T AT AL DN PA
FARMING SYSTEMS RESEARCH (FSR) IN
HONDURAS, 1977-81: A CASE STUDY*
Daniel Gait,l Alvaro Diaz,2 Mario Contreras,3
Frank Peairs, Joshua Posner,5 and Franklin Rosales6
IProduction Economist, University of California, Davis
2Agricultural Consultant, Tegucigalpa, Honduras
3Pathologist, Panamerican School of Agriculture, Honduras
SEntomologist and Training Specialist, CATIE, Turrialba, Costa Rica
5Agronomist, USAID Casamance Project, Dakar, Senegal
Plant Breeder, IICA, San Jose, Costa Rica
*This paper is published by the Department of Agricultural Economics, Michigan
State University under the "Alternative Rural Development Strategies" Cooperative
Agreement AID/ta-CA-3, U.S. Agency for International Development, Office of Multi-
Sectoral Development, Bureau of Science and Technology, Washington, D.C.
MSU IS AN AFFIRMATIVE ACTION/EQUAL OPPORTUNITY INSTITUTION
MSU INTERNATIONAL DEVELOPMENT PAPERS
Eric W. Crawford, Carl K. Eicher, and Carl Liedholm, Co-Editors
The MSU International Development Paper series is designed to further the
comparative analysis of international development activities in Africa, Latin America,
Asia, and the Near East. The papers report research findings on historical, as well as
contemporary, international development programs. The series includes papers on a wide
range of topics, such as alternative rural development strategies; nonfarm employment
and small-scale industry; housing and construction; farming and marketing systems; food
and nutrition policy analysis; economics of rice production in West Africa; technological
change, employment, and income distribution; computer techniques for farm and market-
ing surveys; and farming systems research. While the papers mainly convey the research
findings of MSU faculty and visiting scholars, a few papers will be published by
researchers and policy makers working together with MSU scholars on research and action
programs in the field.
The papers are aimed at teachers, researchers, policy makers, donor agencies, and
international development practitioners. Selected papers will be translated into French,
Spanish, or Arabic. A list of available papers and their prices may be obtained from:
MSU International Development Papers
Department of Agricultural Economics
Michigan State University
East Lansing, Michigan 48824, U.S.A.
Individuals and institutions in Third World countries may receive copies free of charge.
Farming Systems Research (FSR) in Honduras,
1977-81: A Case Study
Daniel Gait, Alvaro Diaz, Mario Contreras,
Frank Peairs, Joshua Posner,
and Franklin Rosales
Working Paper No. 1 1982
Many people were instrumental in assisting the Honduran FSR effort, and others
encouraged the preparation of this case study. It would not be possible to acknowledge all
the significant contributions to FSR in Honduras, but the authors would like to single out
the following for their special contributions:
Dr. Ernest Sprague, Director of the CIMMYT maize program, for his vision and
persistence in encouraging meaningful multidisciplinary research.
Lic. Rafael Callejas and Efraim Diaz Arrivillaga, respectively, Minister and Vice-
Minister of Natural Resources in Honduras, who believed that multidisciplinary team
research was necessary for Honduras in 1977.
Arturo Galo Galo and Ivan Madrid, successive heads of the Directorate General of
Agricultural Operations, Ministry of Natural Resources, Honduras, for their administra-
tive and philosophical support.
The FSR trainees and assistants who assisted in introducing the FSR philosophy to
various regions of Honduras.
The Honduran farmers who offered their land for farm trials, and who patiently
provided us with a constant source of practical agronomic and socioeconomic information.
Carl Eicher of Michigan State University for reviving and supporting the idea of
documenting the Honduran FSR experience.
Joseph Beausoleil and Rollo Ehrich of USAID/Washington for arranging the neces-
sary funding to carry out field research to update the study and to see it through the
Robert Hudgens, Gordon Appleby, and Fernando Fernandez of the USAID evaluation
team, who evaluated the USAID Honduras Agricultural Research Project No. 522-0139.
Finally, Mike Weber of Michigan State University, for his encouragement, editing,
and general assistance.
Now that farming systems research (FSR) in the Third World is five years old, it's
time to take stock of attempts to move FSR from the International Agricultural Research
Centers to national agricultural research systems. Although the International Agricultur-
al Research Centers have made important contributions to FSR methodology, it is well
known that researchers in these centers enjoy a measure of financial support which is
seldom matched by researchers in national agricultural research systems.
The Department of Agricultural Economics at Michigan State University has
requested several FSR research teams to document their experience and the problems
encountered in introducing FSR in national Agricultural Research Systems. This paper by
Dan Gait et al. presents an assessment of the problems and the achievements in
introducing FSR in the national agricultural research system in Honduras. Papers on
introducing FSR in Thailand, the Philippines, and Indonesia are in preparation and will be
published in 1983.
Carl K. Eicher
Professor of Agricultural Economics
Interamerican Development Bank
National Development Bank
Agricultural Center of Technology and Education (Costa Rica)
International Center for Tropical Agriculture (Colombia)
Canadian Institute of International Development
International Center for Corn and Wheat Improvement (Mexico)
International Center for Potatoes (Peru)
Central Unit (of PNIA FSR, Honduras)
University Center of the Litoral Atlantic Region (La Ceiba, Honduras)
General Directorate of Agricultural Development
Cooperative Service for Rural Development
General Directorate of Agricultural Operations
Panamerican Agricultural School (Zamorano, Honduras)
National Agricultural School (Olancho, Honduras)
Experimental Variety Trial (CIMMYT)
Food and Agriculture Organization (United Nations)
Farm Systems Research
International Agricultural Development Service
International Bank for Reconstruction and Development (World Bank)
Institute of Agricultural Science and Technology (Guatemala)
International Development Asociation (IBRD)
International Development Research Council (Canada)
Interamerican Institute of Agricultural Sciences (South and Central America)
National Agrarian Reform
International Progeny Testing Trial (CIMMYT)
Office of International Cooperation and Development (USDA)
National Agricultural Research Program
Corn and Bean Project
Ministry of Natural Resources
Interamerican Technical Service for Agricultural Cooperation
United States Agency for International Development
United State Department of Agriculture
TABLE OF CONTENTS
I. HISTORY OF AGRICULTURAL RESEARCH IN HONDURAS 1
Background of Agricultural Research Policy 2
The FSR Team and Its Objectives 4
The Administrative Organization of PNIA 5
II. ORGANIZING FSR IN HONDURAS 6
Restructuring PNIA 6
PNIA National Headquarters 6
FSR Central Unit (CU) 7
FSR Regional Units 7
FSR Organization in Honduras 7
Creation of the FS Central Unit (CU) 9
Duties of CU Staff 9
General Responsibilities 9
Integration of Other Researchers Into the CU 10
Crystallization of CU Research Priorities 10
History of the CU 11
Summary of FSR Organization 12
III. DEVELOPMENT AND IMPLEMENTATION OF FSR 14
Selecting Regions and Farmers, and Design of Farm Trials 14
Regional Reconnaissance (Sondeo) 14
La Paz 16
El Rosario 16
San Jeronimo 16
Formal Questionnaire (Encuesta) 18
Design of Farm Trials (Ensayos de Finca) 18
Selecting Experimental Sites for Farm Trials 19
Selection of Collaborating Farmers 20
Assigning FSR Teams to Homogeneous Zones 20
Field Supplies and Seeds for Farm Trials 21
Farm Trials and Farm Record Keeping 21
Methodology of Implementing Farm Trials 21
Monitoring and Observing Trials 22
Planning Field Trials, Second Half of Season
(Postrera), 1978 22
Analysis of Field Trial Results 23
Redefinition of Farm Trials 23
Implementing A Farm Record Keeping System 23
Other Questionnaires 26
Farming Systems in Choluteca 26
Sesame Cropping Systems of Choluteca 26
Land Preparation in La Paz 27
Soil Conservation in El Rosario 27
Bean Systems and Problems in Danli and Olancho 27
Training Honduran FSR Teams 28
Results of In-Service Training, 1978-80 29
Positive Aspects of FSR In-Service Training 29
Experience Working with Farmers 29
Practical Experience 29
Extending FSR to Other Regions 31
IV. KEY PROBLEMS IN ORGANIZING AND INTRODUCING FSR 32
Problems of Introducing FSR to Personnel of
PNIA and Other Institutions 32
Philosophical Difficulties 32
Problems of Linking-Up with Other Units
of the Ministry and Outside Institutions 34
Extension and Research Linkages 34
The CU and PNIA Headquarters 35
PNIA and Sector Planning 35
PNIA and SAID 35
PNIA and Other External Institutions 36
PNIA and Regional Directors 36
Problems of Acquiring Sufficient Resources
to Implement FSR 36
Budgetary Restrictions 36
Logistical Difficulties 37
Support Facilities, Supplies and Vehicles 37
Personnel Issues 38
V. CONCLUSIONS AND RECOMMENDATIONS 39
Conclusions from the Honduran FSR Experience 39
Recommendations for FSR in Other Countries 39
General Political and Administrative Issues 39
Specific FSR Integration Issues 40
(1) Homogeneous Zone Surveys 40
(2) Formal Farmer Questionnaires 40
(3) Distribution of FSR Manpower and Responsibilities 40
(4) Farm Trial Design and Installation 41
(5) Harvesting Farm Trials 42
(6) Analysis of Trials 42
(7) Farm Record Keeping 42
(8) Training 42
(9) Personnel Issues 43
(10) Linkages 43
APPENDIX: VIVID FSR EXPERIENCES 44
LIST OF TABLES AND FIGURES
1. Government Grain Production Targets for 1978
and Production Estimates, 1978-81 3
2. Evolution of Honduran FSR Methodology, 1978-80 14
3. Characterization of FSR Zones, Comayagua Region
of Honduras, 1978 16
4. Types of FSR Trials Conducted in the Comayagua Region,
By Zone, 1978-80 23
5. Short Courses and Seminars Presented to FSR Trainees,
1. Administrative Regions of Honduras 8
I. HISTORY OF AGRICULTURAL RESEARCH IN HONDURAS
An important date in the origin of the idea for multidisciplinary on-farm research in
Honduras (hereafter referred to as farming systems research, or FSR) is September, 1976.
