CASSAVA-BASED CROPPING SYSTEMS RESEARCH Contributions fkom the First Annual Meeting of the Collaborative Group in Cassava-based Cropping Systems Research Theme: "Linking similar environmentsw Ibadan, 16-19 November 1987 ORGANIZED BY ?HE RESOURCE AND CROP MANAGEMENT PROGRAM INTERNATIONAL INSTITUTE OF TROPICAL AGRICULTURE Ibadan. Nigeria Acknowledgement is made of the technical and editorial preparations of the material in this volume by Humphrey C. Ezumah, Agronomist in the Resource and Crop Management Program and David S. Osiru, Crop Physiologist in the Root, Tuber and Plantain Improvement Program, on behalf of the Cassava-based Systems Working Group, IITA. O 1988 International Institute of Tropical Agriculture Oyo Road. PMB 5320 Ibadan, Nigeria TELEPHONE (022) 400300-4003 14 TELEX TDS IBA NG 203 1 1 (Box 0 15) CABLE TROPFOUND IKEJA FACSIMILE 234- 1-669 185 Printed by Intec Printers Limited, Ibadan CONTENTS Session P Introduction .................................................. Extracts from introductory remarks L.D. Stgel, D.S.C. Spencer and S.K. Hahn .................. Cassava-based cropping systems research collaboration H.C. Ezumah Session 11: "Linking similar environments" Characterizing the crop environment: a n agroclimatic perspective .................................................................. T.L. Luwson Physiological considerations for tuber yield improvement in cassava (Manihot esculenta Crantzl ..................... 0.0. Akinyernjiu and A.S. Adegoroye Linking similar crop environments: relevant agronomic data ................................................................................. H.C. Ezumah and M.P. Gichuru Collecting and interpreting relevant data required to link research results from varying environments: ............................................................... an economist's perspective F.L. Nweke Processing, utilization and nutritional linkages for cassava based systems in various environments ............................ N.D. Hahn Session IIE Diagnostic survey reports Cassava in the farming systems of Cameroon's high-rainfall coast ........................................................................... S. W. Almy and M.T. Besong Diagnostic survey of cassava-based cropping systems in two ecological zones of Bas-Zaire ............................................ 74 O.A. Osiname, C. Bar-ilett, N. Mbulu, L. Simba a n d K. Landu Performance of improved IITA cassava, Manihot ........................................................ esculenta Crank, at farm level 83 F.I. Nweke , H.C. Ezumah and D.S.C. Spencer Session IV The performance of cassava with other staples in intercrops in Cameroon ............................................................... 91 T.J. Arnbe, S.N. Lyonga, A.A. Agboola and S.K. Hahn Effects of fertilizer and time of introducing cassava on the performance of yam-maize-cassava intercrop: 1. Evaluation of the biological yield of the component crops ............................................................................... 98 R.P.A. Unamma, T.O. Ezulike and A. Udealor EKects of fertilizer and time of introducing cassava on the performance of yam-maize-cassava intercrop: 2. Land use maximization and monetary ......................... yield performance of yam-maize-cassava mixture 104 R.P.A. Unamma, A. Udealor and F.O. Anuebunwa Cassava-based cropping systems at the National Root Crops Research Institute. Igbariam Substation ............................... 112 J . E. G. Ikeorgu Introduction of cassava through maize in a humid environment ..................................................................................... 118 J.B. Oyedokun, T.A. Akinlosotr~ and M. Omidiji Economics of fertilizer application by different methods in a cassava-maize intercrop system ................................ 124 F.I. Nweke . H.C. Ezumah and M. Agu Maize variety and population in a cassava-maize intercrop ........................................................................................... 132 J . Arihur. H.C. E z u m a h and E.V. Doku ........................................ Cassava-groundnut intercropping in Zaire 144 N.B. Lutaladio, F.E. Brockrnan K.B. Landu, T.A.T. Wahua and S.K. H a h n Session V Response of a cassava-maize intercrop to nitrogen in ........................................................... two-year sequential cropping 161 H.C. Ezumah, F. Nweke , N.D. Kalabare and A. Karunwi Testing the feasibility of associating cassava and other food crops in oil palm intercropping systems ........................ 176 I.I. Onwubuya and F.K. E n e h Trends in cassava production in the Abakaliki area ............................. of Anambra State and implications for farmers 195 E.C. Okorji a n d 0. Okereke On-farm performance of improved cassava varieties ......................................................................... in Imo State, Nigeria 206 P.S.O. Okoli Session VI The performance of six cultivars of white yam derived from three sources and evaluated across three zones ........................................................................... in southern Nigeria 2 15 C.L.A. Asadu, H.C. Ezumah, F.I. Nweke and F.O.R. Akamigbo Maize-groundnut rotation in lowland southwestern Cameroon .......................................................................................... 225 F. C. Poubom (nee Ngundaml Effect of fertilizer and time of interplanting maize ................................ on the performance of a yam-maize intercrop 232 R.P.A. Unamma. G.C. Orkwor and M. Ibedu Methods for determining leaf area in some crop plants ..................... 237 O.T. Edje a n d D.S.O. Osiru Post-harvest factors of cassava processing and utilization ....................................................................................... 246 N.D. Hahn Annexes 1. Opening remarks ............................................................................ 259 B.N. Okigbo 2. The role of agroecological characterization in West and Central Africa ................................................................ 264 A. Goldman 3. List of participants ......................................................................... 272 Session I Introduction EXTRACTS FROM INTRODUCTORY REMARKS: a summary L.D. Stifel Director General, IITA Participants should address themselves to (a) the paradox of abundance and starvation existing at the same time in most of sub-saharan Africa; (b) the establishment of sustainable systems of agriculture; (c) the problem of food availability and (d) the development of a sound resource- based system of crop production. The participants should therefore regard themselves a s partners in the development of viable systems. The International Institute of Tropical Agriculture (IITA), on its part, is committed to the development of crop varieties that are easily adaptable to the range of ecological zones in West and Central Africa. Efforts have been made to devise methods of integrating resource and crop management research with commodity improvement research. Scientists in the Resource and Crop Management Program liaise in their research with a corresponding commodity program, eg, cassava-based systems with the root and tuber improvement program. IITA hopes that through such working arrangements, which will be reviewed from tirne to tirne, research will be better focused to address the relevant, urgent problems of alleviating starvation from insufficient production of its essential mandate crops. The Resource and Crop Management Program o f IITA D.S.C. Spencer, Director, RCMP The goals of the Resource and Crop Management Program (RCMP) of IITA are to develop economically and ecologically stable farming systems through efficient management of such resources a s soil. water, energy, crop planting materials and vegetation: and to increase resource productivity to meet future demand. Program strategies center on: 1. Analyzing and understanding physical. biological and other factors affecting degradation. 2. Devising resource management tactics that reduce degradation. 3. Identifying useful innovations that can be adapted to farmers' conditions. Major program achievements include the Alley Cropping system. Macuna and other live-mulches, and fertilizer use management. Future strategies of the resource management research include: 4. Using principles identified in the past; eg, live-mulch, prevention of erosion and refining them for on-farm use by extension personnel. 5. Developing and strengthening programs in acid soils in humid forest ecology for which there are few activities. 6. Conducting base-line data studies in the inland valleys for lowland rice production. The second major area of RCMP is the setting up of crop-systems- based working groups. Three of them have already been established: for cassava-based cropping systems, maize-based cropping systems and rice-based cropping systems. The cassava-based systems collaborative research group was previously called root-crop-based collaborative group. Following the IITA strategic study, the Group is emphasizing cassava-based cropping systems; hence the need to adjust its name. Collaboration will continue with national programs where yams and other tuber crops are very important, because research thrusts on cassava are systems-oriented. the geographical area of major focus remains West and Central Africa. In all the activities of the RCMP, attention is given to post- harvest problems such as processing. The Root and Tuber Improvement Program of nTA, S.K. Hahn, Director. TRIP The Root and Tuber Improvement Program (TRIP) now places major emphasis on cassava. Research on cocoyams and other tubers is for the national program. TRIP will concentrate on cassava improvement, plantain research and post-harvest technology, which is aimed at designing and testing of post-harvest equipment. Most of the major cassava-producing countries have received improved materials from TRIP either in tissue culture or seed form. Among improved cassava varieties gaining popularity are TMS 30572. whose cuttings are sold by the roadside, and TMS 4(2) 1425, a sweet type. Those materials are high yielding and produce much dry matter. TMNS 4(2)14-5 is pounded ltke yam. Improved cassava and trained personnel are therefore widely available ion many African countries for collaborative research in cassava-based systems across diverse environments. CASSAVA-BASED CROPPING SYSTEMS RESEARCH COLLABORATION H.C. Ezumah Cassava is the most important root crop grown in Africa. Others are yam (Dioscorea spp.). sweet ~ o t a t o (Ipornoea batatas) and cocoyams (Xanthosoma sagitt$oliurn and Colocasia esculental. These crops had. until recently, received little attention research and related policy. yet they are important because: (a) They provide on the average 16 percent of the calorie needs of the entire African population. 03) They supply almost all the energy needs of an estimated 165 million Africans who live in the humid regions. (c) In specific zones of some countries in West, Equatorial and Central Africa, root and tuber crops account for over 60 percent of peoples' caloric intake. They are also sources of animal feed and industrial raw materials and can be enriched and processed into convenient foods. The potentials of root crops at present under-exploited. can be better utilized by more rapid research to develop component technologies and evolving methods to accelerate flow of contributes over 80 percent by weight of root crops produced in Africa in 1984 (FA0 1984). Background. Interaction among scientists. through organized research collaboration, can facilitate sharing of information, reduce duplications and save scarce resources. As operational methods for a farming systems approach to research become clearer. the need to provide improved component technologies applicable to specific problems identified during on-farm research activities is accentuated. Confidence in the suitability of improved component technologies for specific environments of application Given the diversity of environments for agricultural research and the location specificity of farming systems research, it is not practicable to develop technologies in all environments of application, but a step towards making technologies suitable to wide range of environments is achievable through collaboration with researchers working at several strategic locations. Purpose of research collaboration The purpose is to generate technologies which will enable development of improved economic and ecologically sustainable cassava cropping systems for humid Africa with emphasis on West and Central Africa. Specific objectives of research collaboration 1. To bring together scientists interested in cassava research systems in order to: (a) develop technologies suitable or potentially adaptable for cassava-based cropping systems (b) facilitate interactions between IITA and national research institutions in problem identification and planning of experimentation for developing improved technologies appropriate for cassava-based systems. 2. To facilitate rapid dissemination of improved technologies to various research centres, 3. To identify which data are required in linking similar production environments and develop methods for their collection. analysis and interpretation. The above objectives and the operating procedures were discussed and endorsed during an inaugural meeting held at IITA in Februa~y 1986. Participants comprised seven scientists from: Ahmadu Bello University, Zaria National Root Crop Research Institute. Umudike: Institute for Agricultural Research and Training. Ibadan University of Nigeria. Nsukka: Imo State Agricultural Development Project. Owerri: Shell Company, Warri -in addition to IITA, the convenor. The location of these institutions across the major ecological zones of humid tropics (perhumid, humid. humid/subhumid, transitional and subhumid) permits research to be conducted at sites having major soils of interest (Alfisols. Ultisols and Oxisols) ranging in annual rainfall from about 800mm to over 2,200mm per year. During the meeting, diagnostic survey results emanating from primary and secondary data available for each experimental site were discussed. constraints were identified, researchable topics were discussed and rational experimental plans were made for each of seven locations. Research covered topics on development of component technologies and subsystems and methods of disseminating these technologies to other research stations. At a follow-up meeting held in July 1987, detailed discussion took place on research methods, and on the need for clear understanding of research environments a s a base for linking results to similar environments. Ghana. Zaire, Nigeria, Cameroon and Republic of Benin were represented at the July meeting. Some available component technologies Improved cassava varieties capable of yielding over 30 tons per hectare in 12 months, such as Tms 30572, are available at IRA. This and other cultivars have shown high levels of resistance to mosaic disease and bacterial blight and tolerance to mealybug. High-yielding virus and w e d resistant varieties of sweet potato (25 to 35 t/ha in 4 months) are also available. The mini- and micro-sett seed yam production technology, originally developed by the National Root Crops Research Institute and improved and popularized in conjunction with IITA. provides an excellent example of inter-institutional collaboration in technology development and transfer. That technology is capable of reducing the once-prohibitive expenses in yam planting material inn which 30 percent of tuber produced is saved for planting to less than 5 percent. With increasing environmental hazards, eg, drought. African countries have faced food shortages in recent years. Some of the root and tuber crops, particularly cassava, are tolerant of drought and are, in fact. extensively grown whenever risks are anticipated. The root-crop-based system is normally associated in multiple cropping systems with cereals [maize and rice) and grain legumes (cowpeas and soybeans). Improved. high-yielding, rice and cowpea varieties. resistant to different environmental stresses, are available in Nigeriaat IITA, at the Institute of Agricultural Research and Training. at the National Cereal Research and at the Institute of Agricultural Research, Ahmadu Bello University. Other research centres in Africa have also developed improved component technologies. In addition to improved crop varieties. technologies developed by farming systems researchers aim at continuous and sustained productivity of soils with little degradation. Multiple cropping systems. weed management, alley cropping, reduced tillage. mulch farming and land clearing methods are among areas in which technologies are available for further location-specific research. Mechanisms are required to accelerate accessibility to these technologies and their components to research centres in Africa, which will also develop them further as subsystems to suit their specific needs in root crops research. This constitutes the thrust of the collaborative research management. Operation of the research group The Convenor of the cassava-based cropping systems collaboration is IITA st Ibadan. During inaugural meetings held in February and July 1986. the specific operating procedures were discussedas follows: Discipltnary composition. Since the research group involves interactions of scientists in base data analysis, research station studies and on-farm research, it requires multi-disciplinary teams comprising a core of aeronomists. economists and soil scientists. Other disci~lines will be co-opted as theneed arises. Ecology. Most of the tropical root and tuber crops in Africa are grown in the humid and subhumid zones. Those are the ecologies of major mandate for IITA. Countries with these ecological zones in which cassava is important will be encouraged to join the research grouos. which will expand by gradual degrees. Research activities. Drawing from the existing body of knowledge in cassava-based research, emphasis will be on: (a) development of relevant component technologies at research stations or other environments similarly controlled; (b) dissemination of these technologies to research centres. Achievement of the objectives will require clear understanding of the research environments. The biophysical and socio-economic environments and resources will be determined from secondary data, records of on-farm research team or rapid survey by the research teams. Monitoring of the biophysical environment. particularly radiation. soil and rainfall will be routinely carried out. Data will also be collected on crops responses to treatments and management inputs. Experiments. Cassava-based experiments will be designed, discussed during the annual workshops and approved for execution. Varietal trials and cropping patterns of various designs, especially the dominant multiple cropping systems. will feature prominently. Annual meeting. The yearly meeting, prior to cropping season, will provide an opportunity for presentation of research results of the preceding year and plans for the following season. The venue of the annual meeting will be rotated among member countries and/or institutions represented in the collaborative group. The representative in the group or country hosting the next meeting becomes the chairperson for the year. Resource persons. Funds will be set aside or solicited to invite highly experienced resource persons to the meeting. Who is to be invited will depend upon the specific problem area wewant to emphasize. eg, soil classification. Instrumental in weather monitoring, economic parameters in interpretation of cropping pattern experiments. among others. Visiting the trials. Visits by all members may not be feasible because of limited funds. The Convenor should organize one monitoring visit during the growing season. Opportunities for Training. On-the-job training will be given to technical-level staff to facilitate data collection and care of instruments. Professional-level training on various data processing techniques is offered at IITA. The group members will be encouraged to participate in appropriate programs. Session I1 "Linking similar environments" Papers in this section are followed by a selectwn of comments by participants CHARACTERIZING THE CROP ENVIRONMENT: AN AGROCLIMATIC PERSPECTIVE T.L. L%wson Weather and climate are the most pervasive factors of the crop environment. They restrict the occurrence and choice of species that can grow or be grown in given areas or localities and thus constitute a major determinant of crop distribution over the earth (Bunting et al. 1982). They also determine the species' productivity in those localities over given seasons. Crop yield has, in fact, been defined as the physiological expression of the genetic potential of the crop (Landsberg 19721. The strong dependence of crop physiological processes on the characteristics of the biophysical environment makes a thorough analysis and charac- terization of the environment not only a logical starting point but also a continuing necessity in the rational exploitation of given crops and/or cropping systems. A cropping system may be thought of as an ensemble of activities designed to modify or improve the crop environment to achieve a better, or the best possible. yield using a given natural resource base under given socioeconomic conditions. The role of weather and climate in determining the status of the resource base and the level and timing of human intervention in the production of the crop provides a second level of influence of weather and climate in determining crop productivity. This again creates the need for the assessment and characterization of the agroclimate as a component of the biophysical environment in the exploitation of the crop, or the elaboration or application of the relevant cropping system(s). This paper will, however, focus on the direct rather than the indirect role of the aerial environment. The agroclimatic environment and its characterization Figure 1. derived in part from Hogg (1971). gives a summary representation of the main factors involved in the interaction between the crop and its biophysical environment. Considering crop production in its essence as 'the conversion of solar energy, water and soil nutrients into economic end-products" (Hogg 1971). it is evident why solar energy and its derivatives. air and soil temperature, should constitute key factors in our characterization of the biophysical environment, and in particular the agroclimate. It is, however, rainfall or moisture that is most limiting in crop production and therefore has been and will continue to be used as the primaly factor in the stratification of crop environments. Figure 1. Crop biophysical environment CROP ENVIRONMENT (BIOPHYSICAL ATMOSPHERIC ( AERIAL) EDAPHIC (SOIL) PRIMARY CONTROLS I I I I b 4 1 WATER ENERGY iPRECIP/EVAPOT:SOIL IKWSTURE) PHYSICAL CHEMICAL BIOTIC I I I I - SECONDARY CONTROLS I I ALWECTEDENERGY TOPOGRAPHY (SENSBLE, LATENT) I I I TERTIARY CONTROLS A PEST-DISEASE-WEED OTHER PHENOMENA The moisture balance parameter While rainfall/precipitation constitutes the source of moisture input to the soil or surrace storage system, the climate of any given locality imposes a certain amount of evaporative demand which translates into a continuous oufflow of water from the system as long as water remains available. Thus, although it is easier to use the primary variable, rainfall/precipitation, in the first order classification of the agroclirnate of different localities, it is the water balance (the daerence between rainfall and evaporation/evapotranspiration) or, better still. the moisture balance. where feasible, that has proven more viable for the purpose. The first indication of this was provided in 1948 when Thornthwaite, defying the tradition of defining or equating climatic zones and their limits by empirical matching of temperature and/or precipitation values to apparent boundaries of vegetation types, published his article "An approach toward a rational classification of climate," based primarily on the concept of moisture balance. Contemporaneous or subsequent works by Penman (1948). Turc (1961). Cocheme and Franquin (1967), and Franquin (1969). among others, have demonstrated useful applications of the concept not only in def- and delineating agroclirnatic wnes but also in establishing crop calendars. The viability of water/moisture balance as a basis for these latter purposes in general and for the objectives of the cassava-based cropping systems group in particular has been more than amply demonstrated by a recent study on cropping patterns in West Africa (Dennett et al. 1981) reported by Bunting et al. 11982). This showed that the proportion of arable land in a country devoted to particular crops was related to the average length of the period of the year during which rainfall exceeded potential evaporation IM. in days)-an approximation 'of the period when growth is not likely to be restricted by water shortage" (Bunting et al. 1982). Clearly. the length of the period of positive moisture balance. defined as the period when rainfall is equal to or greater than potential evapotranspiration (Lawson and Juo 1979). will adequately serve our purpose in the broad delineation of a representative domain for research aimed at determining productive root-crop-based cropping systems and analogue areas of application. In the case of cassava, it could even serve as a good indication of the relative performance of the crop in the various wnes, as may be inferred from figures 2 and 3, which show the results of two different studies by this writer. Figure 2. Mean yield of cassava (kg/pl as affected by the duration of favorable moisture cycle Mean fresh tuber yield (kg / p i ) 6.01 1 TMS 30572 - At final harvest (MAR:82) V=- -V At DEC. 2 / 8 1 Sampling T M S 3 0 5 5 5 - At final harvest ( MAR182) W- -d At DEC. 2 /81 Sampling 2 .0 +----p---+ 1.0 0 I ri m irrigation regimes Note on irrigation regimes: In addition to blanket irrigation on 20November and 1 December: Regime I-no supplementary irrigation: Regime 11-supplement irrigation thmugh 31 December: Regime Ill-supplementary weekly irrigation%rEih"g March Figure 9. Relationship between the fresh yield of cassava and the dry season severity index at IITA over a 5ve-year period F r n h yldd ofcouova ( t /ha) 52- 48- Y=-0.3421 + 176.69 . Corr=-0.98 44 - 40 - 3 6 - 3 2 - 28 - 24 - 1 370 390 410 430 450 Dry 8oa8on ( Doc-Fob) Sevrrlty index (nun) Shorter-tern variability in moisture balance Beyond establishing the broad zones of moisture balance defining the mean duration of the cropping cycle in space, time variations in this parameter on a decreasing scale which reflect the year-to-year and inter- and intra-seasonal situations, and which more directly affect the corresponding yields, are crucial and require collation/observation and analysis as both historical and concurrent data in the cropping season. The results provide a basis for modifying the environment to suit the crop or cropping system where possible, but more generally for fine tuning the crop/cropping system to the environment. It is also indis- pensable for sound interpretation and comparison of experimental yield data (Lawson and Juo 1979: Lawson et al. 1979: Lawson 1982a). Potential evapontion/evapotranspiration as a component of the moisture balance is. as observed elsewhere [Lawson et al. 1979). a relatively conservative property of the environment, and much less variable than rainfall. In most practical applications, therefore, the variability in moisture balance is primarily assessed in terms of the variability in rainfall. It is also to be noted that root crops, particularly cassava, yams and sweet potato, to name those of interest in the region, are much more resilient than other crops under conditions of short-term moisture deficits or limited moisture supply [Jones 1961). Except for the demands of associated crops in mixed cropping a rougher time determination of rainfall variability than the weekly period found to be appropriate in this region for other crops, especially cereals, should prove adequate. Finally, as plants gain access to moisture through the soil, any substantial moisture storage capacity found in the soil should be taken into account when the crop moisture regime is assessed. In all cases, observations of both soil moisture and water balance components are desirable. Other aspects of rainfall/precipitation are relevant in the context of the envisaged activities and should be monitored where feasible. These include the rainfall intensity, which largely determines the redis- tribution of the incoming rain at the surface and the amount entering the soil, its erosive capacity, and its indirect effect through its innuence on pest and disease manifestation. Solar energy: light The cardinal role of solar energy, or light, cannot be overstated. It is generally agreed that in the humid and, to a lesser extent, the subhumid tropics, where a large part of the root crops in West and Central Africa are grown, global radiation levels are less than optimal and must therefore be regarded as limiting to growth and productivity in seasonal crops (Lawson et al. 1979. Lawson 1982a). Thus. under non- limiting moisture (and nutrient) conditions, comparatively higher insolation. and/or more efficient intercention of the incident light. drv- . , matter production and yield should increase. With the predonGnance of crop mixtures as the basic cropping pattern in the region of interest, the efficient use of light through the temporal and spatial manipulation of crops and choice of crop varieties in the above circumstances must be regarded as a premium. This has been confirmed by experimental studies [Lawson 1982h; Lawson and Jackai 1983). It is therefore evident that the assessment of the actual light climate at given locations, as well as the quantitative monitoring of the variable on component mixtures in specific experimental conditions is indispensable for the design of efficient cropping systems and for com- parative analysis of experimental results (IITA 1981; Lawson 1982b). Qualitative assessment of the incident light in this situation may also be needed. Accurate measurement of solar radiation however requires fairly expensive instrumentation. Where this is not available, proxy measure- ments with mechanical pyranometers or sunshine recorders can provide limited but useful information. Even in the absence of these, accurate estimates of cloud cover may still serve a purpose. Temperature regimes Air and soil temperatures are generally not limiting for root crops in much of the lowland humid tropics. Supra-optimal tempera- tures can however occur, particularly at the onset of the growing seasons (La1 1974; Monteith 1978; Lawson 1982a). affecting sprouting in yams, for example, and accelerating water loss by cassava cuttings and sweet potato vines. These temperatures need to be monitored. and methods of modification should be incorporated into the relevant cropping systems. As in the case of solar radiation, the temperature regimes on the microscale in crop mixtures may reveal substantial modification of the prevailing mesoscale temperatures monitored in standard screen (IITA 1981: Lawson and Jackai 1986). Systematic monitoring of these tem- perature regimes should also prove useful, in specific cases, in assessing crop performance and incidence of disease and pests (Lawson and Jackai 1986). Humidity In addition to its influence on evapotranspiration rates and the consequent effects on yield potentials (Lawson 1982a). the relative humidity regime in the crop canopy may, like temperature, influence disease and pest incidence in the crop or crop mixture. Parallel obsenration of these related variables can only be useful (Lawson and Jackai 1986; Jackai and Lawson 1986). Wind As an important factor in atmospheric turbulence transport and evapotranspiration losses, the monitoring and assessment of wind speed can assist in estimating potential moisture demand and establishing the moisture balance (Lawson 1982a). Its role in the advection of pests and disease organisms also justifies its measurement where possible. Where wind speeds are such as to induce lodging, there is added reason for monitoring them. Topography The role of topography in the aerial environment is indirect. deriving essentially from modification of the main climatic variables a s a result of changes in elevation and the aspects and slope of the land. Conclusion The profound influence of the environment on distribution of crop species and on the growth and yield of individual crops makes it both pertinent and necessary for the production of these crops or the evolution of cropping systems to occur in the context of a sound assess- ment and understanding of the agroclimatic background of the localities of production or experimentation. The need for adequate monitoring of the important variables for the duration of the relevant studies is self- evident. In any given case, the range of data to be collected will depend on the objectives of the study or the facilities and resources available. REFERENCES Bunting. A.H., M.D. Dennett. J. Elston and C.B. Speed. 1982. Climate and crop distribution. In Food. Nutrition and Climate, ed. Sir Kenneth Blaxter and Leslie Fowden, pp. 43-74. London: Applied Science Pub- lishers Cocheme. J.. and P. Franquin. 1967. An agroclinlatology survey of a semi-arid area in Africa south of the Sahara. Geneva: WMO. Franquin, P. 1969. Analyse agroclimatique en regions tropicales. Saison pluvieuse et saison humide. Application. Cah. ORSTOM Ser. Biol. No. 9. Hogg. W.H. 1971. Regional and local environments. In Potential crop production. a case study, ed. P.F. Wareing and J.P. Cooper. London: Heinemann. IITA. 198 1. IITA Annual Report 1980. Ibadan: IRA. Jackai. L.E.N., and T.L. Lawson. 1986. Insect pest surveys on cowpea (Vigna unguiculata Walp) and the possible effects of some climatic factors on the population trends of the legume pod borer and pod sucking bugs. Paper presented at the WMO/IITA Conference on Agrometeorology and Plant Protection. IITA-BENIN. Cotonou. 7-11 July 1986. People's Republic of Benin. Jones, S.T. 1961. Effect of irrigation at dirferent levels of soil moisture on yield and evapotranspiration rate of sweet potato. Proc. Am. Hart. Soc. 77: 458-462. Lal. R. 1974. Soil temperature relations in tropical Africa and their effects on crop yield. Paper presented at the International Expert consultation on the use of improved technology for food production in rainfed areas of the Tropical Asia, 24 November-13 December 1974. Hyderabad. KhonKaen, Kuala Lumpur. Landsberg. J.J. 1972. Microclimate and the potential productivity of sites. Scientific Horticulture 24: 126-141. Lawson. T.L. 1982a. Climatic factors for higher yields of root crops. Paper presented at the West African Regional Root Crops Workshop. 27 June-2 July 1982. Central Agricultural Research Station. Suakoko. Liberia. Lawson. T.L. 1982b. Light regime and productivity in mixed crops. In IlTA Annual Report 198 1. pp. 3-6. Ibadan: IITA. Lawson, T.L., J.S. Oguntoyinbo, and 0. Ojo. 1979 Agroclimatic conditions of West Africa. Paper presented at the llTA Annual Research Conference on soils and Climatic Resources and Constraints in Relation to Food Crop Production in West Africa, 15-19 October 1979. Ibadan: IITA. Lawson, T.L., and A.S.R Juo. 1979. Climate and soil conditions in the subhumid and semi-arid regions of West Africa with special reference to maize production. Paper presented at the First SAFGRAD Maize Production Workshop, Ouagadougou, Burkina Faso, 20-23 February 1979. Lawson. T.L., and L.E.N Jackai. 1983. Cassava canopy structure in intercropping. In IITAAnnual Report 1982, pp. 144-146 Ibadan: IITA. Lawson,T.L., and L.E.N Jackai. 1986. Microclimate and insect pests population in mono and intercropped cowpea (Vigna unguiculata Walp). Paper presented at the WMO/IITA Conference on Agrometeorology and Plant Protection.. IITA-BENIN. 7-11 July 1986. Cotonou. People's Republic of Benin. Monteith, J.L. 1978. Soil temperature and crop growth in the tropics. In Soil physical properties and crop production in the tropics, ed. R La1 and D.J. Greenland, pp. 249-272. John Wiley, Chichester: John Wiley & Sons. Penman. H.L. 1948. Natural evaporation from open water. bare soil and grass. Proc. Royal Soc. London. A. 193: 120-145. Thomthwaite. C.W. 1948. An approach toward a rational classification of climate. Geogr. Rev. Vol. 38, (1): 54-94. Turc. L. 1961. Evaluation des besoils en eau d'irrigation. evapo- transpiration potentielle. Ann. Agron. 12(1): 13-49. PHYSIOLOGICAL CONSIDERATIONS FOR TUBER YIELD IMPROVEMENT IN CASSAVA (MHOT ESCULENTA CRANTZ) OA. Akinyemiju and AS. Adegoroye Abstract Cassava is an important tropical root crop largely because it is a cheap source of carbohydrate. It is amenable to agronomlc as well as genetic improvement, and it is a crop that is native to the tropics. Much agronomic work has been done on cassava, and even more on improvement studies. However, very few studies have been conducted on the influence of environmental components on the growth and develop- ment of cassava; nor has much been done on the physiological processes of cassava and their interaction with environmental factors in the different ecosystems where cassava is cultivated. In order to achieve a systematic approach to yield asymptote of cassava, it is recommended that studies be carried out on important physiological determinants of yield in cassava. To aid breeding work. a cassava ideotype is proposed. The proposed cassava ideotype should have a 10-month growth cycle in which crops are planted at the onset of rains and harvested towards the end of the following dry season; a single short erect stem; rapid early leaf area development, a maximum leaf area index of 3-4. slow decline of leaf area during senescence and early initiation and cellular development of the tuber. * * * * * * * * * * * * * Cassava is a major tropical root crop, important in supplying the daily carbohydrate needs of the vast populations of the tropical countries of Africa, America and Asia. Its botany is well covered by Cobley (1965) and Purseglove (1968) while its agronomy was comprehensively reviewed by Onwueme (1978). However very little is reported on its physiology. This paper reviews some of the physiological constraints on yield improvement in cassava. It is divided into three sections: the first examines the environmental and developmental components that are considered important for the growth and development of cassava; the second considers the physiological and metabolic processes that lead to tuber initiation and bulking in cassava; and the third proposes a cassava ideotype. Environmental components important for growth and development of mot crops There is little information on the effects of environment on growth and development of any root crops, particularly cassava. Consideration of these environmental eEects will therefore be general. 