Farming systems newsletter

Material Information

Farming systems newsletter
Caption title:
International Maize and Wheat Improvement Center -- Eastern Africa Economics Programme
Place of Publication:
Nairobi Kenya
The Centre
Publication Date:
Copyright Date:
Physical Description:
v. : ill., forms ; 30 cm.


Subjects / Keywords:
Agricultural systems -- Periodicals -- Africa, Eastern ( lcsh )
serial ( sobekcm )
periodical ( marcgt )
Spatial Coverage:


General Note:
Description based on: No. 26 (July-Sept. 1986); title from cover.
Statement of Responsibility:
CIMMYT Eastern Africa Economics Programme, International Maize and Wheat Improvement Centre (CIMMYT).

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
18247451 ( OCLC )
sn 90040819 ( LCCN )

Full Text
No. 34

With Compliments
P.O. Box MIP 154 Mount Pleasant HARARE ZIMBABWE Telephone 303544

Page 1
- Workshop on cereal/legume intercropping ................ 2
- Provisional Training Programme MAMC ..................... 2
- IITA Postdoctoral Fellowships .......................... 5
- Reviews ................................................... 6
- Testing a Package of Innovations around Bako: Its Impact and Implications for Future Research Legeese Dadi and Asfaw Negassa ......................... 8
- Maize Technology Research in Mangwende, a high Potential Communal Area Environment in Zimbabwe Part 1: Developing a Research Agenda Enos Shumba ....................... ...................... 12

Page 2
This workshop is sponsored by CIAT, CIMMYT and the
Government of Malawi and will be held in Lilongwe, Malawi from 23-27 Janauary, 1989. Requests for nominations have been sent to National Programme Directors to whom enquiries should be directed. Approximately 45 participants are expected from within and outside the region.
(PC44 9th January to 10th March)
(PC45 29th May to 28th July)
Continuing our successful management development programme, these two nine week courses are aimed at middle managers working in the agricultural sectors of their countries in government, parastatal and commercial organisations. The programme is designed to help managers achieve their full potential. by challenging them to examine their own attitudes critically, developing their managerial skills and enhancing their awareness of the "manager's world".
This four week course, run in conjunction with Silsoe
College, U.K., is for experienced irrigation managers and
planners. The aim of the course is to encourage those involved with irrigation projects to develop their managerial abilities in order to make more effective use of the resources human,
financial and material available to them. Visits to the many
varied irrigation projects in the area form an integral part of the course.
(CM4 20th March to 14th April)
This four week course has the specific objective of giving middle and senior managers an awareness of how computers can assist in organisations. The course addresses both the technological aspects of computerisation and the implications for management in the introduction of information technology. The course participants will make intensive use of the Centre's microcomputers and the wide variety of application programmes available.

Page 3
(DM2 24th April to 19th May)
A four week course run in conjunction with the Disaster Preparedness Centre of the Cranfield Institute of Technology those with responsibilities for disaster preparedness planning and for disaster management.
(FPM41 7th August to 1st September)
This is a four week course run in conjunction with the Food Studies Group, Oxford, U.K. Its central concern is the long term solution to famine and the course is designed to provide a forum whereby senior people who are directly responsible for food policy can discuss the issues and problems involved. Food studies made in many of the participants' countries provide the material for this course.
A four week course run in conjunction with the Economic Development Institute (EDI) of the World Bank, aims to enhance the skills and knowledge of participants in preparing and analyzing agricultural and rural development and managing such projects through an understanding of the relationships between management issues and project design. The course is designed for experienced managers/officers in ministries of agriculture, planning and related sectors.
(RCM4 11th September to 6th October)
A four week course concentrating on development finance and credit for commercial and Central Bank executives and financial managers in government, parastatal or private sector
organisations. The course is designed to help participants become more effective and responsive to the challenges facing their organisations by creating an awareness of the role of finance and credit in the development process, improving skills and judgement in appraisal techniques and emphasising the importance of the human relations in dealing with clients.
(MRD2 16th October to 15th December)
The aim of this nine week course is to further develop the participants' management skills and awareness, to improve the

Page 4
performance of rural development project in the light of the current "implementation crisis" facing such projects. The course is designed to be of value to personnel from a broad range of experience in different specialisations, all of whom with
responsibilties for rural development in their countries including: extension officers, project planners, administrators and managers, community development officers, health officials, home economists, cooperative managers and educational administrators.
For further details, or early nominations, please contact: The Director, MAMC, P.O. Box 20, MHLUME, Swaziland. Telephone: Swaziland 31133/31334. Telex: 2320 WD.

Page 5
The International Institute of Tropical Agriculture (IITA) provides outstanding scientists the opportunity to start a career in international agricultural research. This program is aimed at encouraging young men and women recently trained in the agricultural sciences to conduct research in association with ITTA scientists on problems of tropical agriculture in Africa.
About IITA:
IITA was established in July, 1987 as the first major African link in an integrated network of international research and training centres located throughout the developing regions of the world. IITA serves as a link in a chain that stretches from basic research laboratories of the first world through national agricultural research and extension agencies, to the ultimate target the African family farmer. Primary research emphasis is on developing productive and sustainable farming systems for the lowland and sub-humid tropics of West and Central Africa.
Research at ITTA is conducted on three interrelated topics, with feedback at all levels:
1) Systems of natural resource management.
2) Improved crop varieties that can stabilize and increase
production with minimal damage to the environment.
3) The products of the first two sets of research converge
in the common goal of the Institute as a whole, the development of sustainable and productive cropping systems compatible with the resources and objectives of
the family farmer.,
The first of these research components is natural resource management. This involves the search for sustainable alternatives to shifting cultivation. The four major commodity improvement programs in the Institute conduct research on cassava, yams and plantains; maize; rice; cowpeas and soybeans.
Principal financing of the Institute is arranged through the Consultative Group on International Agriculture Research (CGIAR). The CGIAR is an informal association of governments, international and regional organizations, and private foundations set up in 1971 and it is dedicated to supporting a system of agricultural research centres around the world.

Page 6
The Fellowship:
The basis for the selection of post-doctoral fellows will be the scientific quality of the proposal and its relevance to the goals of IITA. The competition for fellowships will be open all year, with selection being conducted twice in each year (March and September). Fellowships will be open to applicants world-wide and will initially be tenable for one year, renewable for a second year if the performance of the fellow is satisfactory.
Competition is open to any outstanding scientist (with a Ph.D. or equivalent) recently qualified in any field of research related to IITA's research activities. The terms of the fellowships will include payment of competitive salaries in US dollars, medical and other insurance schemes, housing, and other appropriate benefits.
Interested scientists should send a brief proposal on the research areas they wish to pursue in addition to their curriculum vitae, university transcripts and addresses of referees to the Manager, Human Resources at either:IITA c/o L.W. Lanibourn & Co. Ltd., 26, Dingwall Road, Croydon, Surrey, CR9 3EE, England. OR IITA, PMB 5320, Ibadan, Nigeria.
William M. Rivera and Susan G. Schram, editors, Agricultural Extension Worldwide: Issues, Practices and Emerging Priorities. Croom Helm: London, New York, and Sydney, 1987 (294 pages).
The subtitle describes the three-part presentation of 16 papers by "big name" authors in a colloquium at the Center for International Extension Development, University of Maryland. Attention is immediately invited by the views of distinguished representatives of the major international donor agencies, American and foreign universities, and consultancy organizations. The book is a well-organized storehouse of wide-ranging information, experience, and expert opinion which should appeal to policy makers, development professionals, extensionists, and university scholars. The editors' introduction and epilogue add order, cohesion, and a summary with conclusions.
1986 Supplement: Publications of the International Agricultural Research and Development Centers. The International Rice Research Institute: Los Banos, Laguna, Philippines, 1986 (167 pages).