During this month, a group of Cornell/CIMMYT/Rockefeller Foundation multidisciplinary
graduate studentsI met in San Miguel de Allende, Mexico, to produce a final team
report.2 At one of the sessions, Mario Contreras explained his new job in agricultural
research with the National Program of Agricultural Research (PNIA) in the Ministry of
Natural Resources (SRN) of Honduras. In the SRN, disenchantment with traditional
research had led to plans to establish a multidisciplinary research group to help implement
a reorganization of agricultural research. As Contreras described plans in Honduras,
several other team members expressed their interest in continuing interdisciplinary
research together after the completion of their graduate training.
Interviews were arranged by Contreras between the interested team members and
Honduran officials in the SRN. During these interviews, the students and the representa-
tives of the SRN reached a general philosophical agreement concerning the multidisciplin-
ary, on-farm research focus needed in Honduras. In March, 1977, Contreras arranged to
offer formal contracts to the three team members who had been interviewed in Honduras.
Peairs and Gait accepted their contract offers and, when Rodriguez accepted another
position, his contract was offered to, and accepted by, Joshua Posner.3
The student research team was composed of six Cornell Ph.D. candidates,
representing six agricultural science disciplines. The students were: 1) Mario Contreras,
pathologist from Honduras; 2) Daniel Gait, agricultural economist from the USA;
3) Samuel Muchena, plant breeder from Zimbabwe-Rhodesia; 4)Khalid Nor, biometri-
cian/statistician from Malaysia; 5) Frank Peairs, entomologist from the USA; and 6) Mario
Rodriguez, agronomist from Colombia. Each student spent approximately 18 months--
three cropping cycles--in Mexico conducting thesis research on tropical and semi-tropical
corn cultivars under the auspices of Cornell, CIMMYT (International Center for Corn and
Wheat Improvement) and the Rockefeller Foundation. The research effort was interdisci-
plinary, with frequent team meetings to trade ideas and problems and to discuss individual
and collaborative research field trials.
Contreras, M. R., et al., "An Interdisciplinary Approach to International Agricul-
tural Training: The Cornell-CIMMYT Graduate Student Team Report." Cornell Interna-
tional Mimeograph 59, Dec. 1977. See also Whyte, William F., Participating Approaches
to Agricultural Research and Development: A State-of-the-Art Paper, ARE No. 1.
Ithaca, New York: Rural Development Center, Cornell University. May 1981.
3posner is an American agronomist who, in early 1977, had five years of experience
in western African agriculture and was scheduled to complete his Ph.D. requirements at
Cornell at about the same time as the rest of the team members.
Background of Agricultural Research Policy
In 1976, the Minister and Vice-Minister of the SRN agreed that the traditional
research-extension model was not working effectively in Honduras. Since 1966, per
hectare yields of the basic grains in Honduras have either been stable or falling. Honduras
has gone from being a major net exporter of corn and beans in Central America to a heavy
importer of corn and a marginal importer of beans by 1980-81. Imports of corn have been
considerable since 1975, averaging slightly less than 23,000 metric tons per year. During
this six-year period, total metric tons of the four basic grains imported averaged 27,285
per year. Only sorghum was a net marginal export at 1,327 metric tons per year.
Table 1 documents the Honduran government's targets of production for the four
basic grains in 1978, and the actual, estimated and forecast production levels of these
grains from 1978 to 1981. With the exception of rice, production levels were considerably
lower than target levels. Since production levels of corn, rice and sorghum have dropped,
while that of beans has risen only slightly since 1978, the gaps between government
production targets and actual production of these grains have widened. Moreover,
between 1966 and 1980, the Honduran general price index has risen 232%, and general
food prices have increased 257% (USAID, 1981). More hillsides and other marginal lands
have been coming under cultivation to keep pace with demand, as productivity has not.
These problems were addressed by both the SRN and the national 5-year plan (1978-
82). The national plan stressed (1) increased research on the small- to medium-sized
farmers in the independent and reformed sectors, a group which produces the majority of
the basic grains (corn, beans, rice and sorghum);1 and (2) continued research emphasis on
basic grains with the medium-to long-range emphasis gradually shifting to vegetables,
oilseed crops, and livestock as grain output increased (IADS, 1978).
Farms in the 5-35 hectare size are estimated to sell about 50% of their basic grains
to feed other sectors of the population, while farmers operating less than 5 hectares sell
approximately 25-33% of their production (USAID, 1981). Thus, the dual national goals of
helping the majority of both rural and urban poor can be met by placing increased
emphasis on research for small- to medium-sized farmers.
Farms of 0-20 hectares make up 88% of total Honduran production units, but only
27% of the farming acreage. This same group of farms produces 69% of the corn, 72% of
the beans, 55% of the rice, and 77% of the sorghum in Honduras (Agricultural Census of
Honduras, 1974). In terms of all cultivated crop area in Honduras, corn covers 50.6%,
beans 10.6%, sorghum 8.7% and rice 2.3%. These four basic grains together account for
72.2% of all cultivated crop area (Compendio Agropecuario, 1977).
Government Grain Production Targets
for 1978 and Production Estimates, 1978-81
(1,000 Metric Tons)
1978 1979 1980 1981
Target Actual Actual Estimate Forecast
Corn 472 417 343 358 400
Rice 30 32 32 26 27
Beans 56.2 35 38 38 42
Sorghum 55.9 42 37 34 36
Source: USAID. 1981. "Evaluation Team Report." Evaluation of USAID Honduras
Agricultural Research Project No. 522-0139 with The National Agricultural
Research Program (Preliminary Draft, 5/29/81).
The FSR Team and Its Objectives
The Ministry of Natural Resources (SRN) recruited a multidisciplinary research
group--a Farming Systems Team--to help carry out a reorganization of agricultural
research. The Farming Systems Team consisted of six scientists: (1) a Honduran plant
pathologist who led the effort to reorganize the agricultural research system; (2) a
Honduran plant breeder who headed the implementing and support unit in the field (known
as the FSR Central Unit, or CU); (3) an entomologist from the U.S.; (4) a forage breeder
from Uruguay who coordinated training; (5) an agricultural economist from the U.S.; and
(6) an agronomist from the U.S. Each contract was concluded directly between the
individual researcher and the Government of Honduras, and covered a period of one year.
