Cassava is produced under diverse ecological conditions and production systems. Although this diversity makes generalization difficult, this paper attempts to provide an overview of recent developments in the area of cassava agronomy. Cassava is produced mainly by small farmers in multiple purpose and complex production systems. This complexity at farm level increases the risk of generalization since agronomic practices tend to be site specific.
Only a few countries and notably Thailand, produce cassava primarily as a single crop over relatively extensive areas of land. Monocropped cassava is rare in the rest of the world.
The available information about modern cassava technology has improved considerably in recent years with the advent of international cassava information networks. Unfortunately, information from some areas of the world where cassava is an important food crop, has not kept pace with recent improvement in communications. This is particularly the case in some African and Southeast Asia countries.
Despite the diversity of physical environments in which cassava is cultivated, most of the information relevant to cassava agronomy has been generated on experimental stations, and is frequently delivered without on farm validation.
Although cassava is in a more favourable position than other roots crops fewer resources are allocated to cassava research at both national and international levels than to other crops, particularly grains (Cock, 1985). This lack of resources negatively affects technology development and transfer to farmers. Under the present economic crisis facing Third World countries, underfunding for research, development and transfer of technology for cassava, will continue to limit the development and application of better agronomic practices.
In this article a brief overview of the most important cassava based cropping systems in different ecological areas is presented, together with relevant recent research results that could lead to improvements in cassava production.
CASSAVA CROPPING SYSTEMS
In the Low Humid Tropics (0–3 mo/yr with less than 60 mm rainfall) cassava is produced year round in a multiplicity of cropping systems. Land clearing and soil preparation demand high inputs of hand labour due to the vigorous growth of native vegetation in this ecosystem, therefore, weed control is the most labour demanding cultural practice, after crop establishment. Declining soil fertility and root rots limit cassava yields when production continues on the same plot.
In the Semi arid Tropics (7–9 dry mo/yr) cassava, develops under water stress during most of the growing period. Uncertainty about the onset and end of the rainy season negatively affects hand labour planning and allocation, concentrating peaks in labour demand before and during planting. After one or two years of drought the availability of planting material becomes critical. Similarly, mite attack can be devastating in conjunction with prolonged periods of drought. Nevertheless, cassava is one of the few crops that can produce reasonable yields under these extreme conditions.
In the Seasonally Dry Tropics (4–6 dry mo/yr) cassava is usually produced in combination with several other crops. Timing of land preparation and weed control during the establishment phase of the crop are critical to achieve high yields. Marketing considerations, and particularly prices for fresh roots or dry chips, determine the level and quality of farmer management of cassava production.
In the Subtropics, cassava's growing season is defined by temperature rather than precipitation. In the Southern Hemisphere cassava stakes are cut at the onset of the winter, stored, and planted as soon as the temperature rises again. Early and late cassava varieties are commonly intercropped or relay cropped with several other species.
COMPONENTS OF IMPROVED TECHNOLOGY
Although one of the most energy demanding activities for cassava production, land preparation has received comparatively little research attention, and only a few practical recommendations are available.
In areas at risk from flooding, soil should be ridged before planting. However, although fresh root yields in ridged and unridged plantings in short term experiments, are not consistently different, it is generally accepted that ridging is beneficial in these types of environments (Lozano, 1987; Rodriguez, 1990).
The performance of cassava under different tillage systems is rather site specific. No-tillage, reduced tillage and conventional tillage have been tested in different ecosystems with variable results in terms of yield (Ofori, 1973). Soil preparation is more important for effective weed control than as a means to improve the microenvironment for root bulking. For example, acceptable yields are obtained by small farmers using zero-tillage, whilst no significant differences in yields have been obtained between “conventional” tillage (plough and harrow) and different forms of reduced tillage. Deep soil preparation does not result in better yields, except under very specific circumstances. The most common form of reduced tillage is the local removal of soil around the area where the cassava stake is to be planted. This practice, although rather primitive, (Okigbo and Greenland, 1976) is still widely used today in different parts of the world.
Mulch effectively increases yields when zero-tillage is used (Hulugalle and Opera-Nadi, 1987), though in semi arid zones, mulching is more important as a means to conserve water than to improve the soil condition around the root system.
Because of its ability to grow in poor soils, cassava is frequently cultivated on steep land, which because of the slow initial growth of cassava can result in soil erosion during the first three months after planting. The development of improved technology to reduce erosion is one of the most serious challenges facing cassava research today.