Page 7
This is a valuable, conveniently-arranged compilation of new files not included in the larger 1985 catalog.
R. G. Lowe, Agricultural Revolution in Africa? Macmillan:
London, 1986 (295 pages).
The question mark is an important part of the title, as is the subtitle: "Impediments to Change and Implications for
Farming, for Education and for Society." The author draws heavily on his experience in West African countries, particularly Nigeria as perhaps the first "to experience the trauma of agrarian revolution," switching from peasant to commercial farming. Insights, laced with history and scholarship, are sometimes acerbic, as in the analysis of universities and their uses and abuses.
David Vance Youmans, Extension Education Impacts of Farming Systems Research in Lesotho. Research Institute for Education Planning, University of the Orange Free State: Bloomfontein, South Africa, 1986 (293 pages).
The work (in typescript form) by a U.S. extensionist working in Lesotho concludes that extension education has an appropriate and essential place in all stages of FSR "if desired change is to take place and endure." This conclusion is bolstered by many data and analyses. On FSR itself in Africa, the conclusion is that it assumes "too much with regard to the homogeneity of small farmers, their access to resources, their likemindedness, and their propensity toward adoption of 'packages' or integrated
farming schemes." Instead, farmers are heterogenous, quite unlike in behaviour, and adopt innovations, not in packages, but in
pieces on the basis of values, risks, costs, and personal inclinations. (Part of these research results were presented in the November, 1986 issue of INTERPAKS INTERCHANGE.)
Patricia Garrett, Jorge Uquillas and Carolyn Campbell,Interview Guide for the Regional Analysis of Farming Systems. Program in International Agriculture, Cornell University: Ithaca, New York, 1987 (90 pages).
This publication concerns the "diagnostic phase of farming systems research ... intended to identify targets of opportunity for research and extension." It is an "interview guide a prepared to help multidisciplinary teams analyze a region and identify needs for the development and dissemination of appropriate technologies."

Page 8
By Legeese Dadi and Asfaw Negassa
Assistant Research Officers, Bako Research Centre, P.O. Box 3, Bako, Ethiopia.
In order to feed the rapidly increasing rural and urban population, sustained agricultural production is required. However, in developing countries the food production growth rate lags far below the population growth rate because of various natural, socio-economic and institutional factors under which the farmers are operating. In view of these problems there is an urgent need to improve the agricultural sector, especially small farmer productivity. Improving productivity of small farmers, where resources are often limited, is very difficult. This is because small farmers are operating under different natural circumstances (climate, soils) as well as economic circumstances (resource levels, extension, access to inputs). Because of the variable circumstances faced by small farmers, a given technology does not work equally for all farmers in a region. At the same time it is impossible to design technologies for each farmer separately. Hence a well focused and target oriented research and extension program which seeks to design technologies which work more or less equally for a group of farmers facing similar circmstances (target groups) is needed to improve the productivity of small farmers.
At Bako research centre many technologies have been developed in the past 20 years. Efforts were made to transfer the technologies to farmers in order to increase productivity at the farmer level. In line with this a package testing program was launched around Bako to assess and evaluate the performance of the research results on farmers fields.
Components of the package were selected and formulated for four crops by multidisciplinary research teams. The packages were tested for seven years at least, at 5-10 location each year. For the first two years the package testing was carried out on individual farmers fields. Then it was moved to producer cooperative farms and communal fields belonging to peasant associations.
Improved varieties of maize, sorghum, peppers and sweet potato were tested with their corresponding selected improved management practices. The improved management practices included adequate land preparation, fertilizer rate and type, planting method and

Page 9
time, seed rate and recommended crop protection practices.
Inputs required were made available by researchers and farmers. Researchers provided technical advice, improved seeds, planting materials and insecticides without payment. Land, labour, draft power and fertilizer were provided by farmers.
The package performed well on farmers fields. On average yields of 4.5 t/ha from maize, 2.9 t/ha from sorghum, 1.6 t/ha form pepper and 16.1 t/ha from sweet potato were obtained. Maize, sorghum, and pepper yields were increased by 137%, 93%, and 167% over farmer yields. However, the additional costs to attain these yield levels were found to be high.
Out of the package, the variety was widely accepted around Bako by producer cooperative farmers and individual farmers. Almost all producer cooperative farmers around Bako accepted the improved maize variety. Most of the individual farmers acquired seeds of the new variety by exchanging local seeds for the improved variety seeds from test farmers. Few farmers purchased new seeds from the market. Pepper and sweet potato varieties introduced were also accepted by some producer cooperative farmers and individual farmers. At present, the planting of maize in rows is being accepted because of advantages such as higher yields, ease of weeding, etc. Nearly all producer cooperatives and 34 percent of individual farmers plant their maize in rows.
Though improved varieties and row planting of maize were generally accepted by farmers, some of the package components were either modified or totally rejected. These included fertilizer rate and type, weeding time and frequency and the insecticide recommendation for termite and maize stalk-borer control. Legitimate reasons for this were given by farmers. First, the recommended fertilizer rate and type for a hectare of land needed more than 80 percent of a typical farmer's annual cash income. Fertilizer application for small farmers is a risky undertaking because of the high risk of crop failure. Farmers try and avoid this risk by applying little or no fertiliser. Second, the weeding recommendation was rejected because it demanded an additional 28 man-days/ha of labour at a time of high demand for labour during June-August. Around Bako at this time, teff and noug land preparation and planting, pepper transplanting and weeding overlap with maize and sorghum weeding. *Farmers were not able to provide the extra labour for maize and sorghum weeding during this period to conform with the current recommendation. Third, the insecticides were costly and not available. Insecticide applicators were also not available. The application method recommended for DDT (putting a pinch of DDT on each plant) was very tedious and impractical.