1. Light intensity High light intensity (3.000-16.000 lx) increases initiation and early growth of the tuber, early attainment of maximum stem length. and senescence. Total dry matter and distribution of dry matter to tubers should increase at high light intensities, but final tuber weight will be limited because of earlier leaf senescence (Wilson 1977). In the humid tropics, cassava will be expected to be higher-yielding during the late season than in the wet season because of partial cloud cover and reduced light intensity in the wet season (Onwueme 1978). 2. Day length The effect of day length on growth and development is particu- larly important in tropical tuber crops such as cassava. Tuberization in cassava is promoted under short-day conditions and reduced by day lengths greater than 10-12 hours. The crop is therefore most successfully cultivated between latitudes 15"N and 15"s (Njoku 1963. Onwueme 1978). However, varieties have been identified at IITA (1982) that are classified as short-day or long-day plants. 3. Temperature Seasonal variations in temperature are small in the tropics, so the effects of temperature on growth and development in tropical root crops like cassava are less relevant. However, little work has been done on the effects of temperature on other tropical root crops. Kim (1961) established that low night-time temperatures increased tuber weight in sweet potato. Tuber development in sweet potato was more rapid at 25°C than at 30°C and no tubers were formed at 10°C and 15°C [Spence and Humphries 1972). 4. Evapotranspiration Few studies if any have been done on the relationship between tuber yield and potential evapotranspiration in tropical root crops (Wilson 1977). However. yields increase from less than 5 to more than 10 tons/ha as potential transpiration increases from 2 to 6cm. A linear relationship between tuber yield and evapotranspiration has been reported for Solanurn potato (Wilson 1977). 5. Water Water supply is probably the most critical environmental factor affecting the growth and development of tropical root crops since precipitation is subject to seasonal and diurnal fluctuations. It is par- ticularly important for the establishment of cassava and other tropical root crops because of the methods of vegetative propagation used in their cultivation. For further growth of the crops. water requirements are diEerent. Except at planting, cassava can withstand prolonged periods of drought and is a valuable crop in regions of low or uncertain rainfall (300-500mm) (Onwueme 1978). Optimal water requirement is, however. in the region of 1000-1500mm of well-distributed precipitation (Kay 1973). Inadequate and/or irregular water supply leads not only to poor yields but also to malformed tubers. A better understanding is still needed of the relationship between tuber yield and water supply in cassava. 6. o a w The deleterious effects of waterlogged soils on tuberization in tropical root crops are due to restriction of the oxygen supply shown to be critical for tuberlzation in sweet potato (Wilson 1977). On poorly drained soils, cassava root growth is poor, and the tuber-to-shoot ratio is considerably reduced. The poor soil aeration under such conditions causes the few tubers formed to rot readily (Onwueme 1978). 7. Soil Physical Conditions and Nutrient Supply The best soil for cassava cultivation is a light, sandy loam of medium fertility: but tropical root crops can be cultivated in heavy clays with proper management. In crops in which the yield organ (ie, the tuber) must grow against the resistance of compacted soil, soil physical conditions become important for high yield. However, very little work has been done on the interrelationship between physical properties of soils and tuberization of tropical root crops. Under conditions of very high fertility, cassava tends to produce excess vegetation at the expense of tuber formation (Onwueme 1978). Yields on many soils are apparently limited by lack of adequate potassium: when the potassium level in the soil is low the response of cassava to N fertilization is poor (IITA 1982, Onwueme 1978). However, little information exists on the best times to apply fertilizer to cassava, although it is more appropriately applied at planting (Onwueme 1978). Physiological processes leading to tuber yield In cassava Shoot and tuber ontogeny. photosynthesis and protein synthesis are among the most important physiological and metabolic processes that contribute to growth and development, and are therefore important in the productivity of tropical root crops. 1. Shoot ontogeny Shoot ontogeny in root crops can be separated into the following stages according to Wilson (1977): (a) germination (sprouting) to produce a seedling (b) period of juvenile growth (c) period of growth of the mature plant to maximum leaf area (dl leaf senescence. Besides the report that juvenile leaves of cassava are high in anthocyanin content, the effects of the environment on shoot ontogeny and indeed on ontogeny per se have not been critically examined in most tropical root crops (Wilson 1977). 2. Tuber ontogeny The morphogenetic processes of tuber initiation and tuber growth are different and distinguishable from one another. Some external factors affecting tuberization in tropical root crops have been reported (Wilson 1977). The only environmental factor categorically shown to have an effect on tuber initiation is photoperiod in root crops sensitive to day length. The cellular evenls leading to tuberization in root crops other than Solanurn potato and sweet potato have not been thoroughly studied. However, the anatomy of tuberization in cassava has been described (Doku 1970). Light intensity, temperature. water supply and mineral nutrients apparently operate through their influence on tuber growth rather than on tuber initiation. The effects of the various environmental factors on tuber initiation and the cellular component of tuber growth are mediated through hormonal responses. Several plant growth substances have been shown to be involved in tuber growth. These hormones include auxin, gibberellin, cytokinins, ethylene and abscisic acid (Wilson 1977). However, the exact nature of the plant growth substance involved in tuber formation has not been identified. What is known is that the inhibiting effects of light and low oxygen supply on sweet potato tuber growth and development are mediated through inhibition of meristematic activity in tubers (Milthorpe 1967). Similar studies on cassava tuberization have not been reported. 3. Photosynthesis There is a close interrelationship between the parameters of dry matter accumulation in cassava. For example Enyi (1972) established a direct relationship between LAI [Leaf Area Index) and tuber bulking in cassava. Since accumulation of dry matter depends on photosynthesis, it is important to examine quantitative interactions between the environment and photosynthesis. It is generally believed that only a very small percentage of incident radiation is used for dry-matter production. For example in maize it is estimated that only about 3 percent of incident radiation is used during the active growing period of a crop that has a vertical display of its leaves and C4 photosynthetic efficiency. A major problem in crop production, therefore. is increasing the efficiency of the conversion of radiant energy into dry matter. The optimal light interception has been estimated to occur in most root crops at W = 3 (Wilson 1977). Rapid increase in leaf area during early growth would therefore increase the efficiency of light interception. Very little work has been done to eluci- date the interaction of LAI and photosynthesis in cassava. However, in sweet potato it has been established that the quantitative yield limiting process is tuberization rather than photosynthesis and hence the external factors which influence tuberization, eg. water supply and oxygen, are likely to have considerable effect on yield [Wilson 1974). 4. Protein Synthesis Protein synthesis involved in cell division and expansion is important in tropical root crops. It is known that abundant N promotes excess foliage at the expense of tuberization. Besides this fact. little is known of the role of nitrate reductase in root crop metabolism. Clearly, an enormous amount of information is lacking both on the environmental factors affecting growth and tuber yield in cassava and on the physiology of the growth and development of cassava in particular and other tropical root crops in general. These are virgin areas for root crop physiologists. In the next section. evidence will be adduced to show that very little permanent improvement in tuber yield of tropical root crops can be achieved without first understanding the physiology of the growth and development of crops. Cassava Ideotype Classical plant breeding methods rely on two approaches: (1) defect elimination in breeding programs for disease resistance and [ii) selec- tion of individual plants for high yield (Donald 1968). The gross shortcoming here is that no reference is made to designated physiological or morphological characters which determine such high yield. For example, in cassava improvement at IITA the mosaic virus resistant variety is expected to out-yield the non-resistant varieties. Leaf area seems to be a factor in this. However, as has previously been highlighted, the relationship between leaf area production and tuber yield in cassava is not known. In cowpea (Vigna unguiculata [L.] Walp). for instance, it !s known that a 25 percent defoliation will not lead to any significant yield reduction (Ezedinma 1973; Akinghohungbe 1980). The obvious drawback therefore in the use of the two classical breeding approaches is that the methodology does not generate any information on the plant characters associated with improved crop performance. Therefore improvement in crop performance to the asymptotic limit of the species by systematically optimizing the determinants of yield (or disease resistance) in new cultivars cannot be obtained [Donald 1968). Donald (1968) proposed an approach to systematic improvement in crop productivity to the species asymptote through breeding for model plants or ideotypes. A crop ideotype was defined a s a biological model which is expected to perform or behave in a predictable manner within a defined environment and to yield a greater quantity or quality of useful produce than existing varieties when developed as a cultivar. Wilson (1977) then concluded that an ideotype therefore presupposes detailed and accurate knowledge of the range of anatomical, morphological. physiological and biochemical characters existing in varieties and/or cultivars within a species. Data on the performance of the cultivars in the ecosystem for which they are intended needs to be compared with data obtained on a variety of ecosystems. Such a model was charac- terized for cereals [Donald 1968). However. no such model has been developed for cassava or any other tropical root crop. The absence of such a model may be due in part to the lack of basic physiological information as well as to the diversity of the habitats in the species. The implication is that a model will have to be developed for each tropical root crop. The ideotype of cassava wffl be considered. 1. Crop Growlh Cycle The wet season of the humid tropics is characterized in general by high precipitation but low incident radiation: this may lead to reduced rates of photosynthesis, reduced l e d density, increased stem length. reduced individual leaf area and consequently reduced availability of assimilates for storage. The dry season, on the other hand, apart from the lack of moisture, provides conditions of soil oxygen and nutrient supply, light intensity, day length, and night temperature which are more favorable for growth (Wilson 1977). Therefore one basic wony in a long-season crop like cassava is the development of leaf area (W) in relation to the optimal conditions of growth. The choice of crop growth cycle should be such as to maximize cassava tuber productivity. A 10- month growth cycle in which the crop is planted at the onset of the wet season and harvested towards the end of the following dry season will be ideal for cassava. 2. Stem habit In cassava, as in most root crops, the stem has the following functions: (i) support, exposure and display of leaves; (ii) transport of assimilate from leaves to tubers: (iii) storage in the underground stem or root; and (iv) genesis and attachment of tubers (Wilson 1977). The important considerations here, in contrast to cereal crops where the stem is involved in aerial support for the yield organ, are the leaf display and transport of assirnilate, which require opposite characteristics. In cassava, a single short erect stem rather than many long stems would seem to be an advantage for efficient utilization of environmental resources (Enyi 1972). Another consideration in the stem habit is the potential Zor mechanization in cassava, when shallow compact rooting may be advantageous over deep dSuse rooting. 3. Production and Disbibution of Assimilate Some characteristics worthy of consideration in assimilate production and distribution in cassava and other tropical root crops are: (a) rapid early leaf area development to W = 1 for effective light interception during early crowth, (bl development of leaf area to a maximum W = 3-4, and slow turnover of leaves after maxlmum W has been attained, (c) slow decline of l e d area during leaf senescence, since this period often coincides with maximum rates of tuber bulking, (d) early initiation and cellular development of tubers to create an active tuber sink for assimilate [Wilson 1977). These characteristics are considered important physiological determinants of yield in cassava and any improvement based on them is likely to lead to a systematic approach to the yield asymptote of cassava. However these characters have not been seriously investigated in cassava. In order to establish the potential of each character, it must be closely studied, taking into account the environment for which the ideo- type is being designed. In Nigeria, for example, IITA has several reports showing that improved cassava clones out-yielded local standard varieties by 2 to 18 tons, primarily because they are resistant to diseases and have benefited from improved cultural practice (IlTA Annual Reports. 1975. 1976, 1977. 1980. 1982). But in a recent study conducted at Obafemi Awolowo Uni- versity. Ile-Ife. Alimi and Akinyemiju (1987) o b s e ~ e d that the "gari" processed out of a local cassava variety. Odongbo, was as good as or better than that from the improved varieties of IITA (TMS 30572) and IAR&T. though the tuber yield in Odongbo was signiilcantly lower than that from the improved varieties. In this example, Odongbo, a native unimproved cultivar, is susceptible to various foliar diseases [mosaic disease and bacterial blight). the leaf area production is much less than the improved varieties and also it usually has few stem branches which are considerably taller than the improved varieties. The contradiction in cassava improvement program revealed in this study underscores the significance of detailed studies of the physiological determinants of growth in order to systematically improve the quality and yield potential of cassava to the asymptote level within our environment. REFERENCES Akingbohungbe, A.E. 1980. Artificial defoliation of cowpea (Vigna unguiculata [L.] Walp) cv. Ife Brown to simulate insect damage: effects on crop performance. Ife J. Agric. 2. 17-25. Alimi. J.T. and O.A. Akinyemiju. 1987. Economics of cassava production and processing. Ife J. Agric. 8 (1 & 2). In press. Cobley. L.S. 1965. An introduction to the botany of tropical crops. London: Longmans, Green and Co. Doku, E.V. 1969. Cassava in Ghana. Accra: Ghana University Press. Donald. C.M. 1968. The breeding of crop ideotypes. Euphytica 17: 385- 4a3. Enyi, B.A.C. 1972. Effect of shoot number and time of planting on growth, development and yield of cassava (Manihot esculenta Crantz). J. Hort. Sci. 47: 457-466. Ezedinma, F.O.C. 1973. Effects of defoliation and topping on semi- upright cowpeas (Vfgna unguiculata [L.] Walp) in a humid tropical environment. IITA 1975. Annual Report 1975. A 1976. Annual Report 1976. IlTA 1977. Annual Report 1977. IlTA 1981. Annual Report 1980. IlTA 1982. Annual Report 1981. Kay. D.E. 1973. Crop and Product Digest 2, Root Crops. Tropical Products London: Institute, Foreign and Commonwealth Office. ODA. Mflthorpe, F.L. 1967. Some physiological principles determining the yield of root crops. In Proceedings of the International Symposium on Tropical Root Crops, University of the West Indies, St. Augustine. Trinidad. 2-8 April 1967, ed. A Tai. W.B. Charles. E.F. Iton. P.H. Haynes and K A Leslie, U.1-11.19. Njoku, A. 1963. The propagation of yams (Dioscorea spp.) by vine cuttings. J. West Ak. Sci. Assoc. 8: 29-32. Onwueme. I.C. 1978. The tropical tuber crops: yams, cassava, sweet potato, cocoyarns. New York: John Wiley & Sons. Purseglove. J.W. 1968. Tropical Crops. Dicotyledons 1. London: Long- man.. Spence. J.A. and E.C. Humphries. 1972. Effect of moisture supply, root temperature and growth regulators on photosynthesis of isolated leaves of sweet potato (Ipomoea batatas). Ann. Bot. (London) (N.S.) 36: 115-121. Wilson, LA. 1977. Root crops. In Ecophysiology of tropical crops, ed. P.T. Alvim and T.T. Kozlowski. London: Academic Press. Wilson, L.A. 1974. Improvement and development of tropical root crops. In Interaction of agriculture with food science, ed. R. MacIntyre. CIDR- 033e. LINKING SIMILAR CROP ENVIRONMENTS: RELEVANT AGRONOMIC DATA H.C. Ezumah and M.P. Gichuru Abstract Network experiments at carefully selected sites where important features of both environment and crop performance are measured are useful in the interpretation of and extrapolation from the climatic, edaphic and socioeconomic factors that determine crop adaptability, distribution and productivity. Important edaphic factors are: (i) chemical and physical properties of soil, including inherent soil fertility; (iil erosion hazard: (iiil topography: and (iv) water deficiency or excess and other crop-related factors. These are among the most important causes of site- specific crop responses to given agronomic management practices. They may to some extent determine agronomic practices, including planting dates, growth cycles and planting patterns, and consequently yield and yield components. The importance of careful data collection to quantify the edaphic factors and effects of related agronomic practices is emphasized. *I* ...******* Although crops may grow over wide areas, geographic segregation over a range of environments is common (Martin and Leonard 1964). The degree of adaptation to environments of growth is determined by normal growth and high level of biological yields. Factors influencing localization include climate, topography. soil characteristics. insect pests. plant diseases and economic conditions Ide Vries 1963). A further illustration of the overriding importance of climatic factors in crop adaptation and distribution is provided by Good ( I 953). who states that 'Plant distribution is 1) primarily controlled by the distribution of climate, 2) seconddy controlled by edaphic factors, and 3) great movement of flora have taken place in the past and are continuing." Climate comprises temperature, moisture, light and wind. The edaphic factors are parent materials, soil and physiography. Since the manifestation of these edaphic factors is influenced by climate, they are secondary. However. soil factors determine whether a plant or crop is actually found in an area, and in what abundance. Thus every plant species exists and reproduces successfully only within a definite range of climatic and edaphic conditions, a clear understanding of which will facilitate extrapolation of cropping systems results across similar environments. A network of experiments located at carefully-selected repre- sentative sites at which important features of the environment are measured will be useful in interpreting and extrapolating results. This is especially the case in farming systems research which aims to develop location-specific packages of farming practices. Results from a given site should be applicable to areas with similar environments. The key tasks are. therefore, (a) to collect environmental data in specific experimental areas. and (b) to determine the minimum set of data essential to interpret site-specific results and extrapolate from them to other environments. Relevant data Climatic and economic conditions and their influence on crop plants are discussed in the papers of Lawson. Akinyemiju. Nweke, and Hahn. Salient data generally required to describe crop edaphic environments include: 1. Physical features: -location: longitude, latitude, elevation: -physiography: land form and land type: sloping, flat. or rolling; slope direction; angle of slope: position of land type on land form. 2. Soil classification information: the underlying assumption in soil fertility evaluation is that fertilizer recommendations and agronomic management practices are site-specific. Differences in soil properties are a main reason for this site-specificity. A logical conclusion is that soil fertility characterization must be closely related to soil survey and classification, but soil survey groups gather only a fraction of the information needed by soil fertility specialists. Although soil survey reports can be of value in determining the general nutrient deficiencies and average fertilizer requirements of a region, they cannot serve as a basis for fertilizer management recommendations for a specific field. Therefore. both pedon description and on-site characterization of the soil are essential. In order to correlate crop response to these factors, a pre-plant topsoil sample (0- 15cm) should be analyzed for organic C. total N. pH, extractable P. exchangeable bases, and acidity. These are indicators of soil reaction and fertility for any given site. Any soil amendments applied. including source and amount. should be recorded clearly (see table 1). 3 Land clearing: type of vegetation. method of land clearing (mechanized or manual), and residue management (whether burned or left on soil surface). 4. Cropping history: cropping history and fallow management sequences as far back as obtainable but at least for five years. 5. Experimental design and treatments: The experimental design used and rationale for choice of design should be stated clearly. Other information required includes: -treatment factors -factor levels -number of replications -land area and plot size -o ther resources (labor. chemicals, etc.). 6. Land preparation and soil amendment: tillage information indicates type of tillage (if any)-hoe tilled. no-till, tractor tilled with harrowing and diskingand depth of tillage: how residues and fertilizers are handled, whether incorporated or surface-applied; and to what extent native vegetation is destroyed. 7. Plot Size: plot size should be in meters: treatments should be labeled. Border and interplot areas should be sketched and labeled. 8. Planting: rainfall is the most important determinant of planting time in the tropics. The objective is to enable a crop to develop when rains are available. Therefore. timing of planting should be matched with the length of the growing season and the duration of effective rainfall. Plant density and spacing (inter- and intra-row spacing). should be recorded; optimum density gives the highest crop yield. 9. Crop variety: the crop and variety use should be indicated. A brief history of the source is helpful. 10. Pest and disease control: insect and disease problems sometimes determine the success or failure of a crop. Any disease or pest attack should be recorded along with control measures taken. Weeds should be included. even though they are not always seen as problems. 11. Crop growth data: at specific growth stages it may be necessary to obtain information on such growth parameters as: a) date to 50 percent emergence (seedlings) b) date to 50 percent tillers (leaf stage) C) date to 50 percent anthesis d) number of apical branches at certain periods e) other information related to harvest samples. such a s biomass at given growth stages. 12. Harvest information: the final yield data. an integrated value of all the factors that have influenced the crop during the season up to harvest (thus Yield = f [climate, sofl, crop variety management. insects. diseases, etc.1). This final product (output) wfll determine the crop production enterprise. Extreme care should be taken in recording the data as follows. Area hawested: m2 Plant population in area harvested: plants per m2 Biomass [top): g/m2 or &/plot. Pod weight (soybean, cowpeas): g/m2 or &/plot Grain weight or root (tuberons/wt): g/m2 or &/plot Seed or root number: seed/m2 or roots/plot Panicle number (sorghum, rice): ~anic les / rn~ Ear number (maize): ears/m2 Pod number (soybeans, cowpea): pods/m2 Table 1. Common commercial fertilizers Element Source Remark Nitrogen Sodium Nitrate Ammonium sulfate Urea Phosphorus Single superphosphate Treble superphosphate Ammonium phosphate Potassium Potassium chloride Potassium sulfate Mixed (contain 15-15-15 two or more of the above elements) Soil tends to be alkaline Soil tends to be acidic May volatilize Fixation Fixation Soil tends to be acidic; fixation most common in Nigeria Table 2. Environmental requirements of crops Edaphic (soil) factors Physical properties Soil reaction chemical properties (soil pH) Soil particle size distribution-soil texture Nutrient reserve Bulk density Exchangeable basesNa, K, Ca, Mg Cation exchange capacity Water-holding capacity/ Exchangeable acidity infiltration capacity Base saturations Soil temperature Available P Soil N Micronutrient elements - Mn, Fe, S may also d e c t crop performance. Soil organic matter REFERENCES Akinyemiju. Y.. and A.S. Adegoroye. 1987. Physiological considerations for tuber yield improvement in cassava (Manihot esculenta Crantz). Cassava-Based Systems Research Collaborative Group (CBSRC) report: this volume. de Vries. B.A. 1963. The physics of plant environments. In Environmental control of plant growth, ed. L.T. Evans, pp. 5-21. London: Academic hess . Good, R. 1953. Geography of the flowering plants. London: Longmans, Green and Co. Hahn, N.D. 1987. Processing, utilization and nutritional linkages for cassava based systems in various environments. Cassava-Based Systems Research Collaborative Group (CBSRC) report: this volume. Lawson, T.L. 1987. Characterizing the crop environment: a n agroclimatic perspective. Cassava-Based Systems Research Collaborative Group (CBSRC) report: this volume. Martin, J.H. & W.H. Leonard. 1964. Principles of fieldcrop production. New York: Macmillan. See pp. 15-50. Nweke, F.I. 1987. Collecting and interpreting relevant data required to link research results from varying environments: an economist's perspective. Cassava-Based Systems Research Collaborative Group (CBSRC) report: this volume. Wflsie. C.P. 1962. Crop Adaptation and Crop Distribution. London: W. H. Freeman & Co. Oboll. H.O.N. 1978. A new outline geography of West Africa. Lagos: Harrap and African University Press. 143pp. COLLECTING AND INTERPRETING RELEVANT DATA REQUIRED TO LINK RESEARCH RESULTS FROM VARYING ENVIRONMENTS: AN ECONOMIST'S PERSPECTIVE F. I. Nweke One way to link research results from varying environments is comparison of conclusions reached on the basis of data collected and interpreted. From the economist's perspective, environmental factors that could vary from place to place and from time to time include farm, household, infrastructure, market, and public policy. Farm Factors Among farm-level information important to economists in our type of investigations are farm and non-farm enterprises adopted, farm size, yield attained, availability and use of farm labor. level of commer- cialization of production, and inputs employed. Virtually all the smallholders in IITA mandate regions engage in multiple farm enterprise. They keep livestock and grow crops, and also grow tree crops and different arable crops both for home consumption and for sale. Most would also engage in non-farm activities. This diversification is due to the high risk in tropical agriculture, to family food security needs, and to the low cash income from farming. Farm size. even crop farm sue alone, is a difficult concept to define because it depends on enterprise, level of capitalization. and other factors. For the present purpose, farm sue should be defined as area cropped in one farming year. Where fields are fragmented, it includes all the different fields cultivated. Definition of cassava farm size is particularly problematic because cassava has no planting or harvesting season. Care should therefore be taken to ensure that cassava farm size is not overestimated by double counting. Only fields planted within a 12-calendar-month period should be counted, whether they are immature, mature, or even harvested. Farm size should not be determined by questioning the farmers. since their estimates may be subject to wide margins of error even in locations where there are local measures. Field measurements with compass. chain tape, and ranging poles are not expensive, and computer programes can compute the areas from such measurements. Knowledge of field size is important for farm budgeting. In one case, a team of investigators using USAID funding collected large amounts of farm input data over a period of 12 months from a large sample of fields. The data included fertilizers. seeds and labor by cost route approach. but the information was useless because field size was not obtained. One ton of fertilizer, or lOkg of millet seed, or 100 mandays of labor per year in field X is meaningless information in itself. To be useful, the information must be given in units per unit area such as one ton of fertilizer per ha, lOkg of millet seed per ha, and 100 mandays per ha. The need for field size measurement may not be critical if the objective is just yield determination and if such yield determination is done by standard methods of yield sampling. In taking cassava yield samples, care should be taken to ensure that fields of different ages and cassava varieties are not mixed up. It is necessaly that the age at which a yield sample is taken be that at which most of the cassava is harvested by the farmers. Sometimes, however, this may not be possible because of sample size limitations. Most economists would agree that farmers should not be paid for scientific information because such pay would introduce various types of bias and can make scientific investigations expensive. However, some consider that it is unfair to the smallholder, who may not have more than one hectare of cassava for an entire year, to take two or three yield samples of 40 square meters each from his farm without compensation, especially when the farmer is not ready to harvest. I am personally concerned that the investigator should pay for such harvest at the market rate. Whether he should take the harvest after the payment will depend on the cost of redisposing of it. Both assistance from family and hired hands are labor forces on the farm. In predominantly farming areas, household size tends to be large, especially among polygamous people. such as the Tiv people of Nigeria. who may be able to increase farm labor by increasing the number of wives per household. Hired labor is more often used by relatively large smallholders who may produce more for sale than for home consumption. Although farm labor is the most critical item of cost for the smallholder, it is also the most expensive farm information to secure accurately. Farm labor collected by the cost route method for one year for a reasonably large sample would take several years to analyze even with modem computers. Farmers know peak labor demand periods, and the most labor- intensive farm operations; for such purposes, labor information collection by the cost-route method is not necessary. More precise labor information is. however, necessary for farm budgeting or for production function analysis. As a short cut, instead of following selected farmers and fields over the entire crop season in a cost-route approach, one could take labor measurements wherever. whenever, and by whomever a required operation is observed. Cassava is particularly amenable to this methodology since it has virtually no limited planting or harvesting season. It should be remembered, of course, that harvesting is more labor-intensive in the dry than in the wet season. Knowledge of the degree of commercialization helps predict potentials for adoption of new technologies, especially those technologies that require purchased inputs. There is published evidence that farmers in general are more willing to adopt new technologies for production of crops for sale than for family food security. Household factors Household information of value to economists on the type of studies that we conduct includes size, composition. religion, income, and tribal origin of the household as well as age, education, sex, social standing. and secondary occupation of household head and spouse or spouses. Household size and composition with respect to age and gender of members provide indications of the labor available to the household as well as its consumption and expenditure needs. Religion and tribal origin could also influence labor availability and use. Among certain Muslim sects, women can leave the residence only at night. The contribution to farm labor of women thus restricted would be, in most cases, limited to crop processing at home. There is abundant evidence that among some tribes, farm operations and even farm crops are divided along gender lines. Household income, age. education and social status of the household head determine access to material inputs from public sources and to extension advice. These could also determine consumption and expenditure habits. Age of household head could, to some degree, be negatively related to access to material inputs and extension advice and hence to adoption of new technologies. Household income information is virtually impossible lo obtain directly with a reasonable degree of accuracy because most people are not able to estimate their income accurately: and even if they can, they may be reluctant to disclose it. A proxy often used is household expenditure. To obtain a reasonably accurate estimate of household expenditure, information must be sought on individual items as disaggregated as possible, even when the period of recall is short. Age is daicult to ascertain in the absence of birth records. Where an age grade institution exists. however, age can be approximated by ascertaining the respondent's cohort and the range of the age of his cohort. Precise information about level of education is obtained by determining the number of years the respondent spent in formal school irrespective of whether the years were spent repeating the same classes. Social status is often constructed as an index based on possession of items of social value or prestige in the area such as modem housing. furniture, or vehicles. A simple index could be the sum of the market values of such items. Infrastructure Factors The presence or absence of certain rural infrastructures, such a s transportation, health, water, industries, credit, and extension. is often used by economists to explain farm situations in study areas. Rural transportation facilities include rural road network and its linkage with urban centres, and transportation vehicles. Their availability would determine the extent of the available market for the farm products as well as access to some farm inputs. The presence of rural health, water and certain industrial facilities would influence farm labor availability in a farming area. When a woman or her baby is sick, she is not available for farm labor. The presence of pipe-borne water or of grain or cassava grinding machinery would reduce the amount of time women and children spend on fetching water or milling food crops. Time thus saved would probably be diverted to farm work. The presence or absence of modem health facilities could also affect the amount of cash available for farm investment. Where there are no modem health facilities, rural people often spend more money on native medicine and fortune-telling than others spend on modem medicine where there is such a facility. Rural credit also influences farm investment. Most of the time, traditional credit is all that is available to the smallholders. This system of credit is much more expensive than modem credit systems. The distance of the extension office or farm service centre, the farmer-to-extension-worker ratio, mobility, knowledgeability, and general working incentive of the extension workers influence adoption of new technologies and availability of publicly provided farm inputs. Market Factors If asked why they engage in farming, most smallholders in Africa would say it is to feed their households and to earn cash income. Cash-income earning depends on certain market factors, especially the size of the market for the farm produce in question as well as the cost of purchased inputs. Farm produce may have a limited market if it is consumed mainly by low-income groups or if it cannot be moved at a reasonable cost from the producing to consuming centers as a consequence of inadequate transportation, processing, or packaging facilities. Symptoms of low market availability for produce are relatively low market prices. and seasonal or cyclical fluctuations in the price. Cassava, especially in certain forms, is an example of a commodity which has a limited market because it is consumed by low-income groups. For example. in 1986. in Bendel State. Nigeria, the price on a dry-matter basis was W0.76 per kilogram for cassava as against 812.29 per kilogram for cowpea. The price of gari. the major commercial form of cassava, exhibited substantial year-to-year fluctuations between 1968 and 1985 in Bendel State, as shown in figure 1. Expansion in production results in steep price declines because the available market is not able to absorb the additional production. Plantain is another commodity that has a limited market because it cannot be moved at a reasonable transportation cost from producing to consuming centers. I t exhibits steep seasonal price fluctuations in producing areas. as shown in figure 2, and wide spatial price Werences between producing and consuming centres. Figure 1. Real price of gari (at 1960 level) in Ooia LO& Bendel State. 