leaves. Farmers have long experience in growing cassava, and have tried to improve their practices to maintain high cassava yields year after year, such as the making of contour ridges when growing cassava on sloping land, or intercropping cassava with peanut or black bean. But the yield of cassava was still low, because of the lack of good planting methods to control soil erosion, as well as the use of unbalanced fertilizer applications; they also lack good varieties and knowledge about how to use peanut residues as green manure for cassava. A lot of research has been done on various aspects of agriculture including cassava, but few of these technologies have been put into practice, or they are not as efficient as the farmers' practices. One of the most limiting factors in transferring new technologies to farmers is that most research has been conducted at research stations under very good management, very different from the real farm conditions. On the other hand, for farmers to know how to apply the new technologies, they need to do them first themselves and to evaluate them according to their own criteria; or they want to observe how their neighbors apply the new technologies first. Given this fact, it is necessary to change the methods used in research and the transfer of technologies, in order to adapt to farmer's conditions. Farmer Participatory Research (FPR) in Vietnam has been carried out since 1994 in collaboration with CIAT, with the objective of improving the adoption of soil conservation practices in cassava fields. FPR PROJECT IN VIETNAM Two villages in Pho Yen district of Bac Thai province, and one village in Thanh Hoa district of Vinh Phu province were pre-selected as the most suitable pilot sites for implementing the FPR methodology. To impliment the FPR methodology the following activities were carried out: 1 . Diagnostic surveys were conducted in the pilot sites using RRA and PRA methods in order to better understand the existing farming practices and to hear the farmers' opinions about how to solve their limiting factors. 2. Demonstration plots were established at the Agro-Forestry College of Thai Nguyen University in Bac Thai, in order to show farmers a large number of management options that can be used to reduce erosion. 3. Four types of FPR trials were conducted by farmers on their own fields: - Soils erosion control by using different soil/crop management practices - Cassava intercropping with peanuts, black beans etc. 456 - Variety testing - Fertilization 4. Farmers field days were organized at harvesting time to discuss and evaluate the results of the FPR trials, as well as to identify the most promising treatments for next year. Some Initial Results of the FPR Project 1. Diagnostic Surveys at Pilot Sites. Using RRA and PRA methods, we surveyed two potential pilot sites in 1994, i.e., Pho Yen district of Bac Thai province and Thanh Hoa district of Vinh Phu province; these were selected as being most suitable for the FPR project. In 1995, another site in Luong Son district of Hoa Binh province was selected. Table 3 shows that the climate in all three sites is subtropical with an average rainfall of 1500-2100 mm/year and considerable monthly fluctuations; the highest rainfall occurs during the summer months of June to September and almost 80% of total annual rainfall is concentrated in this period. This is one of the main causes for severe soil erosion on steep lands. Cassava is grown mainly on sloping land in the selected areas. In Pho Yen district, cassava is grown on gentle slopes with a light textured sandy loam soil. This soil is very susceptible to erosion by water. Thus, the soil has become depleted of nutrients, especially potassium. Although cassava is grown on much steeper land in Thanh Hoa and Luong Son districts, the soil texture is a heavy clay, which is more resistent to erosion than the light soil in Pho Yen district. Another problem is that the eroded soil is deposited in the rice fields at the bottom of the hill where it causes a reduction in rice yields. In both Thanh Hoa and Pho Yen districts, cassava occupies a rather important position among field crops in terms of the income of farm households (Table 1). It is usually ranked second, after rice, in terms of area and income. The data of the RRA (Tables 1 to 4) have shown that cassava yields in the pilot sites are rather low (8-15 t/ha) for the following reasons: 1 . Cassava is grown mainly on sloping land with such practices as contour ridging and intercropping with peanut (or black bean). Cassava has a slow initial growth and the plants do not cover the soil enough during the rainy months to prevent the direct impact of rain drops on the soil surface. The run-off water carries away both nutrients and soil. Therefore, cassava yields have decreased to only 3-4 t/ha in some plots after 10-15 years of continuous cassava production. 2. On average, fertilizers applied to cassava (Table 3) consists of only about 5-10 t/ha of organic (mainly pig) manure, together with 20-50 kg N, 10-50 P205 and 0-160 K2O/ha. Farmers also lack experience about how to use fertilizers and they lack credit to purchase inputs for cassava. The number of households that use chemical fertilizers 457 for cassava was less than 50% of total households interviewed. 3. The local variety Vinh Phu green and a few other exotic varieties, such as Canh Ng and Du varieties, have been grown for a long time without any regeneration practices. 4. Traditionally cassava roots are mainly used for food and animal feed, while some small quantities are sold in the market when money is needed. Cassava is processed into dry chips after manual cutting with a knife; this requires a lot of labor. In Dong Rang village, however, more than 50% of cassava is sold as fresh roots. 5. The demand for cassava products, both for home consumption (mainly pig feeding) and for marketing, seems to be increasing. The greatest difficulty farmers face is the loss of productivity of the soil. This is thus a good opportunity to join with them in conducting on-farm testing of more productive and more sustainable management practices at these three pilot sites. 2. Demonstration Plots on Soil Erosion Control and Farmer's Field Days at the Agro-forestry College. Based on previous research on soil erosion control using various soil/crop management practices, a demonstration plot with 17 treatments was established on about 20% slope in the Agro-forestry College in Thai Nguyen in 1994. These plots gave a good picture of various alternatives for growing cassava on slopes, so farmers could select the most promising treatments during the farmer's field days. The data in Table 5 show the effect of treatments on cassava yield and soil erosion, as well as the economic returns and the farmers' preferences. Treatments with hedgerows of Tephrosia candida, Flemingia congesta and vetiver grass, or the intercropping with black bean and with Tephrosia hedgerows gave very good results with respect to soil eroison control; total soil erosion decreased between 47.7 and 61.9% as compared to the check plot (treatment 1). In treatment 10, with vetiver hedgerows, only 7.25 ton dry soil/ha was lossed, while cassava fresh root yield was maintained at 20.5 t/ha. In treatment 4, with balanced NPK and organic manure application, the cassava yield reached 26.6 ton fresh roots/ha, but dry soil loss was still high at 14.4 t/ha. After the results were discussed with farmers during the field day at the end of Nov, 1994, some treatments were selected for on-farm testing in 1995. In 1995, the same demonstration plots were continued with 16 treatments: treatment 6 was replaced by NPK and intercropping with peanuts, while treatment 15 was replaced by NPK and no tillage. The results of soil loss measurements and cassava yields are given in Table 6. They show the same trend as in 1994. Soil losses in treatments 6. 8, 9, 10, 11 and 13 decreased between 59.8% and 81.3% as compared to the check plot. Highest cassava yields were achieved in treatments 12, 4, 5, 16, 10 and 7. Gross and net income were the highest in treatments 12, followed by treatments 4, 5, 6, 10 and 16. 458 Table 1. Secondary data of Luong Son district of Hoa Binh, Thanh Hoa district of Phu Tho, and Pho Yen district of Thai Nguyen provinces in north Vietnam. Province Hoa Binh Phu Tho Thai Nguyen District Luong Son Thanh Hoa Pho Yen Latitude (°N) 20° 50' 21o30. 21° 20' Altitude (masl) 30-40 80 40 Soils dark red clay typic Paleustult red clay loam while sandy loam Rain fall (mm) 1600-1800 1700-2000 Temperature (°C) 23-25 22-24 Landscape hilly hilly rolling -rice in lowlands in lowlands in lowlands -cassava on <25% slope on <40% slope on <20% slope Main crops (ha) -rice (crop land) 4,690 3,580 7,200 -cassava 838 1,500 650 -sweet potato1' 402 670 1,400 -peanut 798 345 1,212 -maize 680 732 700 -mungbean 370 385 500 -tea 917 1,075 1,181 -forest 13,036 14,250 14,886 Yield (t/ha) -cassava 6.5-7.0 4 6 9.7 -rice 2.3-2.5 2 3 3.2 -sweet potato 5.5-6.0 4.5-5.0 5.5 -peanut 0.8-0.9 0.9-1.0 0.9 -maize 1.5-2.0 1.8-2.2 5.0 -soybean 0.5-0.7 0.6-0.8 0.7 Farm size (m2/fam.) -total 9,500-10,500 2,000-15,000 7,000-11,000 -cassava 1 ,000-2,200 1 ,500-2,000 720-1,080 -rice 600-800 500-5,000 360-6,400 -garden 0-5,000 0-5,000 1,800-5,000 -tea 0-3,200 0-2,500 -sweet potato 200-400 350-400 360-1,080 -trees 300-500 500-600 500-720 "mainly as winter crop on rice land. 459 Table 2. Population, land classification and land use, as determined from Rapid Rural Appraisals (RRAs) conducted in four FPR pilot sites in Vietnam in 1996/97. Province Hoa Binh Phu Tho Thai Nguyen District Luong Son Thanh Hoa Pho Yen Village Dong Rang Phuong Linh Population Tien Phong Dae Son -no. of households (hh) 81 275 -no. of hh. interviewed 40 158 6 35 -ethnic group Muong (90%)/Kinh" Kinh Kinh Kinh -no. of people/hh 6-7 5.2 5.3 6.1 -length of time settled >20 >25 >20 >20 (year) Land characterization -lowlands (ha) 27 86 -upland (ha) 95 (36 ha planted/yr) 29 -cassava (ha) 30 15 50 40 -slopes (%) <25 <40 <20 <10 -soil type, fertility clay, medium fertile clay, low fertility sandy loam low fertility sandy loam low fertility -erosion medium high medium low Land use -uplands cassava, taro, peanut, cassava, peanut, cassava cassava sugarcane, fruits sweet potato peanut sweet potato forestry tea, palm, bamboo sweet potato peanut soybean -lowlands rice (2 crops) rice (1 crop) rice (2 crops) rice (2 crops) peanut, sweet potato, fish in paddy sweet potato sweet potato maize, soybean, fish ponds fish ponds sugarcane (chewing) Area main crops (m2/hh) fish ponds -summer rice 2,500 3,700 4,500 3,100 -spring rice 3,100 2,160 2,460 2,063 -cassava 4,000 720 4,320 977 -sweet potato 300 500 1,080 954 -peanut 350 420 2,040 634 -soybean, mungbean etc 390 385 - 526 -tea 360 400 - - -garden 1,000 9,500 1,800 7,925 . Total farm size 11,940 6,191 Land ownership(%) 100 100 no. of parcels/hh 8 7 Animals buffalo buffalo buffalo buffalo cattle for meat cattle for plowing fish fish (± 100 head) fish fish "Kinh = lowland Vietnamese 460 Table 3. Cropping systems, varieties and agronomic practices, as determined from RRAs conducted in four FPR pilot sites in Vietnam in 1996/97. Province Hoa Binh Phu Tho Thai Nguyen District Luong Son Thanh Hoa Pho Yen Village Dong Rang Phuong Linh Tien Phong Dae Son Cropping system" -upland tea C monocult. C + P or C + B C monocult. or C+T C + P or 2 yr C rot. C-P rotation C monocult. tea, peanut with 2 yr fallow or C-B, C-SP Varieties peanut, maize maize sweet potato sweet potato -rice CR 203, hybrids DT 10, DT 13, DT 10, DT 13 CR203 from China CR203 CR203 DT 10, DT 13 -cassava Vinh Phu, local Vinh Phu, local Vinh Phu Du, Canh Ng Vinh Phu Cassava practices -planting time early March early March Feb/March Feb/March -harvest time Nov/Dec Nov/Dec Nov/Dec Nov/Dec -plant spacing (cm) 100x80 80x80; 80x60 100x50 100x50 -planting method horiz. /inclined horizontal horiz. /inclined horizontal -land preparation buffalo/cattle by hand/cattle buffalo buffalo -weeding 2 times 2 times 2 times 2 times -fertilization basal basal + sidea basal + side" basal + side4) -ridging mounding flat flat flat -mulching rice straw peanut residues peanut residues peanut residues -root chipping hand chipper knife small grater small grater -drying 3-5 days 3-5 days 2-4 days 2-4 days Fertilization -cassava -pig manure (t/ha) 5 5 3-5 8-11 -urea (kg/ha) 0 50-135 83 83-110 -SSP (18% P2Os) (kg/ha) 50-100 0 140 0-280 -KC1 (kg/ha) 0 0 55 0-280 -rice -pig/buffalo manure (t/ha) 5 0 - - -urea (kg/ha) 120-150 80 - - Yield (t/ha) -cassava 10-12 8-15 8.5 8.7 -rice (per crop) 3.3-4.2 4.2 3.0-3.1 2.7-3.0 -taro 1.9-2.2 - - - -sweet potato - - 8.0 3.3 -peanut 0.8-1.2 0.5-1.1 1.4 1.3 pigs (kg live weight/year) 100-120 - - - C=cassava, P= peanut, B=black bean, T=taro, M=maize C + P=cassava and peanut intercropped; C-P=cassava and peanut in rotation urea at 2 MAP urea when 5-10 cm tall; NPK +FYM when 20 cm tall NPK when 30 cm tall; hill up 461 Table 4. Labor use, crop utilization and farm income, as determined from RRAs conducted in four FPR pilot sites in Vietnam in 1996/97. Province Hoa Binh Phu Tho Thai Nguyen District Luong Son Thanh Hoa Pho Yen Village Village Dong Rang Phuong Linh Tien Phong Dae Son Labor (mandavs/ha) cassava rice cassava rice cassava rice cassava rice -land prepartion 30 30 44 35 28 30 27 29 -planting 20 30 45 25 20 25 20 25 -weeding 45 40 35 35 30 40 30 40 -irrigation - 10 35 - 10 - 10 -fertilization 10 20 20 10 17 12 16 12 -harvest 30 35 50 35 36 32 34 32 -chipping/drying 45 - 45 -thrashing 20 28 50 28 48 27 Total 180 185 239 203 181 177 175 175 Prices (done/ke) -fresh cassava 300-400 500 400 -rice 1,500 1,500 1,500 -peanut 3,500 3,300 3,500 -urea 3,000 3,100 2,700 -SSP 850 780 700 -KC1 2,500 2,700 2,500 Utilization (%) -rice 100% hh . use 100% hh. use 100% hh. use 100% hh . use -cassava -fresh roots 60% (60% to market) 15% (5% market) 20% (all home) 17% (all home) -dry chips 40% (pig feeding) 85% (5% market) 80 (10% market) 83 (5% market) -stems firewood firewood firewood -leaves fish or pigs fish or pigs Market use of cassava -fresh starch, maltose, alcohol alcohol - ■ -dry chips human food, pig feed pig feed alcohol pig feed pig feed Income (mil, dong/hh./year) -crops 3.5-4.5 5.4 -pigs, chickens, ducks 1.5-2.0 -others 1.0-1.5 1.5-2.5 6.1 3.1 0.13 08 7.4 3.4 3.2 2.3 3.0 1.1 0.2 3.5 Total 6.0-8.0 7-8 Costs and income (mil, dong/ha/yr) Gross income crops 3.7 Production costs crops 2.5 Net income crops 1 .2 0.134 Note: 1US$ is about 11.000 dong in 1996/97 •a S I e v ■o U I a v. 88 a D. I C o C 1 3 I c U S—' in 1 o.s a. £ .s III I■w mo^oomON- m m CMM» ' i i i oo 00 ' oo oo r- NON i Qs f-~ Os O m i f- Os i — 00 o — «NNnNNNNNN m C*^ <*s. C*) r*^ C^ ^ M M 00 >0 oor~-sosOON■oooooooo ooomTj-r~-Ttr-)' b J? SP " ■a a- wsb<»o 8 5 § S -5 S Z & O § « rSi '« J■ * > It, > X>,■«' o S ^. r-i e o •- A A A A A A A a. a- a- D- o. cl, a, zzzzzzz m NO r~ 00 Os o — Tj- m so r~ oo oo oo •O T3 T3 888 m o m p.i — - :3 S *© *© oo >n Tt ■* ao N ts*0 ■* m « « n n inN-xi.din^ri(-i«-• r*- ob ri v> >d onovtinminininvtinvtoininin o — -ir4-H^-i — -i — — — d — ~ — *oooooo'*©Tt'*N©©©©N* oo g in rn t— m r4 — o. r- 2 8 00 in 3 o oo mn mm oo N in on r-t^no(s—■ m © t^ r- — woo©t--© — -H-H .fr o .2 ■Ezuu+^ ex 3 w-> u. >js "e£^ c ° « c o E * -S S !S if if us" ii U U U U z US US US Z^^tuZZZZZZZZZZZZ 464 During the farmers' field day, farmers from the pilot sites had the opportunity to see, discuss and select the most appropriate treatments. Most of the farmers based their choice on two main criteria: the quantity of soil lossed by erosion and cassava yield. The largest number of farmers selected treatments 9, 6 and 10. Although treatment 12 produced the highest net income, only 32.4% of farmers considered this treatment very good. 3. Farmer Participatory Research (FPR) Trials in 1995. Four different components of technologies were tested on farmers' fields: - erosion control - intercropping systems - fertilization - varieties As a result of the farmers' field day held at the Agro-forestry College in 1994, farmers selected the treatments that they wanted to test on their own fields. A total of 35 farmers volunteered to conduct FPR trials: 18 in Tien Phong and seven in Dacson villages of Pho Yen district, and ten in Phuong Linh village of Thanh Hoa district. 3.1. Results ofsoil erosion control trials On sandy loam soils of two villages in Pho Yen district, having slopes ranging from 10 to 25%, the results of six FPR trials show the same trends. Treatments 4 and 5 reduced soil losses between 51.1 and 59.9% as compared to the farmers' practices of growing cassava in monoculture. The reduced erosion is due to the hedgerows and the intercrops. However, the cassava yield was only 10.5 t/ha in treatment 4 and 12.1 t/ha in treatment 5, which are lower than that of the farmers' practice. But these treatments also produced between 0.15 to 0.3 t/ha of peanuts (in dry pods), so the net income of treatment 5 was among the highest while the cost was not as high as that of the farmers practice (Table 7). In Phuong Linh village of Thanh Hoa district in Vinh Phu province, ten farmers participated in the project to conduct trials on their own fields with slopes ranging from 32 to 45%. The results, shown in Table 8, indicate that all treatments with hedgerows reduced soil erosion between 17 and 21% as compared to the check plot, while the cassava yields were between 3 and 29% higher than in the check plot (treatment 2). A simple economic analysis of the trial indicate that treatment 6 produced the highest net income of 23.57 mil. dong/ha, while the farmers practice in treatment 2 produced only 11.13 mil. dong/ha, which is less than half of that of treatment 6. After discussing the results of the FPR trials, nearly all farmers in Thanh Hoa and in Pho Yen selected the treatment of cassava intercropped with peanut, with NPK application and with vetiver grass hedgerows as the best management practice. jgI o c s cI I c B s n> -9 j 11 e o .2 «Hh ■B J an BQ ■a 1 1 Go r-" D. M 1 S -a 00 C O s m o o m c o rl r- oo 00 tN * an n rl ss 00 00 r- g 00 o oo * r- to Tt * 00 c> 3 8 —i oor~ oo oc ri 2•J •> » « 8.s -2 90 ui£ iM H H o r-1 §8 ip in Oj O h; n « o * >d r- od -h d —. —. —c —. —. N r» r-; f> -- ci Tt On pi ci 't 6 in n N inID IT> 1D ■*Tt•<* ■* in © r- r- cs t- S v> n on n r-^ ^ M ci ci Oil c o fi ti t; t: t; S 3 a> a> a> a> 3jj C C P4oKo+10 t FYM 13.87 11.10 1.68 9.42 - N8oP4oK8o+ 10 t FYM 15.65 12.52 2.01 10.51 81 "Prices : fresh cassava roots urea (45% N) SSP (17% P2O5) KC1 (60% K2O) pig manure 800d/kg 2,500 d/kg 1,000 d/kg 2,500 d/kg 100,000 d/t 2,Percentage of farmers choosing treatment. 3)On average applied 20.8 t pig manure, 37.1 kg P2Os and 137.5 K2O/ha 469 3.4. Results of variety trials In 1995, two promising clones, i.e. KM60 and CM4955-7, were evaluated on farmers' fields. Table 11 shows that these clones had fresh root yields between 1 1.6 and 28.5% higher than the local check variety, Vinh Phu. Among 18 farmers who evaluated these trials, 61.1% selected KM60 and 77.8% selected CM4955-7 for planting next year. Table 11. Average results of three FPR cassava variety trials conducted in Tien Phong village of Pho Yen district, Bac Thai province of Vietnam in 1995. (million dong/ha) Cassava Variety yield Gross Net Farmer (t/ha) income income" preference2' Vinh Phu 14.30 11.44 9A3 Vl CM4995-7 18.37 14.70 12.69 78 KM-60 15.96 12.77 10.76 61 "Cost of fertilizers + manure = 2.01 million dong/ha. 2)Percentage of farmers preferring the variety. 