Page 10
Farmers practising subsistence agriculture do not always allocate resources to maximize yields. They aim at sustained production to secure a family food supply and will only accept innovations if the yield advantages outweigh the extra costs. Thus package of innovations believed to greatly increase yield might not be accepted by farmers because of their aversion to risk, limited cash and inadequate institutional support. The acceptability of a whole package or its components should not be judged solely from the view point of maximising production. Attention should also be given to the goals farmers have, their capabilities and limitations. Neglecting these factors will result in the modification or rejection of technologies. The fertilizer and crop protection recommendation included in the described package needed excessive cash investment and labour. These were not available and the recommendations were not accepted.
In the light of these problems the Farming Systems Research (FSR) team at Bako research center divided farmers into two target groups, based on their access to inputs, institutional support, crop management and difference in development opportunities. These were producer cooperatives and individual farmers. The team modified technologies to suit the institutional and economic circumstances of each target group. Thus different experiments on fertilizer rates were designed for the two target groups.
Risk analysis, economic analysis and farmer assessment were introduced to better evaluate experimental results. In the past recommendations were entirely based on technical responses neglecting other factors.
To examine why farmers rejected the weed control recommendations, a trial looking at weeding twice (current recommendation) was carried out on farmers fields. It was confirmed that the recommendation needs more labour than the farmers' current practice and that there was no significant difference in yields between recommended and farmers' weeding practices. This information has been fed back to the on-centre researchers. In addition, as a result of the above finding, the FSR team designed two on-farm verification experiments to address better the weed problems in the area. These included a mechanical weeder (wheel hoe) and pre-emergence herbicide as test technologies for maize.
In conclusion, the observed rejection and modification of some of the technologies by farmers were good pointers to the need to refine our research programs. It indicates that research
experiments aimed at the fertilizer rate recommendation need to be based on the target group farmers objectives, resources endowment and level of institutional support. Future research on weeds should take into account the labour shortage that farmers

page 11
face at critical periods.
Finally, we would like to stress' that failures to incorporate important factors which have a bearing on adoption of technologies will prevent the research results from being accepted.

Page 12
Enos M. Shumba
Agronomist, Farming Systems Research Unit, Department of Research and Specialist Services, P.O. Box 8108, Causeway, Harare, Zimbabwe
1.1 Introduction
This working paper presents the results of a three year on-farm trial programme in Zimbabwe.
The trial programme was developed from a study of a communal farming system and an assessment of the major factors limiting productivity of the system. On the basis of this assessment potential technical solutions were listed and the most promising selected for inclusion in an on-farm trial programme.
The first five sections of the paper describe the process of deciding on the content of the on-farm trial programme. The remaining sections (which will be presented in a subsequent edition of this newsletter together with the list of references) describe the design, implementation and analysis of two trials from the overall programme. Particular emphasis is placed on the analysis of the trial data across sites and across seasons. The implications that the results have for economic analysis and the formulation of recommendations is discussed as are the implications of the results for further monitoring and trial work.
1.2 Overview of Zimbabwe's agricultural Sector and the place of FSR
Zimbabwe's agricultural industry has two major sectors: the communal areas, also known as the peasant farming sector, and the commercial sector, each comprising about 40% of the total land area. Most of the communal areas are on infertile sandy soils deficient in organic matter and derived from granite. Seventy-four per cent of the communal areas fall within natural regions IV and V which are characterised by low and erratic rainfall (Vincent and Thomas, 1960). The large scale commercial

Page 13
sector is generally on heavy fertile soils that are suitable for intensive crop production. Communal areas provide a subsistence living for about 75% of the country's population and a social security system for over 90% of the community. However, with respect to agricultural performance, these areas have contributed until after independence only 20% of the total marketed produce compared to 80% from the commercial sector. Although there was a 96% increase in the value of communal area agricultural production between 1966 and 1975 (Stubbs, 1977), the increase has been largely due to an increase in area cultivated and to a lesser extent to increased yields per unit area. Yields per unit area are extremely low compared to yields attained in the commercial sector where high-yielding technologies are widely employed (Tattersfield, 1982). All available data suggest a declining availability of food in the communal areas. In 1962, the estimated availability of maize (the major starch staple) was 160 kgs per capita and by 1977 it had fallen to 105 kgs, a figure well below the national average of 175 kgs per capita for the same year. Why productivity per unit area and the per capita availability of the major starch staple in the communal sector have remained poor when improved varieties and production practices have been available is a question that needs urgent investigation.
Three explanations are commonly given:
(a) the available improved practices are not appropriate to
farmers needs or resources;
(b) the available improved practices are appropriate but
have not been effectively disseminated;
(c) the available improved practices are appropriate and
have been well disseminated but farmers are not
sufficiently motivated to utilize them.
It is against this background that the Department of Research and Specialist Services (DR & SS) recognised that more agricultural research, particularly on-farm research, was needed to develop innovations suitable to specific communal area environments and thereby increase agricultural production and productivity in these areas. Since Independence in 1980, a number of institutes and stations in DR & SS including the Agronomy Institute, Chiredzi Research Station, Crop Breeding Institute, Plant Protection Research Institute and Soil and Chemistry Research Institute have initiated on-farm trials in communal areas. Coupled with this expansion in the amount of research carried out on-farm, the FSR programme was established as an additional research tool to foster farmer involvement in the research process.

Page 14
1.3 The position of FSR in DR&SS and the role of IARCs
The history of the FSR programme dates back to 1981 when DR&SS, with technical and financial support from CIMMYT, established an FSR team in the Agronomy Institute of the Crop Research Division. This team, representatives from from other institutes in DR&SS, the Department of Agricultural and Technical Services (AGRITEX), CIMMYT and the Department of Land Management at the University of Zimbabwe undertook a diagnostic survey in the Chivi district (a low potential, low rainfall communal area). This was followed by on-farm experiments in the 1981/82 season. In 1982 CIMMYT assisted the team to mount another survey in Mangwende (a high potential communal area) which was followed by three seasons of on-farm trials and a Technology Assessment survey conducted in August 1984. Results from this work in Mangwende form the basis of this report.
In 1982, with the assistance of ILCA, the Livestock and Pasture Research Division of DR&SS designed a farming systems project to carry out research on livestock production in the communal areas and obtained financial assistance from IDRC. However this project did not start until February, 1984 when the project advisor arrived. Thereupon the Directorate of DR&SS decided to form one FSR Unit to conduct research on both crop and livestock production rather than having separate units under two different divisions. Under the reorganised setup in 1984, FSR is an autonomous unit with a core team at the Harare Research Station (DR&SS Headquarters) guiding and supporting two small implementation teams, one in each'research area. The Deputy Director, who is also overall coordinator of all DR&SS on-farm research activities, is the coordinator of the FSR unit.
According to the FSRU Annual Report (1985) the five member core team is responsible for designing or adapting research strategies, methodologies and programmes of work and for guiding and supporting the field teams in the implementation. The specific duties of field teams include selecting suitable
research sites and farmers, conducting farming systems surveys and monitoring studies and implementing research trials as well as continuous liaison with farmers to obtain feedback on proposed interventions.
1.4 Objectives of the work and methods employed
The work described in this report is based on the assumption that the principal reason for low levels of maize production in the communal areas is the lack of improved maize technology appropriate to the circumstances of communal area farmers. The report presents the results of a study aimed at identifying improved maize production technologies that will be acceptable to most farmers in the study area and lead to increased yields and