1968-85 #/ml 5 Year Figure 2. Indice- of retail market prices /mt at Umuagwo village market and plantain bunch yield in UmuaggPo. January-December 1985 Percent -, ; 1 ! ( 4 2mt /ha /mth = l o o % ) 0 - A J F M A M J J A S O N D Months 4 2 Price information needs for our type of study would therefore include time series. if possible monthly, and cross-sectional prices not only for cassava but for its major substitutes and complements. In most countries, a notable exception being Cameroon, price information is routinely collected over time and space. In this case, an investigator usually need not collect price data just as he need not collect most agroclimatic information directly. The wage rate is another important item of market information. It may vary significantly with location, gender, farm enterprise and operation, and sometimes with season, but rarely with fanner in one location. Policy factor Public policy with respect to exchange rate can affect aggregate market availability and hence product prices. It can affect availability and prices of such imported inputs as fertilizers, herbicides, other pesticides, farm machinery and appliances, and fuel. Government policy on urban minimum wage rate could result in artificially high farm wage rates leading to high production costs and reduced produce markets. The consequence of negative public farm policies in these areas would be a reduction in incentive to adopt improved technologies. Direct governmental production in agriculture affects the availability and efficiency of use of important farm resources. It would be interesting to see how much of available prime farm land, labor, and other farm resources are employed by the CDC and how much is left for private farmers in Cameroon. The iterature is loaded with evidence of inefficient use of resources including land. labor. and extension resources in direct governmental production in agriculture. The case of Ghana in the 1960s is a classic example. Grouping for emciency in data collection To collect the required data efficiently. the information needed should be listed for each objective and then grouped according to source, as follows: 1. Primary information which the investigator should obtain by direct observation or measurement without asking the respondent. This includes crops grown, crop mixtures adopted, farm size, yield attained, and certain items required to construct index of social standing of the household such as the type of house, household furniture, transportation vehicle owned, as well a s some rural infrastructural facilities. A young PhD holder once interviewed a farmer who was staking yam in his yam+cassava+maize+vegetable mixed-crop field. When this professional asked the farmer what crops he grew, the farmer retorted with You've got eyes; can't you see with them?" 2. Primary information which the investigator cannot obtain by direct observations or measurement but is common to all respondents in a location. Examples are certain infrastructural facilities, farm Wage rates, sources of hired labor, and frequency of market meetings. Inexperienced colleagues often make such mistakes as asking every sample farmer in a village what the wage rate was or when the rainy season set in the previous year. 3. Primary information that the investigator cannot obtain by direct observation or measurement and that varies with respondent in a location. This includes most data at the farm and household levels. 4. Information obtainable irom secondary sources. This includes all time series data such as price, weather, and import and export information, that are routinely collected by public agencies. In Nigeria. such agencies are the Federal Office of Statistics, the state ministries of Finance and Economic Development, the Central Bank, the Meteorology Department. the Department of Customs and Excise. research institutes and universities. This type of information can also be obtained internationally: FAO, USDA. WHO. UNICEF. for example, also publish secondary data of interest for investigations in agricultural economics. The first three groups of data suggest that, for collection of primary information, three different types of data collection schedule may be prepared: one for direct observations and measurements, one for group or village interviews. and the third for individual respondent interview. All the schedules should be labeled with field, farmer, and village numbers. Sampling The unit of investigation could be the field. as in the case of yield sample, the household, or even the village. depending on the objective of the investigation. Sample size should not be too large. or analysis will become unwieldy and final result will be delayed. The spread or concentration of the sample would depend on the importance of different types of information in the investigation. For example, if the main focus is on the farm or household factors, then the sample could involve many farmers in a few locations. If, on the other hand, the main focus is on the variables which are common to farmers in a location but vary from location to location. then the sample should involve many locations with few farmers in one location. Analysis and Conclusion A wide range of economic analyses, including adoption, supply or demand function, sensitivity or risk. farm or household budgeting. market integration, and marketing margins. are possible with the type of data discussed above in combination with some agroclimatic and agronomic data discussed in other papers. Therefore, it will be possible to compare different environments with respect to adoption of any given technology, efllciency of use of resources in production of a given crop, its market potentials, the relative riskiness of production of a crop or use of a new farm technology, the emciency of the marketing system for a crop, with respect to potential income distribution effect of wide adoption of a new technology which could result in significant increase in the output of the commodity of interest. In other words. with the type of data discussed above, it is possible to predict and compare potentials of new farm technologies in different environments. REFERENCES Grilliches, Z. 1957. Hybrid corn: an exploration in the economics of technological change. Econometrics, 25: 4. Oko~ji. E.O. 1983. Consequences on agricultural productivity of crop stereotyping along sex lines: a case study of four villages in Abakaliki area of Anambra State. MSc thesis, University of Nigeria, Nsukka. Nweke. F.I. 1978. Direct governmental production in agriculture in Ghana: consequences for food production and consumption: 1960-66 and 1967-75. Food Policy. 1: 2. PROCESSING. UTILIZATION AND NUTRITIONAL LINKAGES FOR CASSAVA-BASED SYSTEMS IN VARIOUS ENVIRONMENTS N.D. Hahn During the 1980s. cassava has been receiving increasing attention from international groups. UNICEF has cited it as a crop for 'household food security" as indicated in a policy shift from primarily health and irnrnu- nization programs to improved cropping systems and the use of 'social mobilization" in introducing improved cassava varieties and food products. A US$40 million program on cassava multiplication, with support from the International Fund for Agricultural Development (IFAD) and the World Bank, was started this past year in Nigeria. In August 1987, the International Food Policy Research Institute (IFPRI) organized a workshop in Washington. D.C.. on Trends and Prospects of Cassava in the Third World" to round off this interest in cassava by international donors. Some skepticism has been voiced concerning this reported atten- tion to cassava research. Many observers wonder whether it is because the crop does not compete with U.S. and European produce and is thus safe for further development. More positive opinions stress that cassava has been grossly underresearched and underdeveloped, and that approp- riate processing, improved storage, and inter-country marketing have vast potential for the urban African population. The research concen- tration has so far been on the production aspects. The purpose of this paper is to highlight the linkages among utilization, processing, and nutrition, and the reasons why these factors must be reincorporated as an integral part of cassava research in varying environments. Recognizing and reducing the enormous crop losses that occur between harvesting and final use can sienificantlv contribute to improving the supply of a&icultural products ab&e and f& beyond what may be achieved by increased primary production (Booth 1974). Cassava possesses many merits as an insurance crop, but it also presents constraints, particularly as an energy food. Losses during storage are high and the crop is highly perishable. In addition, the arduous processing necessary requires much labor, particularly female. Roots and tubers: food consumption in A€rica For countries such as Central African Republic, Congo. Mozambique and Zaire. cassava provides 70 percent of the caloric intake. and an average total of 407.4kg is consumed annually per inhabitant (Dorosh 1987: Gebremeskel and Oyewole 1987). Cassava ranks far above any other roots and tubers or cereals in consumption both in these four countries. which constitute Group I in table 1, and in the countries of Group 11. Table 1. Staple food consumption (kg per inhabitant) in sub-Saharan Africa. 1981-83. by group Group I Group I1 Group 111 Total kg per inhabitant Roots: total 427.2 234.9 43.1 182.4 Cassava 407.4 123.0 21.3 117.8 Yam 6.6 72.4 3.5 36.8 Sweet potato 6.6 20.3 5.0 12.5 Others 6.6 19.2 13.3 15.3 Plantain Cereals % equivalent calories Roots: total 74 43 8 36 Cassava 70 22 4 24 Yam 1 14 4 7 Sweet potato 2 3 1 2 Others 1 4 3 2 Plantain Cereals Source FAO. Note: These countries are grou d as follows: Group I: Central African ~epuKic . Congo. Mozambique, Zaire. Group n: Angola, Benin. Burundi. Cameroon, Comores, Guinea, Gabon, Ghana, CBte d'lvoire. Nigeria, Rwanda, Tanzania. Togo. Uganda. Group i1I: Botswana. Burkina Faso, C a r Verde, Chad, Ethiopia, Gambia. Guinea, Guinea Bissau. Kenya. Lesotho, Liberia. Ma a ascar. Malawi. Mali, Mauritania, Mauritius. Namibia. Ni er, Reunion. SBo Tome an8 Principe, Senegal, Seychelles, Sierra Leone, Somalia, Su$an, Swaziland. Zambia. Zimbabwe. Cassava accounts for about 1,200 calories per capita per day (about half of total calories) in Zaire and the Congo and more than 900 calories per capita per day in the Central African Republic. On the basis of FA0 Food Balance Sheets, it is estimated that 'there are 40 million people living in Central Africa and Mozambique whose average daily cassava consumption exceeds 600 calories per day. and another 120 million people (throughout Africa1 whose average daily cassava consumption exceeds 200 calories per day" (Dorosh 1987). Cassava as a staple Cassava has some distinct disadvantages. The protein content is only 1 percent of fresh weight and 3 percent on a dry matter basis. This protein content compares unfavorably with other roots and tubers: white potato has 9 percent. sweet potato 4.3 percent and yam 8.7 percent protein. Compare this with the 40 percent protein content of soybeans. Cassava has a moisture content of between 60 and 70 percent, which increases transport costs, and a very short post-harvest storage life. The losses have been estimated at 14 to 75 percent (Janssen and Wheatley 1985). A more conservative estimate suggests that around 25 percent of all perishable food crops harvested are lost before they are consumed. Nevertheless, the potential long-term ground storability of cassava is a distinct advantage. Deterioration, manifest in loss of quality and quantity, results from pathological, physiological, or mechanical damage (Booth 1973). Because cassava can be stored for only two or three days after harvest, it is left in the ground until needed and then is consumed or processed immediately. According to calculations given by Ingram and Humphries (19721, if only half the global cassava crop is left in the ground for as little as two months longer than necessary, more than 8 percent of the total area planted to the crop is unnecessarily occupied, assuming a 12-month growing season. Thus, on a global basis of just over 9 million hectares cropped with cassava, about three quarters of a million hectares of agricultural land are withheld from alternative production. Cassava is usually available all year round thanks to this practice of "storage avoidance" (Intermediate Technology Dwelopment Group 1987). Cassava nutritional drawbacks are its low protein content, low energy density, and potential toxic effects from the natural content of cyanide-yielding compounds (Jaynes 1987). The first of these can be effectively counteracted with protein-rich supplementary food and the second with energy-dense supplementaly food (Rosling 1987). As reported by Dr. Jesse Jaynes, the cassava root contains 30-40 percent dry matter, composed mainly of starch and sugar. which is found in a higher proportion than in most other roots and tubers. Thus cassava is an admirable source of calories, but its low protein content and the extremely poor quality of the protein it does contain make it an incomplete food (see table 2). Table 2. Dry matter. carbohydrates. and protein content of root and tuber crops Crop Dry matter Carbohydrate Protein (%I (% dry matter) (% dry matter) Cassava 37.5 92.5 White potato 22.0 85.9 Sweet potato 30.0 91.0 Yam 27.6 87.3 Taro 27.5 84.4 Sounx: Jaynes 1987 Figure 1 shows that cassava is extremely deficient in certain essential amino acids. It has been difficult to produce significant increases in the essential amino acid content of cassava by means of classic plant-breeding approaches (Jaynes 1987). Rosling (1987) showed that toxicity in cassava is caused by the poison cyanide (prussic acid), which has the simple chemical structure HCN. Toxic effects occur when cyanide is liberated from a more complex chemical compound called linamarin. Dietary cyanide exposure from cassava will result from consumption of insufficiently-processed roots. probably from liberation of cyanide in the gut from ingested linamarin. The human body has a fairly effective thiocyanate (SCN). The substrate for this reaction is sulfur IS) originating from proteins in the diet; so if the protein intake is adequate. the human body can withstand moderate cyanide exposure without any symptoms of accumulated effects. But cassava has a low protein content. and, especially during droughts. poor families will also have a low intake of protein-rich supplementary food. The toxic effects of ingested cyanide may thus be aggravated by a low sulfur intake. Figure 1. Essential amino acid yield in the amount of cassava necessary to yield 25 percent of the caloric requirement of a 20-kg child (about 325 gl EAA yield in 25% caloric amount of cassava Valine Tryptophan Threonine Phe/Tyr Met/Cys Lysine Leucine Isoleucine % minimum daily requirement Rosling explains the incidences of goiter and cretinism (a form of mental retardation) as caused by iodine deficiency. These disorders can be considerably aggravated by continuous dietary cyanide exposure from insufficiently-processed cassava. This effect is caused by the detoxi- cation product thiocyanate which interferes negatively with iodine up- take in the thyroid gland. Paralysis of both legs, caused by permanent damage to the spinal cord, has been associated with a combination of a high cyanide and a low sulfur intake from diets dominated by insuf- ficiently-processed cassava and lacking protein-rich supplementary food. The disease, named Epidemic Spastic Paraparesis, has in the last decade crippled 5.000- 10,000 women and children during periods of food shortages in cassava-dominated areas in Zaire, Tanzania, and Mozam- bique. For lack of other foods, the affected families had to consume newly-harvested cassava roots without processing normally: more than one week is required to remove cyanide effectively (Rosling 1987). The historical association of cassava consumption with kwashiorkor in weaned infants in tropical Africa and Brazil rests on the very low protein content and the frequently simultaneous absence from the diet of satisfactory sources of supplementary protein. It is essential to understand the cyanogenic effects of poorly processed cassava to ensure proper processing for the crop's develooment. Nevertheless. as a household food securitv croD. the advaniages of cassava production and use certainly oGtweigh the disadvantages. The following factors are put forward for consideration as nationar collaborators initiate a program on cassava post-harvest technologies and utilization. Factors to consider when collecting and interpreting relevant data related to cassava utilization, processing and nutrition 1. Socioeconomic-nutritional base-line surveys that determine the importance and trends in cassava production and use within production system. Particular attention should here be focused on the methods of cassava preparation for household use: market sales for feed, food, or industrial use; prices of cassava products vis-a-vis other competing crop products; measured nutritional value of various cassava-based foods and potential for introducing new cassava-based foods, both for household and commercial use. 2. On-farm research to include a nutritional and health assessment component for cassava-based regions. The most common sets of indicators are mortality statistics and anthropometric measurements. Keep in mind that children from birth to three years suffering from third-degree malnutrition have been found to have mortality rates 6 to 20 times as high as children of normal weight. For anthropometric indices. measurements of weight and height (or supine length for children under two years) are the most sensitive indicators of the nutritional status of infants and young children. In addition, arm circumference can be used for assessment of nutritional status indepen- dent of age between six months and four to five years [Austin 1981). Other assessments which non-nutritionists can make include aclinical assessments of hair, eyes, and indications of edema. The most certain assessment is laboratory biochemical analysis of bodily fluids. 3. Itl terms of the quality of the improved cassava varieties, it will be necessary to determine whether they speed up or slow down processing time and marketing. Data will need to be collected on time allocation and labor use at various stages of cassava development. Careful consideration needs to be given to women's labor and women's response to the improved varieties. Some women's groups in Oyo and Ondo States, Nigeria, have complained that the larger tubers are more difficult to peel and to market and that they require longer frying because of the high water content. 4. Medical and health-related research needs to be developed on the dangers, particularly to pregnant women and children who inhale fumes during the fiying of gari. 5. What are the richer energy foods that can be introduced into a cassava-based system? The potential agronomic and nutritional advantages of introducing soybeans should be considered. 6. Collection and analysis of samples of processed cassava from processing centers to determine the overall quality of product, storage capabilities and mechanical damage on the improved tubers. Research should be done in local traditional settings rather than in sophisticated laboratories. 7. Extent of bitter and sweet varieties in a production system and the incorporation of both varieties into any new system with com- plementary research on the acceptability of new food products. Oben and Menz have concluded that 'the potential benefits from the breeding of improved low cyanide cassava varieties in Nigeria are extremely high relative to the cost" (1980). Their survey indicates the relative impor- tance of sweet cassava in various regions of Nigeria. One of the primary objectives of the IITA-UNICEF Program on Household Food Security and Nutrition is the development of cassava- based foods from sweet varieties. Primarily indigenous sweet varieties are used and 44 new food products have been developed. 8. A South-South Exchange on the introduction and testing at rural household level of new food products based on Asian preparation techniques and use. 9. Improved efficiency in machinery. particularly for the proces- sing of cassava into gari to cut down on the time women spend in proces- sing. A 1986 IITA sunrey by Oyewole indicates that a power grater can reduce the time needed to grate 140kg of tubers from 6 hours to 20 minutes. The development of machinery to cut down on labor requirements is essential. Given the hours spent, particularly by women, cassava is not a low-input crop. A study comparing the percentage of labor input con- tributions in cassava processing and utilization in five Nigerian states Table 3. Relative importance of sweet cassava in various regions of Nigeria State Sweet cassava Percentages Percentagea as a percent- of farmers of farmers agea of total growing growing both cassava grown only sweet sweet and (by area) cassava bitter cassava Anambra 0 Bendel 6 ogun 21 Kaduna 97 Note: a l n a @en year found that women contributed 82 percent of the total requirement (Ikpi et al. 1987). A 1986 IITA-UNICEF Study in Oyo State. Nigeria, indicates that with the introduction and use of new cassava-processing equipment. women can save considerable time. For instance, one processing hour on a machine saves women 21 hours' work each week. Given the average amount of cassava processed by a household in a year in the Oyo State areas surveyed, with appropriate cassava processing equipment, each family could save an average of 441 hours of work (Ikpi et al. 1986). 10. National research on genetic engineering should be initiated to modify the essential amino acid composition of cassava and thus increase its nutritive value. Priority research attention should be given to supplementing the existing proteins of cassava with new synthetic proteins with a high content of essential amino acids. REFERENCES Austin, J . E. 1981. Nutrition intervention in developing countries. Prepared by the Harvard Institute for International Development for the Office of Nutrition. U.S. Agency for International Development. Booth. R. H. 1974. Post harvest deterioration of tropical root crops: Losses and their control. Tropical Science 16(2): 49-643. Boccas. Bernard. 1987. Cassava. staple food crop of prime importance in the tropics. The Courier, 101: January-February 1987. Cooke, R. D., J.E. Richard, and A.K. Thompson. 1985. Nutritional as- pects of cassava storage and processing. Tropical Development and Research Institute, VIIth Symposium of the International Society for Tropical Roots Crops IISTRC). 1-6 July 1985, Guadeloupe. Dorosh. P. 1987 Economics of cassava in Africa: an overview Paper prepared for workshop on Trends and Prospects of Cassava in the Third World. IFPRI, 10-12 August 1987, Washington, D.C. Ekpere, J. A,, E.A. Ikpi, G. Gleason, and T. Gebremeskel. 1986. The place of cassava in Nigeria's food security, rural nutrition and farm income generation: a situation analysis for Oyo State. Nigeria. IITA-UNICEF Consultation on Promotion of Household Food Production and Nutrition, 2-8 March 1986. IITA, Ibadan. p. 26. Gebremeskel, T.. and D.B. Oyewole. 1987. Cassava in Africa and the world trends of vital statistics. Socio-Economic Unit. IITA, January, 1987. Ikpi. A. E.. T. Gebremeskel, N.D. Hahn. H.C. &urnah, and J.A. Ekpere. 1986. Cassava: A crop for household food security: A 1986 situation analysis for Oyo local government area, Nigeria. IITA-UNICEF Collaborative Program on Promotion of Household Food Production and Nutrition. Socio-Economic Unit. June 1986. IITA, Ibadan. Ingram, J. S.. and J.R.O. Humphries. 1972. Cassava storage: a review. Trop. Sci. 14. 131-148. Intermediate Technology Development Group. 1987. Root crops processing: Food Cycle Technology Source Book. United Nations Development Fund for Women. Janssen. W., and C. WheaUey. 1985. Urban cassava markets: the impact of fresh root storage. Food Policy, August 1985: 265-77. Jaynes. J. 1987. Genetic engineering of root, tuber and leguminous crops for increased nutritional value and improved disease resistance. Unpublished document, University of Louisiana. Jaynes, J., and J. Dodds. 1987. Synthetic genes make better potatoes. New Scientist. 17 September 1987: 62-64. Oben, D. H.. and K. M. Menz. 1980. Prospects for low cyanide cassava in Nigeria. Ibadan: IITA. Rosling. Hans. 1987. Cassava toxicity and food security: a review of health effects of cyanide exposure from cassava and ways to prevent them. UNICEF Program on Household Food Security and Nutrition. Appendix I. Stagen of cassava proceuing Dried cassava chips I Cassava P~e led Sieved Roasted Gari Cassava tevmed Pelletized - I Fermented Soaked Grated and D F U ~ U Dewatered ,"ice Boled extracted L-+rtstawh decanted off Y Eaten as Vegetable w Tapioca SnackFoods 90- UNIFEMIWAFI October 1987 Appendix 11. Source of calories (%I by commodity group (19771 Source of All Develo~inP market economies calories developing countries Africa Latin Near Far America East East Vegetable products Animal products Cereals Wheat Rice Maize Millet and sorghum Roots and tubers Sugars and honey Pulses Nuts and oilseeds Vegetables Fruit Meat, eggs, fish, milk Oils and fats Miscellaneous - - Total Kcal/day 2 260 2 205 2 557 2 620 2028 Kichard Longhurst and Michael Li ton, Secondary food crops and the reduction of seasonal food insecurity: the role of agricugural research. IFPRI/FAO/AID workshop on -Seasonal Causes of Household Food Insecurity: Policy Implications and Research Needs," 10-13 December 1985, Annapolis, Maryland. Appendix III. Source of calories (0.6) by season in three Zaria villages in northern Nigeria. 1970-71 Food Group Apr- Jun- Aug- Oct- Dec- Feb- May Jul Sep Nov J a n Mar Cereals Cereal products Starchy roots Milk Meat Poultry, fish. eggs Seeds. nuts, legumes Fats and oils Vegetables, fresh Vegetables, dry Fruits Sugar, sweets Salt. spices Snacks. Misc. Total calories intake Kcal 2457 2311 2456 2274 1951 2137 So- Calculated from E.B. Slmrnons, Calorie and Pmteln Intakes in Three Villages of 7 a r h Pmvlnce, May 1970June 1971. Samaru Miscellaneous Paper 55. Ahmadu Bello Universl@. Zarla, (19761, 11 and 129. Richard bnghurs t and Michael Li ton. Secondary food cmps and the reduction of seasanal fwd Insecurity: the role of a cugural research. IFPRI/FAO/AID warksho on -Seasonal Causes of Household Food ? nsecurity: Policy ImpUcaUons and ~ e s e a r c h Reeds, 10-13 December 1985. Annapolis. Maryland. Appendix N. Percentage labor input contribution of females and males in cassava production and processing activities Activity (1) (2) (3) (1) (1) Approx. Anambra Bendel Benue Cross Oyo average River for Nigeria Field preparation 20 80 60 40 25 75 30 70 35 65 34 66 Planting 90 10 80 20 75 25 70 30 70 30 77 33 Weeding 90 10 100 - 75 25 90 10 75 25 86 14 Harvesting 80 20 90 10 75 25 80 20 60 40 77 23 Processing 100 - 100 - 100 - 100 - 100 - 100 - Storage 100 - 100 - 100 - 100 - 100 - 100 - Marketing 100 - 100 - 100 - 100 - 100 - 100 - Overall Average 83 17 90 10 79 21 81 19 77 23 82 18 Soum J. A. Ekpere. A. E. Ikpi, G. Gleason. T. Gebxmeskel, IITA-UNICEF Consultation on Promotion of Household Food Production and Nutrition, 2-8 March 1986. IITA. Ibadan, the place of cassava in Ni eria's food security, rural nutrition and farm income generation: A situation analysis for $0 State. Nigeria. Calculated and compiled by Dr A. Ik i Senior Lecturer. Agricultural Economics De artment. Universiq of lbadan from (17 bngaing field surveys in Anambra, Crass River an: Ovo State. 1985-86: 121 Cram 0. Udele .Ruura1 women in aorirultural marketing--case ~~~~~. - ~ ~ ~ . ,-, - - ~ ~ - ~~ ~~~ ~ ---- ~ ----- ~ studvbf cassava in Isoke Local Government Area. Bendel State: MSc thesis ~niv&i&af lbadsl, for the Bendel State R res and (3) Mary E. Buxfisher and Nadhc R Horenstein 'sex mles in the Nigerian Tiv fann euseholdsm for tbe Benue State Rgurcs (1981). COMMENTS ON THE THEME TOPIC : "LINKING SIMILAR -0NMENTS" 1. H.J. W. Mutsaers The emphasis in this section was "similar environment". The first objective should be to ascertain how similar. how different and what key factors are responsible and how the information can be gathered. The papers provided too much data. there should be a minimum set of data perhaps at Werent levels of the environments, eg. broad zones within which selected regions are identified for research with which specific research sites are identined. 2. J. Smith The first question should be an identification of purpose of data collected. it could be for production-function analysis, adoption studies etc. The research objectives determine which data to emphasvx. The paper did not address comparison across countries, given different currencies. What exchange rate should be used to compare data meeting terms? One may examine possibflities with shadow exchange rates, ratios of wage rate to price of commodity. If the wage rate is low compared with price of cassava, there will be more incentive to produce cassava. Cassava is increasingly becoming an industrial, not just a food, crop. There is a need to emphasize end use. Flexibility of harvest time affects end product quality and quantity, eg, water content and garification rate. Most products derived from cassava are affected by pre-hanrest factors. 4. D.S. C. Spencer While noting the danger in oversimplification in data collection. especially when research/survey objectives are not fully observed, the set of data suggested in the papers is too broad. Groups may define a minimum data set for activity groups, for example, in the following areas: [a) Constraints analysis (b) Experimental station studies [c) Research managed studies [d) On-farm studies The objective should be to ascertain how similar or dilferent various research sites are, and what minimum sets of data are needed to characterize the environments. It is perhaps more practical to consider only agroclimatic data in a broad sense and link socioeconomic data with specific use intended for data collected. Diagnostic survey results presented showed two trends: those based purely on qualitative data and those based on quantitative data as well. For the collaborative research objectives, quantitative data should be routinely collected by the survey group themselves, eg, field area and yield. CASSAVA M THE FARMING SYSTEMS OF CAMEROON'S HIGH-RAINFALL COAST S.W. Almy and M.T. Besong Abstract The Fako Division in the Southwest province of Cameroon extends from sandy loam soils with a mere 1,800mm annual rainfall to the recent volcanic soils of the highest rainfall area in Africa [9.800mm). In both areas food-crop farmers grow cassava, but not a s the only staple. Prolonged rainfall permits two cropping seasons and the sequential cultivation of several major and minor crops, thereby increasing food quantity, quality and security. This report is based on a 1986 agro- socioeconomic survey with accompanying agronomic resarch. Implications of the ongoing release of new high-yielding varieties of cassava into the area are discussed in this context. The Fako Division of the Southwest Province stretches from the coast of Cameroon to the eastern, western, and southern slopes of Mount Cameroon, the tallest peak in West Africa (4,072m) and an active volcano. The larger part of the Division is recent volcanic soil, rich but very rocky. The major cassava-producing area is on the older. sedimentary, sandy loam soils of the Tiko Plain [TP), which receives 1.800-2,800mm of rain annually. The second major area is the volcanic coast (Lower Volcanic, or LV). from the old port of Limbe [Victoria) up to an altitude of 600m. with from 4,800mm of rainfall on the coast to 2,200mm inland. Both regions enjoy seven to eight months a year with more than lOOmm rainfall. Rainfall is unirnodal. A third volcanic area along the western coast of the mountain [West Coast, WC), with few food- crop farmers, receives 5,400-9.800mm of rain a year, with 0-2 months under 100mm. The fourth zone, the Upper Volcanic [over 600m high) grows no cassava. Almost half the agricultural land is in large plantations of rubber, oil-palm, and banana, most belonging to the parastatal Cameroon Development Corporation. The rest is in the hands of small farmers, primarily food-crop growers. There are five principal staples: plantains [Musa L. [Am]). cocoyams [Xanthosoma sagitt$olium), maize (Zea mays). cassava [Manihot esculenta) and taro [Colocasia esculenta), in order of land area occupied. Cassava is the predominant crop on the Tiko plain and West Coast. Fako farmers come from all over southern Cameroon and eastern Nigeria, and confront such a wide variety of soils and topographical and climatic conditions that it is not surprising that fanning methods have not standardized around a single system, even within a village. Previous agronomic and breeding research was based on patterns found in other parts of the country or obsewed at roadsides and among farms near the experimental fields. In order to comprehend the diversity of Fako farming and to locate some modal patterns from which agronomic trials could start to test improvements. a survey was launched in October- November 1986. A partial report, focusing on the place of cassava within the farming systems, is made here. The full report. Farming systems survey of Fako Division. can be obtained from the Testing and Liaison Unit (TLU), IRA-Ekona. Methods A nine-page questionnaire requiring one and a half hours for execution was administered to 124 randomly selected full-time farmers in 18 towns and villages chosen to represent nine tentatively defined agroecological zones of Fako. The questions covered all aspects of the farming system-household composition and labor, land access and use. cash flow and credit, cropping patterns and calendar, crop provenance, field problems. harvesting, storage, marketing, the place of animals, and extension. More emphasis was put on maize since the TLU receives its outside funding from the National Cereals Research and Extension Project (USAID). Enumerators were school-leavers with 0-levels who received three days' training and close supervision. Farms were not directly evaluated except as described below. In order to obtain a picture of the relative importance of the different crops in Fako and its zones, a series of approximations had to be made, both for crop area and production. Fako farmers do not have any measure for land area, so for the first we substituted clearing time. This was checked by measuring one field per village and comparing the averages. This showed that previous status of the field (forest. bush fallow, immediate replanting) dected these times drastically, so any results have to be used with extreme care. Estimates of area within a field occupied by a specific crop were made using a formula based on the farmer's identification of the crop as major or minor to the field, and the number of other major and minor crops in it. From these figures, a very rough calculation was obtained of the percentage of total food-crop area planted in one year (1985-86) occupied by each crop. For production figures, farmers were asked to remember how many hand-trucks. basins or baskets they had taken from any of their farms in the previous two seasons. In the case of crops harvested over several months. they were asked to recall how often they went and what they removed in one trip. These approximations were converted to standard units' of volume, weight, and then price. Neither area nor production figures are very satisfactory, but the latter has only one principal source of u~eliability-the farmers' memories. Such results should only be used for purposes of general comparisons of crop importance. For yield estimates and farm budgeting, intensive measurements must be carried out on a subsample of fanners, as we are now doing for cassava in Fako. Results and Discussion Fako farmers adopt one of two general strategies for their fields. Either they plant many crops in a few fields, or a few crops in many. They range from the 25 percent of farmers who plant only one field, with an average of 5.7 crops. to the 6 percent that plant five fields. each with 1.5 crops (including minor crops with only a few plants per field). Twenty-four percent of the fields in the survey contained only one crop, while another 41 percent contained only one major crop, plus one or more minor ones spotted through the field. But monocropped (and what we call 'effectively" monocropped) fields are smaller: 18 percent of the land is estimated to be in monocropped fields. 40 percent in elTecUve monocrop. and 43 percent in major intercrops. Twenty-seven percent of LV fields were monocropped but only 8 percent of TP and 4 percent of WC fields. In terms of relative crop area, plantains lead with 23 percent followed by cocoyalns (20 percent), maize (18 percent), cassava (15 percent) and taro (8 percent). The largest number of farmers 193 percent) grow maize, but only 55 percent as a major crop, whereas 76 percent grow plantains as a major crop. 60 percent cocoyams. 