4. Farmers' Field Days The first farmers' field day was held for farmers from both pilot sites at the Agro-forestry College in Thai Nguyen, to evaluate the demonstration plots and to discuss the technology components to be tested on their own fields in 1995. The second farmers' field days were held both at the pilot sites and at the Agro- forestry College, one day in each location. A very important aspect of FPR is that farmers, researchers and extensionists work together in conducting the trials. Thus, all of us joined and worked together during harvest, evaluation and the discussion to select the best options and to decide on the most appropriate technologies to be included in the design of the work plan for the next season. Farmers' preferences are presented in the last column of Tables 7, 8, 9, 10 and 11. Based on this, the following technologies were considered as best adapted to the farmers' conditions (Tran Ngoc Ngoan et al., 1995): 1. Cassava intercropped with peanut, with application of 60 kg N + 40 P205 + 120 K20 + 10 ton pig manure/ha, and with vetiver grass hedgerows to reduce soil erosion and increase net income. 2. Cassava intercropped with peanut with a basic fertilization of 10 ton pig manure, 80 kg N + 40 P205 + 80 K20/ha for soil conservation and increasing net income. 3. Both KM60 and CM4955-7 were selected for the next evaluation. 470 5. On-going Work in 1996. These initial results are encouraging more and more farmers to participate in the FPR trials. The number of farmers participating in 1996 increased to 57 households, of which 37 in Pho Yen district, 13 in Thanh Hoa district and 7 in Luong Son district. Also, some treatments have been adjusted in accordance with the farmers' own ideas, such as the use of the new variety KM60 in combination with treatments on soil erosion control. In 1996, the variety trials increased both in number of farmers participating (20 farmers) and in the number of promising clones tested (5 promising clones/trial). CONCLUSIONS After two years of work and some initial results, we have learned a lot about the use of the FPR approach, and farmers also know why we need to work more closely together with them; they are encouraged to contribute their local experiences while working together as a group. FPR is very useful for us, because by conducting this type of research together with farmers on their own fields, other problems can be identified. We can get feedback information about what farmers want to do and what they don't want to do. A strong linkage between researchers, farmers and extentionists in the technology development process was established. The learning from each other, while working closely with farmers, will increase the adoption of new technologies by farmers and will also improve the relevance of the research conducted by us. It is also encouraging the process of farmers learning from farmers. However, while working with farmers we need to learn more about the FPR methodology and improve our skill in working with different groups of farmers. REFERENCES Howeler, R.H. 1981. Mineral Nutrition and Fertilization of cassava (Manihot esculenta Crantz). CIAT Series No. 09 EC-4 Centro International de Agricultural Tropical (CIAT), Cali, Colombia. 52 p. Tran Ngoc Ngoan, Nguyen The Dang, Le Sy Loi and Dinh Ngoc Lan. 1995. Variety selection and technological management of cassava to maintain soil fertility and high income in the mountainous area. Final Report 1991-1995. (in Vietnamese) 471 FARMER PARTICIPATORY RESEARCH IN SOIL MANAGEMENT IN INDONESIA W.H. Utomo1, Suyamto2, H. Santoso1 and A. Sinaga1 ABSTRACT Soil degradation, both due to soil erosion and nutrient removal, is a major problem in cassava fields. Most of the existing management technologies that have been developed have had little success in adoption. Some of the reasons are that the technology developed is technically oriented, based on experiment station research with very little farmer involvement. Research on "Farmer's participation in developing management technologies for cassava- based cropping systems" was carried out in farmers' fields in Wates and Dampit subdistricts of Blitar and Malang districts of East Java, respectively. The research is planned for five years, with the first year activities of : (a) Rapid Rural Appraisal (RRA) and diagnostic surveys for identification of the problems encountered by cassava farmers, and (b) establishment of demonstration plots to test and to show to farmers the management technologies that have been identified and selected by the collaborating farmers. Activities planned for the second and following years include the testing of the most attractive technology options on farmers' fields by farmers themselves. In general, most farmers had already been aware of the problem of soil degradation in their fields, and had tried some management technologies for overcoming this problem. However, since the technology is too complicated, laborious and costly, they were unwilling to adopt. In fact, most farmers were very keen to adopt any attractive new technology, if the technology is simple, does not imply a lot of extra costs, and is easy to be adopted. This was indicated by the number of farmers willing to participate in the second year to test some technologies in their own fields. INTRODUCTION Soil deteriotation and land degradation are common and serious problems in Indonesian cassava fields. Therefore, a lot of people believe that these problems arose from cassava cropping. With its high dry matter yield, cassava removes a lot of plant nutrients from the soil. In addition, a high erosion rate, especially during the early growth phase, is often observed from cassava fields. It is true that the soil in most cassava fields is in very poor condition. However, the soil deterioration is not only due to cassava. In fact, the soil has become in such a critical condition, that cassava is the only crop that can grow and produce a reasonable yield. For example, of the 300,000 ha planted to cassava in East Java more than 80% is in critical condition. Realizing the problem, many soil management technologies have been developed, both to control soil degradation and to improve soil productivity. However, cassava 1 Brawijaya University, Malang, E-Java, Indonesia. 2 Research Institute for Grain Legumes and Tuber Crops (RILET), Malang, E-Java, Indonesia. 472 farmers have hardly adopted these technologies. If they apply the technology, the adoption process takes a very long time, and the application of the technology is only temporary. This is not due to the lack of understanding of soil erosion and soil management. A lot of studies have indicated that in most areas of marginal land, farmers are already well aware of the problems of soil erosion, soil degradation and of ways to overcome these problems (Utomo et al, 1994). However, technology adoption is another matter. In most cases, farmers are reluctant to adopt the technology developed by researchers, because the technology is too costly, too complicated or does not give direct benefits. It seems that most researchers developed the technologies based on their own ideas, without considering the farmers conditions and requirements. In addition, most of the technologies are technically oriented and developed under experimental station conditions. Thus, as stated by Henry (1994), researchers used a top-down approach. Some sociologists and anthropologists (Fujisaka, 1989; Saragih and Tampubolon, 1991) have suggested that to develop a more adoptable technology, we should take into account the farmers' conditions and farmers' requirements. Thus, the technology should be more farmer-oriented. The methodologies used include what are called "On Farm Research" and "Farmer Participatory Research" (FPR). Henry and Hernandez (1994) have shown that FPR is very effective for the dissemination of new cassava varieties in Colombia. The potential of FPR in soil management technology transfer has been discussed extensively by Fujisaka (1991). The work reported here aims to study the effectiveness of FPR for soil management and varietal dissemination to cassava farmers in East Java, Indonesia. The research was carried out at Ringinrejo village, Wates subdistrict of Blitar district, and in Sumbersuko village, Dampit subdistrict of Malang district, and was planned for five years. The report describes the first and second year's results. FARMER PARTICIPATORY RESEARCH The research started with a Rapid Rural Appraisal (RRA) to identify the problems of cassava farmers, and to learn about the farmers' ideas on soil management. Based on this discussion, demonstration plots were established to show a wide range of prospective soil management technologies, after which farmers selected and tried some soil management technologies in their own fields. The project staff helped farmers to conduct this research. Rapid Rural Appraisal (RRA) A Rapid Rural Appraisal was carried out in August 1994 in Blitar and in Sept 1994 in Malang district. A team consisting of two soil scientists, one agronomist, one agricultural socio-economist and one soil conservation extensionist (from Brawijaya University, RILET and the Institute for Soil Conservation and Land Rehabilitation, 473 respectively), interviewed key farmers in the two villages covering the following topics: a. Soil and land conditions b. Agricultural practices (crops and varieties used, spacing, cropping systems, fertilization, etc.) c. Soil conservation and management practices d. Socio-economic conditions The main purpose of the RRA is to learn about the awareness and the perception of the target farmers concerning soil degradation problems, and their understanding and ideas about soil conservation and management practices. To obtain more detailed baseline data, a diagnostic survey was conducted in Sept 1995 in Blitar and in Nov 1995 in Malang. The results of the Rapid Rural Appraisal (Table 1) indicate that most farmers in the study area are aware of the problems they encounter. The farmers know well that their soil is in a very poor condition and a lot of work needs to be done to get a reasonable yield. They are also well aware that soil degradation due to soil erosion is occuring very rapidly. Hence, soil conservation, both the preventing of soil erosion and the improving of soil fertility is essential. The problems are that the land area owned by individual farmers is very small, about 0.40- 1 .0 ha/household (also notice the results of the detailed survey given in Table 2). Hence, the income from farming activities is very low (around Rp 600,000/year/household). With this low income it is imposible for farmers to worry about erosion and to manage the soil more effectively. Actually, the farmers are very eager to practice better soil management in their fields. However, most of the recommended soil management technologies are very expensive and require constant maintenance. In addition, some soil management techniques (such as alley cropping) are complicated and require a lot of time and labor for the technology to work well. At certain times during the year, especially during the dry season and after planting, many people go to the city to earn additional income. Therefore, there is often not enough labor available to look after the soil management technology they want to practice. Although a lot of farmers raise livestock (Table 2), they do not like to include elephant grass barriers in their crop area. It seems this is due to strong competition between the main crop and the elephant grass. Some farmers use leaves of Gliricidia sepiwn which is planted as a fence, for livestock feeding, especially during the dry season. Demonstration Plots Based on the information obtained through the RRA, a set of soil management technologies was tested in the demonstration plots established both in Ringinrejo village, Wates subdistrict of Blitar district, and in Sumbersuko village, Dampit subdistrict of Malang district, both in East Java. 474 The soil management technologies tested in Blitar and Malang are given in Tables 3 and 4, respectively. All experimental activities, including land preparation, planting and the taking of measurements, were done by the farmers, which were coordinated in a farmers' group in each site. These farmers' groups consisted of 33 farmers in Ringinrejo and 23 farmers in Sumbersuko. The data given in Tables 3 and 4 show that some introduced soil management technologies produced better results, both in terms of increasing farmer's income and decreasing soil loss, than the traditional farmer's practice. Even though the alley trees were still small in both sites, the practice of alley cropping was already able to decrease soil losses due to erosion. To obtain the perception of the farmers' group and of their neighbors to the soil management technologies tested, field days were organized on April 28 and Sept 3, 1995, for the Ringinrejo site, and on Oct 24, 1995 for the Sumbersuko site. In addition, to broaden the view of the participating farmers, another field day was held at Jatikerto Experimental Station of Brawijaya University in Jatikerto, Malang. Farmers evaluated and discussed the soil management technologies that have been tested for many years at this experimental station. The opinion and selection priorities of farmers to the soil management technologies are given in Table 5. It seems that the opinion and preference of farmers to any technology is based mainly on what they see. They did not yet know the cassava yield at Jatikerto, but the field showed that cassava planted with Gliricidia contour strips grew very well. Hence, they choose this technology. Although there was clear competition between the elephant grass and cassava, as indicated by poor cassava growth close to the elephant grass strips, this technology is still preferred by some farmers. These farmers argued that they need elephant grass to feed their livestock. Crop competition could be decreased by the application of manure. They also saw that the soil loss in the treatment with elephant grass strips at Jatikerto was very low. Some farmers in Ringinrejo still preferred planting cassava intercropped with maize using the farmer's traditional wide spacing of cassava rows because they can use more space to plant maize and other crops after maize. These farmers use maize as the main family food. 475 Table 1. Results the of Rapid Rural Appraisal (RRA) conducted at Ringinrejo village, Wates subdistrict of Blitar district in East Java, Indonesia, in Aug 1994. Key issues Conditions Farmer perception Limiting factors Soil - Thin top soil with - They understand that soil - Lack of capital underlying limestone should be improved - Difficult to obtain - Low nutrient content (esp. N, manure P, K and some micronutrients) Cropping a.Species b. Variety c. Spacing d. Planting date e. Fertilizer f. Yield Maize as the main crop, cassava as intercrop Maize: local, Arjuno or Hybrids Cassava : local or Faroka Maize: 100x20 cm Cassava: 100x200 cm Start of rainy season Urea Maize: 2-3 t/ha Cassava: 5-8 t/ha - Maize is the main food, cassava is only a security food or for cash - Farmers plant the variety available - Farmers are able to grow a second intercrop after maize Do not apply enough Yields are extremely low High fluctuation of cassava prices Lack of information about new varieties Beginning of rainy season is unpredictable Lack of capital Soil and Land Conservation a. Tree crops - Very few tree crops b. Terracing c. Contouring d. Strip cropping /alley cropping - Improper terracing - Some farmers practice contour ridging - Gliricidea is used as a fence Farmers understood well the importance of soil conservation Technology for soil conservation is too costly and too complicated Gliricidea and elephant grass used as animal feed Lack of capital Lack of labor for maintenance of terraces Farm size - 0.60 ha/household - Too small to obtain enough income Land distribution - Unequal land area - Need a lot of time and labor- Fragmented land holdings Farmers income - Rp 600.000/household/year" Mobility - Many people move to the nearest city temporarily to earn additional income - Limited work opportunity at at the village "1USS = Rp 2100, 476 Table 2. Characteristics of the two project sites. Characteristic Ringinrejo(Blitar) Sumbersuko(Malang) 1. Altitude (m above the sea) 420 500 2. Rainfall (mm/year) -1500 >2000 3. Soil Lithosol (Entisol) Latosol (Oxisol) 4. Topography undulating to hilly undulating to hilly 5. Total area (ha) 2,232 1,057 a. Upland (legal) 552 835 b. Plantation crops 857 - c. Forest 159 - d. Others 664 222 6. Crops (ha) a. Upland rice 115 36 b. Maize 292 425. c. Cassava 145 725. d. Soybean 60 - e. Peanut 30 - 7. Total Population 3,420 1,191 a. Farmers 3,283 641 b. Others 137 550 8. Livestock ownership a. Cow/household 9. Income (US $/h.h./year) a. Farming b. Others 1.36 250 500 0.6 550 300 .a g. 1 i/-, 8 ■V■a. * c ■* o•a S«.- .3 § go e 1 o■o■o c« w 5 .3 cs I -^ 3 OS UJ s J a *d> .k u id .■£ u X!(*. O U fS■o a .11 as ^ *—■o 5 ■ 1 = 5 f 1 3 1 H fit a H moooininininviininm n — — — rit--r-r-r-r-t~r- s n n « in ^ n ot i/"i-<*i/^r"ic>w->->t O p- ri r- « * m n O 00 ^ pi ci n vj in N M t■* n r^ m 8 o oo o o in — c-on o\ in m oo Tt n t—otsv>-Hinr-r-in*rJt~ Nin't^itciciNnnNn a o U U U O •Hrini-vi.dr-ooond^N eg q 7 O It <§■ t 8S — U I H > v u Ca M — -2 j= 3 - Si2'S w <**> oo r-- c> op —■ oc r-io >*-^ '* o o — © cs — ~* -" - N oov">o -H -H ri m rs rn rl rs ri © © Tt Tt *t *t r^ oo *o ri in ri 8 2 - 9 "=. — — 60 . g> io > i I 5 > „ I S 5S J3 cm « c > S 60 6fi ■£ -c c c o o E E c a. ,3,5S 5P •c | | c .So .So ~ -c c E o o o c c <*- 00 o p SI i1 o 2 o s « •S. 5 5 ■ S - ~ 8 i 2 9 -O .5 -c 60 5 ■c >• ■ ed E .a o a. P3 tt W rr > > > "E S 3 ^ it § 3 £ 5 =1 111 1 .s s "O '« ^3 °r 'V c c. c- -5 =3 c a u + + s s + + u u 9 P 4> to en u is 60 -c■o g al ' ie C Si O 3<■> -Sf ■S 3 f-J; 5 + + S S + + u u | 5 s > U M »•& .5 T3 =B § o . c E e!■* s •> u > KI 2 ^ 1 a 2 5. 1 'i o u ^ y 2 o « 8" 0. H * o o *ft- <^ - a ■* o o .. B. 2 S 3 o J3 *> o « ^ E « f u U 8 . H H rl " o Ts - s. " ™ e if * © o - O o N ^ — v-i t- "7"■ O ej 60 O 3 « 2j « 1 s 1 1 19 tu w e 3 P X o c «; 24./kg 340/kg 340/kg 5 **^ 60 .. 2U O . U .S * E E B, a u. 3 E 5 -2 o X c c .5 -a 2 .c E g M II « "O s . 