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. 31

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

Page 1
- Data Collection and Analysis Workshop
(for social scientists) ................................. 2
- Workshop on Crop/Livestock Interactions ................. 2
- On-Farm Research with a Farming Systems Perspective:
Experimental Phase Workshop ............................. 2
- Training Course on Agroforestry Research for Development. 2
- Agricultural and Rural Project Management .................. 3
- Abstracts and Intercropping ............................. 4
- Outline and Notes for Preparing Project Proposals
for On-Farm Trials
Steve Franzel and Ann Stroud ................................... 6
Richard Edwards ......................................... 11
- CIMMYT ................................................. 23
- ICRISAT ................................................ 23
- CIP .................................................... 24
- ICIPE ........ ...................................................... 25
- ICARDA ................................................. 25
- IRRI .......... ............................ ............................ 26
- IFPRI ................................................... 27

Page 2
This workshop will take place on the ILCA Campus from April 18 to 29, 1988.
Contact: Dr. P. Anandajayasekeram, CIMMYT, P.O. Box 25171, Nairobi, Kenya.
This workshop will take place in Zimbabwe and will be hosted by the Department of Research and Specialist Services.
Contact: Dr. A. Low, CIMMYT, P.O. Box MP154, Mount
Pleasant, Harare, Zimbabwe.
This is the second phase of the regular regional training workshop run annualy in cooperation with the University of Zimbabwe. It deals with setting up and implementation of on-farm trials. It will take place at the University of Zimbabwe, from Aug 22 to Sept 10, 1988.
Contact: Dr. M. Rukuni, Faculty of Agriculture, P.O. Box MP167, Mount Pleasant, Harare, Zimbabwe.
May 9 27, 1988.
The International Council for Research in Agroforestry is organizing a three-week training course with the support of the "Direct Aid to Educational Institutions/Organizations (DSO)"
Programme funded by the Royal Netherlands Government.
The course aims at providing research scientists and
development planners with conceptual and methodological training as well as practical experience on subjects related to agroforestry; mainly on how to undertake the interdisciplinary identification, design and implementation of research projects leading to the development of locally adapted agroforestry systems and technologies adoptable by farmers. The training course will consist of lectures, workshops, group discussions, field exercises and independent study.

Page 3
Support is available for about 15 participants, who will be selected on a competitive basis. The sponsorship will cover travel to and from Nairobi, accomodation and a modest allowance for incidental expenses from the funds provided by the D.S.O. This will not cover travel and accomodation for the participants' families.
The training course is intended mainly for research scientists and development planners responsible for agroforestry research, rural development projects or programmes. Candidates should be citizens of a developing country and presently employed by a national institution. They must hold a degree, or equivalent qualification in agriculture, forestry, livestock sciences, or economics, or a related discipl.ine. It is desirable that they also have three years' post-qualification work experience. It is a pre-requisite that candidates wishing to be considered for sponsorship by ICRAF should be nominated by their employing national institutions. Proficiency in the English language is essential.
Contact: The Principal Training Officer, P.O. Box 30677, Nairobi, Kenya. Telex: 22048 ICRAF Nairobi, Telefax: 521001, Cable: ICRAF, Telephone: 521450.
April 18 July 8, 1988.
The overall objective of the course is to help participants to build upon their management experience and develop those areas of management expertise which are particularly relevant to the task of project management in an agricultural/rural context.
While participants are expected to play an active role in planning their own learning process, scheduled course activities will cover the following broad areas:
- the interpretation of project plans
- the techniques for planning project implementation
- tenders, contracts and procurement
- organizational design, project team formation, staff
recruitment, training and development
- project monitoring, the design and operation of management
information systems, resource planning and control
- staff management and interpersonal skills financial planning, management and control
- project management in specific project contexts
Participants will be expected to have had practical

Page 4
experience in agricultural/rural development and to have responsibilities, current or intended, in the implementation and management of specific agricultural and/or rural development projects.
Applicants will normally be expected to have a university degree or equivalent qualification in a relevant subject. It is also desirable that they should have several years of relevant work experience. All applicants should have a sufficient knowledge of written and spoken English to enable them to follow lectures and participate in discussions without difficulty.
The fee of the course is 4,200 to cover accomodation as well as tuition, but not meals. A daily provision for
subsistence and incidental costs of 15 17 is recommended. In addition, participants and funding agencies should note that a further 300 is required for study visits to other parts of
Britain, and a book allowance of 70 is recommended. No accomodation is available for families.
Most overseas visitors to Britain are liable to meet the costs of medical treatment other than that resulting from
accidents and emergencies. The British Council makes its own arrangements for those participants whom it is sponsoring, as well as for those participants sponsored by agencies on whose behalf it acts. For all other participants an additional
insurance premium (which was 9 in 1986/87) will be added by the University to invoices, and this must be paid.
There will be approximately 25 places. In view of the likely competition for places, it is important that application be made as early as possible.
Contact: The Course Director, Course No 109, Project
Planning Centre, University of Bradford, Bradford, West Yorkshire BD7 1DP, United Kingdom.
One of the major activities of the GTZ Project on
Intercropping is to make available information on intercropping and relevant subjects, such as traditional farming systems, agroforestry and farming systems research, to rural development projects of the GTZ and to other interested institutions. A state-of-knowledge report on intercropping was prepared in the years 1980-82 (GTZ-Schriftenreihe No 137) where the available literature and other additional information on intercropping,