It was generally understood that the moral commitment from each side would be for a
five-year period to allow for continuity, implementation of research reorganization, and
sufficient time for on-farm research to begin to demonstrate some advantages over
The FSR Team identified four strategies to implement the reorganized agricultural
research system. They included:
(1) farmer-oriented, inter-disciplinary research;
(2) strengthening of the national research station network;
(3) manpower development program; and
(4) exploitation of opportunities to link with other agricultural research and
development institutions engaged in activities complementary to those of
PNIA (IADS, 1978).
Farming Systems Research (FSR) was viewed as the methodology which would comple-
ment traditional on-station research and be useful in discovering the constraints on small
farmer production at the farm level.
From the viewpoint of the SRN, the advantages of FSR were:
(1) researchers having first-hand familiarity with farmer's conditions and prob-
(2) productivity constraints could be resolved regionally through an interchange
between the experiment station and farmers;
(3) integration of socio-economic considerations into agricultural research and
development to avoid recommendation of technologies with little chance of
successful adoption by farmers;
(4) increased confidence in long-run research planning (e.g., the need for soil
conservation experiments) and early warning of new problems (e.g., the
increasing incidence of sorghum downy mildrew, Sclerospora sorgi); and
(5) the generation of improved linkages between extension and research.
The Administrative Organization of PNIA
Agricultural research in Honduras has progressed, since the 1950s, from the Inter-
American Technical Service for Agricultural Cooperation (STICA), through the Coopera-
tive Service for Rural Development (DESARRURAL) which was included initially (in 1974)
in the newly-formed Directorate General of Agricultural Development (DESAGRO),
finally into PNIA in the General Directorate of Agricultural Operations (DGOA).
A partial flow diagram of the structure of the Ministry of Natural Resources (SRN)
necessary for historical background is presented as Figure 1. Three levels of decision-
making affect research: 1) Ministerial, 2) National Programs in the DGOA and the
decentralized Regional Directors, and 3) The Regional Research Coordinators.
Institutionally, a major problem that was recognized from the outset of the FSR
effort was that both the DGOA and the regional directors were at the same administra-
tive level and that PNIA leadership did not have full control over its regional research
The operation of PNIA as a national entity is complicated by
the MRN organizational and administrative structure. The head of
PNIA has limited authority to implement approved programs be-
cause he does not control regional expenditures of PNIA's budget.
The Regional Directors control these, and thus can name and
manage research personnel, including their travel both within and
outside their region. This makes it impossible to have a strong
coordinated national research program. Further complicating the
issue, is that the different regional directors control the regional
research budgets to various degrees. (IADS, 1978)
This difference between technical and administrative linkages and, thus, loyalities (the
former national and the latter regional) of research staff, was never fully resolved.
H. ORGANIZING FSR IN HONDURAS
The FSR research unit was originally located in Tegucigalpa, the capital. An IADS
mission assisted the FSR unit in preparing a basic document (IADS, 1978) outlining the
reorganization of research in PNIA. The document provided the justification, objectives,
goals, budgets, and manpower needs to implement FSR in Honduras from 1978-83. The
IADS report considered three basic sources of funding to implement FSR: (1) Honduran
SRN budget appropriations for agricultural research, (2) a USAID Grant earmarked for
reorganization and strengthening of PNIA, and (3) an IBRD/IDA Loan to also assist in
reorganization and strengthening PNIA.1 The IADS Report formed the basis of a
subsequent USAID Project Paper, the result of which was a grant of approximately $1.9
million to the Government of Honduras, administered through PNIA. Together with the
expected IDA loan, which was to be approximately $2.0 million, a total of $3.9 million in
external funds was available to the SRN to assist in funding the reorganization of PNIA
research through 1983.
To instill a more national approach to research focused on the basic grains at the
farm level, a change in the research organization was proposed for 1978. This
reorganization consisted of three levels: (1) the national headquarters in Tegucigalpa;
(2) the Central Unit (CU) for FSR methodology development and training; and (3) the
PNIA National Headquarters
The headquarters was to contain the head of research and an administrative
assistant. This small unit was to:
(1) represent research at the national level;
(2) assume overall responsibility for achievement of program objectives;
(3) coordinate other domestic and foreign institutions and organizations with
agricultural research interests in Honduras;
(4) give ultimate decisions and supervision on research activities; and
(5) provide administrative and other support services (IADS, 1978).
A key to the reorganization would have provided the head of research with control of
budget and personnel so that regional research expenditures and personnel changes would
require PNIA approval.
IThese funds were not provided solely for FSR. There were also appropriations to
build up the experiment stations and crop commodity projects so that they could better
backstop the future FSR teams.
FSR Central Unit (CU)
The FSR Central Unit (CU), located from 1978-1980 in Comayagua, consisted of
technical experts brought together to develop the FSR methodology for Honduras and
provide training for a critical mass of Honduran agronomists in FSR research. The CU
was located at the research station in Comayagua, to allow the CU team members to
conduct both farm and station research. The original goals of the CU were to:
(1) develop and implement FSR methodology for use in the regional units;
(2) provide in-service training in FSR;
(3) conduct research to support regional research efforts;
(4) coordinate and supervise PNIA research on a national basis; and
(5) provide technical assessment of PNIA research (IADS, 1978).
FSR Regional Units
The multidisciplinary regional FSR teams were the cornerstones of Honduran FSR.
One regional unit would be trained by the CU each year, and eventually at least one team
would be located in each of the seven administrative regions. Flexibility of composition
was stressed, so that the disciplinary make-up of a given team could reflect regional
priorities. The regional FSR teams were to:
(1) identify target farmer groups through surveys;
(2) collect agro-economic information on the predominant farming systems) of a
(3) identify production constraints;
(4) introduce improvements in farming systems; and
(5) document farmer acceptance of improved systems (IADS, 1978).
FSR Organization in Honduras
The FSR team met ten times between August and the arrival of the IADS
assessment team in October of 1977. Some of the ideas and concepts which emerged from
these discussions were incorporated into the document, "Agricultural Research in Hon-
duras" (IADS, 1978). Regions 1, 3 and 6 were visited in August and September, and the
reorganization proposals were presented to local PNIA staff. During October, the IADS
mission arrived, and similar trips were also made to regions 1, 2, 5 and 6 (see Figure 1).
At this time, the first disagreements surfaced between the relationship of existing
research activities and the newly-proposed FSR activities. The debate between commodi-
ty program research and FSR continues today. The existing head of PNIA and one plant
breeder objected to FSR on the grounds that it would take too much power away from
crop projects and concentrate it in the new Central Unit. The difference in opinion
centered on the decision to refocus research on existing conditions of the Honduran
farmer. The reorganizers of research insisted on a farm-oriented research program, with
the greatest concentration of manpower at the regional level.
FIGURE 1: ADMINISTRATIVE REGIONS OF HONDURAS
\ REGION 4 LITORALL ATLANTIC)
LA CEIBA ,
REGION 3 (NORTH) )
SSAN PEDRO SULA
-, 1 (REGION 5 (EAST)
0 -V OLANCHO
REGION 7 > ., --
-(WEST) < ( T E ^<
W RTl REGION 2
SA NTA (CENTRAL-WEST)
COPAN DE COMAYAGUA REGION 6 -j -
Creation of the FS Central Unit (CU)
The location of the CU was a major topic of discussion among different authorities
in the Ministry, members of the Program and other advisors. Three locations were
considered: (1) Region 7, the region of greatest poverty; (2) Region 3, a region of high
agricultural potential, close proximity to Regions 7 and 4 and a major city conducive to
family life; and (3) Region 2, a region close to the capital, geographically near the center
of the country and possessing the necessary research infrastructure. Several international
experts recommended that the CU be located in the headquarters (in Tegucigalpa). Since
SRN policy at the time was to promote regionalization by discouraging the addition of
agricultural research staff to Tegucigalpa, Region 2--Comayagua--was selected for
location of the Farming Systems CU.