55 percent cassava, and 26 percent taro. Eighteen percent of cassava, 29 percent of plantains and about a third of other major crops are found in fields with two to three other major crops. Maize is most often intercropped with cassava (at twice its frequency with any other crop) especially in the Tiko plain, but cassava is equally often intercropped with cocoyams or plantains. Cocoyams and plantains are almost always found together, and often with taro (major or minor) as well. Production estimates, which are probably more reliable, give more weight to plantains and less to maize (see tables in Appendix). In terms of energy provided, plantains provide 35 percent of the total kilo- Help with weight price and volume equivalencies came from J. Wuloh of the Ekona ROTREP m ect, P Koil of Ekona Plantains, and M. Besong's1986 Food price survey of Fako Division YnL-Ekina). calories, cocoyams 10 percent, cassava 41 percent, maize 8 percent, and taro 4 percent. But in market values, plantains give a full 49 percent of food-farm income. cocoyams 18 percent. cassava 17 percent, taro 7 percent and maize only 6 percent. Plantain prices have been going up for some years, both from internal pressures (increasing borer and nematode infestation, according to farmers' reports) and external ones (destruction of Gabon's plantain production by black Sigatoka disease. and entry into the Cameroonian market]. Cocoyam prices have been rocketing (tripling in ten years) because of the root rot (Pythium) disease now spread throughout Cameroon. As these are the preferred staples. there has not been simple displacement of dem.and to other crops (although taro seems to be progressively invading cocoyam fields). Local cassava is basically of two types: (i) white cassava, which is too high in hydrocyanic acid ("bitter") to eat raw or boiled, but has to be processed into water 'fufu" (by peeling, soaking for three days, grating, drying one day, sifting and mixfng with water) or sometimes "gari" (peeling, grating. squeezing one to two days in a bag, sifting and roasting); and (ii) a lower yielding reddish tuber which can be eaten boiled (either as tuber or pounded fuful. In the Fako survey, 73 percent eat pounded fufu. and only 54 percent water fufu and 36 percent gari. But only water fufu and gari can be kept long enough to market at a distance. Cassava leaves are not a major dietary item. Except on the west coast, planting usually occurs in late March to April and late August to September, later and earlier respectively in the lower-rainfall areas. The July-August rains are too heavy to allow crop establishment. Three-quarters of all cassava fields are planted first season. Uusally all crops are planted together, or within a week of each other. The maize, often with a minor crop of "egusi" melon as well, is harvested and the cassava left to take over the field and provide shade to a cocoyam intercrop. Second season plantings usually include a minor crop of groundnuts. From field observations, plantains intercropped with cassava as major crops are usually scattered in clumps or segregated in one part of the field. Weeding ceases at four months after planting for 57 percent of farmers, and by six months for 84 percent. Local varieties have sparse canopies and are planted at low densities (about 6-8.00pph) because of the intercropping strategy. The fields thus begin to relapse into heavy weed fallows even before harvest starts. Cassava is usually harvested at 12-18 months. Although the average harvesting duration is six months, because of variations in planting and harvest commencement, at least 28 percent of Fako cassava farmers are harvesting every month in the year. The scarcity months (18-36 percent of farmers harvesting] are in July or November, with 45 percent or more harvesting most of the rest of the year. The July to September harvests are impeded in the major processing zone (TP) by swollen rivers cutting people off from their farms. Processing also suffers in this season from competition with cocoa and coffee harvest labor needs. The cassava yield study still in progress shows that cassava in 12 months has yields ranging from 4.6 to 23.8 tons/ha. with a mean of 11.6 tons. Eighteen-month yields range from 6.0 to 22.9 tons/ha. with a mean of 14.8 tons. Within this framework. the IRA/IITA/Gatsby Foundation cassava project, based in IRA-Njombe (Littoral Province), is now introducing several new high-yielding cassava varieties (at present denominated 8017. 8034 and 8061). All three are moderately bitter and have a white tuber. They were selected on the worst (loamy sand) soils to be found in the coastal lowlands. to ensure high performance under poor conditions. Controlled on-farm testing under intercrop and low- management conditions began in early 1987 in Fako and Mem Divisions, under the TLU. Sample harvests of 8061 distributed to farmers for in 1986 give yields at 12 months ranging from 14.6 to 39 tons/ha. with a mean of 28 tons. Apart from yield differences, the new varieties diirer from local ones by closing canopy completely (at 10.000 pph) by four months. Apparently they do not degenerate faster than the local varieties stored in the soil past their peak I15 months). The demand for these varieties. which started to be distributed to farmers in small quantities in 1985, is growing rapidly, and coming from far beyond the environs of Ekona and Njombe. It is highly probable that within another five to ten years they will be found in the fields of most farmers in the majority of the villages of the coastal lowlands. Let us speculate on what this means for Fako farming systems. First, if farmers were simply to replace their present cassava with the new varieties, they would harvest a lot more cassava. This would increase processing time which might be acceptable in the first season but most farmers already have major processing tasks with cocoa or coffee in August to November. Thus they would probably reduce the area planted to cassava that season. Alternatively, an increase in cassava production might provide a boost to the introduction of processing technology, if demand increases correspondingly. Second, important secondary crops might be eliminated by the new cassava. The 1987 TLU on-farm trials have indicated that the maize yields of the new CMS8501 Cl ES are not afrected by type of cassava, whether local or any of the three new varieties (at 10.000 pph cassava and 30.000 pph maize), but this maize variety is somewhat earlier than most local ones. Egusi melon grows until the fourth month after planting, and may be shaded out, as may the groundnuts usually planted with second-season cassava. North of Fako Division, in Meme, some farmers plant cassava several weeks after the establishment of the companion crops, but it is heavily shaded until their harvest, and yields must be aflected. Melon and groundnut are minor but are high-value and nutritive crops important to farmers, and will have to be put into separate fields. If the new lowlands maize initiative does well, there may be more monocropped cassava fields (or cassava intercropped with cocoyaml and separate plantain-maize-egusi/groundnut fields in the Tiko Plain. This in turn would indicate a reduction in the cassava area and, probably, reduced fallows for the maize fields. (Fallows near TP villages are already low. In the LV, field separation already exist, but the soil is more fertile). Third, there will undoubtedly be more cassava produced in Fako Division. Where will it go? Douala is often mentioned, but no one has studied the market potential for cassava flour or other storable products in Douala and other major urban centers of Cameroon. Already, 81 percent of cassava farmers sell an average of half their c r o p 6 5 percent as raw tubers and 30 percent as gari-at local markets. Douala truckers come to buy up plantains, cocoyams ar~d green ma* in season, but not cassava as yet. The Cameroon Development Corporation feeds its huge body of workers on rice, maize and beans, because these crops store better in their warehouses. Is there a market? Conclusion Cassava is secondary in importance in Fako Division to plantains and cocoyams, but both the latter face serious production constraints that are increasing their scarcity. IRA is now widely introducing varieties that could multiply local cassava production two to three times. Farmers are likely to respond by reducing the area planted to cassava [especially if processing technology is not improved), separating cassava from seasonal intercrop fields. and increasing overall cassava production. A market will have to be created outside the farming villages, either among the wage-workers of the Division, in Douala. or abroad. Appendk Tables on average household production by crop in Fako Assumptions for Tables: Plantains: 15 &/bunch. 100 CFAF/kg. 73% edible matter Cocoyams: 27 kg/&. 185 CFAF/kg Cassava: 54 @/bag. 50 CFAF/kg (as tuber) Taro: 25 kg/bag, 140 CFAF/kg Maize: 21 kg (dried shelled) per bag of cobs. 195 CFAF per kg (dried shelled) if sold green Yams: 54 kglbag. 200 CFAF/kg Groundnuts: 15 &/bag (unshelled), 340 CFAF/kg; 73% edible matter. The bag used is the 50-kg N-P-K 20-10-10 fertilizer bag Table 1. Production (in kg) of seven crops per Fako fanning household inl986-86 Crop Zone Fako LV Plantains 7 835 9 130 5 835 5 515 845 Cocoyams 1 585 1 725 270 5 690 265 Cassava 5 585 4345 12 615 0 2 265 Taro 815 1110 60 480 200 Maize 485 650 795 135 60 Yams 165 90 85 1 550 115 Groundnuts 30 35 15 0 10 Table 2. Production (in '000 CFAF) of seven crops per Fako farming household in1985-86 Crop Zone - Fako LV TP W WC Plantains 784 913 584 552 85 Cocoyams 293 319 50 1053 49 Cassava 279 217 63 1 0 133 Taro 114 155 5 67 28 Maize 95 127 155 26 12 Yams 33 18 17 310 23 Groundnuts 10 12 5 0 3 Table 3. Production [in '000 Kcall of seven crops per Fako farming household in1985-86l Crop Zone Fako LV TP W WC Plantains 7 320 8 530 5 450 5 155 790 Cocoyams 2 110 2 295 360 7 570 350 Cassava 8 545 6650 19300 0 4 080 Taro 920 1255 70 540 225 Maize 1 760 2 360 2 885 490 220 Yams 170 95 90 1600 120 Groundnuts 125 150 65 0 40 Note: 1 Assuming insi@rance of minor crops Table 4. Relative contribution of seven crops to total lose income in Fako in 1985-86l Crop Zone Fako LV TP UV WC (%I (%I ( O h ) ( O h 1 ( O h ) Plantains 49 52 40 27 26 Cocoyams 18 18 3 52 15 Cassava 17 12 44 0 40 Taro 7 9 1 3 8 Maize 6 7 11 1 4 Yams 2 1 1 15 7 Groundnuts 1 1 - 0 1 Note: 1 Assuming insignificance of minor aps Table 5. Relative contribution of seven crops to total lose energy in Fako in 1985-813~ Crop Zone Fako LV TP UV WC ( O h ) (%I (%I ( O h ) (%I Plantains 35 40 19 34 14 Cocoyams 10 11 1 49 6 Cassava 41 3 1 68 0 70 Taro 4 6 - 4 4 Maize 8 11 10 3 4 Yams 1 1 - 10 2 Groundnuts 1 1 - 0 1 Note: 1 Conversion data fmm Westphal et al. 1985. Cultures m k r e s troplcales avec dfkrence spfflale au Camemun, pp. 96, 170, 302,422. W-n, the Netherlands, Pudoc. DIAGNOSTIC SURVEY OF CASSAVA-BASED CROPPING SYSTEMS lN ?WO ECOLOGICAL ZONES OF BAS-ZAIRE O A Osiname, C. Bartlett. N. Mbulu, L. S h b f ~ and K. Landu Abstract Cassava-based cropping systems were studied in separate exploratory surveys of two ecological zones. the forest over deep sand (forest arenoferrals) of Kasangulu zone and the savanna over clayey ferrisols around MVuazi, in Bas-Zaire. In both regions the cropping system consists of one year of intercropped cassava followed by a fallow. The main association crops are maize. melons and cowpeas in the forest zone. and groundnuts and pigeonpeas in the savanna zone. In both regions the optimum planting dates of association crops are obse~ved with greater respect than are those of cassava, which is planted as long as there is enough soil moisture for sprouting. The principal agronomic constraints in both regions are low soil fertility, unimproved variety. and disease. The intervention most desired by farmers is a high-yielding. erect, drought- and disease- tolerant cassava variety with tubers that retain their marketability for 18-24 months. Cassava is Zaire's most important food crop. Its tubers furnish about 60 percent of the calorie needs of over 70 percent of the Zairean population. The leaves of cassava. which are consumed as 'pondu." are the most popular leafy vegetable and a good source of protein in the diet. The regions of Bandundu and Bas-Zaire are the main suppliers of cassava products-'cassette." 'pondu" and 'chikwanque" to the urban population of Kinshasa. Although the average tuber yield per hectare is slightly higher in Bandundu (7.0 tons) than in Bas-Zaire (5.0 tons), market preference is for Bas-Zaire products, which are whiter. The brownish tint to Bandundu cossettes has been traced to the color of water in the streams in which the tubers are soaked. The proximity of Bas-Zaire to Kinshasa also gives the region a greater share of the pondu market in the capital city. Description of Survey Areas Two sample areas-Kasangulu forest zone and M'vuazi wooded savanna zone-were selected for this exercise. Kasangulu forest zone The area sumeyed fell between 4'35' and 5"7'S latitude and 15"3' and 15"15' E longitude. The altitude is between 450 and 650x11. The terrain is very hilly with V-shaped valleys and slopes sometimes attaining a 40- percent gradient. The villages are located on hill crests; the slopes are cultivated. The valleys are generally too narrow for extensive cultivation. The vegetation is essentially secondary forests of varying ages. depending on the periods they have remained in fallow. About 30 percent of the species in these forests were, however, obsenred to be leguminous. Important leguminous species are Sapiuin cornuturn. Pentaclethra cetueloleama and Millettia lanrenthf. The soils in the area have been described as forest arenoferrals on Kalahari sand (Sys 1972). Depending on the age of the forest fallow. the soils consist of 3-8cm of humified gray-brown loose sand at the surface, and very loose grayish sand subsoil reaching a depth of 150cm in places. This layer is underlain by another very deep fine yellowish-brown sand, equally structureless. The soils are extensively uniform. The soil pH is around 4.0, organic carbon 0.90 percent. and exchangeable Ca, K, and Mg are 1.0, 0.2 and 0.8 meq/ lOOg respectively. They are excessively drained with poor capacity for both nutrient and moisture retention. The sole means of fertility regeneration are the long (10-20 years] forest fallows. The climate is typical "Aw" of Bas-Zaire. The dry season, which has an approximate duration of 130 days, begins generally between 20 and 25 May and lasts untill around September 30. The rainy season is bimodal, with maxima in April and November. There may be a short gap in the rains between January and February. The average annual precipitation is about 1500mm. The average daily air temperature is 25.5'C in the rainy season, with a maximum of 37°C. In the dry season, the average air temperature is 23.5"C with a maximum of 27°C. The mean monthly insolation is between 25 and 60 percent. M'Vuazi ~uoaled savanna zone An area of approximately 25km radius was surveyed, with Mlruazi a s center. The area lies between 5"27' and 5"45'S latitude and 14"45' and 15"lO'E longitude. The altitude ranges from 470m to 750m. The area is part of the formation on Calcitic Schist System. The relief is strongly influenced by the nature of the prevailing rock in the substratum. In the more calcareous formations the valleys are V-shaped and dry. In the regions dominated by schists, the relief is very steep with abrupt slopes of concave shapes, signifying the greater resistance of the schists to weathering. The plateaus and valleys are more extensive and highly cultivated than the Kasangulu forest region. The nonconcave and less steep slopes are equally cultivated.The vegetation in this area consists of herbaceous fallows composed mainly of Irnperata cylindrica, Panicum maximum and Pennisetum purpureum on the plateaus and slopes. These grasses are often associated with Phaseolus lunatus and Mucuna pruriens. Hyperrhenia confinis, occupies regions that are sandy to sandy clay in texture. The alluviums in the valleys that are only periodically inundated are occupied mostly by Andropogon gabonensis. Pennisetum purpureum and Hyperrhenia confmis. These grasses often attain heights of up to 3-4 meters. Sys 11972) classed the soils in the area into two groups: 1. The ferrisols (Paleudults), derived from residues from the alteration of calcareous and schist rocks on the plateaus and slopes. These materials are heavy clays, but are porous. The structure is subangular with excellent water retention capacity. 2. The Alluviums, derived from altered materials transported and deposited by rivers. These alluvial materials are of variable depths. clayey to sandy clay in texture, with good structure, permeability, and aeration. (See also Denisoff and Dawed 1954.) Both groups of soils are often humic to about 30cm in the profile, and show acid conditions: the pH varies from 4.6 at the surface to about 5.1 in the lower horizons. Exchangeable bases are very low: Ca, 1.0 meq/ 100 g K. 0.17 meq/ l00g and Mg 0.8 meq/ 100g. Exchangeable Al is generally about 2.0 meq/ 100 g. Land Preparation Land preparation for cassava and associated crops begins in July. In the forest zone, the secondary forest earmarked for the year's cropping is slashed and allowed to dry through late September before being set on fie. Most of the month of October is spent stacking the wood either for charcoal or for direct sale as fuel in Kinshasa. and for construction of heaps. As most of the terrain cultivated is very hilly. the farmers start work from the lower part of the farm and work uphill. During heap construction, the fine and medium roots around the heaps are removed. According to the farmers, these roots can damage cassava tubers if they are left within the heaps. In the savanna zone, land preparation may start with burning of the grassland, followed by the construction of ridges which on plateaus and slopes are generally constructed along the slope irrespective of the gradient. Sometimes unburnt grasses are lined up and buried under the ridges. The ridges are commonly about 1.5 to 2m apart and may be as wide as 60-70cm at the top. A more ingenious method of land preparation is the 'Mafuku" Dried grass. cut at the base with some root mat left with soil adhering, is arranged in heaps 1 to 1.5m apart, about 0.75 to 1.0m high, and 70 to 90cm in diameter. The soil between the grass heaps is then loosened by hand hoe and piled over the grass heaps until the vegetation is part~ally buried but with sufficient air space left to permit slow burning. The mixture of ashes, soil, and partially-burned debris is left until the start of the rains. In some villages, after two or three showers. ridges are constructed in the adjoining space between the Mafuku heaps fornling continuous ridges with the heaps. For further description of Mafukrl land preparation system, see %urnah and Okigbo (1980). Cassava Planting System The cassava planting system is remarkably similar in the six communities surveyed in the forest zone. This similarity is mainly a reflection of the generally uniform soil/vegetation types in the area and method of land preparation. The planting system in the savanna zone, however, offers slight diversity depending on the land preparation method, whether Mafuku or ridges. The main elements in the cassava production systems in the two zones are planting time, stake size, varieties, stake planting (depth, arrangement, population), crop association, weeding, and maintenance of soil fertility. Planting time Cassava planting in Bas-Zaire generally begins in mid-October and continues until the end of the rains in May. If the rains extend beyond May, some farmers in the forest zone will continue planting. All farmers interviewed were aware that October/November plantings gave the best yields, and plantings tend to be concentrated within these two months. One main reason for stretching planting throughout the rainy season seems to be the need for a regular supply of tubers and pondu throughout the year. In the MVuazi savanna zone, farmers seem to observe a break in planting in January and February. the short dry period. Whereas in the forest region farmers clear more land than they can plant a t one time. leaving space for continuous planting, farmers in the savanna have to go through the process of land preparation again for the second season. Planting begins again in March and continues through the end of the rains in May. Varieties Popular varieties being grown in each region appear to be those that have survived the harsh processes of natural selection. In the forest region the common varieties observed in the fields were Mpelo-Longi, Kidombi, Nsubakani and 'six mois". Visual observations on the farms show that these varieties are fairly tolerant to drought. a problem that can be severe in the deep arenoferrals. In the savanna region the common varieties are Mpelo-Longi. Leni. Dinkondo. Mapuata and Mboaki. Considering the soil conditions in the area, these varieties are most likely to be tolerant to high A1 concentration in soil solution, and also perform well under low soil fertility conditions. As many as two to five different cassava varieties were observed per field. The greater number of varieties are more likely to be planted per field in the forest zone, where the unit farm size is larger and planting is spread over a longer period than in the savanna zone. Farmers' reasons for planting more than one variety include: 1. The desire to maintain a mixture of early and late maturing varieties to assure a steady supply of tubers for family consumption and a s a source of cash. 2. The need for a small quantity of sweet cassava for home con- surnption. It was noted that farmers refrain from planting large areas of sweet cassava varieties because they are more oiten stolen. Although there was no preference for any particular variety for early [season A) or late (season B) planting. the farmers in both regions agree on what qualities constitute a good cassava variety: high yields, erect stem. early maturation [but with tubers that retain their quality when the harvest period is prolonged). late flowering. and good pondu and fufu quality. Stake size and stake plnntlng In the forest zone stakes are generally between 15 and 20cm long. In the savanna zone. they are a little longer, ranging from 20 to 40cm. The shorter stake length in the forest zone may be related to the shortage of planting materials common at the onset of the rains, which is the result of loss of planting materials from dry season harvests. In the savanna there is a relationship between stake length and method of plantation. Stake planting offers some interesting points of comparison between forest and savanna. In the forest zone. three or four stakes are planted on the same side of the heap. The stakes are buried completely in horizontal positions 5-10cm into the soil. In the savanna zone stake planting varies with methods of land preparation. Where Mafuku is practiced, as many as six stakes may be inserted on the periphery of the Mafuku mound. The center of the Mafuku itself is reserved for the associated crops. Where the Mafuku mounds are joined by ridging, the stakes are planted on the ridges between the mounds. 0 1 1 the ridges, the stakes are planted about 50cm apart in double rows 60.70cm apart. The stakes, which vary in length between 25 and 40cm, are planted in slanting positions leaving at least 2-3 nodes exposed above ground. Given the variations in number of stakes per hill and spacings used, it was observed that the stake populations in farmers' fields are generally between 15,000 and 20.000 per hectare. After taking into account poor stake sprouting, the actual plant population is nearer 15.000/ha. Crop association The main crops grown in association with cassava in the forest zone are. in order of importance, maize, cowpea, and melon. Occasional stands of water yams, sweet potatoes, or tomatoes may be inserted. Maize is planted in October and late February. When wood clearing and heap construction are completed on time. maize is planted immediately at the base of the heaps. Cassava follows after some days or weeks. Where wood clearing is delayed, maize is planted in late October 1-1.5m apart. 2-4 seeds/pocket, among the partly burned wood. When clearing is completed, the heaps are constructed between the maize stands and stakes introduced. Maize population barely exceeds 10,00O/ha. Melons and cowpeas are more frequently associated with the cassava planted in February/March. Crop associations with cassava are more intensive in the savanna zone than in the forest zone. On the plateaus and slopes, the main food crops in association with cassava are groundnuts. beans, sweet potatoes, and pigeonpea. Where Mafuku is practiced, beans, tomatoes and amaranthus may be planted on the Mafuku mound. Groundnuts are planted about 30cm apart on the ridges between the two rows of cassava stakes. There are usually about three rows of groundnuts on each ridge. The spacing for beans associated with cassava is similar to that of groundnuts. There is no specific or planting pattern for pigeonpeas, although south of Mlruazi the population may be about the same as cassava. Groundnuts and pigeonpeas are usually planted at about the same time a s cassava. Although some farmers intercrop groundnuts with March-planted cassava, most season-B cassava fields are monocropped. Weeds pose a greater problem to cassava in the savanna than in the forest zone. The grass weeds. mostly Imperata and Hyperrhenia, grow rapidly after land preparation, and the first weeding for cassava/groundnut association is needed three to four weeks after planting. A second weeding is done about eight weeks after planting, and a third after the harvest of the groundnuts. If need be, another weeding may be done at the beginning of the dry season. In the forest zone. because the long forest fallow generally suppresses all fast-growing grasses, no weeding is needed until after the maize halvest in April/May. A second weeding may be done during the dry season. Maintenance of soil fertility The main crop rotation in both zones is one crop of cassava followed by a forest or grass fallow. The forest fallow may last 15-20 years. One variation in the forest fallow systems was observed in some villages: the crop planted after the 15-20 year fallow is followed by a short (3-5 years) secondary forest fallow, then a second crop, and finally the long fallow again. In the savanna zone, the grass fallow following the cassava crop, lasts 3-5 years. If Mafuku is to be repeated the old burned spots are avoided. Where ridges are used the new ones are constructed in the furrows of the old. sometimes burying the grass stubble. Agronomic constraints of production The agronomic constraints listed by the farmers and the changes desired are very similar in the two zones sumeyed and are relevant to all the crops grown in both zones. The nature of these constraints, however, varies. The major agronomic constraints are poor soil status, crop varieties, crop management, crop disorders, and weeds (table 1). REFERENCES Denisoff, I. and R Devred. 1954. Carte des sols et de la vegetation du Congo Belge et Ruanda-Umndi. Bruxelles. Ezumah. H.C. and B.N. Okigbo. 1980. Cassava planting systems in Africa. In Cassava cultural practices. Proceedings Workshop on Cassava Cultural Practices. Bahai. Brazil 1980. Ottawa: IDRC. Sys, C. 1972. Caract6risation morphologique et physico-chimique de profiles de I'Afrlque Centrale. Publications de 1'Institut National pour 1'Etude Agronomique du Congo (INEAC). Table 1. Agronomic constraints of cassavrr-based crop production in Bas-%ire Nature of constraint Forest zone Savanna zone - soi l Low fertility, very sandy. acid, poor moisture retention. Supports only one crop of cassava/ maize before revert- ing to fallow. Crop varieties Low-yielding and susceptible to diseases and pests Crop management Lack of respect for optimum planting dates for crops. Im- proper plant density in crop association Crop disorders Premature death of plants caused by CBB and anthrac- nose cause low tuber and pondu yields Weeds Invasion of Chrornolina Low fertility, high exch. Ai. low exch. Ca. K & Mg. Rob- lem of empty kernels in groundnuts. Can support only one crop of cassava/ groundnuts before reverting to fallow Low-yielding and susceptible to diseases and pests Same as in the forest zone Same as in the forest zone lmperata and other grass weeds required frequent Change desired by farmers A mem., of sust,?mulg soil ferblity so that more than one crop of cassava can be grown after each land clearing. High yielding, disease re- sistant varleties of cassava. maize and groundnuts Iriforination on both co111- patibility and optimum crop densities fn cassava associations Varieties resiskint or tolerant to pests and diseases lmproved cultural methods for weed cdorata slows weedingsfor crops.- control other Ulan down forest Mimosa, orianally the use of a hand-hoe regeneration introduced & a falow crop, now poses a serious land-clearing problem PERFORMANCE OF IMPROVED UTA CASSAVA. MANIHOT ESCULENTA CRANTZ. AT FARM LEVEL F.L Nweke, H.C. Ezumah and D.S.C. Spencer To assess the performance of IITA's improved cassava varieties, TMS 30572. TMS 3021 1, and TMS 30555, a survey was conducted in February 1987 in a predominantly cassava-producing area in Nigeria where such improved varieties were observed to be widely grown. The area. Ohosu in Bendel State, lies 6' 25'N latitude and 5"30'E longitude and within the tropical rainforest vegetation zone. Mean annual rainfall is about 2,000mm and sou type is "Acid Sands". Multiple cropping and land fallow are the main features of the cropping pattern. The major intercropping associations with cassava are intercropping with maize and vegetables: intercropping with plan- tains, maize, and vegetables; and intercropping with trees, plantains. maize and vegetables. The introduction of the improved cassava vari- eties has not induced change in the cropping pattern. Yield Attributes of the Improved Cassava Varieties The improved varieties harvested at 12 months yielded 75 percent higher root weight than local varieties. This difference is statistically significant at 1 percent level using the two-tailed 't' test. Since number of plants, shoot fresh weight, and number of roots per hectare are not significantly different between the improved and the local varieties (table I), the dmerence in root yield is attributed to higher bulking capacity by the improved varieties (Hahn 1979). This is reflected in higher root size (by 38 percent), higher harvest index (by 29 percent), higher total biomass (by 32 percent), and higher average root weight per plant (by 101 percent) obtained from the improved than from the local varieties (table 1). Root yield from the improved and from the local varieties were regressed with average yield of the two (the environmental index, el using the model of Eberhart and Russell (1966) and Hildebrand (1984). This analysis, as well as the observed yield distribution frequencies (figure 1). confirm that the improved varieties have higher potential for tuber yield and may consistently outyield the local varieties at the farmers' level of management. The percentage difference in yield of improved over local varieties however decllned by 77.75, 56 and 32 percent at 9. 12. 16 and 18 months respectively. The performance of the improved varieties in terms of root yield is. however. below their potential, based on breeders' expectations (Hahn 1979). The reasons would include suboptimal plant population and production without chemical fertilizers. Nutritional effects of cassavn consumption Infant mortality. observed at the rate of 22 percent of the total sample, occurred in 65 percent of the households. Kwashiorkor symptoms were observed in 85 percent of the children under five years of age. In April, the kwashiorkor symptoms had virtually disappeared. February is the peak of the dry season; vegetables are scarce, and women are very busy hamesting and processing cassava. By April the rains have set in, vegetables have appeared in the fields, and harvesting and processing of cassava has eased off. Economics of use of the improved varieties The total cost of gari production is about 45 percent higher per hectare and 20 percent lower per ton under production with improved than with local varieties. Higher cost per hectare is a result of higher harvesting and processing costs associated with higher yields. while lower cost per ton is because of the higher yields under production with improved compared with local varieties. Net revenue is about 85 percent higher per hectare and about 75 percent higher per ton of gari under production with improved varieties than with local varieties. Production expansion is constrained by shortage of labor and perhaps by aggregate demand limitations. The probability distributions of net revenues and gross margins per ton for production with both the improved and the local varieties (figure 2) are positively skewed, suggesting that under production with either the improved or the local varieties the probabilities of generating positive net revenues are higher than of generating negative. However, the probability of generating positive net revenues is higher under production with improved (75 percent) than with local (54 percent) varieties. In addition, production with improved varieties has lower probabilities of generating high negative net revenues than do local varieties. This is because of lower yield under production with local than with improved varieties and because of the additional negative effects of possible incidence of pests. especially cassava mealybug (CMB). under production with local varieties. The probability of generating a negative gross margin is zero under production with improved varieties and very low under production with local varieties. Areas for further research Further research in the following areas would be likely to lead to fuller realization of the potentials of the improved varieties. IITA's nitrogen management technology needs to be tested in researcher-managed adaptive research to assess its potentials in replenishing the soil nutrients in the cassava-based cropping system. Estimates of marketing margins from cassava products at the processor, transporter, wholesaler and retailer levels are essential to complement this study, which has provided estimates of cassava income at production level. Knowledge of cassava income at these other levels, in case of expanded production based on wide adoption of improved technologies. will help predict the potential effects in terms of income redistribution of the new technologies and incentives to farmers to adopt them. A nutrition study would be necessary to determine the cause, extent, and seasonal nature of the nutrition problem in the area. On-farm adaptive study would be necessary to determine the feasibility of introduction into the farming system of grain legumes, which would be available in the dry season when vegetables are not, and to evaluate the potentials of various IITA cowpea and soybean varieties in the area. Research in mechanization of more of the gari processing operations would. if successful, allow women of the area more time to attend to the nutritional needs of their children. REFERENCES Eberharl. S.A., and W.A. Russell. 1966. Stability parameters for com- paring varieties. Crop Science 6: 36-40. Hahn. S.K.. E.R Terry, K. Leuschner. 1.0. Akobundu. C. Okah and R Lal. 1979. Cassava improvement in Africa. Field Crops Research 2: 193-226. Hildebrand, P.E. 1984. Modified stability analysis of farmer-managed on-farm trials. Agron. J. 76: 271-274. Table 1. Yield of improved and local varieties of cassava from 18 fields surveyed in the Ohosu area. Bendel State of Nigeria. 1987. Yield Improved Local Y ' value for parameter comparison between variety Mean SE Mean SE means Fresh cassava tons/ha 19.6 +3.95 11.2 +3.19 No. of plants UMO)/ha 7.0 +1.70 7.8 +1.25 NS Shoot fresh wt. (tons/ha) 21.9 +8.72 19.3 +7.41 NS No. of roots ('000)/ha 26.0 +9.30 27.8 +7.91 NS Average root Size (kg) 0.77 +0.14 0.56 +0.21 Harvest index 0.49 4.079 0.38 +0.009 Total biomass (tons/ha) 41.0 +12.26 31.2 +10.36 Average root wt. (kg)/plant 2.94 4.83 1.46 4.36 Potential yielda (tons/ha) 29.4 - 14.6 - Source: Field survey. Note: a Potential yleld = leld at recommended population of 10.000/ha, ie, observed kglplant x 10.000 - 1OOO tom,k Figure 1. Root yield response at 12 months of improved and local cassava varieties obtained by farmer. in Ohosu area. Bcndel State, Nigeria. 1987 0 4 8 12 6 20 24 28 ENVIRONMENTAL INDEX, e tans 2. Frequency distribution of root yield (t/ha) obtained by sampled farmera in Ohosu area. Bendel State. Nigerlp. 1987 t Eu L O C A L "= I8 TMS 305 t 2 n = I8 ' I0 10.1-149 15.0~19.9 20 0-24.9 >25 CLASS ( t / h a l Figure 3. Root yields of improved and local cassava varieties at different ages obtained from sample flelds in Ohosu area. Bendel State. Nigeria. 