Page 5
with a focus on West Africa, were reviewed.
This literature collection became the basis of a computerized documentation service that started as an abstract service in 1983. It is intended to ultimately that this service will be able to provide a comprehensive documentation on
ihtercropping and related subjects; not only the latest publications are collected but also older papers that are still of interest. The documentation now comprises more than 800 publications.
The abstracts are set up in the following way:
1. Abstract number
2. Principal key-word: focus is still on intercropping (and
in some cases cropping systems), but traditional farming systems, farming systems research and development and
agroforestry became the second focal point in 1985.
3. Key-words: if relevant, the geographical demarcation
(continent, country) or the agro-ecological zone is
given; the key-words: review, field trial, field study or farm survey indicate the nature of the paper; common
names of field crops are used.
4. Author's name
5. Title in the original language, the English version is
set in brackets.
6. Source of publication.
The abstracts are decided into different sections, which are listed on the first page of each issue. Within each section abstracts are ordered alphabetically by name of author. The
sections are:
a. Field experiments with various crops and intercropping
b. Interactions and competitions between crops, biometrics
(statistical designs and evaluation)
c. Nitrogen fixation and residual effects (also mycorrhiza)
d. Pests, diseases, and weeds in intercropping systems
e. Fertilization and intercropping systems
f. Description of traditional intercropping systems
g. Economic aspects of intercropping systems
h. Traditional farming systems
i. Farming systems research and development
j. Agroforestry
k. Agro-ecology and agro-meteorology in tropical cropping
1. General soil science and soil fertility in tropical
cropping systems
m. Fertilizer use, micronutrients, and rockphosphate in

Page 6
tropical cropping systems
n. Tropical crops
The subject index, based on the key-words, and the crop and geographical indices provide a means of rapidly locating abstracts on specific topics. As there are many cross references, due to the nature of intercropping, the same abstract will appear different times under different subjects.
The author index permits to locate easily all publications of an author. Readers interested in the abstracts are asked to address their request to: Dr. Juergen Canls, Neuenlander Weg 23, 2725 Hemslingen, Federal Rep. of Germany.
Steven Franzel and Ann Stroud
Farmer problem you are addressing. Discuss cause of problem, interactions with the problems, and relative importance of the problem in the farming system. Comment on farmer recognition of problem, the priority they give it and their assessment of the causes. Briefly describe pertinent facts about the farming system which help to explain the problem and causes. Refer to any survey reports that have been done in the experimental area.
Farmer target group and/or agro-ecological zone this trial is aimed at-. The trial may be aimed at other groups as well, e.g., policy makers (e.g., yield loss assessment trial due to aphids) or station researchers (e.g., comparing farmer and recommended weeding practices).
Objective(s) of experiment, including hypothesis(es) being tested. These should directly address the problem stated above. It is the objective primarily exploratory, refinement, or verification?
1. Past and present work on this problem:
a. In the research organization you are working in. Include present recommendations related to the operation under study in the experiment. For variety trials, give present status of the

Page 7
varieties being tested.
b. At other institutions, e.g., state farms, MOA/ADD trials, University trials, etc.
c. Outside of your country, in other national programs, at International Agricultural Research Centers, etc. This is only necessary for technologies which are being tested in your country for the first time.
2. Is this the only trial dealing with the specific problem:
In some cases, a single trial is only one of several related trials which are being done to answer the stated problem. If this is the case, explain how this trial fits into the larger experimentation program.
1. Proposed duration
2. Location:
a. Research center, sub-center, individual farmer, common holding, producer cooperative, or other. b. Number of sites
3. Plot size (including length and width of plots):
Indicate for main plots and sub plots if using a split plot type design.
4. Experimental design.
5. Number of replications per site.
6. Treatments (those factors which you are varying):
a. Briefly explain the rationale behind the choice of treatments/levels. Describe the control or check treatment(s). If the farmer's practice is one of the treatments, describe what this is. By "farmer's treatment"t, do you mean to allow each farmer to manage the control in the manner he likes or do you mean to fix the control at levels which represent the average farmer?

Page 8
b. Include timing and methods for the various activities related to treatment application. For expression of timing of treatments, use internationally recognized stages of growth for the crop you are dealing with wherever possible.
c. Make a table and list the treatments. For treatments that are continuous variables, e.g., fertilizer rates, list the treatments from lowest to highest.
d. Express rates of pesticides and fertilizers in the following manner:
i. Pesticides should be listed using their chemical name followed by their trade name in parentheses, e.g., atrazine + metolachlor (Primagram 500). All rates should be given in kilograms of active ingredient/hectare and stated as such rather than kilograms product/hectare. All formulations used should also be stated, e.g., 500 flowable (FW).
ii. Fertilizer rates should be stated in kilograms of N, P, or K per hectare. Also list the actual material and quantity used, e.g., 200 kilograms diammonium phosphate (DAP) at a rate of 36-92-0 per hectare.
7. Non-experimental variables.
Here, the reader needs to know which of four ways each
non-experimental variable will be treated:I
a. Set by the researcher at a recommended level.
b. Set by the researcher at average farmer level.
c. Set by the researcher at some other level, e.g., fertilizer rate set at an intermediate level between recommended and farmer rates.
d. Determined by each farmer according to his/her own will. This is the approach used in "superimposed" trials.
A table, such as Table 1, showing what is to be done in the