Comayagua is 80 kms from Tegucigalpa by paved highway, has enough building
space, an adjacent experiment station, and has a wide range of different environments and
associated farming systems. In time, even though Tegucigalpa and Comayagua are
relatively near one another and telephone services are fairly good, communication proved
to be less efficient than required between the CU and headquarters. Thus, too much
stress can be placed on regionalization. The FSR implementing unit should be located
physically with those decision-makers responsible for guiding the methodological develop-
ment of FSR and obtaining long-term financing for such methodology. Shortly after
agreement on the location of the CU was reached, a Honduran plant breeder was
nominated to head this unit.l All researchers agreed that a Honduran should also lead the
field FSR effort.
Duties of CU Staff
Once it began operations in Comayagua, the FSR staff in the CU had to
develop the on-farm research methodology, adapting their activities to the prevailing and
perceived institutional framework of the Program. The researchers in the CU also had to
lend technical assistance to researchers in different regions of the country, and
participate actively in training incoming and present research personnel. Furthermore,
the CU had to lend support to the Headquarters in policy development and implementa-
tion, project analyses, coordination with external groups and exploration of new research
Up to this time, all researchers were assigned crop coordinating responsibilities.
Upon appointment as head of the CU, the new head was relieved of these responsibilities
to allow him more time to concentrate on initial CU coordinating duties.
The FSR core group comprising the CU was made up mostly of expatriates.
The professionals divided work mainly along disciplinary lines. However, strong interdisci-
plinary interactions characterized working relations. All researchers were expected to
conduct field work themselves. The CU made a conscious decision that everyone in the
unit would get hands-on experience while learning to set out farm trials. The CU was also
expected to integrate the Comayagua station into their research plans. Thus, a replicate
of each trial was included on-station from the beginning.
The work schedule of the CU researchers included periodic visits to the
different regions to give technical assistance to regional PNIA staff. While this worked
after a fashion, certain regions were much more open to interaction and exchange of ideas
than were others.
While regional and commodity responsibilities were evolving, additional duties
for the CU members were those of traditional disciplinary expertise. For example, in
addition to his responsibility for Regions 5 and 6 and beans, the entomologist had national
responsibility for entomology. Thus, most CU researchers had responsibilities for regions,
commodities and disciplines. It soon became apparent that this arrangement spread
expertise too thin and required too much time, but a satisfactory solution to this dilemma
was never attained.
Integration of Other Researchers Into the CU
The CU sought to prepare young Honduran agronomists to undertake FSR. The
CU evolved from a small core group in 1977 (made up of a plant breeder, entomologist,
economist, and two agronomists) to a unit composed of several discipline specialists, an
in-service training group, an animal science research group and intermediate technology
project staff by 1979. Expatriates assigned to PNIA under different institutional
arrangements were incorporated into the CU. This mechanism integrated personnel from
various externally funded projects into a central support group. The integration of
expatriates working under some international contracts encountered some interinstitu-
tional friction. Usually when this occurred, these institutions had preconceived roles for
their professionals. Integration within PNIA required compromises that were at least
partially acceptable to PNIA leadership.
Crystallization of CU Research Priorities
After one year of field implementation, the CU met and decided on the
following research priorities:
(1) Assistance to, and presence in, the regions
To provide this assistance most efficiently, Regions 2, 4, 5 and 6 were given
priority. Reduced emphasis was given Regions 1, 3 and 7.
(2) Verification of technology (of PROMYFSA)
At this time, PROMYF (the Corn and Bean Project) was incorporated into the
Ministry under the DGOA (General Directorate of Agricultural Operations). PNIA,
and subsequently the CU, was assigned to provide technical assistance to
PROMYFSA technology verification units.l
This phase of FSR retained its high priority.
(4) Documentation of FSR Methodology
This was one phase of FSR which was never satisfactorily completed.2
History of the CU
Frequent group meetings characterized the operation of the CU during 1978.
Monday meetings in Tegucigalpa gave way to Friday meetings in Comayagua. All CU
researchers were expected to attend such weekly planning meetings, unless busy with field
trials. Travel and work plans for the following week were discussed, as was distribution of
work vehicles. In addition to administrative matters, these meetings provided an open
forum for discussion of FSR methodology, planning of seasonal work loads and trials, and
reflection on, and analysis of, the general direction of the FSR focus. Also, discussions
involved relations between PNIA and other programs, such as extension, human resources,
and sector planning.
In addition to the regular Comayagua meetings, sub-groups of CU researchers made
frequent regional visits to promote FSR methodology. This was one of the drawbacks of
the operation of the CU: many more ideas were generated than could be implemented,
and, in addition, too much time was spent in trying to introduce FSR into regions lacking
even the minimum institutional conditions to launch FSR.
One constant and partially deserved criticism of the CU researchers was that they
did not spend enough time interacting with other PNIA researchers. Such interactions
could have been facilitated by: (1) relocation of some heads of commodity projects in
Comayagua; (2) the nomination of counterparts who could have worked closely with each
Incorporation of PROMYF into the Ministry was not accomplished without consid-
erable controversy over FSR methodology. This controversy still surfaces today among
different PNIA researchers.
2The FSR manual prepared by the CU contains a detailed list of procedures and
observations which should have been followed or taken (SRN/PNIA/Unidad Central, 1978).
However, time limitations meant that some of the agreed-upon routine procedures (e.g.,
soil samples and subsequent analyses) or observations (e.g., plant densities) were not taken
at each farm site. Documentation of the FSR practices actually carried out was never
Farming Systems Researcher in the CU; and (3) regular in-service training sessions. Of
these, only the third item was realized.
The joint meetings between the CU staff and the regular PNIA Comayagua
researchers were a waste of time. It would have been much more efficient to have been
able to maintain liaison with PNIA technicians with open attitudes toward the FSR
philosophy. However, PNIA leadership had neither the inclination nor the power to
implement FSR without agreement from regional directors and crop commodity special-
ists. Only greater force from the Head of PNIA on up through the director of DGOA,
including the Minister, would have added the necessary power to accomplish the
administrative changes to implement FSR more smoothly.
The new research areas or disciplines, such as animal science and intermediate
technology, were relatively easy to integrate into the CU as they had played no
traditional role in PNIA. Integrating crop specialists was never accomplished, with the
exception of early corn and sorghum. There were several reasons for this. One was the
reluctance of regional directors to let qualified personnel move to Comayagua. Another
was outright opposition of the older heads of commodity research programs to support the
FSR effort in Comayagua. Thus, in 1979-80, the heads of the maize, rice, and bean
projects participated very little in CU activities. It should be recalled that the original
FSR focus in Honduras was precisely on these basic grains.
By early 1981, the influence of the CU support unit had been significantly reduced in
size. There was a large staff turnover as personnel were removed or reassigned. The in-
service training activity was cancelled for a number of reasons, including the lack of
personnel to carry out the training and the philosophical incompatibility of DGOA
leadership with on-farm research as represented by the CU. The May 1981 evaluation of
the PNIA program recommended that the CU be revitalized and that every effort be made
to continue some form of the in-service FSR training program (USAID, 1981).
Summary of FSR Organization
When FSR is added to an existing on-station research system, it will not automati-
cally be welcomed with open arms. Some opposition is to be expected from those already
established in the agricultural research system. Those introducing FSR should remember
that FSR: (1) has to prove itself anew in any given country or region; (2) may take
resources away from on-going research (e.g., breeding or variety trials on experiment
stations); and (3) has start-up costs which have to be justified (e.g., large recurrent budget
for petrol). In addition, those initiating FSR may have to deal with some of the following
(1) How will the newly-formed FSR units interact with established researchers?
(2) Where will the FSR central support unit be located, and will its training
function cover all regions?
(3) How long should a training session last?
(4) Who should be trained (agronomists, M.S. candidates, social scientists, etc.)?
(5) Should the FSR trained technicians be located in a single region as a regional
team upon successful completion of FSR training, or should they be assigned
individually to regions based on need or other criteria?