2 a « 3 a U S & U S 0. U a e e 2 a s 3 3 3 c c c o O 0 s s s o o o + + + 5 2 2 + + + o o u ■o -a « .3■O 73 -— 0) £ 479 Table 5. Farmer preference and opinion on soil management technologies demonstrated in Ringinrejo (Blitar) and Sumbersuko (Malang) as well as at Jatikerto Experimen Station in Malang, E. Java, Indonesia. Farmei Soil management Net income Soil loss Farmers. opinions preference treatment C000 Rp/ha) (t/ha/year) 1. Gliricidia contour strip 1,509" 39.6" a little"' - Cassava grows very well5' - Soil loss is low - Gliricidia leaves can be used for livestock feed 2. Flemingia contour strip 1,451" 40.3" a little41 - Cassava grows very well" - No competition - Flemingia leaves are good for feeding goats 3. Elephant grass contour 1,737" 42.9" - Altough there was strong strip none"' crop competition, the soil loss is rather low - Elephant grass used as livestock feed 4. Calliandra contour strip 1,574" 38.6" - Calliandra leaves are good for livestock feed - Soil loss is low 5. Cassava + Maize (farmers. practice) 7242) 41.12) - Need a wide cassava spacing to plant more maize 6. Farmers. practice 4722' 47.42> - 7. Farmers' practice 1,219s' 72.3" - "Date are average values of Ringinrejo and Sumbersuko demonstration plots 2>Ringinrejo demonstration plots "Sumbersuko demonstration plots "'Soil loss observed at Jatikerto Experiment Station "The opinions on cassava growth were based on observations made in Jatikerto Experiment Station Participating Farmers' Experiments After evaluating the soil management technologies tested at Ringinrejo and at the Jatikerto Experiment Station, 15 farmers in Ringinrejo participated in the FPR project by testing some selected technologies in their own fields. The technologies tested in the farmers' fields are given in Table 6. Nine farmers conducted erosion control trials and six farmers participated in variety trials. In Sumbersuko, experimental activities in the second year still focussed mainly on the improvement of the demonstration plots. 480 Table 6. Soil management technologies tested by farmers in Ringinrejo in nine erosion control and six variety trials in 1995/96. Soil management Number of technologies farmers" 1. Erosion control trials: - Gliricidia contour strips 7 - Calliandra contour strips 3 - Flemingia contour strips 2 - Leucaena contour strips 1 - Elephant grass contour strips 2 - Intercropped with cowpea 2 - Control without strips 7 2. Variety trials: - Faroka 6( + 6):> - 15/10 5(+D - SM-4772 6(+D - Local(Ketela ijo) l(+4) " Number of farmers including the treatment in their FPR trials 2) The number in brackets shows the number of farmers doing erosion trials in which the variety is included As in the first year all experimental activities were done by the participating farmers. Project staff acted as coordinators and provided materials for the experiments. In Ringinrejo all participating farmers used 5000 kg manure/ha and applied 300-375 kg urea, 150-200 kg TSP and 100 KCl/ha in all treatments. The yield and income of each treatment tested by participating farmers are given in Table 7. Due to high variability between farmers' fields in the eroison trials, a valid comparison between treatments is difficult. From the variety trials it can be concluded (Table 7) that the two introduced varieties, 15/10 and SM-4772, have a good prospect. To discuss the results and obtain the views of the participating farmers, field days were held on August 15, 1996 in Sumbersuko and on September 19, 1996 in Renginrejo. In addition to the participating farmers, some neighbor farmers were also invited and participated in these field days. Farmers discussed the experimental results as well as the future of the project. For the third year, there will be 22 farmers in Ringinrejo and 15 farmers in Sumbersuko participating in the project, testing some selected technologies in their own fields. 481 Table 7. Average yields of cassava and maize as well as farmers' net income in nine FPR erosion and six variety trials conducted by farmers in Ringinrejo in 1995/96. Treatment Yield(t/ha) 1. Erosion control trials: a. Gliricidia strips(Faroka)11 b. Gliricidia strips(15/10) c. Gliricidia strips(SM-4772) d. Calliandra strips(Faroka) e. Calliandra strips( 15/ 10) f. Flemingia strips(Faroka) g. Leucaena strips(Faroka) h. Elephant grass strips(Faroka) i. Cowpea intercrop(Faroka) J. No strips(Faroka) k. No strips (Local/Ketela ijo) 2. Variety trials: a. Faroka b. 15/10 c. SM-4772 d. Local/Ketela ijo Net income (Rp/ha/year)Cassava Maize 5.6 1.1 1,387 8.9 0.6 868 9.0 0.7 731 6.3 2.2 2,453 8.2 1.5 791 5.6 0.9 1,191 2.2 0.8 707 3.4 0.9 1,052 1.5 0.6 520 4.9 1.2 1,778 5.5 1.0 1,307 4.2 1.1 1,075 9.4 1.0 1,238 5.1 0.8 1,084 5.2 1.2 581 "Cassava variety used in treatment REFERENCES Fujisaka, S. 1989. A method for farmer participatory research and technology transfer: Upland soil conservation in the Philippines. Expt. Agric. 25:423-453. Fujisaka, S. 1991. Improving productivity of an upland rice and maize system: Farmer cropping choices or researcher cropping pattern trapezoids. Exp. Agric. 27:253-261. Henry, G. 1994. Importance of the user perspective in agricultural research. Sasakawa Project FPR Workshop, held in Rayong, Thailand. July 24-30, 1994. Henry, G. and L.A. Hernandez. 1994. FPR case study: Cassava variety selection in Colombia's North Coast. Sasakawa Project FPR Workshop, held in Rayong, Thailand. July 24-30, 1994. Saragih, L. and S.M.H. Tampubolon. 1991. Integrated field level participatory policy to promote soil and water conservation programs and project. Soil and Water Conservation of Indonesia. Solo, Indonesia. Utomo, W.H, 1994. Factors Influencing Soil Conservation Technology. Research Report. Dept. of Education, Brawijaya University, Malang, Jakarta, Indonesia. (in Indonesian) 482 FARMERS' PARTICIPATION IN CASSAVA TECHNOLOGY TRANSFER IN THE PHILIPPINES Editha A. Gundaya and Fernando A. Evangelio1 ABSTRACT Experiences gained from past root crop extension activities made PRCRTC realize the importance of clienteles. participation in technology development and transfer. PRCRTC is now trying to adopt a participatory approach in all cassava technology transfer activities. Results of some of these extension activities shows that farmers. involvement in technology testing and modification and in the over-all decision making process of a project, led to the development of technologies that are better suited to the needs of the clienteles and helps to build up their capacity to manage the project on their own. In 1996 a project was initiated that employs farmer participatory research (FPR) methodologies to transfer soil and water conservation technologies to cassava farmers in Bontoc, south Leyte. This is a cassava growing area where cassava production is increasing due to the establishment of a cassava-based feedmill, as well as the entry of San Miguel Corporation, which buys large volumes of dried cassava chips for export. A preliminary survey about the farmers' knowledge, attitude and practices concerning soil conservation has already been conducted with 91 cassava farmers as respondents. The majority (75%) of the farmers realized the damage that would occur on continuously cultivating hilly areas, but only less than 50% are actually trying to control erosion using various methods they learned from different sources. Around 65% of the respondents, however, expressed their interest in learning soil conservation methods that would be more effective than the methods presently used. The information given by the farmers is now used as the basis for conducting other FPR activities in Bontoc. INTRODUCTION The increasing interest in cassava by the business sector offers opportunities to further the development of the cassava industry in the Philippines. At present, the crop is used not only as food by many Filipinos, but also as raw material in the manufacture of several commercial and industrial products, like chips for export, starch, adhesives, binders, feed and various food products. Some private companies are also looking into the possibility of using cassava as raw material in the massive production of alcohol, glucose and sorbitol. Despite these opportunities, the cassava industry in the country has been growing slower than what is desired, because it has been hindered by several problems. These problems include low crop yield, poor processing facilities, low product quality, low product price and lack of domestic markets. The Philippine Root Crops Research and Training Center (PRCRTC) has been supporting the cassava industry in the country through the development and transfer of cassava technologies, such as high-yielding cassava varieties, improved cultural 1 Philippine Root Crops Research and Training Center (PRCRTC), Visayas State College of Agriculture, Baybay, Leyte, Philippines. 483 management practices, cassava storage technologies, processing machines, feeds and different food products. However, the Center's past extension activities had limited impact on the cassava industry, especially on farmers. Acceptance of the technologies introduced was rather low and sustainability of technology adoption, especially among farmers and small processors, was hardly obtained. Several factors caused these problems. One is too much focus on the technology itself, disregarding other factors that greatly influence technology adoption, such as market, the entrepreneurial and organizational capabilities of the technology users and others. Another constraint was the farmers' lack of participation in the technology development and transfer process. Farmers' participation was limited to being mere recipients of the technologies developed by the research institutions, respondents in some baseline surveys, participants in some formal consultation meetings, or cooperators of some on-farm trials, which were basically designed by the researchers. According to Werner (1993), nobody has a better understanding of his different needs and the opportunities his farm offers than the farmer himself. Thus, the farmers' lack of participation in the technology development and transfer process often resulted in technologies that did not correspond with their needs and conditions. With these realizations, PRCRTC is now trying to improve its technology transfer approaches. Efforts are geared towards improving farmers' participation in the technology development and transfer process. PRCRTC is also trying to incorporate market development, organizational/entrepreneurial build-up, linkage establishment and other neccessary components in the Center's technology transfer efforts. All these are being considered in the PRCRTC's current cassava technology tranfer program (funded by PCARRD-DOST), which focus on the expansion of cassava markets through the establishment of integrated cassava projects. This paper will discuss how the farmers are being involved in the implementation of PRCRTC's cassava technology transfer projects in northern Mindanao and in Bontoc, southern Leyte. This will also present the problems encountered and the possible ways of increasing farmers' participation and/or improving the efficiency of project implementation. Moreover, this paper will present the status of the FPR project on soil and water conservation, which PRCRTC has started to conduct late this year as part of its integrated cassava project in Bontoc, southern Leyte. Direction of PRCRTC's Current Cassava Technology Transfer Activities Considering the present status of the cassava industry in the country, PRCRTC's cassava extension activities are geared towards: 1. improving the farmers' cassava production level and processing efficiency in areas with existing cassava markets; and 2. promoting cassava commercialization/market expansion in cassava growing areas which are far from the existing cassava industries to encourage farmers to improve their cassava production level and enable them to increase their incomes. 484 PRCRTC's Ongoing Cassava Technology Transfer Activities The Center currently focuses its technology transfer activities in northern Mindanao and the Visayas. The work in northern Mindanao is done to help the cassava farmers improve their cassava production level and chips processing efficiency. The project in the Visayas, on the other hand, are intended to expand the market for cassava through the establishment of integrated cassava projects that can create sustainable livelihood opportunities for the people. PRCRTC is currently assisting three integrated cassava projects in the Visayas. Of these three projects, the Integrated Cassava-Feedmill- Livestock Project in Bontoc, southern Leyte, was established first and is already operational. Thus, the discussion of PRCRTC's experience in involving farmers in technology transfer in the Visayas will be based on the Bontoc experience. Farmers' Participation of the PRCRTC's Cassava Technology Transfer Activities The cassava technology transfer projects in northern Mindano and in Bontoc, southern Leyte, are implemented by multidisciplinary action teams and in coordination with several agencies, i.e. the Department of Agriculture, Land Bank of the Philippines, Department of Science and Technology, etc. Phasing of activities for each area differed due to the differences in the nature of their problems and concerns. However, the activities followed the same pattern of: 1. micro-level diagnostics (analysis of problems and identification of possible solutions); 2. identification of project intervention and technologies to be introduced; 3. identification of technology transfer strategies to be used; 4. drawing up of detailed implementation plans; 5. actual project implementation/technology transfer; and 6. monitoring and evaluation. In the first four stages, farmers' participation was made possible through joint visits and appraisals of the project sites, consultation meetings, and formal and informal group discussions. * In the actual project implementation/technology transfer, the farmers were asked to participate in the following activities: 1. establishment of demonstration farms; 2. trainings; 3. consultation meetings; 4. evaluation of the chipping machines; and 5. the conduct of simple experiments. The farmers participation in the different technology transfer activities are further explained based on the experiences in the implementation of the projects in northern Mindanao and in Bontoc. I. Cassava Technology Transfer in Northern Mindanao Background information Northern Mindanao is one of the most important cassava producing regions in the country. It is presently the site of rapid expansion of cassava production area due to the presence of a number of starch mills and the increasing chip export activities by some big companies, like San Miguel Corporation (SMC), CAPICAOR and GUANI 485 Marketing. In fact, financial assistance for the growing of cassava has been made available by CAPICOR, the Land Bank of the Philippines, and recently by the SMC Agribusiness Division. Results of the diagnostic surveys revealed that despite these developments many farmers are still complaining about having low income from cassava because of the low yield of their native varieties and the low price offered by the chips exporters. They also complain about the absence of processing machines that can improve their processing efficiency and the quality of their chips. The exporters, on the other hand, are also complaining that the dried cassava chips produced by the small cassava farmers/ processors do not meet the required quality of the importing firms in Europe, resulting in high price discounts imposed on a number of their deliveries. The above situation led PRCRTC to implement an action project that would deliver the needed package of assistance/technology to the people in the area to make cassava production and processing profitable for the farmers and at the same time improve the quality and acceptability of dried chips. The technologies introduced to the farmers include high yielding cassava varieties, improved cultural management practices and chipping machines. In the conduct of the different technology transfer activities, PRCRTC worked with San Miguel Corporation, which had earlier sought the Center's assistance in looking for cassava suppliers, and with other support agencies. During the consultation meetings with the different agencies concerned, it was agreed that SMC would do the legworking and the organization of farmers, Land Bank of the Philippines (LBP) would be tapped to finance the cassava production activities of the farmers, while PRCRTC would provide the needed technical assistance in the propagation of planting materials of the recommended high-yielding cassava varieties, training farmers on the improved cassava cultural practices and chips processing, and in the fabrication of the needed chipping machines. The farmers, on the other hand, were asked to participate in the establishment and management of demonstration/model farms, technical training activities on cassava production and processing, and in the testing/evaluation and possible modifications of the chipping machines. Establishment of model farms To enable the farmers to actually observe the advantages of the high yielding varieties and the improved cultural practices, and to have propagation areas for the recommended high yielding cassava varieties, ten model farms (1.0 ha/site) were established in strategic sites in Misamis Oriental, Bukidnon and Lanao provinces. These model farms were managed by farmers; PRCRTC only provided the needed technical supervision. Lakan, Golden Yellow, VC varieties (1 to 5) and other high-yielding lines were planted in the model farms. The farmers' native varieties and their usual cassava 486 cultural management practices were used as the control treatment. Selection of the best performing varieties were later conducted jointly by the farmers, the PRCRTC staff and San Miguel technicians. Other cassava farmers were also asked to participate in the selection process. Farmers liked all the recommended varieties as these gave yields which were higher than those of their native varieties. Considering chips recovery, Lakan and Golden Yellow were choosen by the farmers because of their high dry matter content and high chips recovery. Training on cassava production and chips processing To reinforce the knowledge gained by the farmers from the demonstration farms, training on cassava production and chips processing was also conducted by technical experts from VISCA. Fabrication and testing of chipping machines To support the chips processing activities of the farmers, SMC is currently fabricating 15 units of pedal-operated chippers using the PRCRTC design. These machines will be distributed to the farmers for field evaluation and for actual use by farmers if found acceptable. SMC also plans to produce motorized chippers using the PRCRTC design. Project status As of 1996, the total cassava expansion area in the three sites already reached 3,000 ha. Although all farmers liked the recommended varieties included in the model farms, Lakan, Golden Yellow and VC-5 were used in the initial cassava production expansion because planting material of these varieties was more readily available. In the farms that use the recommended varieties and the management practices recommended by PRCRTC, the average cassava yield ranged from 24-40 t/ha. Before the introduction of the PRCRTC technology, the farmers were planting native varieties which only gave them an average yield of 6-8 t/ha. Moreover, the chips processing efficiency of the farmers is already improving with the use of the chippers provided by SMC. However, a