Page 9
trial, is useful for explaining exactly how and and what levels non-experimental variables were set.
For example, Table 1 shows that for a maize variety trial, time of land preparation, number of plowings, time of planting, weeding frequency and time of weeding are to be determined individually by the farmers involved. Each of these farmer
practices is at or near the recommended practice. The researcher determined the method of planting (row), spacing and seed rate at the recommended level, even though most farmers broadcast and use a higher seed rate. The researcher also selected the fertilizer rate, which is intermediate between the recommended rate and the rate farmers use.
Table 1 also shows that farmer practices for the first three cultural practices are approximately the same as those of surveyed farmers. Further, the table describes variability in practices among farmers at different sites. For example, the range of number of plowings among farmers hosting the experiment is shown in parentheses following the average number of plowings.
The table should include the important cultural practices involved in the trial. Management practices to consider for all experiments include: time, method, and number of land preparations; planting time and method including seed rate; fertilization rate and method of application; time and method of weeding; thinning; pest control; and lodging prevention practices. Those practices for which there is a significant difference between trial farmers' practice, recommended practice or surveyed farmers' practices need to be included in the table.
8. Data collected.
st all the types of data to be collected during the
trial. For each type, tell when it will be collected, that is, with respect to other operations and stage of growth. For example, it is important to know whether you will take a crop injury rating before or after thinning.
This sort of information can be presented in a list or
table form.
9. Evaluation criteria
a. State criteria by which trial results will be evaluated, including those criteria important to farmers.
b. List types of analyses to be done to assist in evaluating the results of the trial.

Page 10
Non-experimental variables in an on farm maize variety trial
Surveyed Non-experimental
Cultural farmers Recommended variable fixed by
practice practice practice Researcher Farmers Remarks
Time of
land prep. Jan-Apr Jan-Apr --- Jan-Apr
Method of
land prep. Oxen --- --- Oxen
Number of
plowings 3 to 4 4 --- 3.2 (2-4)
Time of Early to Early --- Early to
planting late May May late May
Method of Brdcst(65%) Row Row --- Maize
planting Row pl(35%) plant plant dribbled
Seed rate
(kg/ha) 40-45 25 25 --Row spacing
(cm) 54 (48-60) 75 75 --Plant pop. 48 44 --- 53 Farmers
(thousand (43-61) do not
per ha) thin
(DAP) 40 (0-75) 150 50 --(N/P205) 7/19 27/69 9/23
Fert method
of applic. Bdcst. Bdcst. --- Bdcst.
Weeding 2 times 2 times --- 2 times 1st is
frequency ox cult
2nd hoe
Time of
first 30 days 20 days --- 28 days
weeding (22-35)

Page 11
Richard J.A. Edwards, Agronomist, Adaptive Research Planning Team, Mount Makulu Research Station, Private Bag 7, Chilanga, Zambia.
The theme of this paper is that greater time should be spent by technical scientists in monitoring and understanding events on farmers' fields and that this greatly improves the awareness of farmers' requirements. Furthermore, this will also lead to a refinement of programmes both on the research station and on the farmers' fields. In particular the paper explains how mapping of farmers' activities coupled with agronomic monitoring is a useful form of experimentation.
The emphasis being placed here on the gathering of agronomic information from farmers' fields as well as on the trial plots, the former often being neglected. Thus the essential theme to this paper is that collection of agronomic information, both within the area of the trials and from the rest of the farmers' fields, has a useful role to play.
Agronomic monitoring is the provision of information on the mechanisms by which farmers produce crops and as to how these and other factors influence subsequent crop growth and production.
This does not imply simply strict monitoring of technical parameters of soils and crop growth. Instead this covers a range of issues all of which need to be recorded in some way. This may be quantitative or qualitative information covering observations from the field or from questions asked to the farmers themselves (either using so called indigenous technology or parameters more clearly understood by the scientists). But, most important of all, it is whatever is recorded must be done in a rational and systematic way. All too often such observations are made but tend not be documented and/or attention is focused on peculiar, atypical but dramatic phenomena to the detriment of what is happening as commonplace on farmers'fields.