Page 15
economic returns through more efficient use of inputs.
The specific objectives of the study were to:
i) identify constraints to increased crop productivity and
relate these to the overall Mangwende farming system.
ii) design, test and evaluate alternative crop husbandry
practices on farmers's fields; and
iii) involve the extension staff and farmers in the problem
identification and technology generation activities.
To meet these objectives, a number of hypotheses were tested in on-farm trials. The methodology employed in the research reported was based on:
a) CIMMYT'S "Planning Technologies Appropriate to Farmers"
(Byerlee, et al, 1980). Essential components of this methodology employed in the study included the diagnosis of farmers' problems, farmers' goals and constraints in crop production and the identification of new technologies or strategies to remove or alleviate those
b) The IRRI approach (IRRI, 1977 and Zandstra, 1980) was
adopted in the testing phase;
c) Methodologies employed by Matlon (1985) in his on-farm
experimentation work in West Africa to evaluate trial data and to solicit farmers' views on the technologies under test have been used in the trial evaluation phase.
(also see Harrington, 1979; Kirkby, pers, comm.)
This section gives a general description of the Mangwende communal area, with special emphasis on agro-climatic circumstances that have a strong bearing on maize production in the area.
Mangwende communal area in the Murewa Kubatana district lies about 90 Km north-east of Harare in Mashonaland East province. Mangwende is in natural region II and about 8.9% of Zimbabwe's communal areas lie in this region. Mangwende consists of five wards with a total of 203,000 hectares and a human population of about 30,000 cultivators giving a population density of about 97 persons per square kilometre. With about seven people in the average family this gives 7.2 hectares of land per cultivator. It is not clear from available data what proportion of this land is arable but it could be below 50%. Although 3.4 hectares is the typical average holding per cultivator, some farmers,

Page 16
particularly the young, owned less than one hectare of arable land.
Rainfall and soils are the most important agro-climatic factors. They have a strong bearing on the crop production decisions made by Mangwende farmers.
2.1 Rainfall
Rainfall is the only source of water for field crops in Mangwende. The communal area has a mean annual rainfall of 800 to 950mm. The bulk of this falls in summer. The larger portion of the communal area falls under subregion IIa which enjoys
reliable rainfall, with infrequent dry spells in summer. The northern portion lies in subregion IIb which is subject to more dry spells during the rainy season and/or to a shorter summer season. Equally important from the viewpoint of agricultural production are periods of heavy rainfall which lead to
considerable soil and nutrient losses, coupled with occasional crop damage and waterlogging of vegetable gardens. Farmers
perceive the infrequent mid-season droughts as a serious problem in maize production.
2.2 Soils
Soils in Mangwende are mainly coarse to medium-grained sands
belonging to the paraferrallitic group. These soils show a
marked correlation with the underlying parent material whose weathering is influenced by mean annual rainfall. Granite is the predominant parent material and smaller areas of dolerite also exist. The soils have relatively low clay and low organic matter contents but have appreciable reserves of weatherable minerals. Soil fertility generally increases as rainfall decreases towards the north of the communal area. The light texture and poor
structure of these soils have the following implications:
(i) in fields with poor surface cover there could be a problem
of soil erosion due to surface runoff, and a need for soil
conservation measures;
(ii) the light soil texture leads to a high degree of leaching of nutrients from the topsoil and the need to split the application of highly soluble fertilisers like ammonium

Page 17
(a) Soil texture
Textural class Proportion of sites, %
Medium grained sands (mgs)* 87.0
Medium grained loamy sands (mgLs) 6.5
Medium grained sandy loams (mgSal) 2.6
Medium grained sandy clay loams (mgSaCl) 3.9
*mgs = less than 15% silt + clay: more than 85% sand
LS = 15-20% silt + clay: 30 35% sand
SaL more than 50% sand
SaCL = 20-30% clay: more than 50% sand (Source: Ivy, 1981)
(b)Soil Nutrients
pH N(ppm) P20s K
(CaCl2) (After (Resin (meq %)
incubation) extractable)
Levels considered
reasonable* Above 5.0 Above 30 Above 30 Above 0.1
Level range of 79 sites 4.3-61 12-35 12-35 0.05-0.2
% of sites with
reasonable levels. 27.8 15.2 16.5 98.7
*pH = 5.0 5.5 i.e. "medium acidic", is satisfactory for most
N = 30-40 is medium (Note: No soil has sufficient N in
PzOs = 30-50 is adequate no appreciable response is likely. K20 = 0.1+ is adequate response unlikely.
Source: Chemistry and Soil Research Institure (nd), "A guide to
the meaning of soil analysis".
Part of this general soil information was verified by an analysis of soil samples from 79 randomly selected sites on which crop trials were conducted over the 1982/83 to 1984/85 seasons. Since the trials were not planted on the same site in subsequent
seasons, different sites were sampled for each season. These sites were selected in such a way that they represented the range

Page 18
of soil types found in the district. For each field, 10 subsamples were taken, five from both diagonals of the trial site, using a shovel. The sub samples were then mixed to obtain a composite sample which was submitted to the Soil and Chemistry Research Institute for analysis. The soil analysis results should be interpreted against the following background. First, the sampling was done during drought years leading to a possible over estimation of the soil nutrient status since some of the fertilizer from the previous application was not completely exhausted by the crop. Second, the overall nutrient status of Mangwende soils is generally better than that of most communal areas (see Mashiringwani, 1983), possibly because the farmers are using relatively high levels of inorganic fertilizer to improve the inherently low soil fertility status of these granitic sands. Thus these results cannot be generalized for all communal areas.
The analyses presented in Table 1 depict the following features. First, the majority of the soils have a light texture. Second, the soils are generally low in nitrogen with only 15.2% of the sites recording medium levels. Third, some of the sites have deficient to marginal levels of phosphate with only 16.5% having adequate levels. Fourth, the soils have adequate potassium. Fifth, the pH values are low for most leguminous crops, although the pH is adequate for the maize crop. Thus the low soil fertility assumption described in this report is based on Table 1 and refers to nitrogen and phosphate only since the pH and potash levels were found to be adequate for the maize crop. Based on the soil analyses, the following inferences can be made:
(i) crop yield can be substantially increased by the
application of inorganic and other forms of fertilizer.
(ii) the low levels of nitrogen and the deficient to
marginally deficient levels of phosphate indicate the need for an application of mineral early in crop development to facilitate the rapid proloferation of
(iii) the low pH values noted imply the need for lime
application when growing crops with a high lime
This section summarises the priorities of Mangwende farmers and how they, allocate their resources to meet those priorities. Emphasis is placed on the dominant maize enterprise as a subsistence and cash crop in the farming system.
3.1 Farmers' Priorities