(6) Will the host country government be able to incorporate training costs into its
regular annual budget to provide continuity through time? and
(7) Is the proposed FSR methodology suited to the political, economic, and
institutional constraints of the host country?
Il. DEVELOPMENT AND IMPLEMENTATION OF FSR
The approach to FSR and the methods of implementation have evolved considerably
since 1978. In this chapter we document important elements of the methodology and its
evolution. Table 2 presents an outline of the key stages in the methodology as it was
developed in 1978 and those stages being practiced in 1980. The reader will be referred to
this table throughout the chapter, which has four major sections. Section one examines
the procedures used in selecting regions, farmers and specific problems to be included in
on-farm and on-station trials. Section two considers the experiences of implementing
farm trials and farm record keeping systems. Section three analyzes the results and
efforts to extend FSR to other regions of the country. The final section discusses training
of Honduran researchers.
Selecting Regions and Farmers, and Design of Farm Trials
Regional Reconnaissance (Sondeo)
Field work in FSR began in January, 1978 as CU sub-groups visited the Camayagua
region and farmers in various zones. We selected zones based on political boundaries,
using the municipality as the basic unit. This had the advantage of giving our study zones
meaningful governmental boundaries, a base in census data, and coincided with the
mandates of other governmental agencies (especially extension).
During February and part of March, the FS economist collected census ownership
lists and corresponding parcel maps for those municipalities covered by the National
Agrarian Reform (INA) census of 1974. The agronomist collected available information on
soils and rainfall for the Comayagua valley.
The system of reconnaissance used by the CU was to interview informally one to
five farmers, either in their fields or their homes, in groups of two to three researchers
(Table 2). Interviews lasted approximately one-half hour. Their purpose was to identify
the predominant crop systems, and the spatial relationships of crops within the system.
Drawing a simple map in the dirt as the interview progressed was the best way to obtain
specific agronomic information, such as the distance between rows and hills. This phase
lasted one month. However, the CU had only one work vehicle assigned to it at this time,
meaning that farmers from only one zone could be visited at any given time.
After members of the FSR CU team wrote up trip chronicles of the reconnaissance
surveys, the CU met and decided to interview farmers in four zones: (1) La Paz, La Paz
Department; (2) Flores; (3) El Rosario; and (4) San Jeronimo, all of Comayagua
Evolution of Honduran FSR Methodology, 1978-80
FSR Cropping Season Phases
1. Informal Reconnaissance Survey
2. Formal Diagnostic Survey
3. Farm Trials:
a. Planning and Design
b. Selection of Collaborators
c. Installation of Trials
d. Observation and Management
4. Installation of Farm Records
5. Analysis of:
a. Farm Trials
b. Farm Records
6. Presentation of Results:
a. To Other Researchers
b. To Farmers
c. Selection of Repeat Treatments
1. Formal Reconnaissance Survey
2. Farm Trials:
a. Planning and Design
b. Selection of Collaborators
c. Installation of Trials
d. Observation and Management
3. Analysis of Farm Trials
4. Presentation of Results:
a. To Other Researchers
b. To Farmers
c. Selection of Repeat Treatments
Department. These zones differed in agro-climatic characteristics, predominant cropping
systems and cropping potential (see Table 3).
The following zones were considered for farm trials in 1978-79. Flores was
ultimately rejected, while the other three were selected.
Located some 15 km south of Comayagua, with abundant irrigation, Flores had
the greatest potential for both system changes and yield increases of existing systems.
This zone was rejected because: (1) the number of different cropping systems was very
large (25 systems were identified on the 28 farms surveyed) and selecting the dominant
ones would have been difficult; and (2) it was the first zone to be planted each season, and
the CU realized there would not be enough time to design proper farm trials for the zone.
Located approximately 12 km west of Flores, on the west side of the
Comayagua valley, La Paz has adequate soil moisture and access to irrigation which
differentiate farmers in this zone. Depending on the season, the towns of Cane and
Lejamani were also included in this zone.
This zone, between 20 and 26 km northwest of Comayagua, was a mountainous
area with fairly low soil fertility and a potential erosion problem. Having the lowest
agricultural potential and the least flexibility with regard to alternative cropping systems,
this zone was selected as representing typical Honduran hillside agriculture. Since corn
and beans were the predominant crops, it also fit in with PNIA's basic grains emphasis and
Honduran government priorities.
Located some 35 km north-northeast of Comayagua, this zone is a small valley
at the base of the mountains on the road to La Libertad. The latter is a coffee zone, and
most of the San Jeronimo farmers either have some coffee in the hills, or work in the
coffee harvest.I This zone is intermediate in agricultural potential and cropping systems
complexity (Table 2). With the exception of Flores, San Jeronimo was the most
homogeneous agro-climatic zone of the four. Even so, there were two different main soil
types present which largely explain the location of the rice and corn cropping systems in
One early hypothesis was that basic grains management in the zone might be fairly
slack during the coffee harvest but this hypothesis could not be confirmed.
Characterization of FSR Zones, Comayagua Region of Honduras, 1978
Average No. of
Is Irri- Percent Farm Existing In Order of Importance, In Order of Importance, the
gation a Average Size, Cropping the 3 Predominant Cropping 3 Most Important Problems
Zone Possibility? Slope Ha. Systems Systems Identified Were: Listed by the Farmers Were:
Flores Year-round 0-2 5.0 25 1) Corn in spring followed 1) Fall armyworm in corn
by corn intercropped 2) Bird damage in corn
with beans in fall 3) Slugs (Babosa) in beans
2) Tomatoes in spring
3) Three tied for 3rd:
a) Corn monoculture
b) Rice monoculture
c) Corn in spring followed
by tomatoes in fall
La Paz Supplemental 0-2 10.6 19 1) Corn intercropped with 1) Fall armyworm in corn
sorghum in the spring 2) Bird damage in sorghum
2) Corn monoculture 3) Bird damage in corn
3) Cassava monoculture
El Rosario No 25-30 3.6 3 1) Corn in spring followed 1) Slugs (Babosa) in beans
by beans in fall 2) Two tied:
2) Corn and sorghum inter- a) Fall armyworm in corn
cropped in spring b) Bird damage in sorghum
3) Corn monoculture
San 3eronimo Yes, but cur- 0-2 4.5 10 1) Rice monoculture 1) Fall armyworm in corn
rently there 2) Corn monoculture 2) Two tied:
is almost none 3) Corn intercropped with a) Bird damage in corn
beans b) Carapacho in rice
aAs identified from 28, 28, 28, and 27 questionnaires for Flores, La Paz, El Rosario, and San 3eronimo, respectively.
Formal Questionnaire (Encuesta)
With assistance from the rest of the CU, the FS economist and the first CATIE
outreach agronomist prepared the initial formal questionnaire (Table 2) to elicit more
specific information about the predominant cropping systems.
The initial questionnaire was pre-tested in one day using a sample of 20 farmers in
the highlands surrounding La Esperanza (Comayagua Department), as interest in FSR had
already been expressed by researchers there. The enumerating teams, composed of one
researcher and one extension agent, discussed the results on the second day, and several
questions were modified or eliminated.
The questionnaire was then modified. Several regions contributed the necessary
manpower and vehicles to facilitate in administering the modified questionnaire to the
four previously identified zones in the Comayagua Region. During four days of farmer
interviews, 111 questionnaires were completed and later summarized to aid in the design
of the farm trials during the first cropping season. Some of the disadvantages
encountered in administering the first questionnaire were:
(1) It was done by novice interviewers;
(2) The CU had not reached internal agreement about which types of information
should be gathered;
(3) Some traditional PNIA researchers resented the questionnaire process, as they
thought it was a step backward in agricultural research in Honduras;
(4) Although invited to participate, extension workers could not participate
because of scheduling conflicts;
(5) The lists of farmers in three of the zones--drawn by random sample from 1974
property owner's lists--were more trouble than they were worth because many
on the list were owners of convenience or because the parcels had changed
hands. However, interviews based on random personal contacts (in El Rosario)
may have provided a sample biased toward the "better" farmers, as some
respondents would point the interview group toward Don Juan, because "he's a
good farmer"; and
(6) Tabulation of much of the questionnaire was done one weekend by two
researchers and not simultaneously in the evenings by the enumerating teams.