farmers' evaluation of the field performance and life span of the machines still needs to be done. II. Cassava Technology Transfer in Bontoc, Southern Leyte Background information Results of the micro-level diagnostics conducted in Bontoc revealed that the municipality had a big unutilized hilly area suited for cassava production. Upland farmers were growing cassava because it was one of the few food crops that could thrive under the marginal conditions of the hilly area. However, the crop was grown only in 487 small patches not exceeding 0.25 ha. This was because cassava utilization in the area was limited to food as a supplement for rice or maize, as animal feed, or was sold as fresh roots to the markets in the neighboring towns. The farmers did not process the crop into chips or other products. The average yield of cassava in the area was low (3-5 t/ha). Many farmers were using native varieties like Imelda, Pulutan, Makan and others. Regarding livestock raising activities, it was found that the animals raised by the farmers include carabao and native pigs and chicken. Only very few farmers raised cattle and hybrid pigs. The other farmers said piglets of hybrid pigs were expensive and not readily available in the area. The meat demand in Bontoc and the neighboring towns, however, was high. This demand was satisfied by importing pigs from Davao. The overall situation of the cassava production and animal raising activities in Bontoc offered some opportunities for the development of integrated projects involving the improvement of cassava production, establishment of cassava processing projects to expand utilization and markets for the crop, and the development of the swine industry. The technologies and expertise needed for the establishment of these projects were available at PRCRTC-ViSCA. A series of consultation meetings with the cooperator, a big cooperative known as the Bontoc Multipurpose Cooperative (BCCI), were conducted to identify the specific projects to be implemented in Bontoc. The BCCI and PRCRTC representatives later decided to initiate an integrated project having the following components: 1 . cassava production and chips processing; 2. cassava-based feedmill; and 3. pig (100-sow level) and poultry raising. The general concept of the project was that the farmer-members of BCCI would be taught to improve their production level by planting high-yielding cassava varieties recommended by PRCRTC. They would also be taught how to process cassavas into dried chips using the PRCRTC-developed chippers. To serve as the main market of the farmers' chips, a cassava-based feedmill would be established. The piggery and poultry projects would serve as sure markets for the feeds produced by the feedmill. The piggery and poultry projects were also envisioned to satisfy the demand for meat by the people in Bontoc and the neighboring towns in southern Leyte. The cassava farmers would be encouraged to regularly supply dried chips to the feedmill to keep it operating. In short, the integrated project had been conceived to generate employment and increase the farmers' income from cassava production. Cassava production and processing These components of the integrated project were established first to assure that the feedmill would have a continuous supply of cassava chips once it started operation. The introduction/transfer of the high yielding varieties, the improved cassava cultural practices and the chipping machines was done through the establishment of demonstration 488 farms, training, and fabrication and evaluation of the chipping machines. Demonstrationfarms To showcase the cassava production technology developed by PRCRTC, a demonstration farm containing the PRCRTC recommended varieties (Golden Yellow, Lakan and the edible VC varieties) was set up in Bontoc. The demo farm, which was managed by a farmer, was intended not only to show the farmers the advantages of using the new varieties and the improved cultural practices, but also to propagate cassava planting materials, which would be used by the farmers in their cassava production activities. Evaluation of the advantages of the improved cassava production practices were later conducted by the farmers together with the PRCRTC and BCCI staff. The performance of cassava in the demo farm was compared with cassava in the farmer's field (using native varieties and traditional practices). The farmers observed that the high-yielding varieties grown using the recommended practices had yields which were more than twice the yield of the native varieties. Training To reinforce the knowlege gained by the farmers from the demo farm, PRCRTC also conducted training activities on "Cassava Production" and "Chips Processing" Farmers and technicians participated in these training activities. During the training, the farmers were taught about the high-yielding cassava varieties, the improved cultural practices for cassava, and the procedures in processing dried cassava chips, including the use of the PRCRTC-developed chippers. The farmers expressed their apprehension about the profitability of processing dried chips from cassava. After the discussion on the cost and returns of chips processing, most participants were convinced to produce and process cassava into chips for the BCCI feedmill. Cassava Production To identify farmer-cooperators for the cassava production and processing project, the BCCI officials together with the PRCRTC staff met with the BCCI farmer-members to inform them about the integrated cassava project. Several farmers expressed their interest in participating in the cassava production and processing components of the integrated project. The farmers, however, expressed their concern about the lack of capital to expand their cassava production areas. Thus, BCCI decided to extend cassava production loans to their members. The loanable amount was set at P6,000 per hectare. Fabrication and Testing of Chipping Machines To facilitate the cassava chipping operations, PRCRTC introduced to BCCI the pedal-operated chipper which could chip 400-500 kg of cassava roots per hour. BCCI 489 ordered eight of these machines from PRCRTC. Construction expenses were shouldered by BCCI. The machines were stationed in strategic places in the cassava-growing village in Bontoc. The farmers took turns in using the machines at a minimal fee for maintenance purposes. Farmers' initial feedback about the machines was positive. They said the machines were many times more efficient than the bolo or knife. However, some problems regarding the machines may occur later. Thus, long-term evaluation of the machines' field performance and life span will have to be conducted by the project staff together with the farmer-processors. Feedmill establishment Building construction and equipment acquisition The feedmill was established after the farmers assured BCCI of their full support through a constant supply of cassava chips. The building, which cost P800,000 was constructed using BCCI's own money. The feedmill equipment, on the other hand, was acquired by BCCI through DOST's financial support (loan grant). Training on feedformulation BCCI members who were identified to manage the feedmill component of the integrated project were trained by PRCRTC-ViSCA on the formulation of cassava-based feeds and on feedmill management. Hands-on training was done at the ViSCA Feedmill. Piggery establishment Construction of the BCCI' piggery was also done simultaneously with the construction of the feedmill building. Financing came from the Development Bank of the Philippines (P 2.5 million) and from the Land Bank of the Philippines-Maasin branch (P 1.2 million). Status of the Bontoc Project Cassava Production The status of BCCI's cassava production project as of the 1995/96 cropping season is summarized in Table 1. There were 52 farmers who participated in the BCCI's cassava production program. All availed themselves of the loan extended by the cooperative. These farmers planted cassava before, but only in an area of less than 0.25 ha. When BCCI started its integrated cassava program, these farmers increased the size of the area planted to cassava to not less than one ha. The total area planted to cassava by the BCCI farmer-members who availed themselves of the cassava loan was 64 ha. The varieties used were Golden Yellow and Lakan because according to them, these varieties had high chips recovery and dried at a shorter period of time. Yield sampling from different farmers' fields were conducted by the farmers and the PRCRTC staff to determine the average yield of the cassava planted using the 490 recommended cultural practices. The average yield of the Golden Yellow and Lakan varieties was 20 t/ha. This is already a very big improvement over the yields of the native varieties (3-5 t/ha). Chips Processing A total of eight pedal chippers are currently used by the farmers to process chips for the feedmill. According to the farmers, these units are not enough, especially during the peak harvest season when many of them would like to use the machines. They also reported that the chipper blades will not last long if used to process big and over-mature roots. BCCI ordered chipper blades from PRCRTC to replace the destroyed ones. The PRCRTC project staff delivered the needed blades but they plan to train some BCCI staff on the repair and maintenance of the processing machines so that they would not have to rely anymore on the Center if repairs are needed. Another problem reported by the farmers in relation to their chipping operations is the difficulty of drying the chips due to unpredictable weather and the lack of drying areas. Solution to this problem still needs to be discussed among the project participants. Table 1. Cassava production status before and during the implementation of the BCCI cassava project in Bontoc, southern Leyte, Philippines. When Project Before was implemented BCCI (1995/96 cropping Parameter Project season) Number of participating farmers (BCCI members) Total area planted to cassava (ha) under the BCCI cassava program Varieties used Average area planted per farmer (ha) Average yield (t/ha) 52 64 Imelda, Golden Yellow Makan, Lakan Pulutan 0.25 1.0 3-5 20 491 Chips Marketing The chips processed by the BCCI members were bought by BCCI at P3.00/kg (farm gate price) for the cassava-based feedmill. From Feb 1996 (when the BCCI feedmill started operating) to Sept 1996, BCCI was able to buy a total of approximately 200 tons of chips from its farmer members. About 61 % of these chips have been used by BCCI for the formulation of cassava-based feeds, while the remaining 39% have been sold at P4.00/kg to other chip users, like the ViSCA Feedmill, Biliran Feedmill, Placer Feedmill and the San Miguel Corporation. Forty one farmer-cooperators have already sold cassava chips to BCCI (at P3.00/kg). During a survey conducted in the middle of 1996, 38 farmer-cooperators claimed to have benefitted from the venture, while 13 did not mention any specific benefit (Table 2). Among the benefits mentioned were higher profit and the availability of money to buy better food (rice and others), pay debts, pay children's school fees and buy medicine. However, when asked about the specific amount of the profit, only 19 farmers were able to give figures, the other 22 said they did not keep records of their expenses and profits. The profits reported ranged from PI,000 to PI 0,000 per hectare (Table 3). It was observed, however, that those who reported higher profit were those who maintained the cleanliness of their farms and followed most of the recommended cassava cultural practices. Table 2. Benefits derived by BCCI farmer cooperators from processing and selling of dried cassava chips to BCCI (n = 41) in Bontoc, southern Leyte, Philippines. Benefits Frequency Big profit 4 Helped to buy better food 1 1 Helped during hardships (buy medicine, pay debts pay children's school fees) 13 No response 13 Feedmill operation Feedmill operations formally started in Feb 1996. The cassava-based feed formulations produced include pig starter, pig grower and pig finisher feeds. The feedmill is currently supplying cassava-based feeds not only to the BCCI's piggery but also to the other swine raisers in the municipality and the other neighboring towns. The 492 BCCI's major feed markets, aside from its own piggery, are its members in Bontoc, Liloan, Malitbog and Sogod. These people can buy feed at cheaper prices (8% mark up) than non-members (10% mark-up). Per suggestion of the PRCRTC staff, the coop management is currently hiring new staff to take charge of the marketing of feeds. This staff will be responsible for planning the appropriate marketing strategies to increase the sales of the cassava-based feeds. PRCRTC-ViSCA, will assist in market development through the production of promotional materials, i.e. brochures/leaflets and posters on the proper use and advantages of cassava-based feeds. Table 3. Amount of profit derived from chips processing and selling activities (n=41) in Bontoc, Southern Leyte, Philippines. Profit Range" Frequency P 1,000 -P 5,000 16 P 6,000 - P 10,000 3 Cannot estimate/no record 22 Total 41 " 1 US$ is approx. 25 pesos BCCI Piggery The piggery building was completed in Jan 1996. At present it has four boars (Duroc and Large White) and 60 sows (Large White and Hypor) most of which are pregnant. Additional animals will be procured early next year. Evaluation of the cassava-based feeds' effect on the performance of the animals i.e. weight gain, carcass quality, etc., still needs to be done. III. FPR on Soil and Water Conservation in Bontoc Background information With the ongoing processing and marketing activities in northen Mindanao and Bontoc, there is a great possibility that the cassava production areas in the two sites will expand. In Bontoc, BCCI has been urged by SMC to become one of its cassava chips suppliers. The BCCI staff who manages the cassava project, however, said that at this time, they are apprehensive to enter into an agreement with San Miguel because even at 493 the present scale of the integrated project, BCCI has already encountered some difficulties in the project. But even if SMC will not enter into the picture, there is still a great possibility for area expansion because the farmers are opening more hilly areas to plant cassava for the BCCI feedmill. A positive development which would promote the expansion of cassava production in Bontoc is the DA-LGU's grant of a tractor to BCCI to support cassava production. The tractor is now ready for use. The Department of Agriculture (DA) and BCCI will make a memorandum of agreement regarding the use of the tractor by the BCCI members (especially on rentals). In anticipation of the opening of so much hilly areas for cassava production, the PRCRTC and DA staff and the BCCI officers and members, during the consultative meeting held in June 1996, agreed on the importance of introducing soil and water conservation (SWC) technologies to the cassava farmers. Preliminary survey. A survey with 91 farmer-respondents was conducted in the cassava-growing villages in Bontoc to monitor the initial results of the integrated cassava project and to gather initial information about SWC practices among the farmers in the area. Of the 91 farmers respondents, 75% realized the damage that would occur in continuously cultivating hilly areas due to soil erosion. However, only 50% were trying to control soil erosion using their own methods, i.e. plowing deep along the contours or fallowing. Nevertheless, 60% of the respondents were interested to learn SWC techniques that are more effective than their present practice. In line with PRCRTC's current efforts to encourage farmers' participation in technology development and transfer, the project staff will be transfering soil and water conservation technologies to the farmers in Bontoc using the FPR approach. A formal survey will still need to be conducted to gather more information needed for the planning and implementation of the FPR activities. Establishment of demonstration farms. An erosion control trial established earlier by PRCRTC in a village near ViSCA does not exist anymore because the owner need the land back for other purposes. Thus, PRCRTC is now planning to set up demo plots showing the different SWC techniques in Bontoc. However, several arrangements have to be made before these demo plots can be established. In the meantime, two one-hectare cassava demo farms using vetiver grass as contour hedgerows have been established in Feb 1996. One demo farm was set up in a moderately sloping cassava farm in Pamahawan village, while the other was set up in a steeper farm in Mahayahay village. As of Oct 1996, the vetiver grass hedgerows in the two demo farms had 100% survival. 494 CONCLUSIONS In general, the cassava-based projects in northern Mindanao and Bontoc still need to be strengthened. A lot of activities, like the expansion of area planted to cassava, protecting the cassava area from destruction through appropriate soil and water conservation techniques, strengthening market linkages for chips, promotion of the cassava-based feeds (in the case of Bontoc), and other related activities have to be continued. However, preliminary results show that the projects have already provided some benefits to the intended clienteles. In northern Mindanao, the use of high-yielding varieties and the improved cultural practices increased the yield of cassava from 6-8 t/ha to 24-40 t/ha. Also, the introduction of the chipping machines improved the chips processing efficiency of the farmers. For the upland cassava farmers in Bontoc, the cassava production and chips processing activities enabled them to earn considerable profits, which they used to buy better food and medicine and to pay some of their financial obligations. The cassava chips trading activity of BCCI added income to the cooperative. Some of the swine raisers in the area were also able to buy cheaper but good quality feeds from the BCCI feedmill. Several factors contributed to the initial success of these projects. One is the farmers' involvement in all phases of project implementation. Some Observable Results of Farmers Participation in Technology Transfer Activities As a result of the increased farmers' participation in the technology transfer process, the following happened: 1. Identification of real farmers' problems, i.e. lack of cassava markets for the farmers in Bontoc; and low yield, low chip prices and lack of processing machines for the farmers in northern Mindanao. 2. Identification of solutions/project interventions that would likely fit their conditions, i.e. implementation of an integrated project that created additional markets for cassava in Bontoc; and the introduction of high-yielding varieties and chips processing machines acceptable to farmers in northen Mindanao. 3. Awareness of the opportunities offered by cassava, i.e. other uses aside from food. 4. Farmers' increasing participation in project activities, i.e. as suppliers of cassava chips to the feedmill and buyers of feeds from the same feedmill. 