page 19
The first priority of farmers in Mangwende is ti satisfy the basic food needs of their family. This is achieved by growing a mix of crops that ensure an adequate supply of basic staples and relishes throughout the year. Maize, the major starch staple, occupies about 70% of the arable land. Finger millet is rarely used as a maize substitute while rice is only eaten on special occasions. Garden vegetables constitute the main relish throughout the year and are supplemented by okra, nyevhe (a vegetable crop used as a relish), fresh pumpkin and cowpea leaves in summer while dried cowpea and pumpkin leaves are eaten in winter. The second priority of Mangwende farmers is to seek cash to meet other basic needs. The major ones are food, clothes, school fees and the purchase or hire of resource inputs for agricultural activities of the following year. In the 1981/82 season maize was the most important cash source followed by garden vegetables and off-farm income.
3.2 The Dominance of Maize in the System
The importance of maize as a food and cash source in the Mangwende farming system is illustrated by the larger area, planted to the crop and better husbandry practices employed. These are reflected in higher crop yields compared to other crops. In addition to planting maize to a larger area, farmers also stagger their maize crop to assure that they at least harvest some grain in the event of a drought (Table 2). As a consequence, maize yielded better and resulted in fewer total crop failures compared to other crops. More farmers winter ploughed, planted earlier, applied manure and inorganic fertilizer, weeded twice and more timely in their maize crop compared to the other crops (Table 3).
Although there has been marked maize yield increases over the past ten years through improved husbandry practices, the large yield difference between lower and higher' producing farmers (Table 4) suggests that there is still room for increasing current crop yields. The identification of factors contributing to the yield differential would facilitate the design of technical interventions that can improve the overall performance of maize in the system.
This section discusses the major technical constraints to maize production in Mangwende. Current farmer management strategies to overcome some of these constraints are also discussed. Research opportunities coming out of these key constraints and aimed at reinforcing strategies already adopted by farmers are presented in section 5.

Crop Maize Broundnuts Fingersillet Sunflowers
*- -------------- --- ---------------------------------------------
Proportion of farmers involved, I In a 761 96,7 84,2 63,2 25,0
Average area planted, ha. 2,0 0,33 0,35
Number of crop planUngs Up to 3 1 1 1
Proportion of farmers reporting total
crop failure, I
- lIt Planting 6,17 17,2 14,6 15,8
- 2nd Planting 1,4 -- 3rd Plahting 7,1
Average yield, tWha. 3,4 0,71 0,86 0,64
Total yield, toos 6,8 0,23 0,30
.Proportion of crop sold 80, 42,2 *
------ ------------ -----~
* Relative proportions of "es and retentions not quantified
Sources Technology Assessment Survey (1984)
------------------------------ -------------------- --------------------------------------------------CROP PROPORTION OF FARMERS INVOLVED IN THE FOLLOWING ACTIVITIES In a 761
ist Plant. 98,7 62,7 33,3 33,3 60,0 6,7 93,3 97,4 9,6 66,2 53,7 41,? 4,0
2nd Plant. 92,1 54,3 20,0 4,3 50,0 44,3 1,4 91,4 97,2 100,0 62,3 15,9 57,0 24,6 1,4
3rd Plant. 36,8 46,4 7, 23,0 53,9 23,1 92,9 92,9 96,3 44,4 7,4 25,9 51,8 14,8
BROUNINUTS 84,2 53,1 1,6 19,6 58,3 22,1 26,6 26,6 96,9 56,2 29,17 51,6 12,5 7,2
FIN6ERMILLET 63,2 66,7 -0 56,2 31,2 10,4 2,1 18,7 10,4 97,9 46,9 29,8 17,0 6,4
SUNFLOWER 25,0 52,6 0 31,6 47,4 21,0 31,6 5,3 89,5 21,0 11,8 35,3 52,9
... ..--... ....-----------------------
Soprcpi Technology Assessment Survey (1984)
--- ---------~.- - --------------------- ---- ---- - ---- ---------MAIZE I FARMERS YIELD RANGE, t/ha. (I responses) YIELD AVERAGE
PLANTINS BSERVE 0 9,2-1,( 1,2-2,2 2,3-3,3 3,4-4,4 4,5-5,5 5,6-6,6 ABOVE 6,6 PER GROWER It/ha
-------------------------- --------------- ------- ----*--------------------- --***First 98,7 6,7 22,7 24,0 13,3 21,3 5,3 2,7 4,0 2,67
Second 92,1 1,4 8,6 10,0 21,4 34,3 5,7 9,6 10,0 3,8
Third 36,3 7,1 7,1 17,9 28,6 0 25,0 10,7 3,6 3,70
-- --------- -----------1 I
* The poor start to the 189314 rainy season and a mid seas drought partly explains why the first saize planting was outyielded by subsequent plantings.
Source: Technology Assessment Survey (1964)

Page 21
During the 1982 diagnostic survey, insufficient draught power and low soil fertility (nitrogen and phosphate deficiencies) were identified as major constraints to crop production in Mangwende. Increased shortage of land in the face of rising human population growth is reducing farmers' ability to improve draught power access by increasing herd size and to restore soil fertility by fallowing fields.
Labour and cash flow problems, as well as pests and diseases are important within season constraints. A major labour bottleneck was identified at planting and first weeding time in maize. Most farmers faced severe cash flow problems during most of the year due to seasonality of their incomes from produce sales. Pre- and post-harvest crop losses from pests and diseases were important in maize, groundnut, sunflower and garden vegetables. A detailed description of these and other factors has been given in separate DR&SS publications (see Shumba, 1985 and FSR Unit Annual Report 1985, in press). For purposes of this report we concentrate on a few basic technical constraints that have a direct bearing on maize production and productivity.
The results of a multiple regression analysis of survey data collected from Mangwende in 1982 suggested that about 45% of the variation in maize production between households can be explained in terms of the five variables in the following linear regression equation.:
REGRESSION on Maize yield (tonnes/ha)
Independent Variable b t
1. Manure application
(l=no, 2=yes) 0.43 1.75*
2. Planting date
(days from October 1st) -0.02 -3.04**
3. Fertiliser application rate
(l=<375D:250AN; 2=375D+250AN
3=>375D+250AN)# 1.08 6.17***
4. Time of compound application
(1=after emergence
2=at planting) 0.65 1.85*

Page 22
Independent Variable b t
5. First weeding time
(days from planting) -0.02 -1.92*
N = 78 R2 = .45 DW = 1.95##
*p<0.10 **p<0.05 ***p<0.01 ****p<0.001
# D Fertiliser Compound 'D', AN = Ammonium Nitrate
## DW Durbin-Watson statistic
These data indicate that maize yields are influenced by:
(a) manure and planting date (related to cattle ownership
and access to draft power)
(b) fertiliser rates and management and,
(c) timeliness of weeding.
This analysis confirmed the impressions obtained from earlier survey work which suggested that Mangwende farmers had achieved a relatively high level of management in their maize crop by the use of hybrid seed, fertilizer and other improved husbandry practices. However, insufficient draught power and low soil fertility were identified as major constraints to crop production in the area. In addition, a labour bottleneck was identified at planting and weeding time. The remaining sections of this paper consider these issues in detail.
4.1 Insufficient draught power
In 1975, 73% of the sample farmers (n=30) owned cattle compared to 46% in 1982. Average herd size (cows and oxen only) fell from 6.86 to 3.29 over the same period. The data also indicate a fall of 55% in the total number of oxen available to sample and 49% in the number of cows. Even more drastic was the fall of 70% in the number of owners with four or more oxen enough for a ploughing team. This trend was mainlyattributed to cattle deaths during the liberation war (1975 1980) because of suspended cattle dipping and uncontrolled cattle movement.