Contour ridges alone, or in combination with live barriers and zerotillage provide effective erosion control under experimental conditions in Latin America and Asia, though, the acceptance of this technology by farmers has not yet been evaluated (CIAT 1989, 1990).
The literature abounds with recommendations related to planting material. Cutting size, planting position, phytosanitary treatment and other topics have been widely researched around the world. Most of these results, obtained on experimental stations, are published with relatively poor descriptions of the conditions under which the experiments were conducted. The applicability of the recommendations to farm conditions is often unclear or highly site specific.
In most production systems, vigorous stakes selected from the middle portion of the basal branches of fertilized mother plants result in better yields. A basic formula for phytosanitary treatment of cuttings is available for use by farmers (CIAT, 1985). Storage of planting material for up to 4 months before planting can be achieved by placing them in the shade and burying the tips of the stakes in the soil (CIAT, 1990).
As with many other crops, planting densities and spatial arrangements currently used by farmers are determined by several factors, not necessarily related to yield. The availability of planting and land preparation implements, the practice of intercropping, weed incidence, water holding capacity of the soil, and market considerations are among the most important of these factors (Norman, 1979).
Increases in planting density normally result in higher yields of smaller roots, a greater labour requirement for weed control (during the establishment phase of the crop) and intense use of available planting material (Cock 1978).
Due to the American origins of cassava, more genetic diversity is present in the Neotropics. New cultivar development in the rest of the world depends heavily on the availability of germplasm obtained from Latin America.
In Latin America most small farmers generally grow more than one variety simultaneously. Some phenotypes tend to be more commonly cultivated than others in individual production zones, though in some production areas, near well-defined markets, single varieties are grown.
Brazil is by far the largest cassava producing country in Latin America, accounting for almost 70% of total production. In the Humid Wet Tropics of Northern Brazil, the cultivars IM-158, IM-168, IM-175 and BGM-021 have been recently released due to their tolerance to Phytopthora and Fusarium spp. (Fukuda 1990; CIAT 1990). The cultivars IAC-12-829 and IAC-576-70 were released by the Instituto Agronomico de Campinas in southeast Brazil. In the South sub-tropical region, the cultivars Aipim, Pioniera and Gigante, released by national institutions, continue to be widely cultivated. No improved varieties are presently cultivated by farmers in semi-arid northeastern Brazil, the largest production area of the country, containing almost 58% of the total area planted to cassava (Fukuda, 1990).
Colombia is the second largest cassava producer in Latin America. The variety Manihoica P-12 has been released in the North Coastal area of the country, whilst varieties CG 1141-1 and CM 3306-4 are in the pipeline for release in 1991 after evaluation by 400 small farmers (Lopez et al., 1987). Varieties ICA-Sebucan and ICA-Catumare were released in 1990 for the Llanos ecosystem (Rodriguez and Hershey, 1989). Other cultivars such as Manihoica P-13 (HMC-1) have been released in the Valle Department.
Cassava is very important in Cuba where the early CIAT cultivar CMC-40, of Brazilian origin, is grown together with the intermediate local selection CEMSA and the traditional late variety Señorita, to guarantee cassava availability in the market during most of the year. Recent releases such as CEMSA 5–19, CEMSA 74–6329 and Jaguey Dulce are of more restricted ecological adaptation (Rodriguez, 1990).
Paraguay is the largest per capita consumer of fresh cassava roots in Latin America. Among several local cultivars with excellent agronomic characteristics, Meza-i has been recently recommended by the extension service (SEAG, 1989).
The varieties Dayana in Panama (Chavez, 1990) and MCol 2205 in Ecuador (Hinostroza, 1990) are recent releases, which are just beginning to be cultivated by farmers.
The national institutions of Thailand and Indonesia release more improved cassava cultivars than other Asian countries. In Thailand, Rayong 1, probably the world most successful cassava cultivar, served as parental material for the development of Rayong 2, released in 1984, and Rayong 3. The latter has a very high dry matter content and, although only recently released, is already extensively cultivated (Sinthuprama et al. 1987; CIAT 1990).
In Indonesia, Adira 1 is widely cultivated by small farmers due to its low HCN content and ability to grow in intercropping systems. Adira 4, with a slightly higher HCN content (90 ppm), was released in 1986 for industrial purposes and is now cultivated on almost 20,000 ha, mainly in Sumatra (Soenarjo et al., 1987; CIAT, 1990).