Thus, there was no indication that certain questions were systematically being
slighted until it was too late to remedy the situation.
Design of Farm Trials (Ensayos de Finca)
The CU then discussed which zones to work in and what types of trials should be
conducted in each zone (Table 3). The preliminary results of the questionnaire were
discussed for the first time. During this discussion, some researchers maintained that
farm variability, being greater than experiment station variability, required a greater
number of trials for statistical representability, while others contended that the research
group was too small to be able to manage such a large number of trials. A compromise
was reached between these two points of view and zonal farm trials were designed.
Meanwhile, zonal selection was based less on agronomic, economic or social
importance, and more on geographic diversity and contrasts which would help the
researchers refine the FSR methodology and provide a variety of work experiences for in-
service trainees. A major problem encountered during this phase was that the question-
naire summary did not adequately characterize each zone. For example, in La Paz, three
distinct spatial arrangements of corn and sorghum were recorded. Trials were designed
around these three and a fourth spatial arrangement from Nicaragua. However, it was
later discovered that time of planting was more critical than spatial arrangements.
Planting time was initially overlooked because cattle production was not given adequate
priority in the cropping system. However, time of planting was determined by the
farmer's favoring either forage or grain production.
The CU concluded that the formal questionnaire phase was too general and too
inflexible to refine the impressions of the reconnaissance survey meaningfully.
Selecting Experimental Sites for Farm Trials
While designing farm trials, any FSR team should consider how they will resolve the
following issues to make best use of their given manpower and mobility constraints:
(1) How many zones can the team cover adequately per region?
(2) Within each homogeneous zone, how many farms (sites) are required to
represent the underlying zonal variability?
(3) Once the number of farms/zone has been determined, how many trials would
the FSR team recommend per site?
(4) Given the distance between FSR headquarters and the zones, and the geo-
graphical distribution of farms within the zones, how many trials/farm can the
FSR team manage comfortably?
(5) If the number of trials determined by (3) above is greater than that from (4)
above, the team should defer the extra trials until a later season or until more
manpower is available.
(6) The FSR team must verify that the number of replicates/trial, and the number
of treatments (plots)/replicate does not exceed the size of the smallest area
borrowed from a collaborating farmer in the zone.
(7) The maximum amount of area a team should borrow from a given farmer,
particularly when FSR has no history in the region, is 10 percent of his
We established at least five experimental sites (farms) in each zone. Each site
contained two to four different on-farm experiments or trials. The zones were reasonably
well defined and small enough so that five sites were judged adequate to insure coverage
of the within-zone variability. Between 10 and 20 individual trials on five or more farms
was about the maximum load that a conscientious FSR team of two to three researchers
could manage with one vehicle. Of utmost importance in site location is site description.
Thus, five farms within a zone are adequate only if the farms are consciously selected to
represent the variation in topography, soil type and farm management prevalent in the
We borrowed 1,000-2,000 m2 of land from each cooperating farmer. This area
permitted us to plant two to four trials of three to six treatments (plots) each. An
individual trial averaged 500-600 m2 in area. Plot size varied but for corn and sorghum
we used approximately 25 m2 and rice and beans, 15 m2. Most experiments had three to
four repetitions, as well as border plots and alleyways to facilitate entry and good note
taking, but some had as few as two repetitions. In areas with both Spring and Fall
plantings, the latter had fewer experiments so that the team was not overwhelmed when
Spring harvest overlapped Fall planting in November or December.
The philosophy of the CU was to plant at least one variety trial each season,
enabling farmers and researchers to see a wide range of newly-released materials. These
trials also allowed the farmers to begin thinking about new cropping calendars: calendars
based on disease resistant varieties or new shorter cycle varieties. The remaining two or
three experiments were usually based on questionnaire data analysis and focused on plant
protection (weed control, insect control) or agronomy (plant spacing, date of planting,
Selection of Collaborating Farmers
Potential collaborators were contacted with the help of extension in the zones of La
Paz and San Jeronimo, and by the researchers themselves in El Rosario, a zone with no
extension service representation. In the three zones it was not too difficult to identify
the major within-zone variation and select cooperators whose farms covered this
Criteria used for selecting collaborating farmers were: (1) The farmer operated a
small-to-medium sized farm; (2) he was planning to plant most of his land to the
predominant farming system in the zone; and (3) he had at least enough land so that a
total loss of the trials would not cause him great economic hardship.
Assigning FSR Teams to Homogeneous Zones
Three teams of two researchers each were formed to conduct the farm trials in the
three zones. By the summer of 1978, each zone had at least two researchers sharing
responsibility for the farm trials. A work vehicle was assigned to each of the three
As agronomy students in their final year of study at CURLA (University Center of
the Litoral Atlantic Region) became available, they were recruited into the CU by the
Honduran FSR leadership. These students generally spent 10 months as in-service
trainees, and each was assigned to work in one of the three zones. From the beginning,
the trainees were considered co-workers on the FSR teams. Each was expected to devote
60-75 percent of his time to field work and FSR application, and to develop a meaningful,
FSR-related, senior thesis topic.
Field Supplies and Seeds for Farm Trials
The next step was to assemble experimental inputs (seeds, fertilizer, insecticides,
herbicides) and support supplies (vehicle, tape, shovels, machetes, sprayers, tags, etc.) so
that trial planting was rapid and professional (Table 3). In accumulating supplies and
materials, funding from IDRC (International Development Research Council of Canada)
and CATIE (Agricultural Center of Technology and Education) was crucial to initial FSR
research.' A small fund was made available to the CU so that input purchases could be
made quickly from local stores. Grant or loan availability at the local level is often a
truer measure of effectiveness than its amount.
Farm Trials and Farm Record Keeping
Methodology of Implementing Farm Trials
Planting the first crops in the FSR associations began in El Rosario on the 2nd
of May, 1978, and ended more than a month later, on the 27th of June, 1978, in a farmer's
field in the zone of San Jeronimo. Altogether, 67 trials were planted on 23 farm sites in
the three zones. The total number of parcels observed during 1978 was 660.
Whenever possible, the FSR team would lay out the trial and then monitor the
farmer and his helpers to make sure they planted the trial correctly. Some trials had to
be planted directly by the FSR team, because: (1) two collaborating farmers were
planting the same day; and/or (2) the farmer did not wait for the arrival of the FSR team
before planting his own portion of his field. Participation of farmers in planting ranged
from none to significant.
1The first CATIE outreach agronomist set up a petty cash fund for small item
purchases and work vehicle maintenance. He also made his own vehicle available for work
(as did the FSR training specialist).
Monitoring and Observing Trials
Many crops emerged in early trials before other trials were even planted.
Thus, the FSR teams made independent emergence observations before the CU met to
standardize measurement and monitoring procedures. In addition to observations taken on
the crops during the season (plant counts for density; disease evaluation; etc.) several
farmers in each zone were provided with rain guages. They recorded daily rainfall data
which were collected about once a month by a team member on a standardized form.
The harvest should be carried out with the farmer's assistance. Coordinating a
harvest with the farmer is perhaps even more difficult than coordinating planting. Where
two or more crops are grown in association, they are seldom harvested at the same time.
This means scheduling at least two harvest dates per trial site. In addition, proper harvest
timing, equipment and measurements (both before and after harvest), are critical in
assuring continuity of worthwhile data.
Planning Field Trials, Second Half of Season (Postrera), 1978
All farm trials in the fall of 1978 contained legumes (common beans, cowpea
and mung bean). Cultivars were selected based on continuous personal contact with the
bean project. These fall trials were planned during a CU meeting the last week of August.
At the same time, station trials were proposed on the control of nutsedge, minimum
tillage, and control of Apion (Apion godmani Wagn.) and Empoasca (Empoasca kraemeri).
While there was some informal discussion on how to integrate FSR and station
trials, the CU never determined how best to utilize an experiment station to complement
farm trials. One objective of the station trials was to provide in-service trainees greater
exposure to different types of research trials. However, these trials, especially during the
1979 season, diluted the manpower of the FSR teams. Some CU members felt that more
of the station effort should have been dedicated to a better understanding of the local
farming systems. It was also easier and less of a risk for the trainees to dedicate more
time to thesis topics on-station than on-farm. However, one of the unique advantages of
FSR over traditional research is that it can provide guidance to students for thesis topics
which are relevant to the predominant crop systems of a homogeneous zone.