Page 12
Farming systems research, in general, involves a flexible system of descriptive surveys conducted by social and technical scientists. Once such information is collected and the problems or areas of potential improvement are diagnosed, technical solutions are tested on farmers' fields to enable their realistic screening.
Much of the information is socio-ecQnomic. This, though essential in setting priorities, is only one part of the picture. More detail, describing the agronomics, is required if the commodity scientist is to make a contribution. Little of the information from the informal and formal surveys is of great use in formulating technologies (but tremendous value in setting the direction of that formulation). To date, much of the response from commodity and specialist researchers has come from actual feed back on a particular material offered by them for testing (varieties of crops in particular).
A more thorough understanding of the causative factors reducing yield on farmers' fields is required to enable the commodity team to direct their efforts more purposefully. An understanding of the mechanics of how the farmers are carrying out their arable practices together with how this' influences yield is necessary to complement the socio-economic information collected. The tendency of the descriptive work to concentrate on single visit surveys, primarily collecting quantitative data, discourages the elaboration of quantitative agtonomic data around the priority development areas. It is these data that the
commodity teams are requesting, once priority areas have been identified, to enable the evolution of adequate solutions.
The technical input into the teams in the form-of trials is clearly the central role of a provincial team's work. It is
obviously necessary, therefore, to ensure that the way in which the technical trials work is implemented is carefully thought out so that the programme is both pertinent to the priority issues but also at the same time carried out in a realistic way. Simply planting a trial on a farmers' fields does not necessarily constitute farming systems research nor imply testing under the farmer managed conditions. It is important that trials are implemented which do look at the conditions and management that a technology or material will be subject to once adopted by a farmer. At the same time, recognition of any additional inputs required for the technology is also important. However, there is a tendency to concentrate solely on the area of a farmers' field where a trial has been planted and to ignore of what is actually

Page 13
hapenning elsewhere on the farmers' fields. As a consequence much valuable information on the technology under investigation is lost. Trials should be regarded as a useful way of providing an entry point on to farmers' fields to clarify and refine the nature of a programme.
Utilizing the variability of practices on farmers' fields as means of informal experimentation is extremely attractive. By having a picture of what the farmers have done where, coupled with collection of certain sets of data, it is possible to analyze data to examine certain hypotheses. Furthermore, providing descriptive information with concrete data around a particular issue makes it much easier to interact with commodity or specialist scientists. For example in one area of Lusaka Province in the Valley areas, residual and seepage moisture coupled with warm or hot ambient air temperatures mean maize is planted sequentially throughout the season. By mapping the farmers' operations it was possible to study the build up of various diseases, in particular maize streak virus. With this knowledge it is possible to say to the maize breeders that there is a particular "hot spot" which can be of use both for testing material and for the collection of germplasm. (There is the possibility that resistant material may be common in the area since it was one of the first areas for maize to be grown in Zambia when Portuguese settlers moved up the Zambezi river).
Collection of such data is generally rare for a variety of reasons:
1. The tendency of the descriptive work, to date, to
concentrate on single visit surveys, primarily collecting qualitative data, discourages the elaboration of detailed agronomic data around priority development areas. It is these data that the commodity teams are requesting, once priority areas have been identified, to enable the
evolution of adequate solutions.
2. To date, the Zambain Adaptive Research Planning Team
(ZARPT) has been following, more or less, the procedures laid down by CIMMYT for the implementation of farming systems research work. So far, most provinces have spent considerable time in zoning exercises followed by one-off informal and formal surveys of particular areas. More detailed case study type surveys which can involve collection of more detailed agronomic information, have not been implemented to any extent. It is only after the ,preliminary surveys have been undertaken that more

Page 14
detailed verification studies can be undertaken as the former provide the direction for these. Similarly,
agronomists frequently find the need for this detail, when experimental control of factors in a trial is minimized. At this stage, interpretation becomes more reliant on detailed knowledge of the variables concerned.
3. Tendency to view the agronomists wrongly as trials
officers. This is partially reinforced by the pressure on agronomists to come up with the proven technical recommendations and by the nature of their training.
This may further be aggravated if insufficient field staff exist for a team. There is need to increase the impact of agronomists in a team (possibly by including at least two agronomist positions in a team and/or ensuring
adequate numbers of field staff for collection of data).
4. Similarly, social scientists (economists, sociologists
and anthropologists) have traditionally been the people with the expertise in designing and implementing surveys.
Though agronomists are involved, to some extent both in planning and actual implementation, their input is less and often on general issues. Interestingly, descriptive studies in Zambia in the early 1900s were often done by ecologists. The contribution that ecologists can make to
farming systems research work should be strengthened.
5. A failure to emphasize that ZARPT has, in essence, two
major clients, firstly farmers and secondly CSRTs the latter is often underplayed. Commodity and specialist scientists working on-station, in general, have a negative attitude towards surveys. This is partly because their usefulness in setting priorities for research with subsequent screening of both available technologies and likely research directives is not as systematically followed as perhaps it should be. The main reason, however, is that surveys provide little hard data around any problem to allow the development of
materials or technologies.
6. Adaptive research in Zambia is now, more or less, fully
institutionalized. At its inception, much time was spent in ironing out teething problems with regards to how such farming systems research work should be implemented. Now this is less so and the detailed requirements of what is needed to formulate and develop material and technologies
is coming to the forefront.