1887 0 0 9 12 5 I8 27 40 AGE IN MONTHS Figure 4. Robability distributions of net revenue and gross margin under production with improved and with local cassava varieties in Ohosu area' M LOCAL VARIETIES - IMPROVED VARIETIES 0 REVENUE i # I M T ) 8 8 s~uwdp$lmd fiq sjuauulor, Jo tiiwul~~~ls w fiq pamolloja~w uwpas qq~ u! s.ladwd THE PERFORMANCE OF CASSAVA WITH OTHER STAPLES IN INTERCROPS IN CAMEROON T.J. Ambe, S.N. Lyonga, A.A. Agboola and S.K. Hahn Abstract Studies of land productivity in a cassava-based cropping system showed that cassava fresh storage root yield was depressed in association with cocoyams (Xanthosoma sagitt$oIium), sweet potatoes (Ipornoea batatas) and maize (Zea mays) with spatial arrangements of crop stands in the field indicating inter- and intra-specific competition for growth resources. However, the yields of the other crops were also depressed. Ten thousand cassava stands intercropped with 20.000 maize stands and with 9.000 cocoyam stands per hectare respectively gave higher returns per unit area of land than when planted as sole crops. This shows profit realization in intercropping but the correct choice of crop component combinations and optimum populations in association are imperative. A cassava/sweet potato intercrop did not seem profitable with an income equivalent ratio (IEIU of less than one. The major tropical root crops--cassava. yams, cocoyams and sweet potatoes--are widely grown and used mainly as subsistence staples in many parts of the African tropics and subtropics. They are a major source of energy for well over 200 million people in the continent (FA0 1979). Their leaves (except for those of yarns) are often used as vegetables, providing proteins, vitamins and minerals. These root crops are also grown for industrial raw materials and as livestock feed. In Africa, cassava grows from sea level to an altitude of 1.800m (Hahn 1984). If a major food is defined as one providing 50 percent of calorie requirements, it is estimated that cassava could be a major staple food for 420 million people. In most traditional cropping systems, cassava is intercropped with other staples. The other crop components in the intercrop are short-season and early-maturing. They include maize, sweet potatoes. cowpeas and cocoyams. When these are harvested, the cassava is left in the field to mature and hawested later when needed. In Cameroon, peasant farmers grow their cassava in a complex mixed intercropping system. Given the role cassava plays in the nutrition of the people. and. the fact that intercropping is a mainstay in the cropping systems of the peasant farmers, it was necessary to study and quantify the performances of the component crops in the intercrops. to investigate the influence of intercropping cassava with other staples on yields and land productivity and to make recommendations on the appropriate method of crop production for sustained land productivity and increased production. Materials and methods A local white cassava cultivar was intercropped with the Ekona Mixed Color maize variety (EMC], a sweet-potato clone (see table 1). and a local cocoyam cultivar [Xanthasoma sagitt$olium). The land was plowed and harrowed, and the crops were planted on flats. Seven treatments were used: sole cassava, at 10.000 plants per hectare (pph): sole maize at 40,000 pph: sole cocoyams at 10.000 pph: sole sweet potatoes at 30.000 pph: cassava at 10.000 pph plus cocoyams at 9.000 pph: cassava at 10.000 pph plus sweet potatoes at 20,000 pph: and cassava at 10,000 pph plus maize at 20,000 pph (see figure 1). Sole plots of maize and sweet potatoes were cropped for two seasons, ie. March and August each year. Each treatment was planted to four 12-m rows im apart. A randomized complete block ~ e r i m e n t a l design was used with four replications and a plot size of 48 m . The two middle rows were sampled for data collection. Observations were made on plant stand counts at harvest. dry grain. fresh root and tuber yields at harvest. IERs were also extrapolated using local market prices in Cameroon. The experiment was run for three years consecutively. Results The stand counts of crops were not seriously affected because of intercropping at harvest. Cocoyams, sweet potatoes and maize depressed the fresh storage root yield of cassava in association by 15.4 percent. 18.6 percent and 29.5 percent respectively. The total cash return per unit area of land was highest for cassava plus cocoyams followed by cassava plus maize intercrop. These same treatments exhibited the highest IERs of 1.58 and 1.38 respectively. Cassava plus sweet potato gave the lowest cash returns per unit area of land and had the lowest IER of 0.98 (see table 11. Cocoyams were seriously attacked by a fungal disease (Pythium myriotyluml, particularly in the sole cocoyam plots. Table 1. The effect of Intercropping -vs with other 8tnple8 on their mean yieldea per 10 m2 plot Treatment Plant Fresh Cash returns I E R ~ stand storage CFAF count at root, tuber harvest and dly grain weight at per per ha harvest &gl plot '000 Cassava 10.000 plants/ha 9.9 Cocoyam 10.000 plants/ha 8.8 Sweet potato 30.000 plants/ha 22.2 Maize 40,000 plants/ha 36.7 Cassava 10,000 plants/ha 9.9 + Cocoyam 9.000 plants/ha 6.9 Cassava 10.00 plants/ plants/ha 10.0 + Sweet potato 20,000 plants/ha 16.6 Cassava 10.000 plants/ha 10.0 + Maize 20.000 pIants/ha 19.2 Note: a Mean yield data for three y m . b IER = Income equivalent ratio Discussion and conclusion Competition of crops in association does not tend to affect the stand count of crops at hawest. Yields of cassava in association with other crops were depressed indicating that competition exists among the crops for the same available resources. This is seen when the yields of the sole plots of cassava are compared with the yields of cassava in association with other crops. This depression may not necessarily mean unprofitability. The yields of the other component crops in association may more than compensate for the yield depression of cassava. The profitability of this can be determined by the use of land equivalent ratio LER calculations. The LER can only be used if the sole plots and the intercropped plots received the same management level inputs and had the same crop densities per unit area of land. In this case. crop densities varied and the only way to determine profitability was to convert the total yield per plot to one unitary factor. Local market prices were used to convert the yields into monetary values and IERs were calculated. On this basis, cassava-maize and cassava-cocoyam intercropping gave higher income returns per unit area of land than sole cropping. It would appear that when these crops are in a mixture, they exploit the available resources differently, thereby reducing the pressure of competition. There is no doubt that there is some competition among the crops in association but so long as the IER is greater than one, profits can be realized from the combination. Andrews (1972) showed intercropping to be most rewarding when crops make their maximum demands on the environment-soil nutrients, moisture, temperature and light-at different times. This can be seen from the cassava-maize and cassava-cocoyam intercropping in this experiment. The low IER (0.98) for cassava-sweet potato shows that no profit wiII be realized from such intercropping. On the contrary, more income will be realized by the farmer if the component crops are cultivated as sole crops. Since some farmers still intercrop cassava with sweet potatoes, it may not mean that they cannot be compatible but that the correct crop varieties or even population densities or planting sequences have not been used. More investigations are required in these areas. The low yields recorded for the cocoyam emanated from an attack of cocoyam root rot fungal disease (Pythlurn rnyriotylurn). This disease has been responsible for the drop in production of cocoyam in Cameroon. particularly the Xanthosorna species. The sole plots of the cocoyam treatments were more severely attacked than the intercropped plots. Intercropping is advantageous with respect to disease control (Mukiibi 1976). Since the root rot disease of cocoyam is soilbome. cassava intercropped with cocoyam creates barriers between the diseased roots of one plant of cocoyam and another. Although the yields were low. the IER of cassava-cocoyam was greater than one. Karikari (1980) in Ghana showed that a cassava-cocoyam intercrop reduced yields but the LER greater than one (1.2) and was hence profitable. There is no doubt that intercropping is a mainstay in the peasant farming communities. There is an obvious compatibility and complementarity of crops in association. Little information exists on the effects of intra- and inter- specific competition of crops in association. Similarly. there is little information about the effects of soil fertility on the performance of intercrops. More research needs to be done on what happens in the soil when crops are grown in association. Knowledge of this will give guidelines on the choice of crops to be planted in mixtures and to other cultural practices that may go along with it to increase productivity and production. While breeders are breeding species for intercropping, soil and crop agronomists should work together to determine crop species that are compatible and complementary in association for optimum yields. Production inputs such as fertilizers can be tested if the growth resources that are competed for by the two crops are understood. REFERENCES Andrews, D.J. 1972. Intercropping with sorghum in Nigeria. Exp. Agric. 8, 139-150 FAO. 1979. FA0 production yearbook Rome. Hahn. S.K. 1984. Utilization. production constraints and improvement potential of tropical root crops in advancing agricultural production in Africa. In Commonwealth Agricultural Bureau: First ScientiRc Confer- ence. Arusha. Tanzania. 12-18 February, 1984. ed. D.L. Hawksworth. Karlkari. S.K. 1980. Plantain in root-crop farming systems. In Tropical root-crops: production and uses in Africa-Proc. Second Triennial Symposium of the International Society for Tropical Root Crops, Africa Branch, Douala, Cameroon. 14-19August 1983. IDRC. Ottawa. Ont. 1984. 231 PP. Muklibi, J. 1976. Possible relationship between intercropping and plant disease problems in Uganda. In Intercropping in semi-arid areas, ed. Monyo. Ker and Campbell. Report of a symposium held at the Faculty of Agriculture, Forestry and Veterinary Science. University of Dar-es- Salaam, Morogoro, Tanzania. 10-12 May 1976. EFFECTS OF FERTILIZER AND TIME OF INTRODUCING CASSAVA ON THE PERFORMANCE OF YAM-MAIZECASSAVA INTERCROP: 1. EVAI,UATION OF THE BIOLOGICAL YIELDS OF THE COMPONENT CROPS R . P A Unamma. T.O. Ezulike and A Udealor Abstract The trial was conducted in the 1986/87 cropping season on a tropical rainforest acid sandy loam soil. The effects of the time of introducing cassava through yam (Dioscorea rotundata cv. Nwopoko or Abii) intercropped with maize (Zea Mays cv. TZSR-W), and the application of 15-15-15 NPK fertilizer 400 kg/ha, 56 days after planting (dap) yam/maize, were evaluated. Sole Nwopoko yields of 15 t/ha and 15.4 t /ha with and without fertilizer respectively were significantly better than corresponding yields from all the crop mixtures under comparison. Sole Abii that received fertilizer gave the third best yield (11 t/ha). Without fertilizer, the sole Abii yield (5.3 t/ha) was comparable to those of intercropped Nwopoko (3.1 t/ha) in which cassava was introduced at 0 dap. without fertilizer or intercropped Nwopoko (4.7 t/ha) with fertilizer in which cassava was interplanted at 56 dap, respectively. The latter yields were comparable to that of intercropped Abii where cassava was introduced at 28 dap (6.1 t/ha) and fertilizer applied. Without fertilizer. the maize component yield (2 t/ha) was unaffected by the time of introducing cassava. the yam cultivar used and intercropping. Fertilizer application raised the maize yield to more 3 t/ha in both the mfxtures and when grown alone. Sole cassava with fertilizer gave a yield (32 t/ha) that was significantly greater than any of the other treatments under comparison. Without fertilizer, the sole cassava gave a yield (27 t/ha) that was comparable to the Nwopoko carrying mixture in which fertilizer was not applied, or the Abii mixture with fertilizer but with the cassava introduced at the same time as the other two components. Introducing the cassava component at the same time as the other two components favored cassava productivity whether fertilizer was applied or not. Over 80 percent of the farmers in the southeastern agroecological zone of Nigeria invariably grow their crops in mixtures (Okigbo 1978; Unamma et al. 1985). In some parts of this area. for male farmers whose crop mixtures are yam based, yams are the major crop. Women are allowed to put in maize in very light populations (often below 3.000 stands/ha) and to introduce cassava at 56 or more dap. This reasonably well-defined practice is based on farmers' experience over the years of working in a system that focused on the interdependencies of the components of this cropping pattern within the farmers' control. and how these factors interacted with the physical. biological and socioeconomic environment beyond their control. The farmers' practice has stabilized. Any alternative interventions must be technically feasible, economically viable. socioculturally acceptable and superior to their current practice. Farmers' cropping arrangement and spacing used to be haphazard and suboptimal, resulting in yields more than 50 percent below the potential productivities of the mixture (Unamma et al. 1985). Intercropping is a farming practice that has in recent times attracted the attention of agronomists as a means of improving land usage. Intercropping involves studying the farmers' mixed cropping patterns and practices and evolving more scientifically arranged planting patterns that aim at increasing output per unit area as well as giving more money to farmers at the time they need it most. There are plenty of commodities-research results on yams, but they are not used by farmers since invanably they are irrelevant to their conditions and outside their capabilities (Eze 1981: Unamma et al. 1985). These commodity research findings need to be tailored to fit farmers' needs and capabilities. The objective of this experiment was to examine the productivity of an improved cassava and maize variety when intercropped with one or the other of two white yam cultivars. Materials and methods The experiment was conducted in the 1986/87 cropping season and located on acid sandy loam soils at the National Root Crops Research Institute's main research farm in the tropical rainforest zone of Nigeria. The site was plowed, harrowed and formed into ridges lOOcm apart. Yam and maize were planted the same day in all the plots during the middle of April 1986. The introduction of the cassava component was carried out at 0. 28 and 56 dap. Setts of the yams, each weighing about 150g, were planted one sett per hole at lOOcm apart IlO.OOO/ha) along the crests of the ridges. Cassava cuttings (six nodes length), interplanted at three-quarters length and buried at an angle of about 45". were planted l00cm apart (10.000/ha) along the crest of the ridges. alternating with the yams at 50cm distance. Maize, at three seeds per hole, was also planted lO0cm apart but staggered on both sides of the ridges (double row) and between the yam and cassava stands diagonally on opposite sides of the ridges so that after yam, maize followed before cassava and on the diagonally opposite side of the same ridge the maize followed the cassava. The maize was thinned and/or supplied to 2 plants/stand at 14 dap (leafing maize plant densities at 40,00O/ha). while the yam and cassava components were supplied at 2 1 dap. The yam vines were trained at the rate of two opposite stands from two adjacent ridges per stake of about 2.5m above the ground surface. Fertilizer (15-15-15 NPK) was side-drilled along the ridges to all the plots at 800 kg/ha. 56 dap. The cassava component was introduced into the plots containing each of the two white yam cultivars, Nwopoko and AbLi intercropped with rnalze. at 0.28 or 56 dap. The sole components of each of the three crops in the mixture were included a s treatments to facilitate evaluation of LER, for the various combinations at the same lwels. Each of the crop combinations was grown both with and without fertilizer application. Malze was hanrested at 112. yam at 224. and cassava at 336 dap. At hanrest, yields were measured for the crops in the middle five ridges. each 4m long, excluding the peripheral stands, and were converted to a per-hectare basis for meaningful comparisons. The treatments (table 1) were arranged in a 3 x 2 x 2 factorial in a randomized complete block design replicated three times. Duncan's new multiple range test was used to compare the mean values of the treatments. Results Table 1 shows that in general yam yleld increased with the delayed introduction of the cassava component whether fertilizer was applied or not. The best time to introduce the cassava was at 56 dap for maximum yield of the fresh tuber of white yam. With and without fertilizer. the fresh tuber yield of sole Nwopoko was better than its highest yield in the mixture by 55 percent and 53 percent respectively. The corresponding yields of sole Abii were 5 percent more and 10 percent less compared with the highest fresh tuber (10.5 t/ha) in the mixture. Efect on the maize component The maize component (grain at 14 percent moisture content) was not affected by the time of introduction of the cassava. Fertilizer application generally doubled the maize yields whether grown sole or a s a mixture (table 1). Emct on the cassava component The cassava component fresh tuber yield decreased with its delayed introduction into the mixture (table 1). In the Nwopoko-based mtxture, the loss in cassava fresh tuber yield from the highest-yielding plot compared with the sole component's fresh tuber yield was 6 percent. if fertilizer was not applied. With fertilizer, the corresponding yield dHerence was 33 percent in favour of the sole cassava component yield (31.5 t/ha). For the Abii combination with and without fertillzer the corresponding differences between the sole cassava and the highest mix- ture cassava tuber yields were 22 percent and 30 percent, respectively. Discussion The yam crop was the most sensitive component of the mixture. It appeared that planting the yam some 56 days ahead of the cassava ensured that by the time the cassava component became established and competitive. the yam plant had itself already become established and less sensitive to the interference of the cassava and maize. The Nwopoko cultivar appeared to be more sensitive (over 50 percent) to the intercropping than the Abii and less responsive to fertilizer application. The cassava component was less suppressed by intercropping than the yam possibiy because it stayed longer in the field. Thus, it had enough time to recover from the suppressive influences of the make and yam when these were withdrawn from the interplant interactions at 112 and 224 dap respectively. The m&e was least affected since at the early stages of growth of the three crops in the mixture (28 days from emergence of the maize plant), when maize is most sensitive to inter- and intra-plant interferences (Nieto. Brondo and Gonzalez 1968). neither the yam nor the cassava had developed sufficiently to have any interference influences on the maize crop. The trial suggested that for optimum productivity of the crops grown in the mixture, the cassava component should generally be introduced at 56 dap when all three crop yields had the fewest reductions simultaneously. Similarly. if maximization of yam production is the goal while intercropping is still desired, the cassava component should be introduced at 56 dap. For maximum yield from the cassava, the three crops should be planted at the same time. REFERENCES Eze. N.O.A. 1981. NAFPP cassava/maize production recommendations. A socio-economic survey of adopters in four states of Nigeria. Paper presented at 5th NAFPP National Cassava Workshop, Umudike. Imo State. 9-13 February 1981. Nieto, J.H. 1938. M A Brondo and J.T Gonzalez. 1968. Critical periods of the crop growth cycle for competition from weeds. PANS [c) 14: 159- 166. Okigbo. B.N. 1978. Cropping systems and related research in Africa. AASA Occasional Publication Series OT1. 8 lpp. Unamma. R P A , S.O. Odurukwe. H.E. Okereke. L.S.O. Ene and 0.0 Okoli. 1985. Farming systems In Nigeria: report of the bench-mark survey of the farming systems of the eastern agricultural zone of Nigeria. Agric. Ext. Res. Liaison Services. National Root Crops Research Institute, Umudike. Umuahia. Nigeria. Unamma, R.P.A.. A. Udealor and F.O. Anuebunwa. 1987. Effects of fertilizer and time of introducing cassava on the performance of yam- maize-cassava intercrop: 2. Land use maximization and monetary yield performance of yam-maize-cassava mixture. Cassava-Based Systems Research Collaborative Group (CBSRC) report: this volume. Table 1. Effects of fertilizer and time of introducing cassava on the biological yields of yam/maiza/cassava intercropping Crop Time of Ferti- Crop yields combination intro- lizer (t/ha) ducing appli- cassava cation (days rate Yam MAaize Cassava after (t/ha) planting) (dad 1 Nwopoko-maize-cassava 0 0 3.lf 1.91cd 25.0bc 2 Nwopoko-maize-cassava 28 0 6.1bc 1.92cd 13.3g 3 Nwopoko-maize-cassava 56 0 4.5cdef 1.94cd 11.0h 4 Nwopoko-maize-cassava 0 800 2.7f 3.33ab 21.0d 5 Nwopoko-maize-cassava 28 800 4.7cdef 3.09ab 15.0f 6 Nwopoko-maize-cassava 56 800 17.0 3.27ab 18.4e 7 Abii-maize-cassava 0 0 3.0f 0.85e 18.8e 8 Abii-maize-cassava 28 0 3.8def 1.14de 15.0f 9 Abii-maize-cassava 56 0 3.9def 1.91ed 11.0h 10 Abii-maize-cassava 0 800 3.2ef 2 . 5 7 ~ 2 4 . 7 ~ 11 Abii-maize-cassava 28 800 6.lbc 2.77abc 14.2fg 12 Abii-maize-cassava 56 800 10.5b 4.14a 21.ld 13 Nwopoko sole 0 15.0a - - 14 Nwopoko sole 800 15.4a - - 15 Abii sole 0 5 . M - - 16 Abii sole 800 11.0b - - 17 Maize sole 0 - 1.61de - 18 Maize sole 800 - 3.06ab - 19 Cassava sole 0 - - 26.7% 20 Cassava sole 800 - - 31.5a Note: In the same mlumn means followed by similar letters are not si ificantly different according to Duncan's new multiple range test a t 5 percent level of probagity. EFFECTS OF FERTILIZER AND TIME OF MTRODUCING CASSAVA ON THE PERFORMANCE OF YAM-MAIZE-CASSAVA - - - INTERCROP: 2. LAND USE MAXIMIZATION AND MONETARY YIELD PERFORMANCE OF YAM-MAIZECASSAVA MIXTURE R. P. A. Unamma. A. Udealor and F.O. Anuebunwa Abstract The experiment. which began in 1986. evaluated the land equivalent ratios (LEN and monetary yield productivity of intercropped cassava (Manihot esculenta cv. TMS 30572). maize (Zea mays cv. TZSR-W), and white yams (Dioscorea rotundata cvs. Nwopoko or Abii) as influenced by the time of introducing cassava into the mixture and whether fertilizer (15:15:15 NF'K at 80 kg/ha) was applied at 21 days after planting idap) yams and maize or not. The trial was conducted under tropical rainforest acid sandy loam (Umudike) conditions. It was beneficial to introduce the cassava component into the mixture either at the same time as the other two components, or at 28 or 56 days after planting, whether fertilizer was applied or not. The LER for the various combinations ranged from 1.547 (when cassava was intro- duced at the same time as yams and maize and no fertilizer applied) to 2.940 (when cassava was introduced at 28 dap yams and mafie, and fertilizer was used) for the Abii-based mixture. For the Nwopoko-based mixture. the LER ranged from 1.926 (cassava introduced at 28 dap and fertilizer applied) to 2.376 (cassava introduced at same time and fertilizer not applied). Monetarily. the best time to introduce cassava into the Nwopoko- based mixture was 28 dap, when money realized was 23 percent more than with the farmers' common practice of interplanting the cassava component at 56 dap yams and maize and not applying fertilizer, which yielded H5.366. With fertilizer, similar amounts were realized whether the cassava was interplanted at the same time as the yam/maize or at 28 dap, with monetary yields being H6.644 and H6.041 naira when cassava was introduced at 0 and 28 days respectively. (i.e. 24 and 13 percent better than the common practice of introducing cassava at 56 dap yam/maize and not applying fertilizer). The Abii-based mixture showed no signiiicant differences among the monetary yields if fertilizer was not used. But when fertilizer was applied. the yield increased from 39 percent over the farmers' common practice when cassava was interplanted at 0 dap, to 133 percent when it was introduced at 56 dap . . * * * * * * * * * * * * The commonest cropping pattern in the eastern agricultural zone of Nigeria involves yams, maize and cassava in a mixture. Depending on the farmers' goals, preferences and resources, other crops, particularly vegetables such as Telfairia occidentalis (Ugu). may be included. The yam or, less frequently, the maize component is usually planted first or both are planted the same day. The cassava is invariably introduced about eight weeks after the yams are planted. in order, say the farmers. to minimize the suppressive influence of the cassava on the yams (Okigbo 1978; Unamma el al. 1985). The farmers' reasoning for introducing the cassava component at eight weeks after yam sounds logical but lacks scientific proof. Moreover, while this practice, which ties up the land and prevents it being used for the next cropping season. was acceptable where land and labor were not limited, it cannot stand the test now that land is becoming limited and planned. intensive, and sustainable cropping systems are being developed for maximum productivity per unit area of land. Therefore. it is necessary to develop alternative practices that will eliminate the disadvantages inherent in the farmers' practices but that the farmers will accept. Materials and methods The materials and methods were the same as those described in the preceding paper ((Unamrna, Ezulike and Udealorl. The experiment was conducted in the 1986/87 cropping season and located on acid sandy loam soils at the National Root Crops Research Institute's main research farm in the tropical rainforest zone of Nigeria. The site had previously carried a maize-cowpea intercrop and was under siam weed (Chromolaena odoratal and guinea grass (Panicum maximum) fallow for two years. The site was plowed, harrowed, and formed into ridges lOOcm apart. Yams and maize were planted the same day in all the plots during the middle of April 1986. The introduction of the cassava component was carried out at 0. 28 and 56 dap. Setts of the yams. each weighing about 150g. were planted one sett per hole at lOOcm apart (10.000/ha) along the crests of the ridges. Cassava cuttings (six nodes length). interplanted at three-quarters length, buried at an angle of about 45" at the appropriate times and at lOOcm apart [10,00O/ha), were planted along the crest of the ridges alternating with the yams at 50cm distance. Maize. at three seeds per hole, was also planted lOOcm apart but staggered on both sides of the ridges (double row) and between the yam and cassava stands diagonally on opposite sides of the ridges so that after yam, make followed before cassava and on the diagonally opposite side of the same ridge the maize followed the cassava. The maize was thinned and/or supplied to two plants/stand at 14 dap, while ihe yam and cassava components were supplied at 21 dap. The yam vines were trained at the rate of two opposite stands from two adjacent ridges per stake of about 2.5m long above the ground. Fertilizer (15-15-15 NPK) was side-drilled along the ridges to all the plots at 800 kg/ha, 56 dap. The cassava component was introduced into the plots containing each of the two white yam cultivars. Nwopoko and Abii intercropped with maize. at either 0, 28 or 56 dap. The sole components of each of the three crops in the mixture were included as treatments to facilitate evaluation of the LER for the various combinations at the same management level. Each of the crop combinations was grown both with and without fertilizer application. Maize was harvested at 112. yam at 224. and cassava at 336 dap. At harvest, yields were measured for the crops in the middle five ridges. each 4m long excluding the peripheral stands, and were converted to a per-hectare basis for meaningful comparisons. LER was used to evaluate the effects of intercropping on the crop mixture with or without fertilizer application. A common unit was obtained for the fresh tuber yield of yam. cassava and dry maize grain (14 percent moisture content) by converting them to their respective monetary values as at the nearest market near the time of harvest. These values were used to compare the results of the treatments' mean values on a per-hectare basis. The treatments were arranged in a 3 x 2 x 2 factorlal in a randomized complete block design replicated three times. Duncan's new multiple range test was used to compare the mean values of the treatments. Results Land equivalent ratio Table 1 suggests that more land will be saved by intercropping either of the two white yam cultivars (Nwopoko and Abii) with maize and cassava than by any o t h e ~ crop combination. For the Nwopoko cv, it appeared that if fertilizer was not applied, it was better to introduce the cassava at 0 or 28 dap, but if fertilizer was applied, the cassava could go in at 0 or 56 dap. With the AbU cv., application of fertilizer improved the LERs, while the time of introducing the cassava component appeared to be best at 28 or 56 dap whether fertilizer was applied or not (table 1). Monetary yield All the sole components except maize out-yielded their respective mixture components, although if fertilizer was not applied, the Abii cv. underyielded some of its yields in the mixture. With or without fertilizer the Nwopoko cv. grown alone gave yields more than double its yields in any of the mixtures, and a 53 and 55 percent greater monetary yield than its highest mixture yield (B13.990). Apparently, the Nwopoko cv. is more adversely airected by inter- cropping than the Abii cv. Without fertilizer, the productivity of Abii grown alone was poorer than some of those of the mixtures. For maximum monetary yield from the mixture, the cassava component was best introduced at 56 dap either of the two white yams and maize if fertilizer was applied. If fertilizer was not applied, the best time to introduce the cassava appeared to be at 28 dap (table 21. Discussion The results of this experiment tended to confirm that the farmers' practice of introducing the cassava component at 56 dap gave more monetary yields than earlier introduction. Therefore, where land is not limited, and farmers have the resources to clear more area, and where the cassava cannot easily be processed and stored, the practice appears to be acceptable. However, where land is limited, this trial suggests that the cassava component could be introduced at 0, 28 or 56 dap to save more land for other purposes. Although the trial suggests that if the amount of income that flows into the farmers' pockets per unit of land were the farmers' only goal, the crops should be grown sole. The farmers take a number of other factors into consideration, some of which are not easily quantifiable. These include the spread of the harvest offered to the farmer in the absence of effective processing and storage facilities, the insurance offered against failure of any of the crops in the mixture had they been planted sole with all the associated production practices, such as land preparation, weeding and time and labor requirements for fertilizer application. We are collecting economic data to enable us to evaluate the economic viability of the practices tested. The fertilizer rate used was based on earlier trials that recommended 800kg 15- 15- 15 NPK applied at 21 dap for yam-maize- cassava intercrop planted at the same time (Nnoke et al. 1987). However. we also took soil samples at planting and at harvest, and these are being analyzed. At the end of this season we shall be in a position to make a more reliable recommendation about the alternative practices under comparison. Table 1. Land equivalent ratio nr, affected by intercropping. ime of introducing cassva. and fertilizer application to yam/maize/cassava intercrop. Umudike 1988 Crop Time of Fertilizer LER combination introducing application cassava, days rate after planting (kg/hal (dap) yams and maize ~ammiri/maize/cassava Agammiri/maize/cassava Agammiri/maize/cassava Agammiri/maize/cassava Nwopoko Nwopoko Agammiri Agammlri Maize Maize Maize Cassava 56 0 28 56 Sole Table 2. Effect of fertilizer and time of cassava introduction on the gross monetary yield of yam/cassava/rnalze intercrop. Crop Timeof Ferti- Crop yield combination introducing lizer CH/ha) cassava appli- days after cation planting rate Yam Maize Cassava Y+M+C yam/maize (kg/ha) (Yl (MI (C) (dap) Nwopoko.maize / cassava 0 0 1,767 1,242 3,500 6,509 11 28 0 3,477 1,248 1,867 6,592 I? 56 0 2,565 1,261 1,54C 5,366 I ? 0 800 1,539 2,165 2,940 6,644 I t 28 800 2,679 2,009 2,105 6,041 I t 56 800 3,990 1,755 2,576 8,321 Agammiri/maize/ cassava 0 0 1,710 552 2,627 4,889 I t 28 0 2,100 741 2,100 5,007 4 t 56 0 2,223 1,242 1,540 5,005 I t 0 800 1,824 1,671 3,453 6,948 I t 28 800 3,477 1,801 1,983 7,261 ?I 56 800 5,985 2,691 2,959 11,675 Nwopoko Sole 0 8,550 - 8,550 Nwopoko ) I 19 800 8,778 - 8,778 Agammiri 11 0 3,021 - - 3.02 1 Agammiri t t 800 6,270 - 6,270 Maize Maize Cassava Cassava Note: Market prices--yam N650, cassava Ed 140. and maize #400 per tan. REFERENCES Nnoke. F.N.. R.P.A. Unamma, L.S.O. Ene and S.O. Odurukwe. 1987. Optimum rate and time of fertilization for yam-maize-cassava intercrops. In Tropical root crops: root crops and the African food crisis, ed. E.R. Terry. M.O. Akoroda and O.B. Arene. Proc. Third Triennial Symposium of the International Society for Tropical Root Crops. Africa Branch. Owen?. 17-23 August 1986. 197 pp. Okigbo. B.N. 1978. Cropping systems and related research in Africa. AAASA. Occasional Publication Series. OTI. 8 1 pp. Unamma. RP.A. T.O. Ezulike and A. Udealor. 1987. Effects of fertilizer and time of introducing cassava on the performance of yam-maize- cassava intercrop: 1. Evaluation of the biological yields of the component crops. Cassava-Based Systems Research Collaborative Group (CBSRC) report: this volume. Unamma. R P A . S.O. Odumkwe. H.E. Okereke, L.S.O. Ene and 0.0. Okoli. 1985. Farming systems in Nigeria: report of the benchmark survey of the farming systems of the eastern agricultural zone of Nigeria. Agric. E k t . & Res. Liaison Services. National Root Crops Research Institute. Umudike, Umuahia, Nigeria. 141 pp. CASSAVA-BASED CROPPING SYSTEMS AT THE NATIONAL ROOT CROPS RESEARCH INSTITUTE. IGaARIAM SUBSTATION Location Igbariam is about 1 lOkm northwest of Enugu and about 40km northeast of Onitsha in Anambra State. Nigeria. It is on latitude 6"20'N and longitude 6"53'E, at an elevation of 160111 above sea level. Igbariarn is at the transition zone between rainforest and Guinea savanna. The vegetation is best described as derived savanna. With a bimodal rainfall pattern, most of which falls during the early growing season, this location receives about 1,600-1.900mm of rainfall annually. Although the soil type is generally referred to a s ultisol, no detailed soil classification and analysis of this rural location have been made. Physical soil analysis, however, shows the soil to be 18-30 percent clay, 10-12 percent silt, and about 60-66 percent sand. The soil has low base status and low CEC. Cassava-based trials The following cassava-based intercropping trials were started in 1986: (a) Effects of fertilizer and time of cassava introduction and maize on economic yields of intercropped yam, maize and cassava in a humid ultisol. (b) Effects of different maize varieties on the yield of cassava/maize intercrops. (c) Preliminary study on the effects of improved cassava varieties on intercrop yields of associated minor crops. Experiment 1 c Effects of fertilizer and time of cassava introduction on the performance of yarn/rnaize/cassaua intercrops This trial is similar to the one at NRCRI. Umudike and differs from it only in location. Details of materials and methods are similar to those given by Unamma. Ezulike and Udealor in a separate report in this volume (see reference). In 1986, the trial began on 22 April. All yams (150 setts) and maize (TZSR-W) together with the first set of cassava (ODAP) were planted. Subsequent cassava introductions were 28 and 56 days later. Compound fertilizer (15-15-15) at the rate of 80 kg/ha was applied to 50 percent of all plots while the others received no fertilizer. Fertilizer was applied four weeks after initial planting. All other cultural practices were carried out as required. Maize was harvested 120 days after planting and oven-dried to 14 percent moisture content. Yams were harvested seven months after planting, while cassava was harvested 12 months after planting. The second year trial was planted later in the year because of the late arrival of the rains. Only the maize from the 1987 trial has been harvested: it has not yet been analyzed. Table 1. Economic yields of intercropped yam, maize and cassava when cassava was introduced at thne sequences in 1986 Crop production Crop yield (t/ha) Yam Maize Cassava 1. Yam/maix - F + cassava 0 dap 2. Yam/maize - F + cassava 28 dap 3. Yam/maize - F + cassava 56 dap 4. Yam/maiz~ + F+ cassava 0 dap 5. Yam/maizx + F+ cassava 28 dap 6. Yam/maize + F+ cassava 56 dap 7. Sole crop without fertilizer 8. Sole crop with fertilizer LSD 10.05) Not-: dap = days d tc r planting. iF = wiUl (+I or wilhout I-) icrlilizer The economic yields of yams, maize and cassava from the 1986 experiment are summarized in table 1. Although only 150g yam setts were planted, yam yields were generally low. There was no significant difference between intercropped yam grown with fertilizer and grown without fertilizer. However, yam grown sole with ferlilizer yielded better than without fertilizer. Maize grown with fertilizer in both sole and mixed plots gave a higher grain yield than without fertilizer. Variations in time of introduction to the yam/maize mixture produced significant yield differences among cassava roots. Intercropping cassava with yam and maize at the time of planting or 28 days later gave significantly higher cassava yields than introduction of cassava 56 days later, whether fertilizer was applied or not. This first-year result shows clearly that fertilizer is essential for yam/maize/cassava intercrops, especially for high yields from the maize component. Planting cassava at the same time or 28 days after planting yam and maize in a yam/maize/cassava mixture gives the highest cassava root yields. Experiment 1 b: Effects oJ'fertilizer and time of intercropping maize on the perjomance of cassaua/maize This experiment recognizes that cassava/maize intercropping is also widely practiced among small-scale farmers in the humid tropics. It is a productive mixture and has lower input requirements than yam/maize or yam/maize/cassava cropping systems. However, in a trial, maize planted at the same time as cassava reduced cassava root yield by 28 percent. This trial was set up to determine the most appropriate time for the maize component to be intercropped with cassava to produce the least interspecific competition and highest combined yield. This result of the 1986 trial is summarized in table 2. The experiment indicated that the best time to introduce maize to cassava is either at the same time the cassava component is being planted or three weeks later. As was noted earlier, lor any productive cassava-based cropping system. application of fertilizer is essential. Although higher cassava root yield was achieved when maize was delayed six weeks, the system was most productive when maize was introduced at three weeks. Experiment II: Effect of d~zerent maize uarieties on the yields of cassaua/maize intercrops Compatibility in cassava/maize intercrops depends largely on low interspecific competition and the ability of the maize component to transmit sumcient light energy to the lower-canopy cassava. Many of the maize varieties currently being associated with cassava have fluffy leaves which shade the cassava. This trial was set up to evaluate the suitability of seven maize varieties for intercropping with cassava. Table 2. Economic yields of intercropped cassava and maize grown with or without fertilizer when mahe was introduced at three sequences in 1986 Crop combination Maize Cassava root yields (t/ha) grains TMS 30572 U41044 Mean 1. Cassava - F+ maize 0 wap 1.28 10.31 7.06 8.69 2. Cassava - F+ maize 3 wap 0.78 8.66 11.33 9.25 3. Cassava - F+ maize 6 wap 0.63 9.46 9.83 9.65 4. Cassava + F+ maize 0 wap 3.03 11.69 11.61 11.65 5. Cassava + F+ maize 3 wap 3.43 13.63 13.61 13.62 6. Cassava + F+ maize 6 wap 0.79 18.34 15.92 17.13 LSD (0.05) 1.63 4.65 3.98 4.23 Note: wap =weeks after planting Seven maize varieties, including a local check, were planted with cassava TMS 30572 on 24 April 1986 in 5m x 8m plots. The trial had an RCB design with three replicates. Sole crops were planted at optimum populations beside the intercropped plots. It was not possible to repeat the trial in 1987 since most of the seeds presenred from the 1986 trial were killed during oven-drying. The trial will be repeated in 1988. Yields from the 1986 trials are presented in table 3. There are indications that TZESR, an early-maturing maize variety, was most compatible with cassava. However, because of the low grain yield potential of TZESR (3.2 t/ha) compared with hybrid maize 8321-8 (7.71 t/ha) and Population 49 (5.51 t/ha). the higher-yielding maize varieties may be more productive in cassava/maize mixtures than TZESR. This will be verified in 1988. Table 3. Economic yields of intercropped cassava and different maize varieties and percent PAR reaching the cassava component at seven wap in 1986 Crop combination Fresh Maize grain (t/ha) % Par cassava reaching sole inter- cassava cropped 7 WaP 1. Sole cassava 18.98a - - 100 2. Cassava + TZESR-W 16.06ab 3.20 2.80 90.3 3. Cassava + 8425-8 11.54cd 6.00 5.70 69.0 4. Cassava t FERKE 81 1 1.69cd 4.99 4.25 64.3 5. Cassava t Pop 49 13.59bcd 5.51 3.37 82.5 6. Cassava t TZSR-W 11.03d 7.46 4.36 63.7 7. Cassava t 832 1-8 15.84b 7.71 5.24 85.2 8. Cassava + local maize 14.28bc 3.28 1.73 85.0 LSD (0.05) 2.98 2.17 1.68 14.8 Notex wap = weeks after planting Par = photosynthetically active radlatian The cassava cultiva~ used was ?MS 30572. Experiment III: Preliminary study of the effects of improved cassava varieties on the yield of associated minor crops There have been reports from small-scale farmers that most of the improved cassava cultivars recommended to them are too shady for their minor crops. A trial involving two-crop mixtures of cassava and maize, egusi melon, okra, groundnuts and cowpeas was set up in May 1986. Because of the failure of egusi melon (due to late planting) and cowpeas (planted too early). it was decided that the trial should be conducted in two seasons. Minor crops usually planted early should be grown with cassava early in the season and those usually grown late should be intercropped late. This will be done in 1988. It is hoped that the 1988 trial will include studies in plant arrangements and populations. REFERENCE Unamma RPA T.O. Ezulike and A Udealor. 