Page 23
Village Total number Proportion of Cattle
of farmers owners, %
Nyamutumbu 193 21.2
Mushaki 56 48.2
Mushinga 150 62.0
Chakanya 32 53.1
Chanetsa 135 74.1
Dzimati 97 44.3
Darangwa 157 24.8
Bere 133 75.2
Chipore 34 .25.0
Chingara 52 38.5
Total 1089 46.0
Two points need to be made on cattle ownership. First, this parameter is not static as witnessed by the trends revealed. Second, the term 'ownership' as used in this report implies that the farmer is responsible for the day to day management of the herd and has ready access to its by products like draught power, manure and milk. The question whether the cattle belonged to him or whether he merely looks after them was not established. This could be important in relation to who makes major decisions on whether to sell or slaughter an animal. Some farmers indicated that they had money to enable them to restock, but animals for purchase were unavailable in the area.
The difference in cattle ownership influenced the cropping system as a whole. Those with cattle had larger arable holdings, better land preparation, earlier planting and weed control, applied manure and achieved better crop yields (Tables 6 and 7). Crop production performance also tended to be closely related to other factors associated with cattle ownership in the area. These included more household members working permanently on the farm (and fewer working off farm), greater involvement in farmer organisations for easier access to credit facilities, less female headed households and more cash income to supplement crop inputs purchased on credit. All these factors help to increase the ability of cattle owners' to manage crops better.

- ------------------------------------- --------Cattle owners Non-cattle owners
Parameter (n = 401 in 401
Z of Sample Z of Sample
farmers average farmers average
Family size (no.1 8,4 6,4
No. working permanently on farm 3,4 2,8
Household head in casual employment 22,7 43,6
Woman head of household 12,0" 30,0
Size of holding (ha.) 3,9 2,9
Area cultivated (ha.) 3,6 2,1
Area cultivated per permanent member (ha.) 1,05 0,75
Farmers leaving fields fallow 67,5 17,5
Farmer organisation membership 82,5 61,5
Main crops and argea oroun, ha.
Hmaize 100 2,5 100 1,3
Groundnuts 92,5 0,7 92,5 0,4
Sunflower 45 0,3 10 0,2
Estimate of income, Z$
Haize sales 92,5 347 75 168
Vegetable sales 80 140 75 84
Groundnut sales 42,5 40 42,5 26
Sunflower sales 17,5 11 5,2 3
Other produce sales 55 19
Off-fare income 30 159 56,4 149
Total annual income 100 752 100 449
Sources Formal Survey (1982)
Proportion Yield
Crop Target of farmers inter Hanure Compost Nonth of No. of Method of Month of Av.I
and Groups cultivating plough Appl. Apple. planting feedings Ist eeding Ist uWeding grover
Planting particular t/
crops Oct. Nov. Dec. Jan. 1 2 ox/hoe hoe Nov. Dec. Jan. Feb. ha.
Ist laize 0 100 82 52 15 54 44 2 0 100 88 59 41 42 54 4 0 3,24
Planting N t100 30 12 28 25 64 11 0 100 085 10 90 26 57 12 5 2,10
Groundnet 0 93 51 0 0 14 62 22 2 100 41 33 67 19 64 17 0 0,74
Planting N 82 10 0 O 7 56 33 .4 100 53 0 100 16 41 43 0 0,54
Sunflower 0 45 25 0 0 0 0 67 33 100 0 88 12 0 0 70. 30 0,65
Planting N 10 0 0 0 0 0 62 28 100 0 0 100 0 0 50 50 0,20
* 0 = cattle owners (n 40) N = non-owners (n = 40)
Sources Formal survey (1982)

Page 25
The fall in herd size and the increase in the number of non-cattle owners are exacerbated by a lack of dry season cattle feed. This results in weak cattle and a reduced draught power pool to service an increasing number of cultivators at the beginning of each cropping season, leading to a general delay in crop establishment. Agricultural research in many countries has shown that crop yields can be seriously depressed if planting is delayed. According to Spear (1968) a 1 to 3 per cent yield reduction per day occurs when maize planting is delayed until mid-November.
The farmers of Mangwende have reacted in a variety of ways to counter the problem of falling animal numbers and a weak draught power pool at the beginning of the season. Cattle owners and a small proportion of non-owners winter plough fields shortly after a crop is harvested when soil is still moist and cattle in good condition. The practice was said to loosen the soil and conserve moisture, thus enabling weak animals to replough the field with ease at the start of the season. The survey showed that it takes 12.7 cattle hours to plough one hectare of non-winter ploughed land compared to 5.6 cattle hours in a winter ploughed field. A small proportion of sample farmers (12%) indicated that they had tried planting directly into winter ploughed land, with or before the first rains, to solve the problem of late access to draught power at the start of the rains. However, 65% of the sample said that the associated increased weed problem discouraged this practice.
Although a four cattle team has traditionally been regarded as the normal size for ploughing purposes, the survey established that only 41% of cattle owners used four animal draught teams compared to 54% who used just two animals due to the fall in herd size. Farmers using two cattle had an average herd size of 8.4 compared to 10.4 for those who used four animal teams. To
counter the problem of falling cattle numbers, 33% of the cattle owners were either substituting for or supplementing their oxen with cows for draught purposes. A few farmers indicated they used bulls. The use of cows for draught has increased since 1975 and appears to be closely related to the fall in cattle numbers in the area. Of the farmers using cows, 14% started before 1970, 53% during 1976 1980 and a further 33% in 1981 and 1982.
Most farmers supplemented veld grazing with residues from maize and groundnut crops. After harvesting, the residues were collected from the fields and stored on raised platforms near cattle kraals. The residues were fed indiscriminately to all classes of cattle by throwing the residues into the kraal at night. The uneaten stalks And haulms were tramped on and became part of the manure. Some farmers stored enough maize stover to last until the rains while most of the groundnut haulms were