In the Philippines, the Philippine Root Crop Research Center released the cultivars Kalabao, Golden Yellow and Colombia in 1980. In 1986, the cultivar CM 323-52 was released under the name UC-1. The University of the Philippines at Los Banos recently released the sweet cultivar Lakan 1 and the bitter cultivars Datu 1 and Sultan 1 (Mariscal, 1987; Carpena, 1987).
In China the local selection SC 205 (South China 205) and the introduced cultivar from Colombia CM 4031-2, are widely cultivated (Lin et al., 1987).
In South Vietnam, the Thai cultivars Rayong 60 and Rayong 1 outyielded local cultivars, and are in the pipeline for immediate release (CIAT, 1990).
Little information about cassava varieties recently released in India and Africa is available. Apparently five high yielding hybrids were released in India around 1987 (Nayar et al., 1987).
Research on cassava intercropping is relatively more recent than other aspects of cassava production. Considerable research effort has been dedicated to gaining a better understanding of interactions between components of crop associations (Leihner 1983). Several publications dealing with intercropped cassava in specific environments, particularly in Asia and Africa, are available, though most of these research results are applicable only in very specific conditions.
Due to the slow initial growth of the crop, Land Equivalent Ratios values above 1 are frequently obtained when cassava is intercropped with short cycle annual crops such as common beans, cowpea or vegetables. With crops such as maize, sorghum or pigeon peas, the ability of cassava to recover a full leaf area after the harvest of the intercrop is the main reason for LER values above 1. Generally, the cultivation of cassava under tree crops does not negatively affect the yield of the trees, and therefore values for LER above 1 are frequently obtained.
Maize is probably the most common annual crop grown in association with cassava. Improved short maize cultivars intercropped with cassava tend to yield more than traditional cultivars and also result in higher cassava yields (CIAT, 1988).
Cassava requires effective weed control, and especially during the establishment phase of the crop, for optimum yields. In traditional agriculture, weeds are controlled through cultural practices such as planting density, the use of vigorous cultivars, intercropping, reduced tillage, cover crops, use of mulches, etc. Most recent research on weed control in cassava has shown that in addition to the application of selective pre-emergence herbicides, at least one hand/hoe weeding is necessary for optimal yields. Among the most researched and recommended preemergence herbicides are fluometuron, diuron and alachlor. Paraquat has also been recommended for post emergence application as a complement to hand/hoe weeding. However, the most widely used herbicide combination for preemergence is probably a tank mix of diuron with alachlor in a variety of doses according to the soil characteristics. This mix is also effective for use with a cassava/maize intercrop when planting of the two crops is either simultaneous or only few days apart (Doll and Piedrahita, 1976; Moody, 1985).
Cassava is grown on a great variety of soils, but is mainly found on ultisols, oxisols and entisols. While the crops grows well with little or no fertilization, it responds well to fertilizer application in infertile soils. The high cost/benefit ratio of cassava fertilization in infertile soils was shown in a series of experiments coordinated by the FAO Fertilizer Program (FAO, 1980).
Cassava responds to P application in infertile oxisols, except in those with high mycorrhizal populations, while N response is found only in sandy soils low in organic matter content (Howeler and Cadavid, 1990).
There is also a marked positive response in root production to applications of K when cassava is grown continuously in the same field for more than 2–3 years (Howeler, 1990a).
In soils with very low levels of available P, high rates of P application are recommended for one or two years in order to increase the available P content in the tissue above the critical level. Since cassava takes up relatively small amounts of P and is highly efficient in P use subsequent P applications can be reduced (Howeler and Cadavid, 1990; Howeler, 1990b). In soils low in organic matter or available N, 50–100 kg N/ha are recommended per crop cycle. In most tropical soils with very low K supplying power, it is recommended to apply at least 100 kg K/ha annually to sustain cassava yields (Howeler and Cadavid, 1990).
Although cassava is frequently considered relatively tolerant to insects and pathogens, its yields are often negatively affected by pests and diseases. In fallow-based agriculture, several cultural practices such as crop rotations and the inclusion of fallow as part of the rotation scheme, help not only in the maintenance of soil fertility, but also in the control of pests, diseases and weeds.
African Cassava Mosaic Virus is considered one of the most serious diseases affecting cassava production in Africa today. It causes serious yield losses in East, West and Central Africa. This insect-transmitted disease is controlled only through the use of healthy planting material in areas where the reinfection rate is slow. The development of resistant varieties is possible by crossing M. esculenta with M. glaziovi but insufficient virus resistant material is currently available to farmers.
Cassava Common Mosaic Virus is an important disease with an unknown vector. The use of “clean” planting material (Lozano 1989) is consequently the only available control measure.