Page 15
Several advantages accrue from such work:
1. Provision of concrete data upon which materials or
technologies can be realistically developed. The
development of more pertinent interventions is also
2. Providing a baseline which commodity and specialist teams
need to evaluate and set their own research directives by: firstly, improving awareness of how farmers are carrying out their cropping practice es and describing how this influences crop growth and yield; secondly, by providing an idea of the importance of a particular phenomena on farmers' fields by detailing its occurrence within and between farmers (and preferably across
3. Allowing for the design of realistic trials and
methodologies for the monitoring and implementation of the trial programme. Design of trials (and methodologies used to monitor phenomena) can be carried out in a realistic way working within the framework of the
farmers' systems.
4. Providing a realistic way of assessing the outcome of
trials in relationship to what the farmers are achieving.
Reliance on farmer control plots frequently is not sound
as they often do not represent true farmer conditions.
5. Enabling the testing of biological assumptions upon which
avenues of research are based.
6. Improving awareness by agronomists as to actual
conditions of farmers' fields. The very nature of a more continuous and systematic following of events means that agronomists (and social scientists) are in a better and more informed position to communicate with a farmer.
Some issues are less apparent from questioning farmers, especially when reliance is placed on one-off surveys and done out of season. Only by being on farmers' fields at
the right time do some issues come to light.
7. Providing a constructive way of bringing together team
members within a team on the one hand. It provides a programme in which it is easier to involve commodity and specialist researchers and thereby improve collaboration
and feedback between the two on the other hand.

Page 16
Liaison with commodity and specialist researchers is a critical issue and misunderstanding has arisen through lack of communication. A prerequisite for effective cooperation between commodity and adaptive scientists is that they have mutual respect for the contribution that one another can make to research process. Commodity
scientists bring their specialist insight of a particular crop to bear whilst the on-farm researcher is able to have an understanding of the broader issues which affect the decisions farmers make about technology adoption e.g.
desired varietal characteristics, management practices, etc. Both perspectives are needed if technology acceptable to farmers is to be produced. To be able to achieve this it is essential that adequate opportunities exist for commodity and adaptive scientists to communicate with one another. Such opportunities include participation in farm surveys, discussion of adaptive and applied research programmes and joint field visits, involving extension workers, to on-farm trials and to
on-station trials.
8. Enabling the use of farmers' practices as a means of
farmer experimentation by studying the range of practices on their fields. Mapping and monitoring of the mechanics of the cropping system enable certain hypotheses to be tested on farmers trials rather than by implementing a trial on phenomena the researcher is interested.
Furthermore this provides a clearer insight into the
likelihood of adoption of a particular technology.
As explained above farmers have evolved practices which are varied and far from static. By mapping farmers operations to build a picture of what farmers have done where, coupled with collection of certain sets of data, it is possible to analyze data to examine certain hypotheses.
Provision of descriptive information on the mechanics of the cropping system can provide information on practices and
technologies recommended which are already in practice. Frequently a new trial will include aspects already being used by farmers in one form or the other. Observing how farmers are
already coping with a particular problem can minimize implementing costly trial programmes. For example in Luapula province of Zambia, sunflower was thought to be a new crop and trials were set up with the crop planted in a particular way closer to that being recommended. It failed while farmers who

Page 17
were growing sunflower themselves obtained adequate yields from planting the crop in a way they had devised themselves. Similarly, in Lusaka Province, land preparation was originally thought to consist of a single ploughing and planting operation (Ihird furrow plough planting) whereas in reality farmers use a r nge of land preparation methods. This became clear when operations were mapped over the season and a picture was built up of what farmers were actually doing. Recently one incoming
project wanted to promote the testing of land levelling to improve seed emergence in the area but it was possible to convince the team that most farmers were already doing this in one form or another (often using a harrow). The important point here is that concrete data on the mechanics of the farmers' cropping systems were available which allowed for an informed dialogue with the team.
Once a clear idea of what is being done where on a farmers' fields over the season is collected, it is possible to collect data around various specific hypotheses and test them. Farmers rarely carry out all their operations at one time; thereby resulting in a diversity of treatments across a field. Once
mapped, each block of different operations and management (which includes differences in the timing of the same operation) can be used to examine certain hypotheses. For example, timeliness of planting, both to individual rains and in relationship to the season overall, can be tested given sufficient number of farmers. Statistical methods used can vary from more complex multivariate regression analysis to more standard analysis of variance of blocks using farmers as blocks and *management within a farmers' fields as treatments. Likewise, the level of land preparation can be subject to similar analyses. On a smaller scale, for example, where plants are b6ing planted in stations on one field, collecting data on seeds planted, plants established enables regression of information against yield.
Despite the attractiveness of doing this, certain points need to be borne in mind. The type of information generated may be "noisy" with high coefficient's of variation (not in itself a bad thing). There needs to be a clear decision to have the social scientists involved (as anthropologists often are when they are working in the area of anthropological ecology and ethnobotany) and to have a carefully planned programme of field visits utilizing primary field staff and senior staff. The methods of analysis of data with high variability are not unfamiliar to ecologists and social scientists.
Probably one of the greatest benefits to come out of this work is sound dialogue with research workers. Once armed with these data, it is possible to convince researchers working on-station how to modify and develop their programmes. Several