Another issue concerning FSR and station trials was the irrelevance of some of
the farm trials to researchers working in the Comayagua experiment station. Replicates
of all trials in El Rosario, San Jeronimo and La Paz were placed on-station. However,
climatic, topographic and agronomic conditions differed between the former two zones
and the station to such an extent that their trials were not really replicates of the farm
trials. In addition, yields in the trials replicating La Paz research were double those of
the average of the farm trials in that zone, severely undermining the contention that
these trials were simply replicates of the La Paz trials.
Analysis of Field Trial Results
During each cropping season, teams kept field notes and a master notebook that
remained in the training center. These sources of information proved very useful when
the experiments were analyzed (Table 3). The analysis began with a simple analysis of
variance, and then, ad hoc across-location comparisons were made to determine which
treatments) consistently out-performed the check throughout the zone.
Unfortunately, not enough time was allowed for this phase of FSR. This phase also
proved to be the most difficult stage for the in-service trainees. They encountered
difficulty both in the mechanics and the interpretation of the analyses.
No formal across-site analysis was done, due partly to the time constraint and partly
to high site and zone heterogeneity. Different parcel sizes were harvest on different
farms, which caused some replicates and/or treatments to be eliminated at some sites.
These field-level modifications were often necessary because the proposed trials were
either too ambitious in scope or too large for certain farms. Finally, no economic
analyses were performed on any 1978 data.
Redefinition of Farm Trials
During the Spring, 1979 Farm Trial Workshop for the second group of trainees,
the major results of the 1978 trials were presented for La Paz, El Rosario, and San
Jeronimo. One of the main conclusions from the 1978 season was that there were too few
CU researchers to supervise the farm trials and strengthen ties to researchers in other
regions. Thus, it was decided that 1979 farm trials would be managed by in-service
trainees. Trials were developed for each zone, and groups of trainees assigned to each
zone. A similar process was used in preceding from the 1979 to 1980 trials. Table 4
provides a summary of the Comayagua Region FSR trials by year and type.
Implementing A Farm Record Keeping System
Questionnaires, regardless of how well-designed, often fail to elicit the responses or
provide the details desired by researchers. This is especially true for certain socioeco-
nomic questions because:
(1) people frequently conceal their economic situation;
(2) a lack of a record-keeping system leads to guesses which cannot be confirmed;
(3) farmers associate questions about economic matters with attempts to increase
taxes, or with the Agrarian Reform program; and
Types of FSR Trials Conducted in the Comayagua Region, By Zone, 1978-80
Zones: La Paz El Rosario San Jeronimo
Corn varieties Corn varieties Rice varieties
((Corn + sorghum) + soil insecticide) Sorghum varieties Rice demonstration lots
(Corn + fertilizer) (Corn + soil insecticide) (Rice + weed control)
Legume varieties Legume varieties (Rice + fertilizer)
Soil insect control
Corn varieties (Corn + fetilizer) (Herbicides + dosage + timing)
(3 corn varieties + 2 bean varieties) Corn varieties: 3 maturities (Fertilizer + dosage + timing)
(Local corn+local sorghum) versus Bean varieties Rice variety trials
(Local corn + improved sorghum) Soil conservation
((Corn+sorghum) + fertilizer)
(Corn varieties + fertilizer)
(Sorthum + spacing + fertilizer +
Livestock trials, including: Corn varieties Corn varieties
-fodder from sugar cane Fertility Rice varieties
-livestock survey (Density + variety + fertilizer) Weed control
-mixed farming systems (Corn + sorghum) Date of planting
Maturity: corn versus bean
(4) recollections of weekly labor use are too often inaccurate for use in
recommending intensive technological changes, especially if such changes
require more labor than is likely to be available in the zone during this time
Farm records were introduced to obtain more detailed information than could be
collected through a formal questionnaire. The farm records were designed to generate
such information as daily labor inputs by field, cost of purchased inputs, etc. During 1978,
a farm record form was prepared by modifying those used at ICTA (Institute of Science
and Technology of Agriculture) in Guatemala to include units of measurement unique to
Honduras. Nine collaborators (four from La Paz, three from El Rosario, and two from San
Jeronimo) were selected to handle farm records. All farmers had CU farm trials, and six
kept rainfall records.
The FS economist was aided in installing records by the training specialist and the
extension agent in La Paz, the extension agent in San Jeronimo and a member of the first
group of trainees in El Rosario. Visits to collaborators were made at intervals of
approximately two weeks to confirm data collection compatibility.
Several major problems were encountered in managing these farmer-maintained
records. These included the following:
(1) The farm record phase received low priority from the CU staff;
(2) CU researchers were spread too thin in managing farm trials to adequately
supervise farm records;
(3) the amount of time necessary to install and monitor a group of farm records
was continuously underestimated;
(4) the lag time between the explanation of how to complete the entries in the
record books and the time when a farmer could actually perform this exercise
reasonably well was longer than anticipated; and
(5) the necessity for farm records was never adequately explained to the FS
The final point was crucial, as goals and objectives of record keeping need to be
specified far in advance of the installation step. An untimely illness in the CU led to the
early demise of the initial record books.
In 1979, the CU used record forms developed for the National Development Bank
(BNF) by an Oklahoma State University/Colorado State University team of agricultural
economists. Only those sections of interest to the research objectives of the CU were
completed. Use of these forms initiated a tentative link between PNIA and BNF, and
lessened duplication of effort.
Assistants were hired to help with record keeping in 1979. Criteria for selection
included literacy, living in the community, sufficient spare time, and an interest in
completing the data collection. One assistant was hired for each zone. Using a mixture
of selection procedures, each collaborator was interviewed once a week by the regional
assistant. Entries maintained included farm labor, input purchases, and grain sales. At
the suggestion of the BNF group, the visits were usually made on either Saturday or
Sunday, when the farmers were most likely to be home. This interview schedule proved to
be quite satisfactory.
Some of the problems encountered in farm record keeping during 1979 were:
(1) delay in contract approval and pay disbursement from PNIA was so long that
one assistant left the project early (after a work stoppage and a subsequent
work slowdown), eliminating the chance to obtain any useful information from
(2) the BNF forms proved to be too complicated, and outlines of simpler forms
were developed in late November, 1979, for use in 1980; and
(3) summaries of costs of farming operations and grain sales were not completed
in time to assist CU research staff design trials.
As a result, the times of peak labor demand were not extracted from the records for use
in modifying trial design for the following season.
In 1978, the CU researchers designed five questionnaires to focus on specific
problems in cropping systems, to give the new trainees some experience in survey
techniques and to design more relevant experiments for the FSR program. They dealt
with the farming systems and sesame cropping systems of Choluteca, land preparation
techniques in semi-arid La Paz, soil erosion problems in steeply sloping El Rosario and
common bean systems and problems in the regions of Danli and Olancho.
Farming Systems in Choluteca
Forty-five farmers were interviewed in five zones by representatives of
research, extension and human resources. The five zones contained between one and five
sub-zones. Between 7 and 11 questionnaires were completed per sub-zone. While the
results were partially analyzed by CU researchers, regional researchers never completed
the analysis. The methodology used in selecting farmers for interviews also detracted
from the validity of the data-gathering process. While farmer interviews in some sub-
zones were conducted at random, other sub-zones interviewed only farmers from the
contact lists of the extension service personnel. Thus, results between sub-zones were not
Sesame Cropping Systems of Choluteca
The bean questionnaire (see Bean Systems and Problems in Danli and Olancho,
which follows) was used to design a questionnaire for studying sesame cropping systems.
1While farm records were also kept during 1980, the record book assistants had not
been paid from December, 1980 until the preparation of this paper in mid-1981.
Suspicious that they may never receive their back pay, they are holding their record books
Twelve farmers were interviewed. Seven of the 12 growers planted sesame in association
with either corn or sorghum, while the sesame project had never conducted a trial in
association with any other crop. These results led to a revision of the objectives of the
sesame project research.