1987. Effect of fertilizer and time of introducing cassava on the performance of yam-maize-cassava intercrop: 1. Evaluation of the Biological yields of the component crops. INTRODUCTION OF CASSAVA THROUGH MAIZE IN A HUMID ENVIRONMENT J.B. Oyedokun, T A Akinlosotu and M.0. Omidiji In the forest zone of southwestern Nigeria. the first planting of cassava (Manihot esculenta) through maize (Zea mays) a t the beginning of the rainy season (14 April 1986) produced significantly higher yields (15.8 t/ha) than the second planting (10.5 t/ha), which was similar to the last planting (9.4 tlha). Yields were adversely affected during critical stages of growth by drought and damage by cane rats and other rodents (37 cassava stands and 20 percent cassava tuber loss). Infestation by cassava mealybug was 19 percent. Fertilizer increased maize yields significantly despite rodent attacks. A dominant simple crop mixture in southwestern Nigeria is cassava (Manihot esculenta) with maize (Zea mays) intercrop, since cassava is tolerant of competition (Omidiji et al. 1983; Ezumah et al. 1986). Cassava is planted by most farmers once maize is established just before it matures, or after it is harvested. This implles that cassava can be planted at any time from the onset to the end of the rains. However, it is necessary to determine the optimum time to introduce cassava through maize for profitable yields with or without fertilizer application. This paper reports an experiment recently conducted with this objective. The experiment was located at Moor Plantation. Ibadan (7'23'N; 3'51'E) in the rainforest zone of southwestern Nigeria. The soils at the research site are Alfisols, ferric luvisols of Ibadan series. These are very dark brown, loamy sand topsoils over brown to strong brown clay loam subsoils with yellowish-brown variegations. Plot size was 8m x 8m with 90cm x 90cm spacing. Three dates (14 AprlI, 5 May and 26 May I9861 of planting two varieties of cassava (local Odongbo and TMS 30572) were factorially combined with and without recommended fertilizer dose. The design was a randomized complete block with four replicates. The fertilizer treatments were: Fo, no fertilizer: and F1. 75kg N. 13. lkg P (30kg P205) and 24.9kg K (30 kg K20) per hectare. N. P and K were applied as 15-15-15 compound fertilizer and urea. by hand along both sides of the ridges at 43 days and 72 days after planting respectively and mixed thoroughly with the soil. The delay in application was due to lack of rainfall during April/May 1986. Plots were hand weeded twice before the cassava canopy had formed. Data collection included precropping soil sampling for chemical analysis. establishment counts, plant growth characteristics, costings of cultural operations. and pests and disease incidence. Results and discussion Early maize grain yields (1 9861 Time of planting cassava or cassava variety did not affect the yields of maize, but fertilizer increased maize yields signiiicantly IP<0.01). 1.93 t/ha. compared with 1.26 t/ha wlthout fertilizer (table 1). Drought in April/May adversely affected maize establishments. Moreover, cane rats (Thryonomys swinderianus) damaged maize plants extensively, reduced yields and increased field variation. Table 2 shows some uniformity in percent damage (transformed data) to maize plants by noxious cane rats, which accounted for a 39.3 percent loss of plants. Cassava tuber yield 11 986/87) Fresh tuber yields of cassava harvested at the end of July 1987 are presented in table 3. The earliest time of planting cassava (14 April 1986) gave the highest yield 115.8 t/ha), which was significantly higher than both 10.5 t /ha for the second planting date and 9.4 t /ha for the third planting date. There were no significant differences between second and third planting dates, cassava varieties or fertilizer treatments. The low yields were partly due to low precipitation during the critical periods of growth. In addition, damage to cassava plants and tubers by cane rats and other rodents accounted for a 36.5 percent loss of plants and a 19.6 percent loss of tuber. while 19 percent of the stands were infested by cassava mealybug (CMB). CMB infestation Occurrence of CMB was influenced by both time of planting cassava and fertilizer treatment (table 4). but the interaction of the two factors was not significant. CMB infestation increased sharply as planting was delayed with the severest occurrence in the last time of planting. Fertilizer application decreased CMB infestation signifi- cantly, though this was not reflected in higher tuber yields. Cane rat damage The level of cane rat damage to cassava plants was influenced significantly only by time of planting (table 5). Cane rat attacks increased as planting was delayed. with the severest incidence in the last time of planting. There was no variation in attacks on cassava tubers by other rodents (table 6). Conclusion This report highlights the importance of site selection for experiments and the necessity for adequate crop protection, especially against vertebrate pests. Yields contained herein could have been at least doubled but for the destructive rodents. REFERENCES Ezumah. H.C.. S.K. Hahn, B.N. Okigbo. andT. Gebremeskel. 1986. Root crops based farming systems research at IITA. Paper presented at Workshop on Farming Systems Research. ICRISAT. India Omidiji. M.O.. J.B. Oyedokun. T.A. Akinlosotu, E.I. Olomu. O.A. Osiname. S.A. Oyeneye, and A.O. Ogunfowora 1983. On-farm adaptive research-Ilugun local government area. Ogun State . Ibadan: FACU. Table 1. Early maize yields (t/ha) under cassava + maize intercrop. IAR&T. 1986 Fertilizer Odongbo IMS 30572 Fo F~ Fo F~ Mean Mean 1.26 1.90 1.27 1.94 1.59 Table 2. Percent damage to maize plants by cane rats Fertilizer Odongbo TMS 30572 Fo F1 Fo F 1 Mean To 46.9 45.5 41.4 45.3 44.8 T1 31.5 34.9 37.4 38.6 35.6 TZ 38.6 42.4 35.2 34.0 37.6 Mean 39.0 40.9 38.0 39.3 39.3 Table 3. Cassava tuber yields (t/ha) under cassava + maize intercrop Fertilizer Odongbo TMS 30572 Fo Fi Fo F1 Mean Tz 8.7 8.8 9.2 10.9 9.4b Mean 11.1 11.9 11.5 13.1 11.9 Table 4. Percent infestation of cassava by cassava mealybug Fertilizer Odongbo TMS 30572 Fo F- 1 Fo F1 Mean To 15.8 11.1 19.0 11.1 14.3b T1 26.0 13.4 18.0 11.3 17.2ab Tz 28.9 19.5 27.2 29.3 26.2a Mean 23.6 14.7 21.4 17.3 19.2 Table 5. Percent damage to cassava by cane rats Fertilizer Odongbo TMS 30572 Fo F1 Fo F1 Mean To 28.6 26.8 37.1 24.4 29.2b TI 34.8 31.0 36.1 49.0 37.7ab T2 47.6 38.4 41.5 42.8 42.6a Mean 37.0 32.1 38.2 38.7 36.5 Table 6 . Percent damage to cassava tuber by rodents Fertilizer Odongbo TMS 30572 Fo F1 Fo *1 Mean To 22.0 20.5 21.7 18.4 20.7 TI 17.8 18.7 18.6 19.8 18.7 Tz 21.9 11.5 23.8 20.1 19.3 Mean 20.6 16.9 21.4 19.4 19.6 ECONOMICS OF FERTILIZER APPLICATION BY DIFFERENT METHODS IN A CASSAVA-MAIZE INTERCROP SYSTEM F.I. Nweke. H.C. Ezumah and M. Agu In 1985 the amount of fertilizers used was equivalent to 0.33kg per ha of arable land and FACU (1986) projected a zero growth rate for fertilizer use in the near future in Nigeria. These data can be explained by the Federal Government not only withdrawing most of its fertilizer subsidy but also devaluing its currency through the introduction of a second-tier foreign-exchange market, thus making fertilizers particularly expensive in Nigeria. Therefore, 'incremental benefits from fertilizers will accrue only through better utilization" (FACU 1986. p. 201. Flink (1982) observed that fertilizer problems extend beyond the correct form and quantity applied to the method of application. Banding is preferred to broadcasting for higher crop yields because nitrogen (N) fertilizers are volatile and if exposed will evaporate, while phosphorus fertilizers are immobile in soil water and therefore require banding to reach plant roots easily (Flink 1982). Consequently. the banding method of fertilizer application is recommended for most crops. However, most smallholders in southeastern Nigeria continue to apply fertilizer by broadcasting and only rarely by banding (Ezeilo 1979). The objective of this study is to determine the rationale behind this practice. The banding method may result not only in higher yields but also in higher labor inputs for fertilizer application and harvesting than broadcasting. It is not clear whether the extra yield justifies the extra labor input. Cassava-maize intercropping is a popular crop combination in the acid soil (pH 4-51 of the Isienu area in southeastern Nigeria where the experiment was conducted. The Isienu area lies within the dry savanna vegetalion zone. Major crops grown are root and tuber crops, especially yam (Dioscorea spp.1. cassava (Manihot esculenta) and cocoyam (Colocasia spp.) as well as maize (Zea mays) and pigeonpea (Cajanus cajan). These are generally grown in various mixtures or relays. For example, a cassava-based crop mixture carrying cassava, maize and a wide range of minor crops is popular, while maize and a wide range of minor crops followed by pigeonpea is a common relay cropping practice. Farms are generally on small plots averaging about 0.05 ha. All operations are performed manually with hand-tools . Materials and methods Three methods of fertilber application were used in a cassava + maize intercrop system. The treatment, which comprised banding, broadcast and no fertilizers (control), was laid out in a randomized complete block design. The effective area of each treatment plot was O.Olha. Before fertilizer application, the land was plowed by tractor, harrowed and ridged across. The ridges were l m apart. Planting took place in April 1985 and March 1986. Two ridges served as discards between replications and around the entire field. A 1-m guard row demarcated the plots and the ridges were tied at 2-m intervals to check on overflow of fertilizer from one treatment to another. An improved maize variety known as Western Yellow (widely accepted locally and provided by the University of Nigeria farm) and IITA's TMS 30572 cassava were the crop varieties used. Both the maize and the cassava were planted on top of the ridge. Cassava was planted at 1-m intervals, giving a population of 10,00O/ha, and maize was planted at 0.3cm. giving a population of 33,30O/ha after thinning. The 30-cm cassava cuttings were planted at an angle of 30" and buried up to the penultimate node. The malze was sown at a depth of 2cm. Fertilizer was applied at 75N: 75K: 40P: 20Mg kg/ha from urea, muriate of potash, single superphosphate. and magnesium sulfate, respectively. These fertilizers were applied in 2-cm bands 4cm from the plants on one side of each row at a depth of2.5cm in the banded plots and broadcast evenly for the other treatment. The fertilizer application, carried out by a single person. was timed. Weeding was done manually with hand-tools twice a year six weeks after planting. Weeding was also done by one person in each experiment and timed. Weeds were collected and weighed wet. Maize was harvested at 110 days and cassava one year after planting and the yields were recorded. The maize yield was converted to a 14 percent moisture level. An analysis ofvariance based on the F-test was used to determine the level of significance of difference among means. Judgment as to whether banding was economical and more efficient than broadcasting was based on marginal benefit to marginal cost ratio analysis. Results and discussion Yields of cassava and maize The effects of the three methods of fertilizer application on maize grain and cassava fresh root yields are shown in table 1. Non-application of fertilizer resulted in a highly significant (P<0.001) reduction of maize grain and cassava root yields during both years (table 1). There were differences between the band and broadcast methods of fertilizer application. A combined analysis of variance of the two-year data for maize and cassava yields gave no significant year effects, nor of year x fertilizer treatment interaction: therefore. both the economic and agronomic interpretations of the data can be based on a combined analysis of the fertilizer treatment effects, which was significant (P<0.001) [table 11. Maize yield without fertilizer was only about 6 percent of the yield obtained with fertilizer. Consequently. it is futile to attempt to grow maize in the Isienu soil environment without the application of fertilizer. Even the cassava yield was adversely dected since only 6 t/ha were produced compared with 17.6 t/ha with fertilizer. The yield without fertilizer was only 34 percent of the yield from fertilized plots. The generaliy poor crop yields in the absence of fertilizer are attributed to the poor soils at Isienu (table 2). They are highly acidic (pH 4.41, low in organic matter (organic carbon 0.80 percent). very low in N and have low exchangeable cations (table 2). Since fertilizers must be used on these soils. the most economic methods are examined. Fertilizer applfcation and weeding labor Table 3 shows that fertilizer application labor was significantly different (P<0.01) between fertflizer banded plots (21.01 man-days/ha) and fertilizer broadcast plots (only 6.41 man-days/ha). However. weeding labor was not different between the fertilizer banded plots (41.18 man-days/ha) and the fertilizer broadcast plots (41.62 man-days/hal. This indicates that under experimental conditions it was not clear whether broadcasting of fertilizers results in more weed growth than banding. Table 3 shows that weed weight was also not different between the fertilizer banded plots (2.5 t/ha) and the fertilizer broadcast plots (2.6 t/ha). Marginal benefit to marginal cost ratio Although the banding method of fertilizer application resulted in significantly higher biological yields of both cassava and maize than the broadcasting method, its superiority depends upon the prices of the intercrop products as against wage rate. This is particularly important in the smallholder cropping system where all farm operations are performed manually; labor is by far the largest cost item and is therefore in most cases the limiting factor. It was assumed that the fertilizer application, weeding and harvesting labor inputs were the only items of cost which would differ between production based on fertilizer application by banding and application by broadcasting. Consequently, only these items were costed at the current local nominal farm wage rate of Bd6.50 per man-day. The products were valued at the current local retail market prices of Pd0.12 per kg for cassava and Bd0.77 per kg for maize. In Table 4. the marginal cost of production based on fertilizer banding over production based on fertilizer broadcasting represents the differences in the costs of fertilizer application and in weeding labor inputs between the two production methods. Similarly, the marginal benefits represent the dirferences in returns from cassava and from maize between the two production methods. The marginal-benefit-to-marginal-cost ratio is 0.96, showing that the production of a cassava-maize intercrop based on the broadcasting method of fertilizer application appears to yield a higher net monetary return than production based on the banding method at the current local farm wage and product price relationships. However, a sensitivity analysis indicates that only marginal changes in wage rate (5 percent reduction), maize price (7 percent increase) and cassava price 18 percent increase] will equalize the marginal costs and the marginal benefits. Conclusions The above analysis suggests that intercrop production based on the broadcasting method of fertilizer application is clearly not superior to production based on the banding method in terms of net monetary return. In spite of this. however, farmers in the study area continue to use the broadcasting method for their produce wen though production using the banding method seems to result in a statistically significant higher biological yield. This is probably because of the opportunity cost of the ditference in labour between the two methods in other activities. The value of labor for other farm and non-farm activities in which the same farmers engage influences their decision with respect to the additional labor input for applying fertilizer to in their crops. Caution must be exercised in accepting the above conclusions since they are not based on real farm situations, although attempts were made to obtain reasonably accurate labor input measurements. The same person was used to apply the fertilizers and to weed all the plots, which removed any differences in working rates. and the person took a long break after every two plots, which reduced any fatigue effects. The inclusion of harvesting labor would have modestly increased the marginal cost of fertilizer banding production over that of broadcasting but this is unlikely to have made any significant difference, if at all, in the conclusions drawn. Table 1. Effects of methods of fertilizer application on maize grain and cassava root yields [tons/ha) under intercropping in Isienu area of southeastern Nigeria during two seasons (years) Fertilizer 1985/86 1986/87 Average treatment Maize Cassava Maize Cassava Maize Cassava No fertilizer 0.08 5.80 0.09 6.20 0.09 6.00 Broadcast 2.00 17.10 2.18 17.60 2.09 17.40 Banded 1.96 16.80 2.32 18.70 2.14 17.80 Mean 1.33 13.20 1.53 14.20 1.44 13.70 LSD 0.05 1.09 5.14 1.17 7.30 0.82 4.05 CV% 33.20 34.74 29.63 31.80 33.96 37.10 Table 2. Chemical and textural characteristi- of Isienu soil (sample = 0 - 15cm layer). Chemical PH 4.40 Organic C [%) 0.80 Total N [%I 0.06 P (ppm) 14.10 Exch. cations Ca 0.25 Mg 0.04 Mn 0.10 K 0.01 Na 0.06 Total acidity 1.37 CEC Meq/ 100 g) 1.82 Textural Sandy loam Sand I%) Silt [%) Clay 1%) Table 3. Effects of methods of fertilizer application on fertilizer application and weeding labor inputs (man-days/ha) and on weed weight (tons/ha) under cassava-maize intercrop in the Isienu area of southeastern Nigeria Ferti- 1985/86 1986/87 Average - lizer treat- ment Fert. Weed- Weed Fert. Weed- Weed Fert. Weed- Weed app. ing wt. app. ing wt. app . ing wt. lab. lab. lab. lab. lab. lab. No ferti- izer Broad- cast Banded Mean LSD 0.05 cv ( O h ) REFERENCES FACU. 1986. Project Report: Multi-State Agricultural Development Project 11, Phase I 1988-1991, Project Document, Niger State Agricultural Development Project. Federal Agricultural Coordinating Unit. Ibadan Fink. Arnold. 1982. Fertilizers and Fertilization: Introduction and Practical Guide to Crop Fertilization. Special Fertilizer Problem. Verlag Chemic. Florida. Ezeilo. W.N.O. 1979. Intercropping with cassava in Afilca. In B. Nestel and M Campbell (Eds.) Intercropping with cassava. Proc. of International Workshop held at Trivandrum, India. 27 Nov-1 Dec. 1978. IDRC - 142 e. 49-%. Nweke. I.. E.C. Okorji, A.U. Oki and C.I. Ezedinrna. 1982. Report on Single Intenriew Management Suwey: Part I. An Analysis of Household Fields and Fertilizer Use for Arable and Tree Crops. Agricultural Projects Monitoring Evaluation and Planning Unit (APMEPU). Kaduna. Nweke. I. 1980. Farm Labour Problems of the Smallholder Cropping System of Southeastern Nigeria. Quarterly Journal of International Agriculture, Vol. 19, No. 2.. Western Germany. MAIZE VARIETY AND POPULATION IN A CASSAVA-MAIZE INTERCROP J. Arthur. H.C. Ezumah and E.V. Doku Intercropping cassava with maize is one of the most popular mixed cropping combinations under rainfed agriculture in the tropics. These two crops of varying growth duration and rhythm supply the bulk of the calorific requirements of the inhabitants of the humid and subhumid tropics (Ezeilo 1979; Moreno and Hart 1979). One practical advantage of intercropping is an increase in land productivity per unit area. Maximization of yield necessitates the use of optimum plant populations of component crops [Baker 1981: Freyman and Venkateswale 1977) and the minimization of above and below ground competition for growth (Trenbath 19761 through the use of suitable varieties and improved crop protection measures. Considerable work has been done on the suitability of improved varieties bred in monocrop situations in intercropping systems. Varieties that perlorm well in a monoculture may not do so in an intercropping system and may adversely affect the growth and yield of the associated crop(s). It has been shown that plant height, maturity period and canopy architecture or maize are important factors in maize-bean mixed cropping systems (Wahua et a1 1981; IlTA 1982). Kang and Wilson (1981) and Guritno (1984) obtained variable results with improved maize varieties in a maize-cassava nlkture of varying populations, probably because of the maize genotype used. The cassava yield was also reduced at higher maize populations: the degree of reduction was related to the associated maize variety. Research workers have attributed this yield reduction of component crops to increased competition for light, moisture and soil nutrient with increasing populations. It is assumed that differences in maize plant height and canopy architecture observed in maize-bean associations result in the varying amounts of light transmitted to the lower canopy in a cassava- maize intercrop system IIITA 1982. 1983). This paper describes attempts to obtain a clearer picture of the plant characteristics necessary for those maize plants intended for use in a cassava-maize intercrop system. The feasibility of intercropping cassava with more maize than the recommended optimum is also reported. Materials and methods Five maize varieties ranging from the short, early-maturing IK84A-2 135 with spreading leaves to the late-maturing, erect-leaved hybrid 1368 x 5012 [table 1). were interplanted within cassava variety TMS 30572 planted at two different spacings. A systematic parallel row arrangement was used within a strip split-plot design. Cassava spacing was laid out in two strips facing each other: maize varieties were randomized within each cassava spacing. The cross plots between the strip plots of cassava spacing and the maize variety were split into five subplots. These were systematically assigned to five maize populations. Treatments are shown in table 2. The sole-crop optimum for maize and cassava at 4,000 plants/ha and 10.000 plants/ha respectively was also planted. Three replications were made. Cassava and maize were planted at the same time in flat alternate rows in May 1986 at IITA. Ibadan. Nigeria. Maize populations were achieved by systematically varying the distance between plants. Within rows all plots were fertilized with a basal appllcation of 300 kg/ha of 15- 15-15 following a side dressing of nitrogen at the rate of 45% urea/ha four weeks after planting. No extra fert i lhr was given to the cassava nor any supplemental water. Hand weeding was carried out twice between maize planting and harvesting and thereafter when necessary. Data on crop performance at eight weeks after planting, four months after planting and seven months after planting and yield of crop economic components were collected and statistically analysed. Yields from the monocrops were used as standardizing factors in computing land equivalent ratios (LER). Results and discussion Eficts on maize Variations in agronomic characteristics in maize were to be expected and are not reported. Results showed that an increasing maize population in the intercrop delayed days to 50 percent silking and increased maize plant height, stem and root lodging. Ear weight decreased with an increasing maize population. Grain yield increased from 10.000 plants/ha. peaked at about 80.000 plants/ha and declined thereafter. The relationship between the maize popualtion and its effect on the crop are given in table 3. The effects of the maize population on agronomic characteristics observed in the study have been reported by Early et al. (19671; Rutger and Crowder (1967) and Dungan. Lang and Pendleton (1958). All indicated that these effects are mainly a result of competition for light, moisture and sofl nutrients. A high incidence of lodging at high populations produced taller plants with weaker stems. Effects on cassava Observations at different cassava growth stages indicated the various effects of maize on the cassava. At all stages, plant height and internode length increased with an increasing maize population in the intercrop. Stem girth, branches and leaves per plant decreased with an increase in maize populations. These results agree with those of Kang and Wflson (1981) who observed that at four to five months after planting. intercropped cassava was etiolated with poor branching and smaller foliage. Enyi (1973) and Hunt. Wholey and Cock (1977) attributed plant height and internode trends observed with increasing population to competition for light. Wahua (1985) also reports that plants produce more leaves and branches when they have adequate supplies of light, nutrient and sofl moisture. Apart from stem girth maize variety did not significantly affect other cassava plant features observed at eight and seven months after planting. At four months after planting. only plant height and internode length were significantly affected by maize variety as shown in tables 4- 6. Trends indicate that plant height and maturity period have an effect on cassava. Cassava intercropped with late maturing maize varieties had larger stems (table 4) a t eight weeks after planting since the late maturing maize varieties were no longer exploiting environmental resources to the maximum, unlike early maturing maize varieties. Consequently. the cassava plants were ensured optimum use of growth factors. The etiolated cassava plants observed under late maturing maize varieties four months after planting (table 5) probably resulted from enforced shade. Findings agree with Wahua (1985) with regard to the responses of cowpea to different maize varieties. The implication of the non-significant effect except for stem girth of maize variety on cassava plant characteristic seven month after planting (table 6) is that the cassava plants had recovered from the effects of the maize varieties. unlike the effects of the maize population which were stfll evident. Tuberous root weight, number of roots per plant, average root weight and root yield per plant decreased linearly with increasing maize population in the intercrop. Diversion of assimilates into tissue synthesis for stem and internode elongation caused by increasing plant population as under shade conditions appear to cause these effects [Hunt. Wholey and Cock 1977). m u r e productivity Land equivalent ratio values indicated the high productivity of the cassava-maize intercrop system. The LER for cassava decreased with increasing maize population in the intercrop (see figure 11. On the basis of maturity, the early maturing varieties. IK84A-2135 and IB84A-203 gave the highest LER. On the basis of plant height. the short maize variety IK84A-2135 again showed its superiority over all other maize varieties (see figure 1). These observations indicated that early maturity and shortness are desirable characteristics for incorporation in maize varieties for use in cassava-maize intercrops. Total LER showed that the optimum population for the short, early-maturing variety is 80,000 plants/ha. This population in the intercrop gave the heighest LER of about 1.90 (see figure 1). The other maize varieties, on the other hand, could be interplanted at an optimum population of 40,000 plants/ha to give the best combination. These recommendations are applicable if natural fertility is supplemented with recommended amounts of fertilizers. REFERENCES Babalola, 0.. and M.E. Akenova. 1981. Intercropping morphologically different types of maize with cowpeas: LER and growth attributes of associated cowpeas. Expl.Agric.. 17: 407-413. Baker. E.F.I. 1981. Population. time and crop mixtures. In Proc. Int. Workshop on Intercropping. p. 52-60. 10-13 Jan. 1979. Hyderabad: ICRISAT Dungan, G.H.. A.L. Lang and J.W. Pendleton. 1958. Corn plant population in relation to soil productivity. Adv. Agron.. 10: 435-473. Early, E.B.. W.D. Melbrath. RD. Self, and RH. Hageman. 1967. Effects of shade applied at different stages of plant development on corn production. Crop Sci. 7: 151-156. Enyi, B.A.C. 1973. Growth rate of three cassava varieties (Manihot esculenta Crantz.) under varying population densities. Ezeilo, W.N.O. 1979. Intercropping with cassava in Africa. In Intercropping with cassava, ed. E. Weber. B. Nestel and M. Campbell, pp. 49-56. Ottawa: IDRC. Freyman, S., and J. Venkateswale. 1977. Intercropping on rainfed red soils ofthe Deccan Plateau, India. Can. J. Plant Sci. 57: 697-705. Guritno. B. 1984. The effects of maize population on the yield of maize and cassava in an intercropping system. Agrlvlta 7(1): 1-6. Hunt. L.A.. D.W. Wholey and J.H. Cock. 1977 Growth physiology of cassava (Manhot esculenta Crantz). Field Crops Abst. 30I2): 77-89. IITA. 1982. Maize/cassava intercropping. 1981 Annual Report. pp. 141- 142. Ibadan: IITA. IlTA. 1983. Maize/cassava intercropping. 1982 Annual Report. Ibadan: IITA. Kang, B.T. & Wilson, G.F. 1981. Effect of maize plant population and nitrogen application on maize-cassava intercrop. In Tropical root crops: research strategies for the 1980s. ed. E.R. Terry, K.A. Oduro and F. Caveness. pp. 129-133. Ottawa: IDRC. Kumar, C.R.M.. and N. Hrishi. 1979. Intercropping with cassava in Kerala State. India. In Intercropping with cassava, ed. E. Weber. B. Nestel and M. Campbell, pp. 31-34. Ottawa: IDRC. Moreno. R.A. & Hart. R.D. 1979. Intercropping with cassava in Central America. In Intercropping with cassava. ed. E. Weber. B. Nestel and M. Campbell. pp. 17-24. Ottawa: IDRC. Rutger. J.N. and L.V. Crowder. 1967. Effect of high density on silage and grain yield of five corn hybrids. Crop Sci. 7: 182- 184. Trenbath. B.R. 1976. Plant interactions in mixed crop communities. In Multiple cropping, ed. RI. Papendick. P.A Sanchez and G.M. Tripleti, pp. 129-169. Spl. Pub. No. 27. American Society of Agronomy. Madison, Wisconsin. Wahua, T.A.T. 1985. Effects of melon (colocynthis vulgaris1 population density on intercropped maize [Zea mays) and melon. Expl. Agric. 21: 281-289. Table 1. Characteristics of maize and cassava test crops studied. Maize variety Characteristics IK84A-2 135 Open-pollinated, short, early maturing with spreading leaves IB84A-203 Open-pollinated, tall, early maturing with spreading leaves TZSR-Y- 1 Open-pollinated, tall, late maturing with spreading leaves 1368 x 5012 Hybrid, tall, late maturing, with erect leaves 9450 x 4001 Hybrid, tall, late maturing with spreading leaves Cassava variety TMS 30572 Low-branching, rapid canopy closure, good recovery from insect attack, high yielding, good gari and fufu characteristics, preferred in the West African subregion to other varieties Table 2. Treatment levels of strip split-plot layout in a systematic parallel row arrangement Factors Cassava spacing-2 levels Maize variety-5 levels Maize population-5 levels (1) lOOcm x lOOcm (ii) 1OOcmx 67an Table 3. Effects of maize population on maize plant characteristics and yield Characteristics Regression equivalent Correlation coefficient (R3 Days to 50% silking Y = 53.6 + 5.03 (10-9 x O.9la Plant height at harvest Y = 172-07 + 1.23 (lo4) x 0.77 Stalk diameter Y = 17.3 - 4.4 (10-5) x Root lodging Ycl = 958 + 0386 x Yc2 = 767 + 0.58 x Stem lodging Ear weight Grain yield Notes: a Significant at 5% level b Significant at 1% level Table 4. Effect of maize variety on the agronomic characteristics of cassava 8 weeks after planting Maize variety Plant height Stem girth Internode Leaves (cm) (mm) length per (mm) plant IK84A-2 135 60.5 9.4 24.8 19.2 IB84A-203 64.5 9.6 30.5 19.7 TZSR-Y- 1 60.4 10.7 28.7 17.4 1368 x 5012 67.8 10.3 28.6 19.8 9450x 4001 65.8 11.0 25.8 19.9 Mean 63.8 10.2 27.7 19.2 LSD 5% ns 0.7 ns ns Table 5. Effect of maize variety on the agronomic characteristics of cassava at maize harvest Maize Plant Stem Internode Branches/ Leaves/ height girth length Per Per variety (cm) (cm) (mm) plants plant IK84A-2 135 98.7 1.5 16.6 3.1 48.6 IB84A-203 108.3 1.6 17.7 3.7 53.7 TZSR-Y- 1 98.8 1.5 18.1 2.3 43.6 1368 x 5012 110.4 1.6 18.3 3.4 52.1 9450 x 4001 103.4 1.5 18.8 2.5 45.1 Mean 103.9 1.5 17.9 3.0 18.6 LSD (5%) 15.3 ns 1.3 ns ns Table 6. Effect of maize variety on the agronomic characteristics of cassava at seven months after planting Maize Plant Stem Internode Branches Leaves height girth length per Per variety (em) (cm) (mm) plant plant IK84A-2 135 119.5 1.83 10.3 10.6 84.5 IB84A-203 132.7 1.87 10.4 12.8 101.5 TZSR-Y- 1 116.1 1.69 12.4 8.3 65.6 1368 x 5012 131.8 1.92 10.2 12.4 95.6 9450x 4001 124.4 1.79 10.7 10.7 88.3 Mean 124.9 1.82 10.8 10.9 87.1 LSD (5%) ns 0.20 11s ns ns CASSAVA-GROUNDNUT INTERCROPPING IN ZAIRE N.B. Lutaladio. FE. Brockman, KB. Landu. TAT. Wahua and S.H. aahn Abstract Field experiments were conducted from 1980 to 1985 in southwest Zaire to assess the productivity of cassava-groundnut intercropping systems and determine the effect of ditferent crop management practices on the yield of the two associated crops. Planting cassava in double rows with groundnuts at 0.4m x 0.2m resulted in the highest yields and the greatest monetary returns. The competitive ability of the two crops depends on their growth habit. planting arrangement and densities, time of planting and soil fertility level. Cassava-groundnut intercropping is practiced in many parts of Central Africa, particularly in south-westem Zaire where it is the most common crop combination. The two crop species are highly compatible and the combination offers many advantages in terms of soil fertility. overall crop yields and the quality of food for human nutrition [Lutaladio. Wahua and Hahn 1987). In Zaire. the two crops are usually planted simultaneously. Cassava is commonly planted on ridges in paired rows with about 0.5m between the two rows and 1.5 to 2m between the rows of adjacent pairs. Groundnut is sown in the wider inter-row space. During the process of varietal improvement, the Zaire national cassava research program [PRONAM) considers it important to determine how newly developed varieties of ditferent plant types react to different crop management factors such as cropping systems. planting arrangements, cassava and groundnut densities and time of planting. The experiments reported here were designed to assess the productivity of the cassava-groundnut intercropping system and determine the effect of different crop management practices on the yield of the two associated crops. Materials and methods From 1980 to 1985, a series of cassava-groundnut intercropping experiments was conducted at Mvuazi station. Zaire (5'27's. 14'54'E. elevation 450m) on an ultisol (sandy clay loam) with a low base saturation and a pH of between 3.5 and 5 (1 soil: 2.5 water). In the 1980-82 experiments. two cassava varieties (an improved standard. 