Bacterial Blight caused by Xanthomonas campestris is another disease of worldwide importance. In addition to the usual control practices cited in the literature, successful control can be obtained through inoculation of planting material with strains of Pseudomonas fluorescens and P. putida (Lozano 1986). P. putida can also be used for the control of Diplodia manihotis, a root rot pathogen (Lozano 1986 and 1988).
Among the important insects pests affecting cassava, the Cassava Hornworm (Erinnyis ello) can be controlled with Trichogramma and Bacillus thuringiensis, but the most promising control practice is application of the hornworm baculovirus (CIAT 1989).
Two species of cassava mealybugs, Phenacocus manihoti and P. herreni, can cause serious yield losses. P. manihoti caused severe yield losses in Africa until the introduction of natural enemies from the Neotropics by IITA and CIAT. Epidinocarsis lopezi, a natural enemy of P. manihoti collected in Paraguay, was released in 1981 in Nigeria and is now established on approximately 750,000 Km2 over a wide range of African ecological zones, helping to maintain low levels of mealybug attach (Bellotti et al. 1987).
The control of the Cassava Green Mite (CGM) in Africa is one of the most serious challenges facing crop protection today. Shipments of natural enemies from the Neotropics to Africa have been made regularly since 1984 as a part of a joint IITA-CIAT biological control effort. Establishment of two species, Typhlodromalus limonicus and Neoseiulus idaeus, has recently been documented in several release sites in West Africa (IITA, 1990). The success of the biological control campaign against CGM in Africa will depend on continued collaboration between international agricultural research centres and national institutions in African countries.
Bellotti, A.C., Braun, A.R., Yan inek, J.S., Herren, H.R. and Neuenschwander, p. 1987. Cassava agroecosystem and the evolution of pest complexes. 11th International Congress of Plant Protection. Oct. 5–9, 1987. Manila, Philippines (In Press).
Carpena, A. 1987. Cassava varietal improvement at the University of The Philippines at Los Banos. In Cassava Breeding and Agronomy in Asia. Proceedings of a Regional Workshop held in Rayon, Thailand, Oct. 26–28, 1987. CIAT/UNDP, p. 55–59.
Chavez, M. 1990. Estado actual del fitomejoramiento de la yuca en Panama. Paper presented at the Reunion Panamerican de Fitomejordores de Yuca. May 21–25. CNPMF/EMBRAPA. Cruz das Almas. BA. Brazil. 10 p.
CIAT (Centro Internacional de Agricultura Tropical). 1989 and 1990. Cassava Program Annual Report. CIAT. Cali, Colombia
CIAT (Centro Internacional de Agricultura Tropical).1985.Yuca: Investigacion, Produccion y Utilizacion. Documento de Trabajo # 50. 656 p.
CIAT (Centro Internacional de Agricultura Tropical). 1988. Cassava Program Annual Report.
Cock, J. 1978.Etapas del crecimiento y desarrollo de la yuca. In Curso de produccion de yuca. CIAT. Cali, Colombia. v 1 p. 1–40. Program Annual Report.
Cock, J. 1985. Cassava. New poten tial for a neglected crop. Westview. 191 p.
Doll, J.D. and Piedrahita, W.C. 1976. Methods of Weed Control in Cassava. Serie EE 21. Centro Internacional de Agricultura Tropical, Cali, Colombia. 12 p.
Food and Agriculture Organization of the United Nations (FAO). 1980. Review of data on response of tropical crops to fertilizers 1961– 1977. 101 p.
Fukuda, W. 1990. Melhoramiento de mandioca (Manihot esculenta Crantz) no Brasil. Paper presented at the Reunion Panamericana de Fitomejoradores de Yuca. May 21-
25 CNPMF/EMBRAPA. Cruz das Almas. BA. Brazil. 23 p.
Hinostroza, F. 1990. El cultivo de la yuca y su mejoramento en Ecuador. Paper presented at the Reunion Panamericana de Fitomejoradores de Yuca. May 21–25. CNPMF/EMBRAPA. Cruz das Almas. Brazil. 21 p.
Howeler, R.H. and Cadavid, L.F. 1990. Short and long term fertility trials in Colombia to determine the nutrient requirements of cassava. Fertilizers Research (In press).
Howeler, R.H. 1990a. Long term effect of cassava cultivation on soil productivity. Field Crops Research (In press).
Howeler, R.H. 1990b. Phosphorus requirements and management of tropical roots and tuber crops. In Proceedings of the Symposium on P Requirements for Sustainable Agriculture in Asia and Oceania. March 6–8 1989. IRRI, Los Banos, Philippines (In press).