Page 26
exhausted by June. Three per cent of the owners fed dried grass to their cattle between April and June and 5% gave them salt around July and August. The latter practice was said to increase the appetite of the animals.
These observations suggest that improvements in the use of the draft power pool will likely have a significant impact on maize production and could be expected to increase crop sales from the district.
4.2 Low soil fertility
As already discussed in Section 2, most of Mangwende soils consist of low levels of nitrogen and low to marginally deficient levels of phosphate. The regression analysis reported earlier confirms the importance of fertiliser use for successful maize production on these soils.
To counter the low soil fertility problems farmers have intensified their crop production by concentrating available inputs on smaller areas. In 1981/82, 67.5% and 77.5% of the
owners and non-owners respectively, left some of their fields uncultivated. Lack of enough fertilizer and poor draught access mainly contributed to this.
Farmers addressed the low fertility situation in several ways (Table 7). Cattle owners and a few non-owners applied kraal manure to their maize fields. Non-owners concentrated on the use of compost to improve the organic matter status of their fields. Due to limitations imposed by few animals per household and the problems involved in making large compost pits, farmers supplemented these nutrient sources with chemical fertilizer. The application rate for the majority of the farmers was 375kg of fertiliser compound 'D' (8N:14P205:7K20) and 250kg ammonium nitrate (34.5%N)/ha. Chemical fertilizers are obtained on short-term loans from the Agricultural Finance Corporation (AFC) or by cash from local stores in Murewa or Harare.
The above rates of fertilizer application provide a total of 116 kgN/ha, 53kg P205/ha and 26kg K20/ha. With these rates, a potential maize yield of 5.0 to 6.2t/ha is expected on soils with an after incubation N of 25ppm. However, actual average yields achieved are around 3.4t/ha although about 10% of the sample farmers fall within the 5.0-6..2t/ha range. It appears that the fertilizer is not being efficiently utilized because of some other limiting factor. This comes out very clearly from the regression analysis in section 4, in which five variables are singled out as explaining about 45% of the variation in maize yields among households (R2=0.69). It is against this background of attempting to cater for most of the other variables limiting

Page 27
maize yields that a comprehensive maize trial programme was designed for Nangwende Communal area.
Use of chemical fertilizer on maize started before 1970 when only 32.4% of the sample farmers said they were applying fertilizer. By 1981, all farmers in Mangwende were fertilizing their maize crop. The trend of fertilizer use is as follows: 32% of the farmers used fertilizer for the first time before before 1970, 26% between 1970 and 1975, 23% between 1975 and 1980, and 19% started using fertiliser in the last two years. By 1982 all farmers were applying a basal dressing to their maize. However 80% of the sampled farmers applied this basal dressing three weeks after planting. Most of those who gave a reason for doing so said it was to avoid fertiliser burn. Others indicated a shortage of labour at planting. Late delivery of fertiliser and copying neighbours were the main other reasons given for late application.
Although other institutes conducting on-farm research trials in Zimbabwe have not experienced 'fertiliser burn' in their trials, this is a serious problem with farmers who because of labour shortages and the need to plant the crop at the earliest opportunity might fail to minimize contact between the seed and fertilizer. This becomes very critical when there is no rain after the crop has been planted into a relatively dry soil. Thus certain practices which appear straightward when implemented by the researcher become more problematic when they are managed by the farmer because of farm level constraints like labour.
On the basis of these observations it would seem that an examination of alternative compound types with higher levels of the nitrogen and phosphorus than compound 'D' (for example, compound 'P' analysis 1ON:18P205:0K20) and management practices that enable farmers to use this expensive input cost effectively would have a good chance of resulting in improved maize yields and profitability.
4.3 Labour bottlenecks
The survey showed that a relatively small proportion of household members work permanently on the farm. Although cattle owners had more adult members working permanently on the farm, the area cultivated per individual was higher at 1.05 ha. in the former group compared to 0.75 ha. for non-owners. Both target groups agreed that the busiest months of the year were November to December which includes the land preparation and weeding period. The labour bottleneck contributed to delayed basal fertilizer application by 23% of those giving a reason for this practice, and to delayed weeding, particularly by non-owners who mainly rely on hand weeding (Table 7).

page 28
Farmers had reacted to the labour shortage in a variety of ways. Some farmers used hired labour to supplement their permanent family labour force of 3.4 members per household for cattle owners and 2.8 members for non-owners. Thirty-nine percent of the cattle owners and 13% of the non-owners hired labour mainly for weeding between November and January at a cost of Z$2.62 per day. A few reported using beer parties ("nhimbes") for land preparation and weeding. The majority mentioned the Group
Development Area (GDA), a farmer organisation as being important during peak labour periods by allowing collective planting, fertilizer application and weeding.
Technical. innovations that reduce or spread the labour bottleneck before or after the November-December peak period are likely to improve overall crop husbandry.
In this section, a number of potential technical. solutions to the identified maize production constraints in Mangwende are suggested. Solutions seen to be compatible with the existing farming systems in Mangwende were selected for inclusion in an on-farm trial programme to be described in the second half of this paper (to appear in a subsequent edition).
5.1 Insufficient draught power
We have seen that reduced cattle numbers per household and poor cattle condition at the start of the rainy season have resulted in insufficient draught power and have contributed to delayed maize crop establishment, hence low yields, particularly for non-owners. Possible strategies to arrest this include the
introduction of alternative draught sources, improving cattle herd productivity and alternative tillage practices using existing draught power more effectively.
5.1.1 The introduction of tractor power
A few farmers reported hiring tractors to plough their fields to enable them to establish an early crop (the cost of this practice was Z$50-60 per ha). Given that tractor hire is twice as
expensive as ox plough hire, its widespread use is only
justifiable if the expected yield increase covers the extra cost and leads to better economic returns. According to Howard
(1980), tractor use, as practised on commercial farms, could be uneconomic for the peasant farmer. The increase in fragmentation of arable holdings as the farmer population increases implies that a tractor would spend more time doing non-productive work.

Page 29
However, there are possibilities of contract ploughing either by private enterprise or co-operative ownership, but managerial problems require to be resolved. Because of the problems mentioned, the possibility of tractor use will not be developed any further in this paper.
5.1.2 Improving cattle herd productivity
The objective here would be to improve the draught capability in the system by manipulating herd composition and increasing herd strength and size by improving the food supply.
(a) Composition of the herd
The survey established that a high proportion of the herd was older than two years and the cows and oxen were in the ratio of about 1:1. Oxen are mainly used for draught
purposes but the use of cows to provide draught was increasing. If cows can provide most of the draught power, oxen could be released for sale and more calves would be born. However, such a solution is long-term and its overall effect on the system unclear. A study on the effect of using cows for draught purposes on milk production and subsequent calving was investigated at the FSR Development
Centre under the auspices of the Agronomy Institute.
(b) Improving the bulk and nutritive value of dry season feed
some farmers reported supplementing veld grazing with maize stover, but this feed source is very low in crude protein.
To improve the nutritive value of winter feed, the
profitability of buying concentrates like urea lick in winter could be investigated. The planting of fodder crops in fenced fields which are left fallow may also improve the winter feed situation. However, since farmers tend to view the cattle component in the short run mainly as an input into crop production the experience has been that they are reluctant to invest cash and other resources into such an
activity which generates few direct benefits.
Better utilization of natural grazing could provide a longer term solution. Attention could be paid to the improvement of the quality of the veld by introducing legumes. However, it is argued that the cost of establishing and maintaining a legume based pasture is usually high and makes such schemes uneconomic unless they are done for fattening purposes.
Communal ownership of grazing land also makes strategies like the establishment of grazing schemes difficult. Points to be borne in mind before embarking on such schemes