5. Feeling of "ownership" by the farmers for their project, i.e. the BCCI feedmill and piggery. 6. Identification of other activities/technology modifications that need to be done, such as FPR in soil and water conservation in Bontoc. Problems in the Implementation of Integrated/Participatory Cassava Projects Despite the positive results of farmers' participation in the cassava technology 495 transfer projects, the following problems were encountered: 1. Coordination among support agencies. The Center, being the lead agency in the conduct of integrated cassava projects, was not able to form a cassava council to oversee the different project activities before project implementation. As a result, the Center staff found it difficult to obtain maximum participation among the different agencies involved in the project. Some targets were not reached because the other agencies were not able to do their parts owing to conflicts with their other responsibilities and priorities. 2. Scale of integrated project is too big for the cooperators. This is true in the Bontoc project. The people in charge of the different project components found the scope of the project too big. They said they could not manage the project if they have to go into cassava area expansion to serve the needs of San Miguel for cassava chips. They said they have to concentrate on solving the problems of the feedmill and livestock projects before they can expand their cassava production to serve the needs of SMC. 3. Lack of capability of the cooperator to handle integrated projects. This could be the reason why the people in charge claimed that the integrated project is too big. The project manager is afraid to take business risks, especially if one has to talk about project expansion. He said he lacks the capability to manage the complexities of the integrated project. 4. Lack of staff to closely monitor the projects. The Center's extension division is generally understaffed. This is the reason why it could not always monitor the activities and solve the problems of the projects. Some technical problems are not addressed right away, such as the problem of lack of dryers. Possible Ways to Improve PRCRTC's Implementation of the Cassava Technology Transfer Activities The following suggestions made by Perez-Crespo (1991) can be used to improve the implementation of the cassava technology transfer projects undertaken by PRCRTC: 1 . Formal organization of a council to support the implementation of the integrated cassava project in Bontoc. The council should be composed of technology generators from PRCRTC-ViSCA, extension workers from PRCRTC-ViSCA and other government agencies offering support to the project, rural development program administrators (LGU officials) and farmers. 2. Keep the integrated project small and simple at first to avoid confusion and to allow the cooperators to learn first about the details of the project before expanding into other ventures. 3. The project should concentrate on giving those aspects direct benefits to farmers. This means that the project should be able to increase the income of the farmers and improve their living conditions. In the case of Bontoc, the project should make sure that the farmers can gain from their cassava production and chips processing venture. 4. Involve farmers in the conduct of small experiments such as FPR on soil and water 496 conservation. This will increase the farmers understanding of the technologies introduced and the likelihood that the technologies will fit the conditions of the farmers, and thus, be adopted on a sustainable basis. REFERENCES Perez-Crespo, C.A. 1991. Integrated Cassava Projects. CIAT, Cali, Colombia. 242 p. Gundaya, E.A., F.A. Evangelio, R.T. Sanico, J.R. Roa, R.R. Orias and M.C.U. Ramirez. 1995. Cassava technology transfer in the Philippines. In: R.H. Howeler (Ed.) Cassava Breeding, Agronomy Research and Technology Transfer in Asia. Proc. 4th Regional Workshop held in Trivandrum, India. Nov. 2-6, 1993. pp. 322-336. Werner, J. 1993. Participatory Development of Agicultural Innovations: Procedures and Methods of On-Farm Research. GTZ, Eschborn, Germany. 251 p. 497 FARMER PARTICIPATORY RESEARCH FOR CASSAVA TECHNOLOGY TRANSFER IN ASIA - CONSTRAINTS AND OPPORTUNITIES Reinhardt H. Howeler1 and Guy Henry2 ABSTRACT Since 1994 a farmer participatory research (FPR) methodology has been used with the objective of enhancing the development and adoption of efficient cassava production practices that will reduce erosion, maintain soil productivity and increase the income of cassava farmers in Asia. This 5-year FPR project, funded by the Nippon Foundation in Japan and coordinated by CIAT, is being executed by national research and extension organizations in Thailand, Indonesia, China and Vietnam. Members of the FPR teams in each of these countries participated in a Workshop in July 1994 in Thailand to become familiar with the FPR philosophy and methodologies. Upon return, they conducted Rapid Rural Appraisals (RRA) in cassava growing regions in their country to select two suitable pilot sites for the project. In addition, they established demonstration plots to show the farmers of the pilot sites a range of management practices to control erosion and increase yield or income. During a field day farmers looked at and discussed the various options and selected 4-5 that were considered most useful for their own conditions in order to try these on their own farms. In 1995 the first FPR trials were set out by fanners on their own fields with help from FPR team members. In erosion control trials they established 2-5 treatments on a uniform slope and constructed sedimentation channels along the lower side of each plot to collect the eroded sediments and measure soil losses due to erosion. In addition, other technology components such as improved varieties, alternative intercrop systems and fertilizer treatments were offered and experimented with by farmers. At the end of the first year, farmers and FPR team members jointly harvested all the plots and calculated cassava and intercrop yields, as well as the amount of soil loss in each treatment. These results were discussed with the participating farmers in order to select the best treatments for the second year of testing in 1996. From the experiences obtained so far we have learned that farmers in the selected sites are interested in the trials, and are adapting and adopting several component technologies. However, the success rate has been varied, especially between different sites and countries. A number of limitations have been identified, both technical, financial, organizational and institutional. This paper assesses the results of the project and identifies and analyzes the various constraints that are currently limiting the project. In addition, new opportunities are proposed that may alleviate the constraints. The analysis is conducted within a framework of how to move the project from the pilot phase to an implementation phase in order to reach a wider audience and obtain greater adoption of the developed technologies. INTRODUCTION Since 1987 the CIAT Cassava Agronomy Program in Asia has coordinated a network of cassava agronomists in various countries in Asia. These national program 1 CIAT Asian Cassava Program, Dept. of Agriculture, Chatuchak, Bangkok.Thailand. 2 Economist, CIRAD-SAR, Proamyl, BP 5035, 34090 Montpellier, CEDEX 1, France; formely CIAT Cassava Program, Apartado Aereo 67-13, Cali, Colombia. 498 scientists have conducted agronomy and soil management experiments in collaboration with CIAT, with major emphasis on soil fertility maintenance and erosion control. Results of these experiments are reported in this and in previous Workshop Proceedings. From these experiments it was concluded that cassava can indeed cause soil nutrient depletion if the crop is grown continuously without application of adequate amounts of nutrients in the form of organic or chemical fertilizers, and that cassava cultivation on slopes may cause serious erosion if the crop is not properly managed. However, research has shown that erosion can be markedly reduced either by common agronomic practices, such as minimum tillage, planting on contour ridges and at relatively close spacing, fertilizer application and intercropping, or by special soil conservation measures, such as application of mulch or the planting of contour barriers of grasses or legumes to reduce raindrop impact on the soil and to slow water run-off down the slope. While most of these practices were found to be effective in controlling erosion, very few cassava farmers are actually using these practices, either because they are unaware of the seriousness of soil erosion, or because they don't know which practice is most effective or most suitable for their own conditions. Since most of these practices require some additional labor or financial inputs, while the benefits tend to be long-term, farmers are seldom interested in practicing soil conservation. From these and other experiences it was concluded that cassava farmers will not adopt more sustainable management practices unless they first become aware of the extent and long-term effect of soil erosion on soil productivity; and secondly, they themselves test and select the most suitable practices under their own conditions. To enhance the adoption of more sustainable practices by cassava farmers in Asia, a new special project, funded by the Nippon Foundation in Japan and executed by CIAT in collaboration with national scientists, was initiated in 1994. The strategy to achieve the objectives was to develop a Farmer Participatory Research (FPR) methodology for testing soil management practices together with farmers in pilot sites in some important cassava growing countries in Asia. Thus, two or more pilot sites were selected in Thailand, Vietnam, Indonesia and China. Research and extension institutions dealing with cassava were invited to join the project and to name an "FPR team" of agronomists and social scientists to collaborate in its execution. In addition to the FPR project, the same or other institutions continued the conducting of collaborative strategic or applied research on cassava agronomic practices in order to identify still more effective technology components to enhance productivity and reduce soil degradation. FPR PROJECT The various activities of the project include the following: 1 . Organize a consortium of collaborating research and extension institutions in each of the four participating countries, and identify the persons that will be directly involved in the project. 499 2. Conduct a one-week Workshop on FPR Methodologies with participation of the four FPR teams. 3. Select potential pilot sites and conduct Rapid Rural Appraisals (RRA) in each site to select at least two appropriate pilot sites in each country. 4. Establish in each country or site Demonstration Plots with a large range of management options. 5. Organize a Farmers' Field Day at the Demonstration Plots to let farmers evaluate and select the best technological options for their own conditions. 6. Select participating farmers and let them choose the type of FPR trials to be conducted at each site and the treatments to be used in each type of trial. 7. Test, evaluate and select the most attractive options with farmers in their own fields. This includes experimentation with practices to control erosion as well as with, varieties, cropping systems and fertilization. 8. Through a reiterative process of trying out, discussing, selecting etc., develop the best package of practices for sustainable cassava production in each site. 9. Try out the best practices in commercial fields. 10. Enhance the testing and adoption of these practices in nearby villages. 1. Collaborating Institutions Table 1 list the various research and extension institutions collaborating in the project. Beside agronomists or soil scientists, attempts were made to include economists or other social scientists in the project. Local extensionists also played an important role in Vietnam and China. 2. Training Workshop on FPR Methodologies In order to become familiar with the FPR approach, FPR team members o*f the collaborating institutes participated in a one-week Workshop, held in Rayong, Thailand, from July 24 to 30, 1994. Besides lectures on the principles of FPR and some of the methodologies used, participants developed a workplan for the project in each country and also practiced interviewing cassava farmers, both in groups and individually. 3. Rapid Rural Appraisals (RRA) Each FPR team conducted RRAs in potential pilot sites by interviewing farmers about general farming conditions in the area, cassava production practices, problems and constraints (including erosion), utilization and marketing. Based on these data, each team selected two sites considered most suitable based on the following criteria - cassava is and will most likely remain an important crop in the area - cassava is grown on slopes - erosion is a serious problem and is perceived as such by the farmers - farmers are interested in participating in the project 500 Table 1. Institutions collaborating with CIAT in the Nippon Foundation Project on Improving Agricultural Sustainability in Asia. Country/Province Institution FPR project Research China-Hainan China-Guangxi China-Guangdong Indonesia-E.Java Indonesia-E.Java Indonesia-W.Java Philippines-Leyte Philippines-Bohol Thailand-Bangkok Tahiland-Bangkok Thailand-Bangkok Thailand-Korat Vietnam-Thai Nguyen Vietnam-Hanoi Vietnam-Ho Chi Minh Chinese Acad. Tropical Agric. Sciences (CATAS) / Guangxi Subtropical Crops Research Institute (GSCRI) Upland Crops Research Institute (UCRI) Brawijaya University (UNIBRAW) / Research Institute for Legumes and Tuber Crops (RILET) / Bogor Research Institute for Food Crops (BORIF) Phil. Root Crops Research and Training Center (PRCRTC) Bohol Experiment Station (BES) Field Crops Research Institute (FCRI) of Dept. of Agriculture / Field Crops Promotion Division of Dept. Agric. Extension / Kasetsart University Thai Tapioca Development Institute / Agro-Forestry College of Thai Nguyen University / National Soil and Fertilizer Institute (NSFI) / Institute of Agric. Sciences (IAS) of South Vietnam 501 Detailed information obtained in each site have already been presented in earlier papers (Zhang Weite et al., Wilawan Vongkasem et al., Nguyen The Dang et al., and Wani Hadi Utomo et al.) in this Proceedings. Table 2 shows a comparative summary of the RRA data collected in the pilot sites selected for the project. This shows that cassava is an important crop in all sites, but is the most important crop in only two of the eight sites. The cassava planted area per household is rather large in Thailand and in Baisha county of Hainan, but very small in most other sites. Farm size is an important determinant in the selection of suitable crops, cropping systems and production practices. The relatively large farm size in Thailand, for example, necessitates partial mechanization and almost precludes the use of intercropping systems. 4. Demonstration Plots In each country FPR team members established demonstration plots with a large number of treatments to be able to show farmers many technological options and their effect on yield, total income, and erosion. Plots were laid out on a uniform slope; along the lower end of each plot a ditch was dug and covered with plastic (Figure 1). Eroded sediments and runoff water would collect in these sedimentation channels. Water was allowed to seep away through small holes made in the plastic, and sediments were collected and weighed several times during the cassava growth cycle. Samples of wet sediments were taken to be dried and weighed in order to calculate soil losses per ha on a dry weight basis. Results of cassava and intercrop yields, gross and net income, as well as soil losses due to erosion in these demonstration plots are presented in earlier papers (Zhang Weite et al., Wilawan Vongkasem et al., Nguyen The Dang et al., and Wani Hadi Utomo et al.) in this Proceedings. 5. Farmers' Field Day Shortly before or during the harvest of cassava, farmers from the selected pilot sites were invited to visit the demonstration plots, with the objective of discussing the pros and cons of each treatment and then to score the treatments in terms of general usefulness, i.e. effectiveness in reducing erosion while maintaining or increasing farmers' income. The occasion was also used to explain clearly the objectives of the project and to inform farmers about new varieties and other new technologies. Table 3 shows the ranking of management practices considered most useful by farmers from seven pilot sites. The treatments in the demonstration plots varied from country to country, but even within the same country farmers' preference varied from site to site. However, in both Thailand and Vietnam, farmers liked the treatment of vetiver grass barriers and decided to try this as one of their treatments in the FPR erosion control trials. In Indonesia, farmers generally preferred intercropping cassava with maize (their traditional practice) and planting Gliricidia sepium or elephant grass as contour barriers, as both can be used as animal feed. 502 6. Selection of Participating Farmers and Treatments for FPR Trials In most cases there was no particular selection of farmers to participate in the project, i.e. those farmers that were interested could participate. Farmers that had sloping land suitable for erosion control trials were encouraged to do those trials, while those with flat land participated in variety, fertilization or intercropping trials. Table 4 shows the types and number of trials conducted in each pilot site in 1995/96 and 1996/97. In Baisha county of Hainan about 37 farmers participated in the first year. They generally had only two treatments per trial, comparing an "improved" practice or variety versus their traditional practice or variety. The large number of farmers participating was difficult to manage, while having only two treatments per trial made it difficult to compare among "improved" treatments. Thus, in the second year, the number of trials in China was reduced while the number of treatments per trial was increased. Moreover, farmers within one site were encouraged to all test the same 4-5 "improved" practices or varieties versus their traditional practice or variety. This way, trial results could be averaged over farms and more reliable results could be obtained to draw conclusions and select the best treatments. In Vietnam the number of participating farmers increased in the second year (1996/97) as farmers saw the benefits of participating in this community effort, especially in obtaining planting material of new varieties. Before starting the trials, participating farmers met in the village with the FPR team members to discuss the type of trials to be conducted and the treatments to be tested. For the erosion control trials farmers discussed the results of the scores given to each treatment during the field day at the demonstration plots. They generally selected 3-4 treatments with high scores, but sometimes they themselves suggested new treatments that seemed of greater benefit for their own particular circumstances. Thus, in Soeng Saang district in Thailand, farmers wanted to test contour barriers of sugarcane (for chewing) as a more useful alternative to king grass, which they had observed in the demonstration plots. Also, some farmers raising silkworms wanted to test hedgerows of mulberry bushes as a means to control erosion while also obtaining benefit from the barrier. In Thailand farmers decided to test 3-4 common treatments as well as one preferred individual treatment, all in comparison with their traditional practice. When farmers test one individually selected treatment, They are encouraged to think for themselves about effective ways to control erosion while at the same time providing additional benefits or income. These farmer innovations can sometimes lead to better and more practical management practices, and also enhances farmers' feelings of empowerment to find their own solutions. This is a fundamental element for success (Ashbyera/., 1997). To compensate for the additional cost of doing the trials, farmers were paid for 503 digging the sedimentation channels and received the plastic for covering the channels. They also received planting material of vetiver grass and new cassava varieties for the trials, as well as seed for intercropping treatments. In Thailand farmers also received one bag of 15-15-15 fertilizers, while in one site in Indonesia the participating farmers as a group received 15 goats. These incentives were kept to a minimum to enhance the feeling that farmers were conducting the trials for their own rather than for the researchers' benefit. 