Page 18
examples serve to illustrate this further. The sorghum
agronomist in Zambia has been able to bring his programme more in line with how farmers are actually planting as it has been possible to show that sorghum is planted at a certain number of seeds per station with a particular spacing per planting per station. Furthermore this was backed up with information on
frequency distributions of this within and across farmers. In another instance, research was under way to validate the benefit of changing the recommended planting of sorghum in rows to one whereby seeds are planted in hills (stations) in the rows. Previous descriptive work had indicated problems with sorghum establishment and that seeds were dribbled in a row. Once
operations were mapped and data systematically collected it became clear that farmers were already doing'this and that other factors should be investigated.
Collection of agronomic data will involve a variety of methodologies, some of which still need to be tested as to their suitability for generating adequate data with limited sources available. Furthermore data requirements can change in the light of greater information and will depend upon the particular problem or issue being studied. It is therefore clearly
difficult to come up with hard and fast rules.
I However, central to the theme is the need to build up over the seasons monitored, a series of field maps for a certain number of farmers in which the mechanics of how and where farmers have carried out their arable operations are indicated. These maps need subsequently to be related to crop performance. This was attempted over the past seasons in Lusaka Province, both in hand hoe farming systems and the oxen ploughing farming systems. It has not been without its problems but has been highly beneficial in providing a sound basis from which a realistic trial programme in future seasons can be planned. Backed up with feedback from people concerning trial implementation under farmer managed conditions, it should be possible to plan a programme taking into account both the needs of farmers and those of researchers working on-station.
The field mapping programme has provided a general picture, of the framework in which farmer carry out the operations. Farmers, for the trial programme, were selected from a cluster of farmers and an attempt was made to bring in a "community" perspective such that the flow of resources between households is more fully understood and which minimize the distances covered. The agronomic monitoring was based around filed maps which allowed for the collection of data from fields in areas where the level of management was known. The more continuous monitoring

Page 19
has made it easy to follow the flow of resource between and within the households.
In the first year a more general picture of how farmers are actually carrying out their crop production should be provided, Subsequent seasons can focus on particular issues indicated from the f irst years work. Once fields are mapped it should be possible to examine certain hypotheses from areas where management is known and analyze the data using regression techniques, at the same time this provides a yard stick to judge the trials by.
In order to avoid collection of "luxury" information, derivation of a simple model would enable on to focus on factors causing yield decline and determine priority areas that need investigating. Information for such a model can be provided by a combination 'of observation of farmer practices and problem identification trials. Using a model will enable further descriptive work to be undertaken and planned which systematically looks at the factors that result in loss of yield potential. The point of working with such a model is to ensure that well defined hypotheses are examined. This should prevent collection of information which has little application. It would discipline any future programme by enabling a more systematic approach to problems on farmers' fields. Once fields are mapped it should be possible to examine certain hypotheses from, areas where management is known and analyze the data using regression techniques. The use of such a model is primarily to rationalize the data to monitor on farmers' fields. Liaison with commodity and specialist researchers to understand what they consider to be the most critical in reducing yiled in farmers' fields is essential. Clearly, it is necessary to rationalize tbe explanatory variables being collected.
Individual components of the model can be broken down to see how farmer management and environmental conditions influence crop growth. Thus the biological parameters determining crop growth from the basis of the postulated model. The socio-economic aspects of the farming system which influence farmer management will manifest themselves through the impact they have on biology of the crop(s) being grown.
Monitoring agronomic parameters from farmers' fields is, liable to considerable confusion due to the range of confounding factors. It is essential that the methodology used to ascertain a particular phenomena on farmers' fields is carefully thought out. Failure to consider the range of factors which influence the data may mean it is not possible to-draw the conclusions from the data as hoped. It should now be possible to implement descriptive and trials programmes which work within the framework

Page 20
of the farmers cropping systems and thereby be more realistic. Much of this work is still in its infancy and a level of methodological investigation is still called for.
A short cut for monitoring the farmers' fields has to some extent been facilitated by carrying out detailed monitoring of trials which were planted largely as farmers would plant the crop. A possible compromise to mapping is to carry out simple farmer field trials under farmer management conditions and placing a lot of-emphasis on agronomic monitoring within these. This could be backed up by general field maps indicating the areas with different management but without any detailed
monitoring of parameters influencing crop growth, apart from informal observations as to the occurrence of particular phenomena. In essence this is how the' programme evolved in Lusaka Province due to severe staff limitations.
Implementing this work is, however, not without its
1. Cost of collecting and analysing data based on continual
case study work however, trials require regular visits from researchers to ensure adequate management so if the
same farmers are used this should not be an issue.
2. Unfamiliar methodologies agronomists tend to be
extensively familiar with factorial analysis of variance but not with sampling and studying populations with high levels of variability. This is where social scientists and ecologists have a major role by refining methodologies which can cope with such complexity on
farmers' fields.
3. Numbers and duties of field staff involved. Clearly if
field staff are only there for trials work there will be inevitably be some sacrifice of trial work unless more
trials assistants are used, each covering fewer farmers.
Several issues arise here:
1. Should the agronomist or social scientist be doing this.
Ideally all team members should be involved and this does offer an attractive way of ensuring that all team members work together'on the priority areas (interestingly, some of the early (1930s) descriptive work in Zambia came from
ecologists and anthropologists).