Page 30
(i) most of the district is already overgrazed and
hence initial destocking is Inecessary before
grazing schemes can be instituted;
(ii) the majority of the present non-cattle owners have
indicated the intention to own cattle in future.
Their cattle will have to be accommodated somehow
even if that implies increasing the stocking rate;
(iii) encouraging the sale of cattle in order to maintain
the correct stocking rate will be difficult since the objective of most farmers is to increase the size of their herds. In some cases the decision to sell is not made by the person who looks after the
The addition of a livestock component into the FSR team in early 1984 allowed some of these issues to be addressed in the FSR work but they are not reported here.
5.1.3 Alternative tillage practice
Such techniques attempt to utilize the existing draught power more effectively. Examples include planting directly into winter-ploughed fields and planting using an ox-drawn ripper tine.
Reduced tillage work done by the Agronomy institute-in Chivi South showed no significant difference in maize yields between a crop planted in winter-ploughed plus late-ploughed field and that planted directly into winter-ploughed lands (Shumba 1984). The implication of this on the availability of draught power at the beginning of the growing season was that farmers who have winter-ploughed their fields can establish their crop early by planting directly into winter-ploughed fields.
However, when this technique was tried in Mangwende, farmers were not impressed because of the weed problem and the labour required to hoe out the planting hills.
With the ox-drawn ripper tine, Grant (1981) showed that yields of 2.5 t/ha could be achieved by early planting maize into ripped furrows compared with 1.0 t/h 'a when planting occured after the first rains using the ox drawn mouldboard plough. Since the tine is easier to pull and works faster than the conventional mouldboard plough it is compatible with the draught shortage situation in the area and is selected as one of the technical interventions for testing on farmers fields.

Page 31
5.2 Poor Soil Fertility and Soil Fertility Management
Strategies addressing the soil fertility problem should attempt to improve the efficiency of fertilizers in currently use, both organic and inorganic, and to increase yields and profits from the maize enterprise. Such strategies could include reducing inorganic fertilizer costs by combining it with organic fertilizer, timely compound fertilizer application and the use of a compound fertilizer whose chemical composition satisfies the nutrient requirements of Mangwende soils.
5.2.1 Cutting inorganic fertilizer costs
This involves investigating levels of organic fertilizer alone or in combination with inorganic fertilizers required to give an economic maize yield. However, the low nutrient quality of communal area manures (Tanner et al 1984) and the differences in materials used and management practices employed in compost making which imply differences in quality make the measurement of responses difficult. This therefore requires more controlled experiments and will not be considered in this report.
5.2.2. Timely compound fertilizer application
Taking cognisance of reasons given for delayed basal fertilizer application by some farmers, timely application could be
facilitated by:
(a) the use of a planter with a fertilizer attachment, or
holing out and fertilizing before the onset of the rains in the case of fenced fields to ease the labour bottleneck
reported at planting time;
(b) the application of compound fertilizer between planting
stations in furrows made by an ox-drawn tine, to eliminate the risk of fertiliser burn, without increasing labour
(0) implementing complementary tillage and compound fertilizer
application strategies such as even ploughing and planting depth, minimising contact between the seed and fertilizercoupled with planting when soil is still moist.
Apart from the planter which is not yet available commercially, the other strategies will be considered in this report.

Page 32
5.2.3 Compound type
Since the soils are deficient in nitrogen and phosphate, but satisfactory in potash, a compound with more of the deficient nutrients will be compared with the currently used one under farmers' field conditions.
5.3 Labour bottlenecks
The peak labour periods identified at planting and weeding times could be flattened out in the following ways:
(a) by investigating the effectiveness and economics of
herbicide use in weed control and the use of a planter with
a fertilizer attachment for basal fertilizer application;
(b) by easing planting and weeding labour requirements through
staggered planting. However, it was noted in the survey that farmers are already staggering their maize plantings for a variety of reasons which include risk aversion, poor access to draught power and labour shortage. Hence the on-farm trials programme only looked at the use of
5.4 Alternative Crop Enterprises
Bearing in mind the reasons given by farmers for staggering their maize planting, being the draught power and labour constraints and the need to spread risk, then system flexibility could be achieved by:
(a) Introducing or intensifying the production of early
maturing and drought tolerant crops as alternatives to late planted maize, thereby improving returns to cash inputs and reducing the dependence on maize. Such crops include
sunflower, soybeans and sorghum.
(b) Intensifying finger millet production resulting in a
possible draught and labour requirement shift since the
crop is normally dry planted.
5.5 Farmer Ranking of Crop Production Constraints
The following responses were given when farmers were asked to list in order of importance their major constraints/problems in crop production (Table 8).

Page 33
TABLE 8: Farmer perceptions of their major crop production problems in Mangwende (n=72).
Major crop production problem Proportion of farmers
mentioned citing it as most
important problem, %
Poor access to draught power 42
Lack of cash for fertilizer purchase 23
Labour at planting and weeding time 17
Drought 12
Small size of holding 3
Transport 3
Source: Technology assessment survey, 1984.
TABLE 9: Approximate cash income and basic needs expenditure 1981/2 (n=80)
Estimated 1981/82 cash earnings: Z$600.50 Expense % Spending Sample average % of cash income
Food 100.0 124 20.75
Clothes 93.5 155 25.50
School fees. 62.5 39 5.85
Crop inputs
(mostly fertilizer) 100.0 73 29.60
Total 491 81.70
Source: Formal Survey (1982)
Although the farmers' perceptions of the basic problems are similar to those described, it is important to note that farmers view the soil fertility problem in terms of the quantity of fertilizer they have to purchase in order to get a good yield. This is reflected in the 1981/82 season expenditure pattern where

Page 34
fertilizer purchases absorbed the largest portion of the farmers' total income (Table 9). This perception of problems in terms of the amount of cash absorbed could partly explain why pests and diseases, although very important and visible problems to the researcher and even to the farmer when reminded, were ranked lower down on the problem list because the cost of chemical to control them represents a very small proportion of the total expenditure. Hence the need for observations rather than relying entirely on the farmer's perceptions of the problem when
conducting surveys with a technology generation focus.
The Newsletter is published quarterly in January, April, June and September. News, comments,letters, reserarch results and opportunities concerning on-farm research in Southern and Eastern Africa will be considered for inclusion.
Contributions should be sent to:
The Editor, CIMMYT, P.O. Box 30727, Lilongwe 3, Malawi (telex: 43055 ROCKFND MI)
The Editor, CIMMYT, P.O. Box MP154, Mount Pleasant, Harare,
Zimbabwe (telex: 2462 CIMMYT ZW)

@ CIMMYT International Maize and Wheat Improvement Center The CIMMYT International Maize and Wheat Improvement Center holds all rights to the source text and shall be considered the copyright holder for the text and images of these publications.
The CIMMYT International Maize and Wheat Improvement Center has made this publication available to the University of Florida, for purposes of digitization and Internet distribution. The CIMMYT International Maize and Wheat Improvement Center reserves all rights to this publication. All uses, excluding those made under "fair use"~ provisions of U.S. Code, Title 17, Section 107 are restricted. Contact the CIMMYT International Maize and Wheat Improvement Center for additional information and permissions.