02864 and a local. Mpelolongi) were intercropped with groundnuts in two planting arrangements (cassava planted in equidistant rows. l m apart or in a double-row arrangement with alternate row spacing of 0.5 and 1.5m) and with two groundnut spacings (0.2m x 0.2m and 0.4m x 0.2m). For comparison. monocrop cassava was grown in the two planting arrangements and monocrop groundnuts at the two spacings. A split-plot design was used with four replications. Cassava cultivars were the main treatments and planting patterns/groundnut spacing the sub-treatments. The sub-plot size was 8m x 6m. All plot yields were expressed on a hectare basis. Mtxture productivity was determined by means of the Land Equivalent Ratio (LEN and gross monetary returns. In another trial carried out between 1982 and 1984 a new improved variety named Kinuani was compared with the local Mpelolongi variety under two cropping systems [monocropped or inter- cropped with groundnuts) at two fertility levels (no fertilizer or 600 kg/ha of 17-17-17) at two densities (10.000 or 20,000 plants/ha) with or without leaf harvesting. A design with two replications was used. The plot slze was 7m x 6m. At harvest. plot yields were expressed on a hectare basis. The 1983-85 experiments were conducted in order to determine the effect of the following factors on the yield of intercropped cassava and groundnuts: cassava plant type (Kinuani-profusely branching, and clone 5052-erect with little branching); planting arrangement (cassava planted in paired rows with 40cm between the two rows of the pairs and 2m between rows of adjacent pairs, and cassava in rows equally spaced at 1.2m): time of cassava planting (with groundnut, and three weeks after groundnut): cassava density (8,333 and 16,667 plants/ha) and groundnut density (83.333 and 166,667 plants/ ha]. A design with two replications was used. Plot size was 7m x 6m. Cassava storage root yield, groundnut yield and LER were determined as described in other experiments. Results and discussion 1980-82 experiments The storage root yield of cassava intercropped with groundnuts was higher when groundnuts were sown at a low plant population (0.4m x 0.2m) than at a higher plant population (0.2m x 0.2m). irrespective of the cassava planting pattern (table 1). This is apparently a result of high interspecific competition in the combination where groundnuts were introduced at 0.2m x 0.2m. The groundnut yield was not significantly affected by the planting patterns of cassava. This seems to result from the fact that the groundnuts were well developed before the cassava could make sufficient use of environmental resources in order to be competitive. Groundnut yield was higher at 0.4m x 0.2m than at 0.2m x 0.2m. as a monocrop and as an intercrop in both cassava planting arrangements (table 2). This is attributed to high plant-to-plant competition in the groundnuts sown at 0.2m x 0.2m. Such a high population of groundnuts could be the cause of excessive mutual shading leading to a reduction in the net assimilation rate I N N and therefore to a reduction in yield. It seems that the population obtained with 0.2m x 0.2m spacing was above the optimum under the conditions of this study. The land equivalent ratios of cassava and groundnut intercrops were all above 1.0, showing that a higher productivity per unit area was obtained by intercropping cassava with groundnuts than by growing the two crops separately (table 3). However, the productivity of the system tended to decrease with the high groundnut plant population (0.2m x 0.2m) perhaps because of the increased competition. The highest returns were obtained from cassava grown in paired rows ir~tercropped with groundnuts at 0.4m x 0.2m. (table 4). It appears that the economic returns from this cassava-groundnut intercropping may valy with changes in planting patterns and plant population which affect the yield of the two crops. 1982-84 Experiments Although cassava storage root yields of both Kinuani and Mpelolongi varieties were not affected by leaf harvesting or by plant density. intercropping with groundnuts reduced the yield of the two cassava varieties by 64 and 33 percent respectively as compared to the monocrop (table 5). The significant difference in the magnitude of the response of the two varieties to intercropping is probably due to the com- paratively slow early growth of Kinuani. When the interaction between variety, cropping system and fertility level was analyzed. it appeared that Kinuani gave a higher yield than Mpelolongi only when grown a s a monocrop at a high fertility level. This confirms the observation that the improved variety Kinuani will perform better than the local variety if planted as a sole crop at a high fertility level ( figure 1). The yield of the groundnut intercrop was also significantly less at a high fertility level while that of the intercropped cassava at the high fertility level was more than double that at the low level (table 6). It seems that at high fertility levels. cassava is more competitive with groundnuts. A point of particular interest is that the application of lime causes a marked increase in groundnut yields (PRONAM 1985). This is receiving further attention since lime is produced in Zaire and could be available at low cost. Figwe 1. Yield of two cassava varieties under two cropping systems at two levels of soil fertility Fresh root yield ( t /ha) 151 Fertility level 0 High Monocrop lntercrop 1983-85 experiments Results show that cassava variety was the only factor with a significant effect on storage root yield and that Kinuani produced a higher yield than clone 5052 (table 7). A similar difference in yield between the two varieties was observed in the monocrop treatments. On the other hand, groundnut density was the only factor wich significantly affected the yield of the intercropped groundnuts. In addition, the time of cassava planting had a significant effect on total productivity per unit area with the LER being higher when cassava was planted at the same time as groundnuts. Delaying cassava planting for three weeks after groundnuts could reduce cassava yield by about 37 percent IPRONAM. 1984). The overall productivity per unit area of intercropped cassava and groundnuts was about 60 percent greater than that attained by growing the two crops in a monoculture. Conclusion Cassava is well suited for intercropping with groundnuts since it does not impose much competition at the beginning of its growth cycle. But the competitive ability of the two crops will depend on their growth habit, planting arrangement and densities. The time of planting and soil fertility level are also important and have biological implications. Planting cassava in double rows with groundnut at 0.4m x 0.2m spacing resulted in the highest yields of cassava and groundnuts in the mlxture and provided the greatest biological efficiency and highest economic returns. However. a single-row ( lm x lm) cassava system could also be used in intercropping without causing significant yield reductions of the associated groundnuts. REFERENCES Lutaladio. N.B. 1986. Planting periods and associated agronomic practices for cassava production in south-westem Zaire. Ibadan: IUP. 374 pp. Lutaladio. N.B.. TAT. Wahua and S.K. Hahn. 1987. Cassava-groundnut intercropping has potential to improve the nutritional quality of the diet in Zaire. In IITA Annual Report and Research Highlights for 1986, pp. 93-95. Ibadan: IlTA PRONAM 1984. Rapport annuel 1983 du Programme national manioc. Dtpartement de l'agrlculture. F&5pubIique du Zaire. PRONAM. 1985. Rapport annuel 1984 du Programme national manioc. Dtpartement de l'Agrfculture. @pubUque du Zaire. Table 1. Storage mot yield of cassava grown alone and intercropped with groundnut in different planting patterns Treatment combinations Yield (t/ha) Cassava DR 18.2 15.8 Cassava SR 17.6 14.6 Cassava DR + groundnut (0.2m x 0.2m) 13.0 13.2 Cassava DR + groundnut (0.4m x 0.2m) 19.1 13.3 Cassava SR + groundnut (0.2m x 0.2m) 12.9 10.4 Cassava SR + groundnut (0.4m x 0.2m) 15.6 12.5 LSD (0.05) 5.0 2.0 Source: Lutaladio 1986. Note: DR = double mws; SR = single rows Table 2. Grain yield of groundnut at two p h t populations grown alone and with casaava in different planting patterns Treatment combinations Grain yield (kglha) 1980-81 1981-82 Groundnut (0.2m x 0.2m) 1,690 1.040 Groundnut (0.4m x 0.2m) 1.730 1.090 Cassava DR + groundnut (0.2m x 0.2m) 1.330 875 Cassava DR + groundnut (0.4m x 0.2m) 1,710 1,005 Cassava SR + groundnut (0.2m x 0.2m) 1.410 835 Cassava SR + groundnut (0.4m x 0.2m) 1.570 1.035 - - LSD (0.05) 250 175 Sourn: Lutaladio 1986. Note DR = double rows: SR = single rows Table 3. Land equivalent ratio [LER) of cassava and groundnut as affected by planting patterns and crop combinations in two cropping years Crop combinations Partial LER Total Cassava Groundnut LER Cassava DR 1.0 - 1.00 Cassava SR 1.0 - 1.00 Groundnut [0.2m x 0.2m) - 1.00 1.00 Groundnut (0.4~1 x 0.2m) - 1.00 1.00 Cassava DR + groundnut (0.2m x 0.2m) 0.72 0.82 1.54 Cassava DR + groundnut [0.4m x 0.2m) 0.87 1.04 1.91 Cassava SR + groundnut 10.2m x 0.2m) 0.77 0.82 1.38 Cassava SR + groundnut (0.4m x 0.2~1) 0.94 0.97 1.91 Note Mean of two cultlvars in two seasons DR = double rows: SR = single mws Table 4. Gross returns of cassava and groundnut as affected by planting patterns and crop combinations in two cropping years Crop Cassava Groundnut Total Yield Gross Yield Gross (t/ha) return (kg/ha) return (Z) (Z) Cassava DR 17.0 34,000 - - 34.000 Cassava SR 16.1 32,200 - - 32,200 Groundnut - - 1,355 20,325 20,325 (0.2m x 0.2m) Groundnut 10.4m x 0.24 - - 1,410 21.150 21,150 Cassava DR + groundnut (0.2m x 0.2m) 12.7 25.400 1.102 16.530 41,930 Cassava DR + groundnut (0.4m x 0.2m) 14.6 29,200 1.332 19.980 49.180 Cassava SR + groundnut (0.2m x 0.2m) 11.5 23,000 1.122 16.830 39.830 Cassava SR + groundnut (0.4m x 0.2m) 14.0 28,000 1,305 19,575 47,575 Source: Lutaladio 1986. Note Mean of hKa cultivars in two seasons DR = double mws; SR = single murs Gross returns: Casava42.CO/kg of storage mob; Groundnut-Z 15.00/kg of grab Table 5. Effect of several agronomic factors on the yield of two cassava varieties Variety Fertility Cropping Plant Leaf Mean level system density harvesting low high mono-inter- 10,000 20.000 with with- /ha /ha out Mpelolongi 7.3 11.2 11.1 7.4 8.9 9.6 9.6 8.9 9.2 Kinuani 6.310.3 12.2 4.4 8.6 8.0 8.1 8.5 8.3 Mean 6.8 10.8 11.7 5.9 8.8 8.8 8.8 8.7 8.8 LSD (5%) for Main effects 1.4 Interactions 1.9 Source: PRONAM. 1984 Nok The variety X cropping system interaction was highly significant Table 6. Effect of level of soil fertility on yield of intercropped cassava and groundnut Fertility Cassava fresh Groundnut fresh level root yield (t/ha) pod yield (kg/ha) Low 3.7 873 High 8.1 519 Source: PRONAM. 1984 Notr The effect of fertility level on yield of both cassava and gmundnut was highly significant Table 7. Maize effects of cassava variety, time of cassava planting . ca-vm planting arrangement. c u u v a density and groundnut density on yield of intercropped cassava and groundnut md on land equivalent ratio (LER) Treatment Cassava fresh Groundnut LER root yield fresh kg/ha t/ha pod yield Cassava variety Kinuani 5052 Time of cassava planting ns ns with groundnuts 5.23 2924 1.32 three weeks afler groundnuts 5.15 2777 1.17 Cassava planting arrangement ns ns equidistant rows 5.01 298 1 1.25 double rows 5.36 2720 1.23 Cassava density 10,000 plants/ha 20,000 plants/ha Groundnut density 63.333 plants/ha 166,666 plants/ha Source: PRONAM 1985. Notes s i g n i h t l y dllferent at 0.05 significantly dllferent at 0.01 ns: not significantly dllferent SUMMARY OF COMMENTS ON TECHNICAL. PAPERS From papers presented on intercropping with cassava, the following points are salient: 1. Intercropping other crops with cassava helps to minimize the output of land resources. 2. A significant difference exists between total yields from fertilized cassava crop mixture and non-fertilized mixtures: hence the use of fertilizer helps to boost yield. 3. Various times of introduction of various crops in cassava-based mixtures were recommended: for instance, maize could be brought in at planting or 28 days after planting cassava. 4. Some cultural methods that could help to control pests were highlighted; for instance, fertilizer application reduces incidence of mealybug on cassava, and intercropping, especially with spreading prostrate crops, reduces weed infestation. 5. The preference of farmers in Nsukka area for the broadcasting rather than the banding fertilizer application method was attributed to the high opportunity cost of the extra labor involved in the banding method. The fanners value such labor highly and use it on other activities. 6. Shortness and earliness were attributes advocated for maize varieties intended to be used in a cassava mixture with a population of 80,000 stands per hectare as appropriate. 7. In a cassava+grcundnut intercrop, the best cropping pattern is determined by varietal growth habit, plant arrangement and density of component crops in the mixture as well as time of planting and soil fertility. It was generally noted that most of the results from the different locations appeared to have similar trends. With standardized methods for various data collection, it may be possible in future to remove the impression that agronomic research results are location-specific. RESPONSE OF A CASSAVA-MAIZE INTERCROP TO NlTROGEN IN TWO-YEAR SEQUENTIAL CROPPING H.C. Ezumah. F.I. Nweke. N.D. Kalabare and A. Kanu~wi Abstract Two open pollinated [early maturing and late) and a hybrid maize intercropped with cassava at three sites, a perhumid ultisol soil. a humid and a subhumid alfisol soil, at five nitrogen levels did not result in reduction of intercropped cassava root yields at humid and subhumid ecologies. Optimum N level for maize yield varied from 80 to 160 kg/ha depending upon site and maize variety. The hybrid maize grown in the subhumid environment required higher N rates than the composites. Cassava gave negative response to N at the perhumid site and only weak response at the humid site but root yield increased with N in the subhumid alfisol soil. Net benefits from intercropping maize with cassava were higher when association was with hybrid maize at the subhumid site. Supplementary nitrogen may not be required in intercropping systems of maize with improved cassava, TMS 30572, since the peak response for the intercrop system averaged across the three ecologies was 120 kg/ha N. Nitrogen applied durlng the first year significantly affected the second year responses for maize. .********I*** The most commonly observed cropping sequence in humid areas of Nigeria following land clearing are those in which nutrient-demanding crops such as maize and yams are followed by less nutrient-demanding crops such as cassava and cocoyams. Vegetables may be included in the cropping cycle. Cassava seems to be displacing other major crops and is rapidly taking over compound farms and less humid areas (Nwosu 1977). Sequences in which cassava and maize are followed by (fb) cassava+ maize fb fallow are now common, particularly in southern Nigeria. southern Ghana, Republic of Benin and southern Togo. A study of the cassava + maize fb cassava + maize fb fallow in 2-year cropping cycles was undertaken in three ecologies, commencing in 1985 to determine the nitrogen response, sustainability and monetary returns from the system. Previous reports have shown that hybrid maize responded to N at lwelsgreater than 220 &/ha [IITA 1983). Materials and Methods The sites. representing the three ecologies, were Warri (perhumid ultisol soil), Okolu (alfisol soil) and Mokwa (subhumid, alfisol soil). The commonest limiting nutrients for plant growth in these environments are usually N and K and sometimes P [Takyi 1976: Agboola and Obigbesan 1974; Ofori 1976; Obigbesan and Fayemi 1976; Okeke et al. 1979; and Okeke et al. 1982). In this study. P and K levels were kept constant at 60 @/ha and N level varied from 20 @/ha to 320 @/ha at rate functions of 20. 2 x 20. 22 x 20.23 x 20, and 24 x 20 kg/ha. N. Three maize varieties, the early maturing (90-day) streak resistant TZESRW. late (1 10- 120 day) TZSRW and a hybrid (832 1 x 180) were intercropped with TMS 30572 cassava at the five N levels in the five sites representing the three ecologies. The three maize varieties and five N levels treatment factors were arranged in randomized complete block at each site. There were four replications per site. The plot size was 6m x 8m. Cassava and maize were planted in alternating rows. Cassava spacing was lm x l m while the maize was spaced at lm x 0.25m for 10.000 and 40.000 plants per ha. respectively. for the cassava and maize in the intercrop systems. All the @/ha of P and of K were applied at planting. Half the N was applied as seedbed and the second half applied four weeks after planting. Both crops (cassava and maize) were planted the same day. Soil samples (0-lOcm depth) were obtained prior to planting at the end of the first year, and at the end of the second year. Yields were obtained at 12 months for cassava roots and when cobs were dry for m e . Maize-grain yield is reported at 14 percent moisture; cassava root is fresh tuber. Statistical analysis A major objective of the trials was to determine the effect on yield of two sequences of cassava + make in each of three sites where N was varied. Therefore the two-year data were combined per site and analyzed as split plot, with the year or season effects as the main plot and the combinations of maize variety + N levels as subplot. The choice of split plot in the combined analysis was rationalized since the treatment factors were not re-randomized during the second year. Significant Year x N interaction indicates influence of first year N application on second year response by test crops. Results and Discussion Maize Grain Yield The analysis of variance showed signflcant effects I1 percent) of maize variety. N level, and season (year) on maize grain yield at each ecology. Interactions of nitrogen and season (year) were also significant (table 11. The hybrid maiz gave a higher yield than the other varieties. The lowest yield was obtained from TZESRW. the early maturing maize variety, while the hybrid maize (8321 x 18) yielded highest, especially in the subhumid (by 62 percent) over TZESRW and humid (by 38 percent) and perhumid (by 50 percent] environments. In the humid alfisol site, the grain yields of late composite ITZSRW) and hybrid maize were not different. Nitrogen eflects on Maize Grain At Warri. (the perhumid environment) and Okolu. (humid), maize grain yield peaked at about 160 kg N/ha and 80 kg N/ha in 1985 and 1986 respectively [figures 1 and 2). At Mokwa, there was no response to N application in 1985 but in 1986, the best maize grain yield was obtained at 160 kg N/ha(flgure 3). Therefore irrespective of ecology, the N level for high maize grain yield under intercropping with cassava was between 80 and160 kg/ha. Weak response during the first year of cropping is attributable to soil N immediately after land clearing. The N level generally recommended for maize in the forest zone ranged from 80-120 kg/ha. (NAFPP 1976; Kang et al. 1980; Ikeorgu 1984). The 1986 rainfall was low, and deviated negatively from usual levels: this could contribute to variances in yields (table 2). Cassava Root Yield Significant effects (P > 0.001) were observed for nitrogen. years (season) and nitrogen x year interaction on cassava root yields (table 3). Maize had no effects on cassava root yield (table 4). Very mild response to N was observed in the 1985 cassava yield but in 1986. increasing N reduced cassava root yield at Warri. the pcrh~lmid envlronment (ligure 4). The hlyhest cassava root yield was obrair~ed at 40-80 ke N/ha at Okolu. the humld environmenl ifleure 51. At Mokwa, however, v e j mild response was observed in 198% with no distinct peak. The 1986 response was rapid up to 120 kg N/ha and tapered off beyond 120 kg N/ha for cassava root (flgure 6). Therefore, the response of intercropped, two sequential cropping of cassava and maize to N varied with ecology-ranging from declining root yields at Warri (perhumid ecology) to distinct peaks at Okolu (humid) and mild response only during the second year at Mokwa (subhumid). Reduced cassava root yield with N fertilization was also reported by Kang and Wilson (1980) in an alflsol soil in southern Nigeria. Cassava + Mafze Economic analysis of nitrogen response in the cassava and maize intercrop involving the determination of the net-benefit analysis from production costs and returns indicated that in Warri, the perhumid environment, negative marginal net-benefits were obtained when nitrogen application was increased from 20 to 40 kg/ha. The highest net benefits were recorded at 20 kg N/ha for all the maize varieties in this ecology irrespective of maize variety (table 51. At Okolu (humid), the highest net benefit was realized at 80 kg N/ha by intercropping TMS 30572 with the three maize varieties. In the perhumid and humid environments, returns from cassava tended to complement lower incomes associated with relatively low yields from maize. In the subhumid environment at Mokwa, the net benefit peaked at 160 kg N/ha for the TMS 30572 + TZESRW intercrop. whereas the response at 320 kg N/ha resulted in highest benefits for the TMS 30572 + TZSRW and TMS 30572 + hybrid intercrops. The inclusion of cost of capital which constituted 40 percent of the cost of fertilizer and application in the variable costs as opportunity cost and risk premium to the farmer for the committed capital did not affect the decisions earlier stated on the net benefits; however, it indicated that the farmers could afford such investments. Although the exact optimum N level for highest monetary returns under cassava + maize intercropping varied with ecology. the peak response for the intercrop system averaged across the three ecologies was 120 kg N/ha, comparing favorably with that required by maize as a monocrop which ranged from 80 kg to 160 kg N/ha. The use of benefit cost ratios analysis as economic indicators showed that the investment is viable with the benefit cost ratios greater than 1 for all the intercrops in all the ecologies (table 61. Conclusions Cassava and maize intercropped across a wide range of environments in humid/subhumid tropical ultisols and alfisols in Southern Nigeria under varying N fertilizer rates show: 1. That maize variety-whether early maturing open. late maturing open or hybrid-had no significant effect on cassava root yield under intercropping. However, the range in maize types was limited to only three at recommended intercrop population of 40.000/ha (IlTA 1984: Kang and Wilson 1980). 2. The best N level for maize yield ranged from 80 @/ha to 320 %/ha irrespective of ecology. It appears some maize types such as the hybrid require higher N rates. 3. From net benefit analysis, it was ascertained that application of N at high rates in the perhumid environment presented by Wani gave negative marginal returns over very little dosage of between 20 and 40 kg/ha. At Okolu. N rates of 80 kg/ha resulted in the highest level of benefit. In the subhumid ecology the benefits reached their peak at 160 kg N/ha for the TMS 30572 + TZESRW intercrop while it was a t 320 kg N/ha for TMS 30572 + 'IZSRW and TMS 30572 + Hybrid intercrops. 4. Significant Y x N interactions for maize and cassava yields indicate that N applied during the first year (1985) influenced response by maize and cassava to N applied during the second year (1986). Second- year responses to N fertilizer were generally higher at the lower N rates and tended to decline as N rates increased. Figure 1. Effect of N on the grain yields of maize in 1985 and 1986 averaged over three maize varieties in a perhumid environment Wanl) Maize groin I kg / ha ) I Figure 2. Effect of N on the grain yields of maize in 1985 and 1986 averaged over three malze varieties in a humid environment (Okolu) 2400 2100 1800 1500 1200 900 600 Maize grain ( kg /ha) I 1 - - - - - Y1985 = 323 + 17.77N-0.0353~'; R =0.98 Y1986 = 1 1 1 1 + 2 0 . 5 0 ~ - 0 . 0 5 7 9 ~ ~ ; R -0.85 - I I I I I 1 I I 1000 I I I I I I I I I 0 4 0 80 120 160 200 240 280 320 Nitrogen ( kg / ha) 0 40 80 120 160 200 240 280 320 Nitrogen (kg /ha) Figure 3. Effect of N on the gain yields of malze in 1985 and 1986 averaged wer three malze varieties in a subhumld environment (Mokwa) Maize grain ( k g / ha) I Figure 4. Effect of N on the root yields of cassava in 1985 and 1986 averaged over three maIze varieties in a perhumid environment Warril 4800 4200 3600 3000 2400 1800 1200 600 Cassava root yield ( t / ha Perhumid Y1985 = 8.241 -0.0017N + 0 . 0 0 0 0 4 7 3 ~ ~ ; R.0.95 Y1986= 18.790-0114N + 0 . 0 0 0 2 5 8 ~ ~ ; Rr0.50 12 1985 10 1986 8 - - - - - Y1986=-197+45,44N-0.0998~; R.0.98 - * '1985= 3146C6.18N. Rz0.72 - I I I I 1 I I I 40 80 120 160 200 240 280 320 6 I I 1 I I I I I 0 40 80 120 160 200 240 280 320 Nitrogen (kg / ha) Nitrogen (kg/ha) Figure 5. Effect of N on the root yields of cassava in 1985 and 1986 averaged over three maize varieties in a humid environment IOkolu) Cassava root 1386 Humid Y1985 = I 170+ 0043N-0.00017Nei R10.89 '1986~ 13.20+0.0918~-0.00034~~; R-0.80 0 I I I I I I I 1 0 40 80 120 160 200 240 280 320 Nitrogen (kg/ha) Figure 6. Effect ofN on the m~t yielcls of cassava in 1985 and 1986 averaged over three maize varleues in a subhumid environment (Mokwa) Cassava root ( tons/ ha ) I I 7.5 I I I I I I I I I 0 40 80 120 160 200 240 280 320 Nitrogen ( kg/ha) Table 1. Summary of analysis of vnrhnce of three maize varieties gmwn with ca.~~ava at flve NIevels in three environment. for 2 years in sequential cropping systems (maize grain yield) Source of variation Warri Okolu Mokwa (perhumid) (humid) subhumid] Maize variety u u ** N levels (N) U .. U Years (YJ ** U U W x N N S NS NS MVxY N S NS N S n x Y U U U MVxNxY N S NS N S Note NS = not significant; *' = slgnlllcant at the 1% level Table 2. Total annual rainfall (mm) at experimental sites Year Warrl Okolu Mokwa Year (perhumid) (humid) (subhumid) mean 1985 2 987 1 735 1 158 1960 1986 2433 1 322 955 1 570 Mean 2 710 1 529 1 073 1 771 Table 3. Summary of analysis of variance of maize varieties grown with cassava at 5ve N levels in three environments for 2 years in sequential cropping systems (cassava yield) Source of variation Warri Okolu Mokwa (perhumid) (humid) subhumid) Maize variety N S NS N S N levels (N) t. U ** Years (YI u u .. MVx N NS NS NxY I I U MVxNxY N S NS NS N o t e NS = not signillcant: ** = signtflcant at 1% level: = signillcant at 5% level Table 4. Effect. of maize variety on cassava root yield at 3 location. Maize variety Warri Okolu Mokwa (perhumid) [humid) (subhumid) TZESRW TZSRW Hybrid SE Mean Cv (%) 18.9 22.8 31.6 Annual rainfall 2 710 1 529 1 073 Table 6. Production costs and returns from cassava/maize intercrop in humid. subhumid end perhumid ecologies Okolu Nitrogen Tms Tms Tms levels 305721 30572/ 30572/ TZESRW TZSRW hybrid Gross field benefit 20 2098.00 2576.50 2771.00 (cassava tdl50/ton. 40 2988.50 3481.50 3107.50 maize B1500/tonl 80 3571.00 3898.00 3947.00 160 3655.50 3779.50 3758.50 320 2264.00 2536.50 2864.50 Total variable costs 20 1369.60 1380.42 1483.26 40 1413.86 1433.71 1518.76 80 1494.65 1507.02 1612.95 160 1611.64 1642.57 1742.57 320 1838.39 1868.87 1971.13 Net benefits 20 728.40 1196.08 1287.74 Net benefits minus 20 713.84 1181.52 1273.18 cost of capital 40 1387.06 2020.99 1561.94 80 2025.07 2339.70 2282.77 160 1943.62 2036.69 1915.69 320 232.09 474.11 699.85 Note: Fertilizer cost = $145/ton ~unsubsidized); cost of capital = 40 percent of the msts of fertilizer and application as opportunity cost and rlsk premium Table 5 (continued) Nitrogen Tms Tms Tms levels 30572/ 30572/ 30572/ TZESRW TZSRW hybrid Gross field benefit 20 (cassava #150/ton. 40 maize #500/ton) 80 160 320 Total variable costs 20 40 80 160 320 Net benefits 20 40 80 160 320 Net benefits minus 20 cost of capital 40 80 160 320 Notr FertUlzer cost = $145/ton [unsubsldh~d): msl 01 wplt;rl = 40 perrent of the msts of lcrtlllxr and application as opponunlty cost and risk premium Table 5 (continued) Mokwa Nitrogen Tms Tms Tms levels 30572/ 30572/ 30572/ TZESRW TZSRW hybrid Gross field benefit 20 (cassava #150/ton. 40 maize #500/ton) 80 160 320 Total variable costs 20 40 80 160 320 Net benefits Net benefits minus 20 645.47 631.93 983.01 cost of capital 40 864.08 1134.10 1267.09 80 901.99 1685.32 2170.53 160 2306.92 2227.06 2180.40 320 2089.99 2543.66 3091.62 Note: FerllUzer mst = $145/ton (unsubsidlmil: mat of capital = 40 percent of the costs of fertilizer and appllcation as opportunity coat and rlsk premium Table 6. Economic indicator from intercropping cassava and maize in different ecologies Ecology Bendit cost ratios (cost = 1.00 Nitrogen Tms 30572/ Tms 30572/ Tms 30572/ levels TZESR TZSRW hybrid Okolu 20 1.53 1.87 1.87 40 2.11 2.43 2.05 80 2.39 2.59 2.45 160 2.27 2.30 2.16 320 1.23 1.36 1.45 Warri 20 1.67 1.72 1.70 40 1.40 1.37 1.41 80 1.2 1.44 1.43 160 1.30 1.52 1.56 320 1.24 1.19 1.34 Mokwa 20 1.48 1.47 1.67 40 1.63 1.81 1.85 80 1.64 2.14 2.36 160 2.48 2.41 2.29 320 2.22 2.45 2.63 REFERENCES Agboola. A.A., and G.O. Ohigbesan. 1974. The response of some improved food crop varieties to fertilizers in the forest zone of Western Nigeria. In Seminar on fertilizer use development in Nigeria. FA0 - AGL/MlSC/76/39/63-77. IITA. 1984. Anual Report 1984. Ibadan: IlTA Nwosu. N.A. 1977. Some indigenous cropping systems of Eastern Nigeria. Proc. 111 International Symposium on Tropical Root Crops. IITA. lbadan. 2-9 December 1973. ed. C.LA Leakey, pp. 293-398. Obigbesan. G.O. 1973. The influence of K nutrition on the yield and chemical composition of some tropical root and tuber crops. In International Potash Institute colloquium, Abidjan. Ivory Coast. 439- 451. Obigbesan. G.O.. and AA.A. Fayemi. 1976. Investigation on Nigerian and tuber crops. Influence of N fertilizer or the yield and chemical composition of two cassava cultivars (Manihot esculenta). J. Agric. Sci. 86 (2): 401-401. Odurukwe. S.O.. and O.B. Arere. 1980. Effect of NPK. fertilizers on cassava bacterial blight and root yield of cassava. Tropical Pest Management 26(4):391-395. Ofori. C.S. 1976. Effect of various N sources on the yield of cassava (Manihot esculenta Crantz). Ghana Journal of Agric. Sci. 9.99-102. Okeke, J.E.. G.D. Obigbesan and B.T. Kang. 1979. Effect of fertilizer application on nutrient concentration and growth relationships in cassava. Manihot spp.. J., Root Crops 11/21 1-7. Okeke. J.E.. B.T. Kang and G.O. Obigbesan. 1982. In Experimental Agriculture 18 (4). 403-41 1. Kang, B.T.. and G.F. Wilson. 1980. Effect of maize plant population and nitrogen application on maize-cassava intercrop. In Triennial Root Crop Symp. of the ISTRC, Ibadan, Nigeria, ed. E.R. Terry. KA Oduro and F. Caveness. IDRC-163e. 129-133. Takyi. S.K. 1974. Fertilizer. planting date and growth period effects on yield of cassava (Manihot esculenta Crantz) in three ecological zones in Ghana. Ghana Journal of Agricultural Science 7(3): 185-190. Wahua. TA.T. 1983. Nutrient uptake by intercropped maize and cowpea and a concept of nutrient supplementation index (NSI). Exp. Agric. 19: 263-275. TESTING THE FEASIBILITY OF ASSOCLATING CASSAVA AM) OTHER FOOD CROPS IN OIL PALM INTERCROPPING SYSTEMS LL Onwubqya and F.K. Eneh Abstract Two trials were set up in 1978 and 1980 to find the best method of intercropping oil palm with conventional food crops. One of the experiments comprised seven treatments of which ofl palm was the main crop. The second experiment also comprised seven treatments which involved only cassava and ofl palm. Results obtained from the first experiment showed that the mean height. girth and leaf number per palm were similar among the treatments with the exception of the oil palm/cassava and the oil palm/plantain mixtures which performed significantly below (P<0.05) the others. Sex ratios for the oil palm were slgniflcantly lower in these two treatments than in the rest. A similar trend was observed for the fresh fruit bunch (ffb) yield per hectare. In the second experiment virtually all the intercropped treatments significantly (P<0.05) outyielded the control treatment (no intercropping). On economic grounds. all the intercropped treatments in the first experiment showed positive net revenues suggesting that their total revenues were large enough to cover the costs involved in intercropping ofl palms with food crops. The ofl palm/cassava system appeared to be the most attractive in terms of oifsetting the establishment cost of the ofl palm. It is concluded that intercropping oil palms with food crops during the early years of ofl palm establishment. apart from having no significant adverse effect on ffb yield, is advantageous in reducing the establishment cost of ofl palms. ** . * ********* The need to intercrop young ofl palms arises from a number of factors of which the most important is the man/land ratio. In sparsely-populated areas in the ofl-palm belt, the need for intercropping ofl palms with food crops may not be very obvious. Intercropping ofl palm in such areas is motivated by economic factors relating to labor-use maximization and to the early food crop income which can be used to offset part of the initial investment in the ofl-palm main crop, or sustain the farmer and his family until the ofl palm begins to generate revenue. In areas where increases in human population have had a depressing effect on the avaflable arable land per man, intercropping is considered solely as a means of maximizing land use. Oil-palm fanners in such land-deficit areas are known to intercrop young oil-palms with food crops. In fact, World-Bank-assisted small-holder oil-palm projects in Nigeria slowed down in some States mainly beczuse oil-palm farmers refused to adhere to sole cropping systems as demanded by the Bank. These farmers insisted on intercropping young oil-palms with annual and short-period perennial food crops a s a means of dealing with the problem of dwindling arable land resulting from population pressure and urbanization. These farmers plant various kinds of food crops a s intercrops in young oil-palm fields at densities which they determine subjectively. In such intercropping systems. husbandry practices may call for increased labor input and the crop associations may intensify or reduce the incidence of pests and crop diseases and there may be possible changes in the soil chemistry as well. All this creates a need for vigorous and intensive research on the intercropping of oil palm and other crops. From the early 1940s to the middle 1960% experiments on the intercropping of food crops with oil palms were embarked upon by the Nigerian Institute for Oil Palm Research (NIFOR) in different locations (Spamaaij 1957). Although the results obtained were consistent in suggesting that oil palms could be intercropped with various food crops in the first few years of establishment, such experiments were not detailed enough to give rise to any recommendation package. However. a method of interplanting oil palms and cocoa has been recommended (Onwubuya. Irerniren and Kolade 1981). Thus. the objective of this study was to examine the merits of intercropping oil palms with food crops. Food crops which were commonly intercropped with oil-palms were selected to fit in with accepted local food cropping practices. Materials and methods Experiment 1 Field 64 at NIFOR's Main Station. near Benin City. Nigeria, was used for the experiment. Tenera oil palms had been previously planted in 1957 in the field which had been opened from a high secondary forest cleared by burning. These palms were felled in December 1977. the undergrowth cut and the area burned in early April 1978. The soil was acid sand with a pH of about 5.6 (Vine 1956). Extension work seeds (Dura x Plsferal used for the experiment were raised in the nursery in May 1977 and transplanted to the field in May 1978 at 8.8m triangular spacing. The area planted was 4.2ha with a total palm population of 630 of which 380 were experimental while 350 were guard palms. The experiment comprised seven treatments as follows: ofl palm intercropped with Puerarin phaselofdes as control ofl palm intercropped with maize and pineapple oil palm intercropped with cassava oil palm intercropped with yam oil palm intercropped with cocoyam oil palm intercropped with plantain ofl palm intercropped with yam, okra and spinach followed by cassava. The experiment was a randomized block design which was replicated four times. The plot size was 670m2. Thus, there were ten experimental palms per plot. A week after the planting of the oil palm, seeds of Pueraria phaseloides whlch were acid scarified were manually planted in Treatment A plots to provide a leguminous cover. Maize and pineapple were planted using zero tillage. The spacings for each of these two crops were 75cm between two rows and 25cm within a row, giving a population of about 3,500 stands per plot planted to alternate with each other. Cassava, yams and cocoyams were planted in their respective plots at l m intervals, each giving a population per plot of 670 stands. Some stands were very close to the palms. Plantains were planted at 3m intervals giving a population per plot of 74 stands. Okra and spinach were randomly planted in the G plots. When yam in Treatment G had advanced in growth, cassava was introduced in August 1978. A local variety of cassava characteriz~d by profuse branching right from the base was used. Replanting of Treatments D and E with yams and cocoyams respectively was carried out again. Measurements of height, girth and leaf number per palm were made at three-monthly intervals. Flowering observations started in September 1979 and the fresh fruit bunch (ffbl yield recording commenced in 1983. Experiment 2 Part of Field 64 was used for the experiment. The old palms were felled in December 1979, and debris packed and burned in March 1980. The area planted to ofl palms was 1.73ha with a total plant population of 266 of which 168 were experimental and 92 guard palms. There were seven treatments as follows: oil palm intercropped with Pueraria phaseoloides (control) local variety of cassava planted at lm intervals TMS 1525 cassava variety planted at lm intervals TMS 3021 1 cassava variety planted at l m intervals local variety planted at 2m intervals TMS 1525 casava planted at 2m intervals TMS 302 11 cassava planted at 2m intervals. TMS 1525 and TMS 30211 are high-yielding varieties that have an erectophile growth pattern and do not usually branch from the base. The three cassava varieties were planted 2m away from palm stands. The experiment was a randomized block design which was replicated three times. The plot size was 0.053ha [530m2). Recording of mean height and girth per palm started in 1980 while yield recording started in 1985. In both experiments all the cultural management practices were adopted. For example, in the control plots, the oil palms were ring weeded and the interlines slashed to knee level. The yam, cocoyam. cassava and pineapple plots were weeded and cultivated. The planted plots were also kept free of weeds but not cultivated. Fertilizer application to ofl palms was carried out on a routine basis. Records of labor and non-labor inputs were used to obtain the total production costs for each treatment over the intercropping period. Input and output prices used reflected the 1985/86 values. These prices were the mean values per kg of the various inputs or outputs obtained at interval times between June 1985 and May 1986. Labor input was costed at six Naira (Bi6.00) or US$1.7 per man-day. Results Experiment 1 Palm growth Tables 1 to 3 show the mean height, girth and leaf number per palm from 1978 to 1981. In each year. there were significant differences (P<0.01) between the mean heights per palm in different treatments. However, the superiority of one treatment over the others was not consistent from year to year. In 1978. Treatment C had the greatest height whfle Treatment A (control) had the least. There was a reversed trend from 1979 when the least height was recorded in Treatment C and the greatest in Treatments G , D and E in 1979, 1980 and 1981 respectively. The differences between the mean girth [circumference of the crown) per palm were also highly significant [P<0,01). Treatment C had the least girth. Differences between the number of leaves produced per plant per year were significant (P<0.05). Evidently more leaves were produced per palm in the intercropped treatments except Treatments C and F than in the control. Flower production The average male and female flower production per palm in each treatment is shown in table 2. Between September and December 1979, male flower production had commenced in all the treatments. The lowest value was recorded in Treatment C. Production of the female flower was low in all the treatments and in C and F, production was non-existent. Table 5 shows the percentage sex ratios obtained in 1980 and 1981. The differences between the treatments were highly significant (P.