Page 21
2. Should commodity and specialist staff o r adaptive
research staff be doing this. Again, both should be involved where possible, if only at the planning stage but the bulk of the field work will clearly fall upon the
adaptive researchers as their brief is to work on-farm.
3. What field staff are required to implement the work. If
the work is left entirely to the trial assistant there
ill be a sacrifice in terms of actual trials that can be
'planted. This can be minimized by varying the level of information collected by concentrating on distinct demarcated plots (trials or otherwise) but at the same time building up a general picture of what farmers have done with sketch maps. The maps are used as a basis to select specific areas and to carry out specific data collection exercises. This is within the capacity of the training of the trial assistants for most issues, given sufficient preliminary training and guidance and backed up with forms and questionnaires. However professional staff should also be involved to a large extent. Use of enumerators together with trial assistants greatly assists this type of work. Furthermore, if the social scientists' programmes use the same farmers and programmes are planned jointly, the work becomes easier
to implement.
It is likely that such work will provide benefits for all systems being studied. Despite this, systems which are less well known with little understanding of the causative factors 'reducing yield would warrant greater levels of input of such work.
Similarly as trials progress to the on-farm stage such work
should be increased. Certainly most teams will have carried out their initial surveys, so more specific studies involving a range factors that are recorded would increase in importance and likely involve the social scientist more. In any arable system it would pay to carry out such studies with mapping and monitoring as a component of the main verification surveys. If this is combined with trials which are judged to be immediately transferable to farmers' fields (best bet technologies!) results are rewarding. The use of trials on farmers' fields as an entry point into the farmers' system for collection of detailed data should be considered as beneficial in achieving good quality descriptive information.

Page 22
Agronomic monitoring, defined here as the provision of information on the mechanisms by which farmers produce crops and as to how these influence subsequent crop growth and production, is a necessary part of the work of farming systems teams. it improves awareness of farmers' actual crop production mechanisms and encourages greater dialogue between team members and
commodity and specialists researchers working on-station. Collection of such data, involving a range of techniques and parameters, can assist in the planning and refinement of a trial programme both on-station and on-farm. At the same time it can also provide information on technologies which farmers are already putting into practice, often in a variety of ways, thereby preventing implementation of a costly on-station programme whilst placing the technology in a more realistic light. Many recommendations are already taken up by farmers but in ways different to what researchers were expecting. More
emphasis should be placed on an agronomic programme which collects data which allow for the development of the technology for a perceived problem or development opportunity. Utilizing this information as a type of informal experimentation means a greater level of confidence in introducing technologies can be achieved. However-, it is not without its problems, in particular the need to develop more unusual (at least for agronomists!) methods of analysis. Ecologists, anthropologists and sociologists have much to contribute in this and a strict interdisciplinary approach needs to be encouraged (rather than just a multi-disciplinary approach).
I would like to thank members of the research branch and extension branch who helped me in my work. The ideas in the
paper are based on my experiences in carrying out farming systems research work. These ideas have, however, been discussed by various people in Zambia and outside of the country. In
particular, I would like to thank Stuart Kean, Alistair Sutherland and Carol Kerven for comments thay have made on my programme.

Page 23
CIMMYT is recruiting for a position as Economist based at its headquarters in Mexico. The economist would provide leadership to a new program on the economics of agricultural research with special focus on returns to plant breeding
research, allocation of research resources and impacts of
agricultural research investments.
The candidate must have a PhD degree in agricultural economics, significant experience working in an agricultural research institution, a capacity to work in a multidisciplinary setting with biological scientists and demonstrated record of professional contributions in the area of the economics of agricultural research. Fluency in English is required and competency in Spanish desirable.
This position is at CIMMYT headquarters, El Batan, Mexico (25 miles outside Mexico City). Some international travel will be required.
Interested candidates should submit their CV and names of three referees to:
Dr. Derek Byerlee Telephone: (905) 761-331
Director, Economics Program Ext. 1211
CIMMYT Telex: 1772023 CIMTME
Apdo. Postal 6-641 Cable: CENCIMMYT
06600 Mexico, D.F., MEXICO
ICRISAT is looking for a Food Technologist to work as a Principal Staff Member of the SADCC/ICRISAT Sorghum and Millets Improvement Program, at Matopos, Near Bulawayo, Zimbabwe. A PhD in food technolology with three or more years .of practical experience is desirable. A minor in plant breeding or previous close working relationship with a crop improvement program is an advantage.
Responsibilities will be:
1. Assist in establishing priorities for both traditional
and commercial uses of the crops.
2. Adapt and develop simple procedures to evaluate a large
number of breeding stocks for trials relative to various

Page 24
priority crop uses. To evaluate breeding stocks of
National and Regional Programs.
3. To interact with both government and private agencies
interested in various aspects of crop utilization.
4. To organize and conduct training to strengthen national
capabilities in the area of food technology.
5. To consult with national programs interested in
establishing their own laboratory capabilities.
If interested, kindly send letters indicating interest with biodata to: Dr. L.R. House, Executive Director, SADCC/ICRISAT Sorghum and Millet Improvement Program, P.O. Box 776, Bulawayo, ZIMBABWE.
The Social Science Department of the International Potato Center (CIP), sponsored by the Consultative Group of International Agricultural Research (CGIAR), announces an immediate opening for an experienced agricultural anthropologist at the center's headquarters in Lima, Peru, to provide anthropological perspectives for agricultural research and training within the context of an International Research Center dedicated to potato and sweet potato improvement in developing countries.
The agricultural anthropologist will be involved in research and training activities throughout the world in collaboration with natural scientists and economists from CIP and national programs. Approximately 90 days of international travel are required annually. Specific responsibilities include the following:
1. Contribute to the setting of CIP's program' s priorities
for potato and sweet potato research and training, by generating relevant information on farmers' and
consumers' needs and resources.
2. Participate on interdisciplinary research teams aimed to
resolve specific constraints to potato and sweet potato
production and use.
3. Provide feedback and the performance, impact, and
consequences of collaborative research and development
4. Help strengthen national systems' capacity to incorporate
client perspectives into their own research and
development programs.
Applicants should have a PhD or equivalent degree in anthropology as well as formal training and practical experience