7. Testing, Evaluation and Selection of Best Options Once the types of trials as well as the treatments were selected, FPR team members helped the farmers to select the most suitable site for each trial, to set out contour lines (for the erosion control trials) and stake out the trials. Especially the first year, team members had to help farmers establish the various treatments. But once established, farmers managed their own trials similar to their other commercial fields. Team members would visit regularly to discuss the progress and try to solve any problems. They would also collect and weigh the eroded sediments in the sedimentation channels, usually once during the cassava growth cycle and again at time of harvest. A very common problem in the erosion control trials was that some trials were not laid out well along contour lines, as most farmers would prefer their trials laid out parallel to roads or the edge of fields. In that case, run-off water would sometimes enter or leave plots through side borders, resulting in unreliable soil loss data. Moreover, if the erosion trials were situated halfway or near the bottom of a slope, run-off water from fields above would enter plots if the diversion ditches along the upper side of plots (Figure 1) were unable to divert all the run-off away from the plots. In that case, large amounts of run-off would enter the plots, causing excessive erosion that was not related to the treatment. As such, the eroded sediments in the ditches did not accurately reflect the effectiveness of the treatment in controlling erosion. At time of harvest, participating farmers and FPR team members would jointly harvest all or part (usually 16-20m2) of each plot to determine cassava root yield. In Thailand, where farmers receive a differential price according to root starch content, the starch content was also determined. In case of intercropping treatments, the yield of the intercrops had been determined earlier at their time of harvest. During the cassava harvest farmers also noted and discussed the amount of sediments in the ditch of each treatment, before those sediments were collected and weighed. One problem observed during the harvest in the first year was that farmers had often planted at irregular spacings, making it very difficult to accurately determine effective plot size and compare yields; in subsequent years farmers were encouraged to use a standard planting distance, usually 1.0x1.0 m, 0.1x0.8 m or 0.8x0.8 m. Another commonly observed problem in estimating yield in plots with contour hedgerows, was that researchers often excluded from their harvested plot any rows bordering those c o•a c CO a S U S > s BOi s CO X) oo c o U c o 00 c o 3 -1 CO O X c > o a. i »*} i* S? w - "i 2 <3 C « 8.2n 9 9 © d c Z oo II H oo c/3 oo c 4> o 6 £ o p. -r 4> — £ 2 .© CO C/3 i u 505 A. Top View I diversion ditch -y - Treatment 1 plot . border" V/////////A S Treatment 2 E o 15m ////////1 Treatment 3 V ///////// / plastic covered channel 0.4x0.4x15m B. Side View plot borders" 0.4mV-. 1) plot border of sheet metal, wood or toll ridge to prevent water. entering or leaving plot*. 2) polyethylene or PVC plastic sheet with small holes in bottom to catch eroded toil sediments but allow run-off water to seep away. Sediments are collected and weighed once a month. Figure 1. Experimental lay-out of simple trials to determine the effect of soil/crop management practices on soil erosion. g E 8 1 3 u 2 a. C I c .© S3 > s IM © at !•* 3 2 1 o T3 c Q | s — CS Tt TO 2 'ca u CQ O X o o 6X) C Tt■ m C*^ — M Tt OQ w J es oc o•o E £ > 1J ^ CO 8 1 "a a5/3 <« z (2 ■c no on On o N VI O m -h — *« t M no r- § J I o 5 t: «C CO H *s w > £ 5 3 o f- On IT> rn ci ci On r^ ci Tt >i-> —■ no —■ t» no m m oo i On c 5 60 i 1 1 1 o 5 t: « W > fc, 5 5 o H 508 hedgerows. As such, they ignored the fact that hedgerows occupy part of the land and may either increase (mulching effect, trapping of water and fertilizer) or decrease (competition effect) the yield of bordering cassava plants. To correct this problem, harvested plots should include the hedgerows approximately in the same proportion they might be found in commercial fields (usually about 10-20% of crop area, depending on slope). On the other hand, if experimental plots have more than the necessary number of hedgerows (usually spaced at 1 m vertical distance between hedges), crop yields are likely to be reduced more than necessary, and farmers may unduly reject such treatments for that reason. Finally, another common problem is that the "farmers' traditional practice" treatment, used as comparison with other "improved" practices, does not reflect actual farming practices. If this treatment is not "real", farmers may reject the results of the whole trial as not accurate. After all trials had been harvested and all data (including yields of intercrops and hedgerows, eroded sediments, etc) collected, FPR team members quickly calculated and tabulated the results, including the gross and net incomes obtained in each treatment, expressed in local currency and units of measurement. These were then presented to and discussed with the farmers. Farmers were asked to score treatments or raise hands for those treatments they preferred, considering both the effectiveness in controlling erosion and producing benefits (higher income or other benefits) to the farmers. In the same meeting farmers were usually asked whether they liked to continue the FPR trials, which types, and which treatments. Farmers usually selected the "best" treatments from the previous cycle, sometimes combined with newly proposed treatments. Thus, "best-bet" treatments were again tested against farmers' traditional practices in a reiterative process to develop a package of management practices, including new varieties, optimum fertilization, productive intercropping systems and possibly hedgerows or other special measures to reduce erosion, that would optimize yields or income while reducing erosion and soil nutrient depletion. 8. Trying out in Practice Usually, after 2-3 years of experimentation in relatively small plots, farmers would be convinced of the usefulness of certain practices, and would be ready to try "their" recommended practices on their own production fields. These practices should be tried for 1-2 years on a relatively small scale, such as 0.1-0.2 ha, to determine whether these practices are truly practical and meeting farmers expectations. If not, they may need to be tested again in a modified version that better corresponds with the conditions in the field. Especially in Thailand, where commercial fields tend to be large, some further modification may be necessary to facilitate mechanized land preparation and harvesting. 509 CONSTRAINTS AND OPPORTUNITIES: WHAT HAVE WE LEARNED? Table 5 shows a comparative summary of the conditions of the FPR trials conducted in the eight pilot sites as well as some of the problem encountered. Specific conditions at the site, the relation between the institutions involved in the project, as well as their relation with local extensionists and village leaders, often have a strong bearing on the success or failure of the project. 1. Conditions at the Pilot Sites Table 5 summarizes some of the pertinent biophysical and socio-economic conditions at the sites. Travel time from the participating institute to the sites varied from less than one to more than five hours by car. Less time required to go to Pho Yen, Luong Son or south Malang facilitated more frequent visits of FPR team members to these sites. The longer distances in Thailand made frequent visits or long stays at the sites impractical. To get the work done quickly and efficiently, the Thai FPR team often brought their own laborers. While certainly efficient, this eliminated the need to involve farmers directly in the work and thus reduced farmers' participation and interest. Although in China the pilot site in Baisha county is only 1 1/2 hours from CATAS, the lack of transport facilities as well as the lack of personnel at CATAS prevented regular visits to that site. For the same reasons, attempts to expand to a second site in Quongzhong county and later in Tunchang county were unsuccessful. Thus, distance to and accessibility of the sites is another important criterion in selecting pilot sites. Cassava was the principal crop only in Soeng Saang and in Malang, and an important secondary crop in the other locations. The more important cassava is in the whole cropping system, the more farmers will be interested in developing practices that will optimize cassava yields and protect the soil's productivity. Similarly, the smaller farm sizes (Table 2) in Vietnam and Indonesia result in more intensive land use management and greater concern for maintaining soil productivity. Also, slopes were quite steep in Thanh Hoa and Luong Son districts in Vietnam, in some areas of Baisha county of China and in Malang district in Indonesia, resulting in serious soil losses due to erosion. This enhanced farmers' concern about the problem. These three aspects influence the degree of farmers' interest and participation in the project. In most sites both men and women participated in the project, but for cultural or religious reasons only men participated in Baisha county in China and in the two sites in Indonesia. This did not seem to have any direct bearing on the interest of farmers or their degree of participation. Of greater importance is the interest of the leaders in the village or of local extensionists. In both sites in Thailand the village leaders were supportive of and actively involved in the project; in Pho Yen in Vietnam and in Blitar in Indonesia this was also the case, while in Baisha county in China a local extensionist 510 was providing good leadership to the project. 2. Institutional Relations In Thailand the project was executed jointly by the Dept. of Agriculture (DOA) and the Dept. of Agricultural Extension (DOAE). Teams from both institutions worked together very effectively in both sites, the DOA providing research expertise and planting material of new varieties, while DOAE provided expertise in extension activities as well as access to a nationwide network of extension offices, which participated in various degrees in the selection of sites, in the conducting of RRAs, and in the execution of the FPR trials. In other countries each site was the responsibility of only one institute. This facilitated decision making and management, but each institution did not benefit much from the expertise or the experience obtained by the other institutes in the country. A sharing of experiences and mutual participation in at least the final field day at harvest, would probably have improved the project execution at some of the sites. In Pho Yen district of Vietnam much of the responsibility for helping farmers establish the trials and the taking of data rested with the local extension office, as well as with 4th-year students from Agro-forestry College of Thai Nguyen University. These students would live, or spend a considerably amount of time, with farmers in the village, which stimulated mutual learning and contributed to a close relationship between farmers and FPR team members. In Blitar district in Indonesia the project was similarly executed by an MSc student who spent a lot of time in the village and build a good relationship with the participating farmers. Where possible, this seems an effective way to improve the quality of the trials and build trust between villagers and government officials. 3. Technical Problems Technical problems in executing the FPR erosion control trials concerned mainly the laying out of the trial along contour lines, preventing water from entering or leaving the plots from fields above or through side borders, and the correct measurement of yields, which should include the effect of contour hedgerows if present. These have already been discussed above. An additional problem is that some plastics used in the sedimentation channels deteriorated very quickly or were stolen. Attempts to substitute plastic sheet with bamboo matting or opened-up plastic fertilizer bags were not very successful. Lining sedimentation channels with split-open bamboo has been successful in the Philippines, but this material is not readily available in other countries. Other technical problems encountered include the poor germination of cassava stakes if they had been stored for too long, and poor germination of some hedgerow or intercrop species due to poor quality seed or unfavorable weather. Furthermore, results of FPR trials were sometimes unreliable due to poor marking of plots, difficulty in calculating effective plot size due to irregular cassava planting distances, varietal mixtures, poor weed control, stealing of crops, and damage of cassava or intercrops by water buffaloes or rats. o T3 C .S Si u E M C u > m .2 a 1 m 1 'S. .SP 1/ T3 C o 6 JS .2 ft. u. •o c c U T3 C f- £ < CC u m o J ^- 73 5 z O 3 E 6i C CQ c c o o f- cI c >■ c c 1/3 g E .^2 -u 73 J2c £Q O ft. o + < O o in < o Q + < O Q X: on < < u 0- U £2 13 Z E .2 E c c T3 N O w 00 & - V C-; - 5 o n V •o c u x 9 3 V 2 «* •o r1 f< ~ o N O E V 00 t oo n "J I/-1 ■i g V — 51 c u .2 '§ « > .2 | 2 £ u > 5 -3 .S c .a &•ill 8 s 2u a •s u > w u — t! illg * r - .-s ■a ft, £ J3 2 G c a -a u s: •s u "3 £ e d u0a 0c '3 T3 5 ■C o C C U I Q £ C 9 Ci >, 00 UJ a. ' H c .a a o — .9 + + ■a + + + 1+ + + + + + + U + + + + + -c a. I- + o c > a. ft. « + + + + + > u u 0. I + + + g 0. 0. o + + + -s + + .s + + + § u u + + + o + -5 u> > > + + ■t -s u ci. c. E. .2(y; 'Sa i2 8 u .5 H ft. o 1U Uu | 1 o • 1 a 3 o c u- U- -J ft. 1 ft. c c E p 512 4. Availability of Attractive Options One of the most important requirements for obtaining farmers' interest in the project is the availability of new and attractive technological options. In most pilot sites farmers were interested in participating not because of their concern for soil erosion, but because of the possibility of obtaining planting material of new higher-yielding varieties. Farmers were primarily interested in testing new varieties, new intercropping systems or fertilizers, because all of these can directly increase their net income. Fortunately, in all participating countries cassava breeders had already developed higher-yielding varieties, which were ready for on-farm testing and distribution. In Indonesia, however, these varieties were only marginally better than the local varieties, while in Vietnam and China the lack of sufficient planting material of new varieties initially limited their use. Once farmers saw the benefits of these new varieties, of better intercropping practices and of correct fertilizer use, they became more enthusiastic about the project, including the development of practices to control erosion. To be acceptable to farmers, erosion control measures must be effective, not require too much additional labor or money, and maintain or increase yield or income. Farmers in Thailand and Vietnam generally selected vetiver grass as the most effective way to reduce erosion, but its adoption on a large scale may still be limited because of unavailability of planting material, high cost of transport and planting, and lack of direct use of the grass except as mulch. For that reason, many farmers in Vietnam prefer contour barriers of Tephrosia candida, which is less effective in controlling erosion, but easier to establish and more useful as a green manure. In Indonesia vetiver grass was not included in the demonstration plots until the second year; once included farmers seem to appreciate its effectiveness in controlling erosion, but adoption will be slow because of lack of planting material and its unsuitability as an animal feed. Intercropping cassava with peanut is a new and very useful technology for farmers in the three sites in Vietnam as well as in Baisha county of China. In some parts of China, intercropping with peanut, however, is impossible due to serious damage caused by rats. In Thailand, plowing with tractor is usually done up-and-down the slope and/or parallel to roads or field borders. Contour hedgerows of any species will interfere with this practice, while curved hedgerows will also interfere with the planting of cassava in straight lines using tight ropes as a guide. These practical problems are likely to limit adoption of contour hedgerows. In fields with uniform and unidirectional slopes, the planting of straight hedgerows or barrier strips across the slope, however, may be acceptable. Intercropping with sweetcorn or pumpkin is rather new in Thailand, but the susceptibility of these crops to drought or excessive moisture will limit their adoption. In Indonesia cassava farmers already use many soil conserving practices, such as terracing (especially in Java), agro-forestry, intercropping, and manure or fertilizer 513 application. Fields tend to be very small and often bordered by hedgerows of grass (mainly elephant grass) or tree legumes (mainly Gliricidia sepium, Leucaena leucocephala, or Calliandra). Some of these essentially function as contour barriers. Farmers have shown little spontaneous interest in the project because of lack of attractive new options, alternatives that are either more effective or more beneficial than what farmers are already doing. The inclusion of vetiver grass in the demonstration plots in Malang showed farmers a highly effective erosion control option, but this technology may still find little adoption because