Hulugalle, R.L., R and Opara-Nadi, O.A. 1987. Management of plant residues for cassava (Manihot esculenta) production in an acid ultisol in southeastern Nigeria. Field Crops Research 16: 1–18.
IITA. 1990. Biological control: a sustainable solution to crop pest problems in Africa. Yaninek, J.S. and Herren, H.R. Eds. IITA, 210 pp.
Leihner, D. 1983. Management and evaluation of intercropping systems with cassava. Centro Internacional de Agricultura Tropical, Cali, Colombia. 70 p.
Lin, X., Shunuan, W. and Tang, X. 1987. Cassava verietal improvement in China. In Cassava Breeding and Agronomy in Asia. Proceedings of a Regional Workshop held in Rayon, Thailand. Oct. 26– 28, 1987. CIAT/UNDP p. 61–68.
Lopez, A.J., Hernandez, L.A., Her shey, C. and Rodriguez, N. 1987. Tres nuevas variedades de yuca para la Costa Atlantica. Seminario de prelanzamiento. Convenio ICACIAT. Mimeo 27 p.
Lozano, C. 1986. Cassava bacterial blight: a manageable disease. Plant Disease 70: 1089–1093.
Lozano, C. 1987. Alternativas para el control de enfermedades en yuca. Paper presented at the Workshop on Interchange of Germplasm, Quarantine and Breeding of Cassava and Sweet Potatoes. Cali, Colombia, June 8–12, 1987. 31 p.
Lozano, C. 1988. Biocontrol of cassava diseases: challenges and scope. Paper presented at the 5th
International Congress of Plant Pathology. Kyoto, Japan. 22 p.
Lozano, C. 1989. Outbreaks of cas sava diseases and losses induced. Tropical Agriculture Research. Series # 22. Japan. p. 113–119.
Mariscal, A. 1987. Cassava varietal improvement at PRCRTC, Philippines. In Cassava Breeding and Agronomy in Asia. Proceedings of a Regional Workshop held in Rayon, Thailand Oct. 26–28, 1987. CIAT/UNDP. p. 43–54.
Moody, K. 1985. Weed control in cassava, a Review. Journal of Plant Protection in the Tropics 2: 27–40.
Nayar, G. G., Nair R. B. and Rajendran, P. G. 1987. Cassava varietal improvement in India. In Cassava Breeding and Agronomy in Asia. Proceedings of a Regional Workshop held in Rayon, Thailand, Oct. 26–28, 1987. CIAT/UNDP. p. 35– 42.
Norman, N.J.T. 1979. Annual Crop ping Systems in the Tropics. An] introduction, Univ. of Florida Press. 276 p.
Ofori, C. S. 1973. The effect of plou ghing and fertilizer application on yield of cassava (Manihot esculenta Crantz). Ghana Journal of Agricultural Sciences 6: 21–24.
Okigbo, B. N. and Greenland, D. J. 1976. Intercropping Systems in Tropical Africa. In Multiple Cropping. ASA Special Publication # 27. p. 63–101.
Rodriguez, S. 1990. Mejoramento de la Yuca en la Republica de Cuba. Paper presented at the Reunion Panamericana de Fitomejoradores de Yuca. May 21–25. CNPMF/-EMBRAPA. Cruz das Almas. B A. Brazil. 16 p.
Rodriguez, N. and Hershey, C. 1989. Tres variedades promisorias de yuca para los Llanos Orientales. ICA CRI-La Libertad - CIAT. Mimeo 27 p.
SEAG (Servicio de Extension Agric ola del Paraguay). 1989. Proyecto de Generacion y Validacion de Tecnologia de Produccion e Industrializacion de la Mandioca. Informe Anual 1989. San Lorenzo, Paraguay. Mimeo 116 p.
Sinthuprama, S., Tiraporn, Ch. and Watanononta, W. 1987. Cassava breeding in Thailand. In Cassava Breeding and Agronomy in Asia. Proceedings of a Regional Workshop held in Rayon, Thailand, Oct. 26–28, 1987. CIAT/UNDP. p. 9– 19.
Soenarjo, R., Poespodarsono, S. and Nugroho, J. 1987. Cassava breeding in Indonesia. In Cassava Breeding and Agronomy in Asia. Proceedings of a Regional Workshop held in Rayon, Thailand, Oct. 26– 28, 1987. CIAT/UNDP. p. 27–33.