Page 25
in agricultural development. The successful applicant is likely to have at least five years of field experience in developing countries and to be fluent in Spanish or French. Ability to work on interdisciplinary teams with natural scientists and economists of various country backgrounds and cultures is essential.
Applicants should send a resume, one recent publication, and names of three references to Douglas Horton, Head, Social Science Department, International Potato Center, Apartado 5969, Lima, PERU.
The Deputy Director acts as the deputy of the Director of the International Centre of Insect Physiology and Ecology
(ICIPE), who is the Chief Scientist and Chief Executive of the Centre. In the absence of the Director from the Centre, he acts in these two capacities. His position is, therefore, located
within the Office of the Director. The Deputy Director has, in addition, overall responsibility for the Research Division of the ICIPE.
The Deputy Director is expected to undertake the following duties, on behalf of and under the general supervision of the
Director of the ICIPE:
1. Coordination of the research activities of the ICIPE,
wherever they may be sited, in cooperation with the
Programme Leaders, Section Heads and Unit Heads.
2. Development of new research activities, in collaboration
with the relevant research staff, and in consultation
with the Planning and Development Unit (PDU).
3. Development of the specific facets of in-house periodical
reviews of ICIPE's research programmes, with sufficient frequency and depth to permit a reasonable assessment of
ICIPE's progress and impact in each field.
Applications to: ICIPE, P.O. Box 30772, Nairobi, KENYA,
quoting vacancy S7/ADVERT/20/597.
The International Center for Agricultural Resaerch in the Dry Areas (ICARDA), with its headquarters in Aleppo, Syria, invites applications for a wheat geneticist in the Cereal
Improvement Program.
The specialist will be expected to:

Page 26
1. Assemble, screen and evaluate germplasm of primitive
forms and wild progenitors of wheat.
2. Identify germplasm with specific traits for utilization
in breeding.
3. Collate, document and organize all relevant information
on germplasm collections.
4. Develop pre-breeding germplasm using inter specific and
inter generic crosses with the aid of conventional and
non-conventional techniques.
5. Participate in the training activities and travel in the
region as required.
Applicants should have a PhD in Plant Biology or Agricultural Sciences with at least five years experience in crop germplasm evaluation. An additional qualification in
cytogenetics of wheat or evaluation and documentation of cereal germplasm will be an advantage. Knowledge in computer analysis of evaluation data is needed.
Candidates should submit two copies of their CV and three letters of reference quoting vacancy No: CIIP/1/87 to: Personnel Officer, ICARDA, P.O. Box 5466, Aleppo, Syria.
The International Rice Research Institute (IRRI) is seeking an Agricultural or Mechanical Engineer with PhD or MSc, with a minimum of 10 years experience, preferably in industry with a strong background in machinery design and a practical knowledge of farm machines and small scale machinery manufacturing techniques. The person must be proficient in oral and written English and have a demonstrated capability in technical administration.
Responsibilities include providing scientific and administrative leadership for the IRRI Agricultural Engineering Department of fifty employees, participating in multidisciplinary research involving other departments of IRRI, and developing links with commercial manufacturers and government organizations in the rice-growing countries. In addition, organization of training programmes will be an important responsibility. Location is at Los Banos, Laguna, Philippines.
Applicants should submit a letter of application, resume, academic transcript and the names of three references to:
Director General, IRRI, P.O. Box 933, Manila, PHILIPPINES

Page 27
The International Food Policy Research Institute (IFPRI) is seeking to employ a Researcher to be stationed at the IFDC Center in Lome, Togo. The Researcher would work primarily on
fertilizer-related issues and on conducting micro-economic level data collection and analysis. The research would involve
interaction with both IFPRI and IFDC Senior Researchers. The position's title is Research Fellow.
The candidate should have a PhD or the equivalent in
Agricultural Economics, or Economics. Strong field experience, particularly in household survey in developing countries (preferably in Africa), is desirable. Fluency in French and
English is required.
Applicants should send two copies of their CV to:
Personnel/Togo, IFPRI, 1776 Massachusetts Ave., N.W., Washington, D.C. 20036, U.S.A.
Notes for contributors
1.) The newsletter is published quarterly in January, April, June and September.
2.) News, comments, letters, research results and opportunities concerning on-farm research in Southern and Eastern Africa will be considered for inclusion in this newsletter.
3.) Contributions should be sent to:
Malcolm Blackie
P.O. Box 30727
Lilongwe 3
MALAWI (Telex 4487 CBS MI)
Steve Waddington
P.O. Box MP154
Mount Pleasant

@ 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.