of other limitations, as discussed above. The challenge for Indonesian team members is to find truly superior alternatives to what farmers are already doing now. 5. Institutionalization of a Participatory Approach in Research and Extension Traditionally, governments in Asia, whether communist or democratic, have all used a top-down approach in the research-extension continuum, i.e. researchers do research, usually at experiment stations, and based on their results make recommendations, which are then extended by extensionists to farmers. Farmers have no input in setting research priorities and they either accept or reject the recommendations of the extensionists. This system works reasonably well for those technologies, like new varieties, that are readily accepted anyway because they increase farmers benefits at little additional cost. Fertilizer practices are also easily accepted if the benefits clearly outway the costs, and if farmers have the resources to buy the fertilizers. However, for knowledge-intensive technologies, like erosion control practices, which are often highly site-specific, the practices recommended in this top- down approach are either not the best for the specific location or require too much additional inputs in comparison to the perceived benefits. The participatory approach used in this FPR project assures that the technologies are better tailored to farmers' needs and to the specific conditions in the pilot sites. Moreover, as farmers are directly involved in the testing and selection of the best practices they are more likely to develop practices that are acceptable and adoptable, thus eventually leading to greater use of sustainable production practices. However, to reach a much wider audience of farmers beyond the pilot sites, either involving them directly in FPR, or at least convincing them of the usefulness of those practices developed by fellow farmers in the same area, it is necessary to involve many more researchers and extensionists in the process. This can only be done if local administrators and policy makers are convinced that FPR is an efficient and effective methodology for the development and transfer of site-specific technologies, and are willing to adopt a more participatory and bottom-up approach within their own institutions. This FPR project on enhancing the sustainability of cassava production in Asia, is a rather insignificant endeavor in the whole realm of things, but can become significant if we can convince policy makers of the effectiveness of the methodologies used, and contribute to the institutionalization of a more participatory 514 approach in national research and extension organizations. Moreover, as evidenced in cassava processing projects in Colombia and Ecuador (Ospina et al., 1996), collaborating farmers themselves, once trained and experienced in FPR in their own sites, are excellent transfer specialists to other farmers, especially when technologies are more complicated, as is the case for processing and erosion control. As is already pointed out, it will take significant human resources to transfer the FPR approach as well as the technologies developed to other villages and regions. Hence, the collaboration of farmers as alternative transfer agents (farmer-to-farmer extension) could be a viable option for the future. REFERENCES Ashby, J. A. 1997. What do we mean by participatory research in agriculture? In: New Frontiers in Participatory Research and Gender Analysis. Proc. Int. Seminar on Participatory Research and Gender Analysis for Technology Development, Cali, Colombia. Sept 9-14, 1996. pp.15 -22. Nguyen The Dang, Tran Ngoc Ngoan, Le Sy Loi, Dinh Ngoc Lan and Thai Phien. 1998. Farmer participatory research in cassava soil management and varietal dissemination in Vietnam. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Ospina, B., S. Poats and G. Henry. 1996. Integrated cassava research and development projects in Colombia, Ecuador and Brazil; an overview of CIAT's experiences. In: D. Dufour, G.M. O'Brien and R. Best (Eds.). Cassava Flour and Starch: Progress in Research and Development. Proc. Workshop held in Cali, Colombia. Jan 1994. CIRAD/CIAT, Cali, Colombia. pp.333- 357. Utomo, W.H., Suyamto, H. Santoso and A. Sinaga. 1998. Farmer participatory research in soil management in Indonesia. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer China. Nov 3-8, 1996. Vongkasem, V., K. Klakhaeng, S. Hemvijit, A. Tongglum, S. Katong and D. Suparhan. 1998. Farmer participatory research in soil management and varietal selection in Thailand. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. Zhang Weite, Lin Xiong, Li Kaimian and Huang Jie. 1998. Farmer participatory research in cassava soil management and varietal dissemination in China. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. 515 CASSAVA BIOTECHNOLOGY NETWORK Ann Marie Thro1 INTRODUCTION The objective of this session is to provide all participants in the Asian Cassava Research Network, regardless of specialization, with a common background for evaluating the usefulness of biotechnology tools for Asian cassava research and development. Specific Objectives -Briefly describe biotechnology tools available for cassava -Review types of research objectives for which each biotechnology tool is appropriate -Consider other factors important for using each tool effectively (time to output, relative costs to be measured, etc) Target Outcomes for Participants and CBN -A common understanding of biotechnology tools and their functions -A shared basis for evaluating whether or not biotechnology tools offer an advantage for developing technical solutions to problems encountered in Asia for cassava -Based on discussions and information during the Cassava Regional Workshop to obtain a information assessment by Workshop participants of the following: * Does biotechnology have any advantages to offer for cassava in Asia? * How? What? Where? * Is biotechnology being used effectively in those situations? What are the elements of practical success? * Where biotechnology could offer advantages but is not being used, what is available and what is missing for effective action? Institutional linkages, information, human resources, financial resources, research policies.. etc? BIOTECHNOLOGY TOOLS FOR CASSAVA I. Micropropagation (cell and tissue culture) Genetic Biotechnologies ("gene technologies"): II. Genetic engineering, or genetic transformation III. Molecular markers Microbial Biotechnology: IV. Fermentation biotechnology Cassava Biotechnology Network (CBN), CIAT, Apartado Aereo 67-13, Cali, Colombia. 516 Each Biotechnology Tool Does Something Different -Molecular markers are a form of information technology; used to enhance breeders' ability to assess and select genetic diversity. -Genetic engineering is a way to extract and insert genes; used to enhance breeders' access to genetic variations or to create new variation. -Tissue culture is a method of vegetative propagation (cleaner, faster,...) -Fermentation biotechnology is a way of changing a substrate; used to give cassava greater value and/or new properties. The nature of the technology determines how it is useful. I. Micropropagation (Cell and Tissue Culture) Used in at least 20 countries for conservation and sharing of cassava genetic diversity Considered "conventional" for cassava by some. Widely used for cassava because: -Cassava varieties must be propagated by cuttings, which can transmit pests and diseases. -Production of disease-free cuttings in tissue culture permits: 1 . Conservation of cassava genetic diversity without long-term build-up of pests and diseases which may occur in field collections. This requires a long-term funding commitment, which is often more problematic than the technology itself. Cryopreservation (freezing): Lower cost, long-term. Has been experimentally successful, but requires pilot testing to assess genetic stability and operating costs. 2. Improved international access to cassava genetic diversity via clean propagules to meet quarantine regulations. 3. Similarly, international sharing of experimental varieties, so breeding programs can use varieties developed in similar environments. This can be managed as shorter-term commitments. Tissue culture facilities can be shared with other crops. 4. Renovation of infested important cultivars Thermotherapy: virus/pest elimination. Effective for yield increases over several years if rate of re-infection is slow. If rate of re-infection is high (rapid), resistant cultivars are required. Careful agronomic management is necesary: Loss of associated beneficial microorganisms, or unusual infection of succulent plantlets just out of tissue culture, can lead to disappointing yield "crashes". 5. Faster multiplication of desired cassava varieties ("in-vitro rapid micropropagation") Potential (biologically-achievable) rates per year: Field multiplication: tens of plants from one plant 517 Single-or two-node cutting multiplication: thousands from one Tissue culture: millions from one Some of the factors affecting success: -Selection of plant materials for tissue culture multiplication. Clones must be well-tested with target users. -Health status of source material. -Post-flask management (hardening, "weaning") Requires investment of experienced supervision, labor, water, materials (screening...). Succulent tissue more susceptible than usual to pests and diseases. -Integration with existing program of field multiplication and distribution Links with agricultural extension, NGOs, private sector. Distribution logistics. Pilot projects -Information on costs and benefits In emergencies when cassava varieties in a region must be changed or restored rapidly due to abrupt social, economic, or biological events. For normal progress in more stable situations Facilities can be shared with other crops -Requires management skills For delivering plantlets on time For efficient use of facilities "Low tech biotech" -A need to develop lower-cost methods for tissue culture In-vitro culture protocols Post-flask management Plantlet distribution Energy-efficient, use more local materials General principals, site-specific aspects Micro-propagation also permits regeneration of plants from single cells, an essential step in genetic engineering. II. Genetic Transformation ("Transgenesis") "Transfer of a piece of DNA into a plant cell, followed by stable integration in the genome" . 518 Used for traits that can be altered by a single-gene, or single-gene steps in multigenic pathways. Increase genetic diversity by inserting new genes into otherwise superior varieties, or, Alter the level of expression of existing genes above or below naturally-occurring levels. Promoters to enhance enzyme activity. Anti-sense forms of existing genes, to inhibit gene action. Examples of types of genes used in genetic transformation: Resistances to viruses and other disease organisms. Enzymes to transform the cassava root into a "biofactory" for biodegradable plastics-polymers or other totally new products. Genetic engineering can be used to create specific mutations, useful for studying complex physiological traits, e.g. could permit the first definitive studies of the ecology and biochemistry of cyanogens in cassava, to permit more effective and better-targeted genetic manipulation of this trait (whether by plant breeding or genetic engineering). Biological requirements 1. Efficient systems(s) for plantlet regeneration from single cells. 2. Efficient system (s) for transformation. 3. Efficient system for selection of transformed from non-transformed cells immediately after transformation. 4. Transformation, selection, and regeration systems must be compatible, i.e. -Transformable cells must be regerable. -Transformation and selection systems must not have too negative an effect on regeneration. 5. Isolated (cloned) genes available: Marker genes. Genes for desired agronomic or quality characteristics. Types of transformation systems either through vectors or by direct gene transfer /. Vectors: usually Agrobacterium Agrobacteria can transfer a piece of DNA from a region of an extra-chromosomal part of their DNA (the tumor-inducing Ti plasmid) to a plant cell. The Ti plasmid contains important regions necessary for successful transfer of DNA to the plant. Transfer DNA (T-DNA), the region actually transferred to the plant cell. The gene to be transferred is inserted into the T-DNA region. 519 Virulence or Vir-region, containing genes which determine whether and how efficiently the T-DNA is transferred to the plant cell. In nature: Several other genes involved in formation and breakdown of opines (sugar amino acids) and so-called Oflc-genes Agrobacterium "breeding" has created strains with additional and more effective Vir genes, capable of transforming additional species, more effciently. Integration of DNA in the genome of the plant by the bacterium occurs at random, although there seems to be preference for certain transcriptionally-active regions. Advantages of Agrobacterium methods More suitable for simple laboratory conditions, require only the equipment and supplies of a normal microbiology and tissue culture laboratory. Possibly fewer stuctural changes or amplifications in the genes inserted, low number of copies inserted compared to direct gene transfer methods. Tend to give more normal gene expression and inheritance 2. Direct gene transfer Examples: - Micro-injection of individual cells with DNA. - Micro-projectile bombardment (biolistics). Gold "bullets" coated with DNA are transferred into plant tissue (intact cells) by high velocity force using a "gene gun". - Electroporation of DNA into protoplasts. Frequencies of success using direct gene transfer range from 0.1% to 10% depending on the plant species and the procedure used. Generally less efficient than Agrobacterium, for which success ranges from 1% to 90%. More costly (specialized equipment and maintenance). However: The only alternative for many monocot species not susceptible to Agrobacterium. Less genotype-dependent than Agrobacterium methods. Factors influencing expression of transgenic traits So-called "position effects", the actual position on the chromosome (hard to influence) Environmental factors (GxE) Structure of gene to be integrated (codon composition, borders, regulatory elements such as promoters, enhancers, introns, etc..) (can be manipulated) 520 Type of gene to be integrated: Genomic sequence from same species: 60-90% of transformants have desired expression Anti-sense sequence (e.g., "turning off' a gene): 0-50% Heterologous sequence (gene from another species): 1-40% When to select and evaluate products of genetic transformation -Before or during regeneration of transformed cells -During tissue-culture growth of transformed plantlets -Based on greenhouse growth and performance -Based on field characteristics and performance -Based on inheritance of traits by offspring (gene can remain present but become silenced, or expression level can change) Field selection and inheritance are the ultimate test; also most expensive; eliminate as much material as possible in earlier stages Genetic engineering of cassava:update Cassava genetic engineering research began in 1987. The only systems successful for regeneration of cassava cells are adversely effected by transformation and selection. Result: progress has been slow and difficult. Breakthroughs in 1996; dramatic progress since last biennial report in 1994. Regeneration of transgenic plants using microbombardment of embryogenic suspension cell cultures (Agri. Univ. Wageningen, Netherlands; ILTAB/ORSTOM, USA/France) Agrobacterium transformation of somatic embryo-derived cotyledons with subsequent regeneration via organogenesis (ETH, Zurich). Wild Agrobacterium strain to infect somatic embryo-derived cotyledons and regeneration through somatic embryogenesis (CIAT). New gene promoters From cassava vein mosaic virus: potential to replace the commonly-used CaMV 35S promoter in cassava transformation (ILTAB); available on request to researchers in cassava growing countries. 521 Possible root-specific promoter (Univ. Newcastle, UK). Next steps 1 . Introduction of genes of agronomic interest a. Genes available: Resistance to viral diseases (CasCMV, ACMV)(Africa) Starch concentration, quality (S.America, Asia) Modification of cyanogen metabolism (Africa, S. America) b. Genes not yet available for: Reduced postharvest perishability of cassava (global) Other... c. Genetic transformation as a basic research tool (global) Mechanisms of insect and disease resistance Nutrient use efficiency Drought tolerance Photosynthesis Cyanogenesis 2. Simplify and improve the new experimental protocols for cassava genetic engineering (repeatability, efficiency). 3. Extend genetic tranformation protocols to a range of important cultivars 4. Continue to identify tissue-specific gene promoters Outlook for the next few years: Improved protocols ready for technology transfer to cassava-growing countries (3 to 5 years) Transgenic cassava with genes of interest, ready for field trials (5 years?) Therefore, planning phase should start now for: -Transfer of improved protocols to additional laboratories -Biosafety -Design of evaluation of transgenic prototypes with applied researchers and cassava users III. Molecular Markers Genetic variation is: The basis of plant taxonomy, classification, germplasm evaluation The basis of crop improvement. What plant breeders do: Detect and measure the genetic variation available and combine it into new varieties by using cycles of crossing, selection, crossing again,... to combine and 522 recombine genes. How plant breeders detect and measure genetic variation: Inferred from the phenotype: measurements of morphology, performance, analyzed via quantitative genetics ("organized ignorance"), or, Direct from the genotype via molecular markers (also requires quantitative analysis!). Molecular marker types Isozymes: limited amount of variation; inexpensive; recommended when sufficiently informative. DNA sequence markers DNA sequence marker types: RFLPs, RAPDs, Microsatellites, AFLPs, etc. ....a rapidly-evolving technology.. Characteristics of different DNA marker types: 1. RFLPs (Restriction fragment length polymorphisms) Uses "restriction" enzymes which cut DNA at specific bases sequences. Resulting fragments separated on a gel by electrophoresis. Genetic variation detected as differences in the size of fragments. Advantage: Very reproducible (reliable) from lab to lab and species to species Specific to particular sequences Disadvantage: Expensive Require radioactive labels (research to develop non-radioactive methods) 2. RAPDs (Random amplified polymorphic DNA) Uses polymerase chain reaction (PCR) which polymerizes nucleotides into polynucleotide chains) and a single small primer (starter sequence), which is likely to occur at random in a genome. Advantage: Easy and quick. Disadvantage: Not always reliable. Not easy to target a specific area of the genome. Can not distinguish heterozygotes from homozygous dominant (less genetic information than RFLPs). 