<0.01). In both years, Treatment E had the highest sex ratios while Treatments C and F had the lowest. Sex ratios obtained from Treatment A (control) in both years was considerably lower than those from the rest of the treatments except C and F and, in 1981, the differences between either E or C and A were significant (P<0.05). Fresh fruit bunch yield [FFB). The fresh fruit bunch [fib) yield per hectare is presented in table 6. In 1983 and 1985. the differences between the treatments were highly significant (P<0.01). The dflerences fell slightly short of signincance (P<0.01) in 1984. The differences between the yields averaged over three years were also highly signincant (Pe0.01). Treatments C and F consistently produced fewer bunches--about 60 percent of the quantity produced by Treatment B which gave the overall highest yield. However. the differences between Treatments A B. D. E and G were not significant (P<0.051. Measurements taken for height, girth and palm leaf number were inconsistent. Furthermore, the initial measurements of these growth parameters were not taken. Hence, their values are not presented. FFB yield. Harvesting of tlb started in 1985. The data for that year are presented in table 7. All the intercropped treatments except Treatment G highly significantly (P<0.011 outyielded the control (no intercropping with food crops). All the intercropped treatments but 6 compared favourably with one another irrespective of the spacing used. Discussion The results obtained so far indicate that intercropping of ofl palms with food crops has no adverse effect on the growth and yield of the former. As a matter of fact, it would appear that intercropping stimulates the growth and yield of oil palms. These findings agree with Toovey (19471, reported in Spamaaij (1957). for oil palm and food crops. Evans (1960) and Andrews (19721, working separately on a mlxture of cereal crops. arrived at a similar conclusion. The ffb yield differences between the control and Treatments B. D. E and G were not significant. However, the differences became more pronounced when compared with C or F. Cassava in C is characterized by fast growth. profuse and violent branching from the base and planting of cassava strictly followed the pattern usually adopted by peasant farmers so that. in most cases, it was very close to the palm stands. This. perhaps. produced a shading effect that affected the early growth and development of the palms. In Treatment G , where oil palms were planted in the first instance with yams there was probably sufficient penetration of light which enhanced the early growth and development of the palms. When cassava was later introduced in August of the same year, the oil palms had already established and growth and development were not impaired. Thus, the flb yield obtained from G was significantly higher than that from C. The implication was that light might have been limiting in Treatment C. This observation was confirmed from the second experiment where the planting arrangement for cassava was such a s to allow a greater amount of sunlight to be intercepted by the oil palms. Thus cassava, whether local or improved, did not adversely affect the yield of the oil palms. Similarly. light was. perhaps. a limiting factor in Treatment F where oil palms were intercropped with plantain. Hence, the yield of oil palms in this treatment was significantly lower than in the control. It was clear that ffb yields from Treatments B, D. E and G were similar to that from A (control) and, in some cases, better. This observation could be attributable to the cultural practices adopted in those treatments. While the leguminous cover and weeds in A were slashed to knee height. B. D. E and G plots were cultivated in the process of weeding. This practice probably increased aeration of the soil. moisture conservation and reduced competition from the weeds. A similar conclusion was drawn by Sparnaaij (1957) to the effect that it was cultivation per se applied to food crops intercropped with oil palms rather than the presence of those crops in the mixture that was responsible for the positive intercropping effect. All the treatments involving intercropping (Treatments B to G) showed positive net revenues at the end of the intercropping period (see table 12). This means that the total food crop revenue resulting from each treatment was large enough to offset the cost involved. The magnitude of the food crop net revenue is also an indication of the extent to which the establishment cost of the oil palms in a given treatment can be covered. The oil palm/cassava system with a net revenue of B15.263.00 per hectare over the period, appeared to be the most attractive in terms of offsetting the establishment cost of the oil palm. This was followed by Treatment B (oil, maize, pineapple mixture) and by E (oil palm/cocoyam). F (oil palm/plantain) and G (oil palm/yam followed by cassava). Treatment D (oil palm/yam) produced the least net revenue over the intercropping period. Yields of the intercropped food crops cannot, however, be considered in isolation from the effects of the intercrops on yields of oil palms. Treatment C. which is the most attractive in terms of revenue, reduced ffb yield by 31 percent or fd1.353.00. However, the resulting net revenue more than compensated for this decrease in ffb yield. Moreover. since the ffb yield can be improved by choosing a good planting time and arrangement for cassava such that shading of the palms will not occur, Treatment C can be recommended on economic grounds. Treatment F, which also significantly depressed oil-palm yields by 37 percent. produced a net revenue that was large enough to offset the loss in the yield of the oil palm main crop. As in C. the ffb can be improved by reducing the plantain population per unit area of land in order to ensure adequate interception of light by the oil palms. Treatments B and F increased the oil-palm yields by 13 and) 6 percent respectively and in addition produced positive net revenues. Thus. an intercropping system involving oil palm/maize/pineapple or oil palm/cocoyam can be regarded as economically sound at the given input and output quantities and prices (see tables 8- 12). Similarly. Treatments D and G can be regarded as economically feasible. Although. they produced lower ffb yields than the control, the differences were not significant (P<0.05). In conclusion, intercropping of oil palms with food crops in the early years of establishment does not adversely affect the oil-palm yield provided that, for some crops like cassava and plantain, planting is such as to allow sufficient interception of light by oil palms. In the case of cassava, relay planting, in which cassava follows later in the season, is suggested. For plantain, planting density per unit area of land should be lower than that recommended. Even with these results, comprehensive agro-economic and crop protection packages for improving oil-palm/food crop farming systems in smallholder agriculture are not yet possible. Thus, new experiments have been set up at NIFOR and elsewhere to test intercrop densities, the role of leguminous cover. weed control measures and the effect of livemulch-food crops in systems where oil-palm or coconut is intercropped with various food crops. Crop pests and diseases are also being hvestigated. The experiments are relatively new so that relevant data are not yet available. REFERENCES Andrews, D.J. 1972. Intercropping with sorghum in Nigeria. Expl. Agric. 80 Evans. A.C. 1960. Studies of intercropping. 1. Mslze or sorghum with groundnut. E. Afr. Agric. For. J. 26. Onwubuya. I.I., (2.0. Irerniren. and J A Kolade. 1981. Astudy of methods of interplanting oil palm and cocoa. In The Ofl Palm in the Eighties, ed. E. Pushparajah and Chew Poh Soon. Vol. 11. 425-433. Spamaaij. L.D. 1957. Mixed cropping in ofl palm cultivation. J. West Afr. lnst. Ofl Palm Res. 2 (7). 244-264. (Toovey. 1947. Seventh Annual Report, 1946-47. Ofl Palm Research Station, Benin City. ) Vine. H. 1956. Studies of sofl proffles at the WAIFOR Main Station and at some other sites of soil palm experiments. J. West Afr. Inst. 0fl Palm Res. 1(4), 8-59. Table 1. Mean height per palm (om) Treatment 1978 1979 1980 1981 Table 2. Mean girth per palm (cm) Treatment A B C D E F G LSD at P. c 0.05 CV Table 3. Mean leaf production per palm Treatment Initial 1978 1979 1980 1981 number A 9.5 B 9.6 C 9.9 D 9.9 E 9.7 F 10.0 G 9.3 LSD AT P. < 0.05 2'1 €'I s.1 €'I E'O 5.1 2.1 q+uour lad q~uour lad a~euraj usaly alew usaly Table 6. Mean % ecx ratio per palm Treatment LSD at P. c 0.05 Table 6. Mean tresh fruit bunch field per hectare (kg1 Treatment 1983 1984 1985 Mean 1983- 1985 A B C D E F G LSD AT P .< 0.05 Table 7. Mean tresh fruit bunch yield per hectare (kg1 in 1986 (2nd experlment] Treatment FFB yield LSD AT P . c 0.05 1,812.4 Tabla 8. Annual labor requirement. in man-days per hectare Treatment 1978 1979 1980 1981 A Oil palm sole B Oil palm Maize Pineapple C Oil palm Cassava D Oil palm Yam E Oil palm Cocoyam F Oil palm Plantain G Oil palm Yam Cassava T ab le 9 . L ab or a n d n o n -l ab or c o s ts o f th e fo od c r o p c o m po ne nt s o f th e in te rc ro pp in g sy st em s T re at m en t La bo r c o st s a t B i6 .0 0 pe r m a n -d ay N on -la bo r c o st s T ot al pr od uc - tio n c o st 19 78 19 79 19 80 19 81 19 78 19 79 19 80 19 81 (B i/ ha ) A O il pa lm so le - - - - - - - - - B O il pa lm M ai ze Pi ne ap pl e 41 4. 00 39 6. 00 39 6. 00 60 .0 0 25 0. 50 30 .0 0 30 .0 0 - 1, 57 6. 80 A C O il pa lm 'D C as sa va 39 0. 00 51 0. 00 39 0. 00 51 0. 00 16 6. 67 - 16 6. 67 - 2. 13 3. 34 D O il pa lm Y am 78 0. 00 78 0. 00 78 0. 00 - 1, 77 9. 20 1. 77 9. 20 1. 77 9. 20 - 7, 67 7. 60 E O il pa lm C oc oy am 35 4. 00 35 4. 00 35 4. 00 - 25 0. 00 25 0. 00 25 0. 00 - 1. 81 2. 00 F O il pa lm P la nt ai n 33 0. 00 28 2. 00 28 2. 00 13 2. 00 55 2. 00 - - - 1, 57 8. 00 G O il ~ a lm y aG C as sa va 1, 17 0. 00 51 0. 00 78 0. 00 - 1, 94 5. 87 - 1, 77 9. 20 - 6, 18 5. 07 Table 10. Food crop yields in kg per ha Treat- Cropping Yield in kg/ha Total ment system yield over the 1978 1979 1980 1981 period A Oil palm sole B Oil palm Maize Pineapple C Oil palm Cassava D Oil palm Yam E Oil palm Cocoyam F Oil plam Plantain G Oil palm Yam Cassava Table 11. Rlce per unit kg of the food crop Crop Price per kg (in Maize Pineapple Cassava Yam 0 .60 Cocoyam 0.45 Plantain 0.50 Source: Obtained Ironfood crop price sulvey canled out at intervals in 1985 and part of 1986. Table 12. Total revenue. total production cost and net revenues at end of intercropping period (1978 - 1981): food crops only Treatment Total Total Net revenue revenue cost of at end of (HI production period (HI A Oil palm sole B Oil ualm ~ a & e Pineapple 6,200.00 1.576.80 4.623.20 C Oil palm Cassava 7,396.50 2.133.34 5,263.16 D Oil palm, Yam E Oil palm Cocoyam 5.336.10 1.812.00 3.524.10 F Oil palm Plantain G Oil palm, Yam Cassava 7,538.55 6,185.07 1,353.48 TRENDS IN CASSAVA PRODUCTION IN THE ABAKALW AREA OF ANAMBRA STATE AND IMPLICATIONS FOR FARMERS E.C. Okorji and 0. Okereke Abstract Cassava, although one of the major staple foodcrops in the Abakaliki area of Anambra State, is not accorded as much attention as yams and rice. especially in terms of farm resource allocation. For instance, not only is the land area allocated to cassava significantly small in relation to other crops but it also more often than not has poor quality soil. This is mainly because cassava is regarded as a women's crop, since cassava products do not feature in any cultural activities (farming festivals, chieftaincy titles, etc.) that accord high social status to men in the area. The size of farm land cultivated per household has increased in response to rising demand for and prices of food products; yam-based crop mixtures (YBCM) and rice are favoured at the expense of cassava- based crop mixtures (CBCM), despite the comparatively high productivity of resources in CBCM. Cassava output, however. has increased. largely as a result of increased cropping density and the introduction of improved cassava varieties. Returns to cassava producers are relatively low because of the limited land area cultivated and the. dominance of low- yielding local varieties as well as the fact that a negligible proportion is processed as "gari" and starch which attracts higher prices per unit weight equivalent than cassava tubers. The mass adoption of improved cassava varieties, use of agro- chemicals and the provision of processing facilities will significantly increase the output, quality. storability and value of cassava products in the study area. Cassava (Manhot spp.1 is one of the major staple foodcrops in Anambra State. In some respect it is a substitute for yams (Dioscorea spp.). Although both cassava and yams are major staples. cassava production is usually not accorded as much attention as yams in terms of farm resource allocation. This situation is not so much based on food security and cash values a s on certain non-quantifiable values where yams, for instance. have a far greater sociocultural significance than cassava among the people (Okorji and Obiechina 1985). With the rising population and the recent ban on rice imports to Nigeria, there has been an increased demand for other staples including cassava products and, as a consequence, food products have risen in price. Changes are expected in the production trends of most staple food crops because of rising demand and prices particularly for such crops as cassava for which there is an increased availability of improved varieties. This paper thus examines the trends in cassava production in response to changes in economic and other factors. and the implications for farmers in the Abakallki area of Anambra State. Nature of data Most of the data used for this paper were obtained from two separate studies conducted in the Abakaliki area of Anambra State during the 1981/82 and 1983/84 farming seasons. The two studies involved sam- ples of 48 and 162 farming households respectively. Questionnaires were given twice weekly to the household heads and their first wives for the 12-month duration of each study. Information was collected on cropping systems, and the sources and pattern of farm resource allocation for different crops. In addition. physical measurements of the sizes of farms. inputs and output etc. were taken. This paper, however, is based largely on information from the 37 farming households that participated in both studies: reference is also made to other recent studies conducted in this area and elsewhere in the state. Cropping systems and resource allocation Cassava is commonly grown in combination with other crops such as yam. cocoyam, maize. legumes and vegetables. There were very few cassava sole crops grown during the survey period. In fields where cassava was grown in combination with yam and other crops. yams were considered the main crop. Such cropping is referred to in this study as a yam-based crop mixture (YBCM). Where there was no yam, cassava was considered the main crop and the cropping was referred to as a cassava- based crop mixture (CBCM). Rice was grown a s a sole crop in the area. There were more and larger fields where cassava was grown in combination with yams and other crops (YBCM) than cassava fields without yams (CBCM). On average, a household cultivated four YBCM fields. one CBCM and two rice fields. Though cassava was grown in YBCM. the cropping densities of it and the other intercrops in such fields were generally very low. Cropping density and output of cassava were significantly higher in CBCM than in YBCM (OkorJi and Obiechina 1985). Greater emphasis is therefore given in this paper to CBCM because such fields give a better representation of cassava production in the area. Land preparation for cassava production started early in the year, but planting began mostly in March and April. Cassava stems were planted on mounds with an average of about four stands per mound in mixed-crop fields. and between five and six stands in the few cassava sole-crop fields; no defined planting distance between crop stands was maintained in the fields. The intercrops in CBCM fields (maize. cocoyams. legumes and vegetables) were planted three to four weeks later, when the cassava had sprouted. and they were also staggered on the mounds with no precise planting pattern. This system differs from that of other areas of southeastern Nigeria, where cassava is grown either on ridges or on flat soil. with fairly uniform spacing. In general. not much attention was given to cropping patterns and other management prac- tices adopted for cassava production in the study area, more for socio- cultural than for economic reasons. The pattern of resource allocation among crops was such that CBCM was least favoured. Table 1 shows resource allocation among crops by household for the 1981/82 and 1983/84 seasons. For each of the two periods surveyed. less than 10 percent of farm land was devoted to CBCM. Uplands and swamp land were available in the study area. Rice and most YBCM fields were located in swamp land whfle CBCM fields were limited to uplands. Swamp land has relatively high fertility because it receives alluvial deposits from river floods. Bachmann (1981) estimated a yield of 7.6 tons per hectare (t/ha) in swamp land and 4.2 t /ha in upland flelds for yams. and 11.6 t/ha in swamp and 9.5 t/ha in upland flelds for cassava in Nteje. Anambra State. Thus. not only was the smallest area of farm land allocated to production of CBCM relative to other crops but it was usually of poor-quality soil. The allocation of labor and capital resources to CBCM followed a similar pattern to land allocation in the study area. This pattern of resource allocation to cassava determined cassava production trends fairly accurately since it related to the value set on cassava by the people. PmductiviQ of cassava enterprise Table 2 shows a summary of cost-return analyses for YBCM, CBCM and rice enterprises for the survey period. CBCM recorded the highest net return per hectare for the two periods compared with other crops in the study area. Resource inputs (land. labor and cash) were most productive in CBCM compared with other crops, suggesting that farmers could perhaps optimize farm returns if the required attention were given to CBCM. A comparison of the cost and return involved in the production of CBCM and sole crop cassava showed that although the output of cassava per hectare was higher in the cassava sole crop, gross and net returns were higher in CBCM. This partly explains why mixed cropping is preferred to sole cropping for cassava production in the area. It is also possible that mixed cropping is preferred because it provides a n insurance against loss of heavy capital and labor input if the the main crop fails (Norman 1971; Nweke 1980). The traditional system of allocating a comparatively small proportion of household land to women also contributes to the dominance of CBCM over cassava sole crops in the study area. At present. land does not constitute a major constraint on farming in the area. but the amount generally devoted to women's crops. of which cassava is the most important. is extremely limited. This stereotyping of food crops in the study area according to sex thus influences the pattern of resource allocation and is an important element to consider in the introduction of modern crop production technology. Trends in Cassava Production in the Study Area Supply has increasingly lagged behind demand for food products, so prices for food products have risen in the study area. For example, price increases of 50 percent were recorded for vegetables. cocoyams and maize; 67 percent for legumes; 75 percent for cassava; 14 percent for rice; and 110 percent for yams in 1983/84. compared with their 1981/82 average market values. In general. the size of farm holding cultivated per household increased mainly because of the increase in demand for and prices of food products. These increases did not. however, achieve the desired impact on the pattern of resource allocation to cassava in the study area during the survey period. Contrary to expectations, the land area allocated to CBCM decreased, while that for YBCM and rice increased, even though resource inputs proved most productive in CBCM. Efforts made by farmers to increase cassava output centered largely on increased cropping density and the use of improved cassava varieties (see table 3). The extent to which either of these contributed to the increased yield recorded in this study was, however. not determined. Improved cassava varieties constituted between 16 and 20 percent of the cassava stems used by the household. Many farmers still have no access to improved cassava varieties, mainly because the quantity provided by the State Ministry of Agriculture and Natural Resources was grossly inadequate, and the varieties available at local markets were sold at exorbitant prices. The bulk of the marketed cassava was sold as tubers in spite of the higher prices per unit weight equivalent of such processed products as gari. The processing of cassava is yet to be popularized in the study area. The higher return from gari than from cassava tubers is, however. inducing more farmers to embark on gari processing which until recently was exclusively undertaken by middlemen from the urban centers. An increased output of cassava will increase the prospects for gari and to some extent starch. since most of the cassava tubers surplus to household requirements would be processed into these forms. Consumption of gari, for example, is not common among rural people, who prefer cassava fufu, tapioca and cassava flour meals in addition to yams. Although a higher percentage increase was recorded per unit price of yams, it seemed that more than a proportionate amount of household farm resources was allocated to this crop. Given that less than 20 percent of the total yam output was sold on average per household, the preference given to yams a had negligible impact on improving farmers' economic position; about 40 percent of the total output of yam was consumed in the home. 30 percent kept for replanting. and about 10 percent for use in cultural celebrations and for gifts. This is even more apparent when it is realized that the cropping densities and hence output of the intercrops in YBCM were restricted in order to minimize competition with yams, the man's crop, even though relatively large proportions of the total intercrop output were marketed by the household. This situation becomes dillicult to justily, considering that such cassava products as gari and starch recorded more than a 150 percent rise in unit prices [see table 4). Implications for farmers Increased cassava output will result in increased sales and revenue for the household. At present cassava output per household in the study area is limited by the constraints imposed on the resource allocation to CBCM from which about 92 percent of the total cassava output/ha is derived. Revenue from cassava and hence farmers' wellare can be improved in the study area, mainly by the mass adoption of high-yielding varieties and improved cultural practices [Gebremeskel et al. 1986). The present yield of about 7.15 t/ha is signilicantly low (see table 5) compared with yields of between 13 to 22 t/ha for dilferent IITA cassava varieties [Ekpere et al. 1986). Increased cassava output will be achieved when farmers in the area are encouraged to apply fertilizer and other agrochemicals to cassava fields; most farmers, for instance, did not apply fertilizer to their yam and cassava fields because they believed that such chemicals were harmful to root and tuber crops and might also be so for human beings who eat them. Increased revenue from cassava products may consequently influence the pattern of resource allocation so that a household may decide to allocate more land and other resources to cassava cropping. This increased revenue may, however, not be sustained since increased output and supply may eventually bring the price of cassava down. Though farmers regulate the supply of cassava to some extent by harvesting at need. the fields will be required for subsequent cultivation and they will be compelled to harvest and sell their cassava; moreover. the opportunity cost of resources invested in unharvested cassava may be higher than the extra revenue from delayed sales. The welfare of cassava producers in the study area could be sustained and possibly improved by developing the different cassava ~roducts . The Drovision of cassava-orocessine facuities should be r embodied in an$ cassava production' packag 9 >8 Humld 7 - 9 6 - 8 Transition - 5 - 6 Subhumid 4.5- 7 4 - 5 Dry subhumid - - sa&Iid 2 - 4.5 2 - 4 Arid < 2 1 - 2 In some cases (eg. Le Houtrorl & Popov). rainfall needs only to exceed 0.5 or 0.35 PET to classify a month as humid. In view of the range of criteria, the choice of which classification to use depends on which boundaries are most convenient and relevant to the specif~c purpose at hand. based analysis).4 The contour lines for this measure are shown in figure 3. The 270-day and 180-day contours appear to correspond most closely with the West African vegetation zones of figure 1, though again there are a number of local discrepancies. A summary of the corresponding values for these moisture variables and the vegetation zones appears in table 2 below. These are based as far as possible on the boundaries shown in figure 1, and it should be remembered that the numerous definitional variations mentioned above imply that any selection is somewhat arbitrary and that the correspondence among different measures is not exact. The degree of discrepancy also increases as one moves away from West Africa. since altitudinal and latitudinal effects have greater influence. As was mcntioned, the main advantages to vegetation-based classification are that it is very amenable to direct observation and that it synthesizes the elTects of both soil and climatic conditions as well as of human activity. Among its main disadvantages, however, are that it is fairly specific to West Africa (and to a somewhat lesser extent to Central Africa) and that it is not associated with a quantitative measure and thus is imprecise and subject to interpretation. Conversely, the ease of quantification and the wider areal applicability are among the main advantages of climate-based classifications. Such measures as total annual precipitation. annual excess of precipitation over P m , number of dry (or wet) months, and length of the annual growing period can be used for classification and comparison of diverse regions, and-of growing current irnportallce-for storage in computerized data bases. This latter capability is likely to play a critical role with the spread of computer mapping techniques. In sum, the following general principles can serve as guidelines for selection of an appropriate basic agroecological classification scheme: (a) When concerned just with West and Central Africa, a vegetation-based classification is usually appropriate. The actual categories will depend on the specific use intended, but the forest. savanna, and Sahelian zones will be central. In some instances categories for the derived savanna, wetland areas. mid- or high altitudes. and/or other subdivisions of any of the wnes may also be included. The growing period is deflned as beginning when rainfall exceeds 0.5 PET and ending when rainfall falls below 0.5 PET plus stored soil moisture (assumed to be lOOmm: FA0 1978). (b) When East Africa and other areas are also included, the best classification is generally one based on moisture balance plus altitude. The categories would be defined by a matrix with altitude in one dimension (lowland, highland, etc.) and moisture balance In the other (humid, subhumid. etc.). Again. the precise number of categories and boundaries need to be defined in relation to the purpose at hand. (c) If areas outside the intertropical zone are included. a latitudinal parameter will also be useful (tropical and subtropical categories often being sufficient). (d) For a computerized data base. continuous quantitative parameters are most useful. Ideally. these should be stored, as far as possible, as disaggregated, non-interpreted point data-eg, mean monthly (or weekly) precipitation. mean high and low temperatures. windspeed, latitude and longitude, altitude, etc. These can then be transformed to find such values as PET and interpolated to provide surface contours. Boundaries of zones can be defined in relation to specific parameters as needed (see Bunting 1987). REFERENCES Bunting, AH.. ed. 1987. Agricultural Environments: Characterization, classification, and mapping. Proceedings of the Rome workshop on agro-ecological characterization, classification and mapping, 14-18 April. 1986. Wallingford, UK: CAB International. Davies. H.RJ. 1973. Tropical Africa: An Atlas for Rural Development. Cardiff: University of Wales Press. Davies, J.A., and P.J. Robinson. 1969. A simple energy balance approach to the moisture balance climatology of Africa. In Environment and Land Use in Africa. ed. M.F Thomas and G.W. Whittington. pp. 23-56. London: Methuen. Food and Agriculture Organization. 1978. Report on the Agro- Ecological Zones Project. Vol. I: Methodology and Results for Africa. Rome: FAO. Food and Agriculture Organization, and United Nations Educational Scientific and Cultural Organization. 1977. Sofl Map of the World. Vol. VI. Africa. Paris: UNESCO. Harrison Church. R.J. 1980. West Africa: A Study of the Environment and of Man's Use of It (8th edition). London: Longman. Keay. R.W.J. 1959. Vegetation Map of Africa South of the Tropic of Cancer. Oxford: Oxford Unversity Press. Kowal. J.M., and A.H. Kassam. 1978. Agricullural Ecology of Savanna: A Study of West Africa. Oxford: Clarendon Press. Lawson, T.L.. J.S. Oguntoyinbo, and 0 . Ojo. 1979. Agroclimatic Conditions of West Alrica. Paper presented at lITA Annual Research Conference on Soil and Climatic Resources and Constraints in Relation to Food Crop Production in West Africa, Oct. 15-19, 1979. Ibadan: IITA. Le Houerou. H.N., and G.F. Popov. 1981. An Eco-Climatic Classification of Intertropical Africa. Rome: FAO. (FA0 Plant Production and Protection Paper. 3 1 .) Morgan, W.B.. and J.C. Pugh. 1969. West Africa. London: Methuen. Papadakis. J. 1966. Crop Ecologic Su~vey in West Afiica. Vol. I ; Vol. 11: Atlas. Rome: FAO. Troll. C. 1965. Seasonal Climates of the Earth. In World Maps of Climatology, ed. E, Rodenwalt and H. Jusatz, pp. 28-40. Berlin: Springer. UNESCO/AETFAT/UNSO. 1981. Vegetation Map of Africa. Paris: UNESCO. AGBOOLA. A. Dept of Agronomy University of Ibadan Ibadan. Nigeria AKINLOSOTU. T. IAR&T. Obafemi Awolowo Universitv PMB 5'029. Mwr Plantation Ibadan, Nigeria AKINYEMIJU. 0. Dept of Plant Science Obafemi Awolowo University Ile-Ife AU;HALI. A.M. lnternatlonal Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria m, S.W. IRA-Ekona PMB 25. BUEA Southwest Province. Cameroon AMBE. T.J. IRA, Ekona PMB 25 BUEA Southwest Province, Cameroon ARTHUR, J. Crop Science Department University of Ghana Legon. Ghana ASADU. C Department of SOU Sdence university of Nigeria. Nsukka Nigeria AWORH. 0. Department of Fmd Technology University of ibadan Ibadan. Nigeria BESONG. M. IRA, Ekona PMB 25. BUEA Southwest F'rovlnce. Cameroon CIMBA. L. PRONAM. Inera M'vuazi, Care Mveke Bas-Zaire, Zaire DASHIELL. K. International Institute of Trovical Aericulture - PMB 5320 Ibadan. Nigeria DOROSH. P. International Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria ENEH. F. ~grlcultural Economics Division NlFOR PMB 1030 Benin City. Nigeria EZEDINMA, C.I. Department of Agricultural Economics University of Nigeria, Nsukka Nigeria EZUMAH. H. International Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria GHAHIG(. K.J.N. University of Cape Coast Cape Coast. Ghana international Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria HAHN, N.D. International Institute of Tro~ical Aericulture PMB 5320 Ibadan. Nigeria International Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria IKEORGU. J. National Root Crops Research Institute, Umudike Igbarlam Substation P.O. Box 142. Urnudike. Nigeria IKPI, A. Universitv of Ibadan Ibadan, ~ ige r i a IWUEICE. C. Federal Agricultural Coordinating Unit (FACU) PMB 1210 Benin City. Nigeria KARUNWI, A. International Institute of T r o ~ i c d Aaiculture PMB 5320 - Ibadan. Nigeria LEMA, RM. Internatlonal Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria LUTAIADIO. N.B. RAV. BP 11635 Kinshasa, Zaire International Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria NDIBAZA. R lnternational Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria NEUENSCHWANDER P. lnternational Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria NGEVE. J. Institute of Agronomic Research Nkolbisson BP 2067 Yaounde. Cameroon NWANA. E. Imo State University Oklgwe, Nigeria lnternakonal Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria OKEKE. J.E. National Root Crops Research Institute. Umudike Umuahia. Nigeria OCUNSUNMI, LD. c/o International Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria OKIGBO. B.N. International Institute of Tro~ical Mriculture v PMB 5320 Ibadan. Nigeria OKOIWI, E.C. Department of Agricultural Economics University of Nigeria. Nsukka Nigeria OLORUNDA. AO. Department of Food Technology -- ~ a c u l l ~ of Technology University of lbadan Ibadan. ~ iger ia OLUBODE. S.O. lnternatlonal Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria OLUKUNLE. E. D e h e n t of Aericultulal Economics ~ n i v e r s l t ~ of ~ i ~ & a . Nsukka Nigeria 40. 0NAEQLU.AO. C/O International lnstitute of Tropical Agriculture PMB 5320 Ibadan, Nigeria lnternational Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria ONWUBUYA. 1.1. Agronomy Division. NlFoR PMB 1030. Benin City Nigeria OPOKU-ASIAMA. Y. School of Agriculture University of Cape Coast Cape Coast, Ghana ORKWOR G. National Root Crops Research Institute. Umudike PMB 7006. Umuahia Nigeria OSINAME. 0. KAV/USAID Kinshasa. Zaire OSIRU. D.S.O. International Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria OYEDOKUN, J.B. IAR&T. Obafemi Awolowo University PMB 5029. Moor Plantation Ibadan. Nigeria POUBOM (nee NGUNDAM). F.C. IRA. kona PMB 25, BUEA Southwest Province. Cameroon SINGH. B.B lnternational Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria 50. SINGH. S.R. International Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria 51. SMITH. J. International Institute of Tropical Agriculture PMB 5320 Ibadan. Nigeria 52. SPENCER D. International Institute of Tropical Agriculture PMB 5320 Ibadan, Nigeria 53. UNAMMtLR National Root Crops Research Institute. Umudike PMB 7006 Umuahia Nigeria 54. WOLDETATIOS. T. IRA. Ekona. PMB 25. BUEA Southwest Province. Cameroon ADDENDA: ANNEX 2 Table 2: Major ecozones and characteristics fn West Africa Zone Number of Mean Growing Main Soils humid annual period months rainfall (days) 1. Forest 7-9+ 1400-4000+ 270-365 Mostly acidic (ultisols (mostly and oxisols): some unlrnodal) nonacid (inceptisols. entisols, vertisols, alA- sols, etc.) 2 Derived 6-7 1300-1500 240-270 Moderately leached Savanna (bimodal. soils (alfisols, some some areas) ultisols, etc.) 3, Southern 5-6 1200-1500 190-240 Mainly alfisols and Guinea (partially related soils: acidic Savanna bimodal) ultisols and d s o l s in some wetter areas: also entisols and vertisols in some areas. 4 Northern 4-5 880-1300 140-200 As above, with greater Guinea [unimodal) proportion of nonacid Savanna alfisols. etc.. and fewer areas of acidic soils. 5 Sudan 2-4 500-880 90-140 Alflsols and some drier Savanna (unitnodal) arldlsols, etc. Sourcee: Papadalds (1966): Davies (19731: Harrlson Church (1980); FA0 (1978): Kowal and Kassam (1978); Lawson (1979); Le Houimu and P o p (1977) Fi g.1 . Ve ge ta tio n Zo ne s in W es t a nd C en tra l A fri ca