3. Microsatellites Uses PCR with long (and therefore more specific) primers. Identifies areas with long tandem (one-after-the-other) repeats of a short DNA sequence. Name refers to appearance of this type of DNA after centrifuging a fragmented sample of total DNA: a microsatellite apart from the rest of the DNA. Advantage: Can target a specific area of the genome. 523 More variability than any other market type—can detect variation among more closely related genotypes than can be distinguished by other marker types. Disadvantage: Expensive. Requires DNA sequence analysis to create the primers. 4. AFLPs (Amplified fragment length polymorphisms) A combination of RFLP and PCR technology: Cut DNA with restriction enzymes. Attach short DNA "adapters" (known primers) to the ends of the fragments. Amplify the fragments using the primers. Advantage: Reproducible (reliable), easy. Disadvantage: Can not easily target a specific area of the genome. Usually does not distinguish heterozygote from homozygous dominant. Basically...differences between molecular marker types are related to how much they cost vs. amount of information and reliability Uses of molecular markers 1. Germplasm screening Measure differences and assay relationships. Easiest with AFLPs or RAPDs. 2. Construction of genetic maps General-purpose tools for gene tagging, quantitative trait analysis, and map-based cloning. 3. Tagging of genes for indirect selection - To find molecular markers closely linked to the gene of interest, especially for difficult- or expensive-to-measure traits (e.g. insect resistance, traits with high GxE), or traits that can be measured only late in the life cycle (e.g. post-harvest perishability). - Then pre-select based on the marker (and select against undesirable linked genes), confirm field phenotype with smaller number. - Increases efficiency of field testing. - Less expensive? depends on the situation and the trait! 4. Quantitative trait analysis - Follow segregation of a large number of RFLP markers in a single cross. - Correlate presence of a certain marker (e.g., chromosome segment) with the quantitative trait. - Assemble genotypes containing multiple favorable chromosome segments (markers) for the quantitative trait (QTLs). 5. Map-based cloning of genes 524 A method of isolating genes for genetic engineering, when there is little information about the gene. 6. Analysis of introgression Increased speed of gene transfer via backcrossing. Select against unfavorable linked genes from donor parent. IV. Traditional and Contemporary Fermentation Biotechnology Used for centuries to preserve and detoxify cassava Improved starter cultures can enhance safety, nutritional value, and consumer appeal of traditional cassava foods. Contemporary fermentation technology: a growth industry that uses cassava substrate to produce a range of products, from animal feeds to industrial enzymes and pharmaceuticals. Need for consumer-oriented market research and product development activity. Microbial biotechnology for waste management Treatment of cassava processing waste water and solids Environmental protection, clean water conservation Detoxification Reduction of biological oxygen demand Value added? 525 Workshop Participants China Zheng Xueqin, Wang Shengxian, Lin Xiong, Zhang Weite, Li Kaimian, Huang Jie, Jiang Changshun, Tang Yuedong and Wu Hao Chinese Academy of Tropical and Agricultural Sciences (CATAS) Baodao Xincun, Danzhou, Hainan, 571737, P.R. China Fax (86-890) 330-0776 or 330-0157 Du Daolin Dept. Biology, Hainan Teachers College Haikou, Hainan 571158, P.R. China Zhuang Zhong Tang and Liu Jin Ping Honghe Animal Husbandry Station of Yunnan Livestock Academy, Science and Technology Service Dept. West suburb, Mengzhe county Yunnan, 661100, P.R. China Tiang Yinong and Li Jun Guangxi Subtropical Crops Research Institute (GSCRI) 22 Yongwu road, Nanning Guangxi, P.R. China Fax (86-771) 280-6902 Yang Mu Cheng Guangxi Science and Technology Information Institute Nanning, Guangxi, P.R. China Liu Yin Fei Liuzhou Science and Technology Information Institute Liuzhou, Guangxi, P.R. China Fang Bai Ping and Fu Guo Hui Upland Crops Research Institute (UCRI) of Guangdong Acad. Agric. Sciences Wushan, Guangzhou, Guangdong, 510640, P.R. China Fax (86-20) 875-03358 He Ya Wen, Ma Guo Hua and Zhu Yi South China Institute of Botany Guangzhou, Guangdong, P.R. China 526 He Lui Zhi Science and Technology Department Haikou, Hainan, P.R. China Ou Ji Chang Planning Department Haikou, Hainan, P.R. China Liang Feng Fei Nankun Starch Factory Nankun, Tunchang County, Hainan, P.R. China Liu Kuan Tang Long-chang Agriculture Development Company Chang-Jiang County, Hainan, P.R. China Colombia Sam Fujisaka, Guy Henry and Ann Marie Thro Centra Internacional de Agricultura Tropical (CIAT) Apartado Aereo 67-13, Cali, Colombia Fax (57-23) 445-0273 India G.T. Kurup, S.G. Nair, C.R. Mohan Kumar, V.P. Potty and S. Ramanathan Central Tuber Crops Research Institute (CTCRI) Thiruvananthapuram, Kerala, 695017, India Fax (91-471) 550-063 S. Thamburaj Horticultural College and Research Institute Tamil Nadu Agricultural University Coimbatore, Tamil Nadu, 6410003, India Fax (91-422) 431-672 Indonesia Yogi Sogito, Soemarjo Poespodarsono, Wani Hadi Utomo and Heru Santoso Brawijaya University Malang, East Java, Indonesia Fax (62-341) 364-487 527 Suyamto H., Director Research Inst, for Legumes and Tuber Crops (RILET) P.O. Box 66, Malang, East Java, Indonesia Fax (62-341) 801-4% J. Wargiono Bogor Biotechnology Institute Jl. Cimanggue 3A Bogor, West Java, Indonesia Fax (62-251) 312-755 Usman Kartawijaya and Palupi Puspitorini P.T. Great Giant Pineapple Coy Jl. KH Moch, Salem 28, Way Lunik Panjang-Bandar Lampung, Lampung, Indonesia Fax (62-721) 31972 or (62-725) 41742 Malaysia Tan Swee Lian Malaysian Agric. Research and Developm. Institute (MARDI) G.P.O. Box 12301, 50774 Kuala Lumpur, Malaysia Fax (60-3) 948-3664 or 948-2961 Philippines Fernando Evangelio, Editha Gundaya and Reynaldo Bergantin Philippines Root Crops Research and Training Center (PRCRTC), 8 Lourdes Street Pasay City 3129, Metro Manila, Philippines Fax (63-2) 588-692 Thailand Chareinsuk Rojanaridpiched and Vicharn Vichukit Faculty of Agriculture, Kasetsart University Chatuchak, Bangkok 10900, Thailand Fax (66-2) 579-8536 Cham Tirapom and Preecha Suriyapan Department of Agriculture Chatuchak, Bangkok 10900, Thailand Fax (66-2) 561-3486, 940-5418 528 Watana Watanonta, Jarungsit Limsila, Anuchit Tongglum, Danai Suparhan and Somphong Katong Rayong Field Crops Research Center Huay Pong, Rayong, Thailand Fax (66-38) 681-515 Wilawan Vongkasem, Kaival Klakhaeng and Somnuek Hemvijit Rice and Field Crops Promotion Div. Dept. of Agricultural Extension (DOAE) Charuchak, Bangkok 10900, Thailand Fax (66-2) 579-2622 Piyawuti Poolsanguan Sri Racha Research Station of KU Sri Racha, Chonburi, Thailand Vanna Changesrisook and Somporn Phongvutiprapan Thai Tapioca Development Institute (TTDI) 1 168/32 Lumpini Tower, 16th floor Rama IV Road, Bangkok 10120, Thailand Fax (66-2) 285-6647 Kazuo Kawano and Reinhardt Howeler CIAT Regional Office, Dept. of Agriculture Chatuchak, Bangkok 10900, Thailand Fax (66-2) 940-5541 Vietnam Truong Van Ho, Trinh Thi Phuong Loan and Hoang Van Tat Root Crops Research Center National Inst. Agric. Sciences (INSA) Van Dien, Thanh Tri, Hanoi, Vietnam Fax (84-4) 253-525 or 613-937 Thai Phien Institute for Soils and Fertilizers Chem, Tu Liem, Hanoi, Vietnam Fax (84-4) 834-3924 529 Tran Ngoc Ngoan, Nguyen The Dang, Le Sy Loi and Dinh Ngoc Lan Agro-forestry College of Thai Nguyen University Thai Nguyen, Bac Thai, Vietnam Fax (84-28) 852-921 Pham Van Bien, Hoang Kim and Nguyen Huu Hy Institute of Agric. Science of South Vietnam (IAS) 121 Nguyen Binh Khiem Street Ho Chi Minn city, Vietnam Fax (84-8) 8297-650 and (84-61) 868-120 Pol Deturck Vietnam-Belgium Project on Sustainable Agric. Developm. in the Uplands of South Vietnam Institute of Agric. Science of South Vietnam (IAS) 121 Nguyen Binh Khiem Street Ho Chi Minn city, Vietnam Fax (84-8) 8297-650 530 APPENDIX Results of Soil Analyses in Asia 1994-1997 R.H. Howeler1 The following tables present the analysis results of soil samples taken in various countries in Asia, mainly in soil fertility maintenance experiments and in FPR trials in farmers fields. To facilitate interpretation of the results, Table 1 indicates the approximate classification of soil chemical characteristic according to the nutritional requirements of cassava. Table 1. Approximate classification of soil chemical characteristics according to the nutritional requirements of cassava. Soil parameter" Very low Low Medium High Very high pH <3.5 3.5-4.5 4.5-7 7-8 >8 Org. matter (%) <1.0 1.0-2.0 2.0-4.0 >4.0 Al-saturation (%) <75 75 -85 >85 Salinity (mmhos/cm) <0.5 0.5- 1.0 >1.0 Na-saturation (%) <2 2- 10 >10 P (Mg/g) <2 2-4 4- 15 >15 K (me/IOOg) <0.10 0.10-0.15 0.15 -0.25 >0.25 Ca (me/IOOg) <0.25 0.25- 1.0 1.0-5.0 >5.0 Mg (me/IOOg) <0.2 0.2-0.4 0.4- 1.0 >1.0 S (Mg/g) <20 20-40 40-70 >70 B (/ig/g) <0.2 0.2-0.5 0.5 - 1.0 1 -2 >2 Cu Ozg/g) <0.1 0.1 -0.3 0.3 - 1.0 1 -5 >5 Mn 0-ig/g) <5 5 - 10 10- 100 100 - 250 >250 Fe (/ig/g) <1 1 - 10 10- 100 > 100 Zn (/ig/g) <0.5 0.5- 1.0 1.0-5.0 5-50 >50 11 pH in H2O; OM by method of Walkley and Black; Al saturation = 100xAl/(Al + Ca + Mg + K) in me/IOOg; P in Bray II; K, Ca, Mg and Na in IN NH4-acetate; S in Ca-phosphate; B in hot water; and Cu, Mn, Fe and Zn in 0.05 N HC1 +0.025 N H2SO4 CIAT Cassava Asian Regional Program, Dept. Agric., Chatuchak, Bangkok, Thailand. 531 Table 2. Soil samples taken in China. Sample no. Sample location and description Date Lab Series Hainan -1 Quongzhong county, Changzhang- area cleared for cassava -2 Baisha county, Kongba village - black soil -3 Baisha county, Kongba village - black soil in fertilizer trial -4 Kongba village - Mr. Pu Yong Chuan, FPR fert. trial check -5 Kongba village - Mr. Tan Jing Zhou, FPR fert. trial check -6 Kongba village - Mr. Tan Jia Chai, FPR fert. trial check -7 Kongba village - Mr. Tan Yin Zhui FPR fert. trial check -8 Kongba village - Mr. Tan Ya Zhui, FPR fert. trial check -9 Kongba village - Mr. Zhou Shao Xiong, FPR fert. trial check -10 Kongba village - Mr. Tan Ming Lei, poor eroded soil -11 Kongba village - Mr. Tan Yen Chai, FPR erosion trial -12 Kongba village - Mr. Zhou Ya Nin, FPR fert. trial check -13 Kongba village - Mr. Tan Ya Tian, FPR fert. trial check -14 Kongba village - Mr. Ma Guo Rong, FPR fert. trial check -15 Kongba village - Mr. Fu Yong Chuan, FPR fert. trial check -16 CATAS - NPK trial, 1995, 4th year, NoP^ -17 CATAS - NPK trial, 1995, 4th year, N3P3K3 -18 CATAS - NPK trial, 1996, 5th year, NoPoKo -19 CATAS - NPK trial, 1996, 5th year, N,P3K3 Guangxi -1 Nanning, NPK trial, 1994, 6th year, N^oK,, -2 Nanning, NPK trial, 1994, 6th year, N3P3K3 -4 Nanning, NPK trial, 1995, 7th year, N^oK, -5 Nanning, NPK trial, 1995, 7th year, N3P3K3 -6 Nanning, NPK trial, 1996, 8th year, NqPoKo -7 Nanning, NPK trial, 1996, 8th year, N3P3K3 Guangdong -1 Huayji, on-farm NPK trial, before planting -2 Gaozhou Agric. College, NPK trial, before planting -3 Yunnan, on-farm NPK trial, before planting Nov 95 S498 Nov 95 S498 Nov 95 S498 Jan 95 S498 Jan 95 S498 Jan 95 S498 Jan 95 S498 Jan 95 S498 Jan 95 S498 Jan 97 S782 Jan 97 S782 Jan 97 S782 Jan 97 S782 Jan 97 S782 Jan 97 S782 Mar 95 S498 Mar 95 S498 Mar 95 S782 Mar 96 S782 Feb 94 S891 Feb 94 S891 Mar 95 S498 Mar 95 S498 Mar 96 S782 Mar 96 S782 Mar 96 S782 Mar 96 S782 Mar 96 S782 C-J IT> u .5 A 3 ■ i ■a ■a >.■g. 1 J3 i "S «S £ | V V <2 A A « N ■c a§ u v 3 a* O « E G j5 ,£ J5 J2 •j 5 5 .j "o 7* ~ -:-:-:-g ^^"g-:-:-: >• "2 ~ "2 "2^ ^ ^ « 5 J2 B sJ .J sJ - B .J B B s«wirji3~~3www*u8oi33s 80\O^NrsO--*aN (N r-■ no oc w, %o o no oc ■ Tt . r- 2 oc in c r-l r* oc 3oc in ™ — O 1 o . o cn OOOOOOOOOOOOOOO — ooooorsoo'tr^\0— oo oo oo r- /-i C~ oc o O o o o O O O O 5 Os nn s CI O CIO *^> ci (-1 CI CI CI Cl m o m oc 0C c- -t CI o — — CI CI — — o O o w, o O CI oc u-. m p- m Os 8 f oc ^> * r*CI O 3C -T OC in i») - CI c m CI CI CI — — —i — rS E J, CI ic, */-. r- oo , o ^<^- in t ^r t * sc in *o -^ N in Tt w*i *o r- rnl Tt r) « o 533 Table 4. Soil samples taken in Indonesia. Sample no. Sample location and description Date Lab Series E-Java -1 Malang, Jatikerto -2 Malang, Jatikerto -3 Blitar, Ringinrejo -4 Blitar, Ringinrejo -5 Blitar, Ringinrejo -6 Blitar, Ringinrejo -7 Blitar, Ringinrejo -8 Malang, Dampit, -9 Malang, Dampit, -10 Malang, Dampit, -1 1 Malang, Dampit, NPK trial, 1995, 8th year, NoP^ NPK trial, 1995, 8th year, N3P3K3 Demonstration plots, black soil + lime , Mr. Ponirin - erosion trial, red clay Mr. Hardi - variety trial, black soil , Mr. Katimin - erosion trial , white purple clay in road cut demonstration plots - T2, severe K def. demonstration plots - T,2, no K deficiency Ir. Noviar erosion trial, severe K def. K trial before planting Oct 95 S498 Oct 95 S498 Febr 96 S498 Febr 96 S498 Febr 96 S498 July 96 S782 July 96 S782 July 96 S782 July 96 S782 July 96 S782 Sept 96 S782 Yogyakarta -1 Playen, farmer's field - Fert x soybean trial Febr 96 S498 Lampung -1 Umas Jaya, NPK trial, 1996, 9th year, Av. N^Ko -2 Umas Jaya, NPK trial, 1996, 9th year, Av. N3P3K3 -3 Tamanbogo, erosion trial, 1994/95, 5th year, T, -4 Tamanbogo, NPK trial, 1994/95, 4th year, N0PoKo -5 Tamanbogo, NPK trial, 1994/95, 4th year, N3P3K3 -6 Tamanbogo, NPK trial, 1995/96, 5th year, N0PoKo -7 Tamanbogo, NPK trial, 1995/96, 5th year, N3P3K3 Febr 96 S498 Febr 96 S498 Dec 94 S782 Dec 94 S782 Dec 94 S782 Sept 95 S782 Sept 95 S782 .3 s a ■§ c .S .-3 8 CO 93 .a u "3 .3 e M "3 u ■c a 2 « 1 * CO x u -o v v I a uA u. 3 c N V CQ "5*3 A U 00E V 5* Ec3 v^ S*1- t* § o. D. E «ooinioinmTtoo-Hvi 5 ^odoobtNonontNtr>tr! d flw"lc>toC>Vir^c>TtO viriririo*r^ood-iri f-l^-h^-h m m n fl 'i •* ~* n n N ■*—•a-—■ tN r- o oo oo r^ 1 no^o^6-H^mri 't t— n N -H o w"> r"-r^r-r-r-oooriooo ~©©©c>©~*c*J-h~H r- 3 —■ m <*©■■» r- r- oo o ooddoooooo © O-hO — OOOTt1/^mTt o — ooooooooo o Qnn©©m©e>-Ho© n n -< wi —■ od 6 ri ri it n *oo*inNOiov>Nno f)»Hqooo«o*Nr- idr-6-nit-cooodoooi ^ •—I Cn r/l (D $ * t 1 t * n n n r» © -H © r) © -H TtinTtvi odoooood6od *rl c> N 't o t^ooTt — 1 ' I ill* ■ I H H i o u >-. >> c c CD CD 00 -O ci en oo ■* T3 C C) ri w"> o\ 43 — vn d ri e> i 1 >d 1 d m m m mm N r- w"> 6 6 d ■ 1 d 1 d O n . 1 ri ri n m m § 00 00 m m o N m m ^ r- n O * 3 r- o o f> o o on O N © O tN — — d o o d S o o oo o r- o * in m ci N n ^ ^ Tt t d d d d d d d U w-i q On oo * r- 00 00 ■"" "* O d d d d s u-i P H"l * «!3 d m r-i * Tt m "^ O m s nd -d m ■* ™ d ri ■o w-^ * o ^O *. —-* ^—< n i i T "? VO r> 0Jl c 3 >, a Eaoo > 2 2CO .3 a a ■S j: 9. £ .a a S 3 8 "a B no M •c u T3 T3 C a. E 08 GO o c a. E rl CI cN CI CO 00 00 00 oo F 00 GO GO GO GO no VO no no nO o O § o\ o OO O O ra o BO E cd >n« g00 G u E o 1 E CO 1cd >* X) V) o VI .a 2 ■g 3 c S 00 3 5s u U > CO iu o a.> c .= « > .5 >^ c > .C is o o 1 u c■5 T3 .32 >n 5 a. '5in cdcd 00 2 > « en en 2* 13 o c cd £ cd IS E Cd O cd M cd O> 00 > 1 a. Cfl c cd 3 en en 3 Cd o s 2o GO u u o u. (-. 00 6* oo ■ scd _M cd s 3 3 o .■> > > •o co a cd cd cfl N * > * _> s"s 03 cd cd GO o -C J= J3 cd cd cd c E E E cd GO _-; M cd cd 1> a. a. a. * 3 . „ ^ E 00 o o § _o iso 3c c c cd c o CQ o o CD cd ffi GO t V o CO .5a.a 3 J3 9. CV £ .s .-3 eg g es JC 0 "3.s C/i JS a. T3 e .3 E ov X s * S 0, v v a 3 u £S c N V CQ 3 A * E o u u v^ E &0- 2 o ' 6 -1 ■* nq „ 1/1 ■* ' ri on r- r| m <> . d CI O 00 rl T♦ N CO 00 CO . 00 o _ nO 00 tN ■* I/-1 c> CI oo ei nO in #* m *-i t» m io 8 cy- nO VO c> VO -h " — ri d VO 5 oo 5 00 ^ (N CO «N wo N oo o ri N -^ ri o r- r- no ■* ■* Tt ■* E cd _o •o T 7 o o CQ 536 Table 8. Soil samples taken in Thailand. Sample no. Sample location and description Date Lab series N.Ratchasima Sra Kaew -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 -14 -15 -1 -2 -3 -4 -5 -6 Soeng Saeng, Mr. Nonglak - FPR erosion trial Soeng Saeng, Mr. Saway - FPR erosion trial Soeng Saeng, Mrs. Kaew - FPR erosion trial Soeng Saeng, Mrs. Thong - FPR erosion trial Soeng Saeng, Mrs. Lamun - FPR erosion trial Soeng Saeng, Mrs. Nuukaan - FPR erosion trial Soeng Saeng, Mrs. Kong - FPR erosion trial Soeng Saeng, Mrs. Wongduean - FPR variety trial Soeng Saeng, Mr. Thongchan - FPR fert. trial check. Banmai Samrong NPK trial, 21st year, 0-0-0 Banmai Samrong NPK trial, tops incorp + 8-8-8 Huay Bong, Daan Khun Thot - TTDI - demonstr. plots Huay Bong, Daan Khun Thot - TTDI-long-term trial 1-1 Huay Bong, Daan Khun Thot - TTDI-long-term trial II- 1 Highway Paakchong-Huay Bong-Field with Fe/Zn def. Wang Nam Yen, Wang Nam Yen, Wang Nam Yen, Wang Nam Yen, Wang Nam Yen, Mr Mr Mr Mr Mr Wang Nam Yen, Mr Sawong - FPR erosion trial Bunlue - FPR erosion trial Udom - FPR erosion trial Prichaa - FPR erosion trial Lek - FPR variety trial Bunlue - FPR erosion trial July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 Mar 96 S-498 July 95 S-498 July 95 S-498 July 95 S-498 July 96 S-782 July 96 S-782 July 96 S-782 July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 July 95 S-498 Dec 95 S-498 Prachin Buri -1 Khaw Hin Sorn-grass barrier trial July 96 S-782 X u H U •a * a V V T3 C« C/3 A A£ 3 u c N V CD ■c u s A U E S g o E O v^ 68 o i a. a E >> -. . . >• caUOcacaOcaca ndm — ndw-)-'tr-;inO;o\mo\<»C>ono\n0O*Tj- n>dooo^- — MNOot**" ONn»r;N*f>tto^4£>ovi cindridd--ddddddd---in^:r-TJ'(:ob — 'tnoin^r-^monmNtN -^-HOqoooorj©OO dddddddd^dddooo l>r-lin©mr-mmoo'tc*i."t-HVOm -* — H Tt * H N ©\ ^; — ooo*oc-Nomnor-onoi— ei-*t*t*****~«~lm****-i © -- Tt HNnnmor^tooOrtNn^in ■ ■ ■ i iTt i i ^- - „ -„ - - - i i i i i i E n <2 •3-3 o in 00 VO ci VO on r- nO - O - m m — ri -*■ *"". m ■* o m n IN m O r- o o O o o o o 00 —1 m in •* r- r- oo in « o*pi on in o m O O o o O O O o o o o o o o Cl o-<* 00 00 OtN ** o o o o o O o 00 r- VO 00 O m 00 00 o Tt ■* nO O o O o o O O CI a Tt Tt i ci Tt -h <> i> —< ■* o o o wo Tt vi wti -t -f in d ci "n r^ m in N on s >• >• 6 C -a c a A 1 , , , o J3 wi W5 i o — 00 CJ ■•t 00 so n O q Os ri m ■O sq os1 ' ri d 1 1 d ri m rl ri 1 <^ r- d d 00 m Os V0 r~ rt i ri in m ri m 8 d riri ■* r4 rvl nsO NO ri ri o o Os NO r* 00 t-; q so — — r- n q ri m os ri n ifl in ~~ -st CO so i—i ob ri sci 1 00 00 ri r~ d m F ri ri o 8 00 00 rl CO00 sO00 r- r- m sO s O 0000 q q m rl os oo ci d — —' d d d d d d d d 1 d d ■ d d d sb OO 00 ri o o so -o so m q ri _H — o r) m t> 00 00 m rl so d r- o\ os sO ri r~ m cn sd ^6 r-i ri •o ri r-> ri •o d d o- 00 Os o ri 00 ~ Zl Tf ri ri 5 ri r- sOCO ■* 00 O OCO Cs r- so r- 00 ■-■ — — — d »H — .—■ d d — d i d O 1 d d d ri ri m rl CO so m os Os o o \0 ^ ri ori 00 r~ Tt m n osrl n m m m n m O O O o o o O o O O o o i o O O O o o o o 00 m Os n 00 m m Os rl — s Osm o s o s n o O mrl SO m m ri oo rl oo r~ m 00 os rn 8 mO g s n mo 8 os r-ri om CI CO o O o O d d d d d d d d d d d d d d d o d d d d d 3 nm so oo■■t 5 rlm r- •* n ri n o CI n o O in O 't o m m m O 6 d d d d d d d d d d d d d d d d d d d 8 o os o os rim o 00 3 00 mm 8 n osr- mr~ rl Os r~ rl ri m — — — * Tt ■* ■* m Tt ■•t ■st CO 't -■t 00 Ml B -C .s Z f2 CQ 0- .5 > E« G >s•^1 03 o ^o^Sireg 'pn ''m ssajj dn pire s\or[ : Suijuuj Suoipuiaq}} imipidy : WK JvDttfdvjQ aufcuSutuireiQ inuoy^ uiooopja^ jiduioj : Sutss30ouj pio/^ jajaMoj-i \\ jpjsquia^ : Suijtpg Bisy joj unuSojj |Ruoj«foy FAfissrr) uopraiiqnd IVI3 V