Land Preparation in La Paz
In La Paz, the land preparation questionnaire, developed and applied by the FSR
agronomist, revealed that farms with less access to irrigation water placed greater
dependence on livestock. This affected the crop residue handling, accessibility of animal
traction, and the type of crops grown. As a result of this survey, the CU focused on
timing of land preparation as a solution to the cultivation bottleneck which normally
occurred with the first May rains, investigating land preparation from March through
June, with oxen and tractors, and with and without herbicides. Plowing the dry soil,
waiting until the rains began and then planting corn with herbicides was an option which
permitted earlier planting, left more time for the Fall relay crop of beans, and was not
Soil Conservation in El Rosario
The purpose of the soil conservation survey in El Rosario was to gain a better
understanding of how farmers perceived their soils, yields, and soil fertility. Also
examined was the time farmers spent in land preparation activities. The results of this
survey indicated that a minimum tillage strategy had little relevance since most of the
stover and forage was either eaten by livestock or burned. Also, the use of living
hedgerows (pineapple, sorghum, lemon grass) on contours made little sense to farmers
renting fields because these fields were heavily grazed by the landowner's herds during the
dry season. As a result, the CU decided to work only with farmers who owned their land
and had it fenced. Fencing allowed farmers to plant fruit trees and sisal on the portions
of their holdings that could be taken out of annual cropping, and construction of contour
walls or live hedges (from locally available materials or plants) on the land remaining in
Bean Systems and Problems in Danli and Olancho
Questions of bean culture practices led to the development of the bean
questionnaire for Danli and Olancho. It was also designed to interest bean researchers in
discovering on-farm problems of bean production. The highlights of the Olancho results
(1) row-to-row and plant-to-plant spacing were quite variable;
(2) earliness and drought tolerance were important plant characteristics;
(3) lodging was recognized but not considered a problem in farmers' varieties;
(4) seed size was found to be more important than color;
(5) the variety planted was either cuarenteno or cincuenteno;
(6) all plantings were by oxen and in monoculture;
(7) the use of insecticide was low; and
(8) storage methods were variable but seemed to include insect control measures.
Being less reliable, the Danli results are not presented. However, the questionnaire
served to show the bean research staff a bean region more important than Danli with
distinct cultural practices and varietal traits. Before the questionnaire was administered,
the Olancho region was not served by the bean program.
Training Honduran FSR Teams
Since a large proportion of agricultural college graduates in Honduras came from an
urban and university background which left them ill-equipped to communicate with
farmers and perform on-farm research, there was need to develop a practical training
program for young researchers in PNIA. The training phase of FSR began informally in
1978 when the first group of three trainees joined the three regional FSR teams.
Formal training began in early 1979, and combined theoretical aspects, specific
themes, seminars and workshops, with the practical training experiences of characterizing
zones and assisting in the design, implementation, management, harvest and analysis of
the farm and station trials. Training lasted 10 months, with 75% of the trainee's time
dedicated to practical field experiences. The training stressed areas which were weakest
in the trainee's previous education: statistical analysis with major emphasis on interpre-
tation of results and drawing conclusions, experimental design, pest control and manage-
ment, economic analysis and technical communication.
The driving philosophy behind training was to encourage young agronomists to study
cropping systems from the farmer's point of view, including control of insects, diseases
and weeds and the relationship between the soil, water and plants. The training program
required each trainee to come in close contact with the problems actually experienced by
Honduran farmers through the study of integrated crop systems, thus giving trainees the
capacity to identify farmer's predominant problems, analyze the existing situation and
suggest possible methods for improving it. Training was considered an essential step in
producing technicians who could understand and carry out FSR in all regions of Honduras.
The detailed objectives of training were to:
(1) Integrate theoretical and practical issues;
(2) Stress the development of independent thought over memorization;
(3) Stimulate imagination and creativity;
(4) Stimulate deeper understanding of how regional and national problems affect
(5) Develop an appreciation of multidisciplinary research and promote the ability
to understand systems of production encompassing biological, agronomic,
cultural, and socioeconomic aspects;
(6) Develop an attitude conducive to team rather than individual research; and
(7) Promote the understanding that PNIA should work with other institutions
involved in regional and national development.
A list of the formal short courses and seminars given to the trainees is provided in
Table 5. Topics ranged from very specific ("environmental effects and stalk rot in corn")
to very general ("regional development"), from one-hour seminars to week-long courses
(short course on Integrated Pest Management); and from office-oriented (short course on
statistical analysis) to field-oriented (short course on weed control).
Results of In-Service Training, 1978-80
During 1978, the ratio of CU FSR professionals to in-service trainees was 2:1, and
there was daily contact leading to assimilation of FSR philosophy and methods. In 1979,
nine trainees completed the program. There was less trainee/researcher contact during
this training season. Problems which arose in 1979 included poorer farm trial manage-
ment and inadequate attention to field books. During 1980, eight of nine trainees
completed the program. The ratio of field work to theory during 1980 was about 60:40.
Although this balance was less than ideal, the trainees spent a great deal of time in
intense field work.
Positive Aspects of FSR In-Service Training
Experience Working with Farmers
Farm level training offered the trainees an opportunity to deal directly with
farmers and to work under farmers' conditions. This experience served to demonstrate
the ability of the farmer to handle advanced technology. The concepts of risk and
environmental variability were grasped very quickly by the trainees, and the farmer's
decision-making rationale was appreciated. Also, the ability of each farmer to integrate
different agronomic, biological, cultural and socioeconomic factors to make his particular
production and marketing system viable was often appreciated fully for the first time.
An important benefit of FSR training revolves around researchers' attitudes
toward farming, farmers, and the role of agriculture in national development. At least
one-half of the trainees were deeply affected by the experience, leading them to
reconsider both their profession and the role of research. This positive change of attitude
toward the agricultural sector of Honduras was a very important by-product of the
training program. In Summary, FSR training effectively allows professional agronomists
to recognize farmer needs and to acquire the ability to improve technology to fit
traditional farming systems' constraints.
Short Courses and Seminars Presented to FSR Trainees, 1978-80
Seminars Short Courses
Environmental effects and stalk rot in corn None given
Insect control in crop systems
Technical tour of ICTA program (Guatemala trip)
A concrete experience of regional planning
A survey of reformed groups: some selected results
General aspects of Sorghum Downy Mildrew (SDM) Statistical analysis
Minimum tillage Selecting clean seed in common beans
Elements of a regional diagnostic survey Applied statistics (workshop)
Agricultural extension Research in agricultural systems of production
Linkages of agrarian politics Weed control
Improved pastures in Honduras IPM (Integrated Pest Management)
Research in soybeans Farm management
Economic analysis of farm trials
Insecticides Introduction to in-service training
Managing technical information Statistical analysis
Regional diagnostic surveys Soil conservation
Pastures and forage crops Farming systems
Sorghum Downy Mildrew (SDM) Dairy cattle
International programs Communicating technical information
Goats Definition of homogeneous zones (workshop)
Rural technology Managing data collection in homogeneous zones
Farm record keeping Developing questionnaires (workshop)
Minimum tillage Administration of, and results tabulated from,
Farming systems in La Esperanza (workshop)
Discussions of farm trials in FSR (workshop)
Scientific photography (workshop)
Planning fall farm trials (workshop)
Extending FSR to Other Regions
Since 1976, the proportion of total experiments conducted by PNIA researchers on
farms has jumped from approximately 12 percent to over 52 percent (USAID, 1981). The
Olancho Region has gone from virtually no agricultural research in 1977 to 161 trials in
1980, of which 61.5 percent were conducted in farmer's fields. Plans for 1981 are for an
even greater on-farm effort. The region, which contains the Guayape Valley Project
(roughly a $9.5 million World Bank loan with about $4 million in Honduran government
counterpart funds), plans to hire two more researchers for the existing zones of
Catacamas and 3uiticalpa, and one each for the new zones of San Francisco de Becerra
and San Francisco de La Paz. The regional PNIA research coordinator, a former CU
trainee, and the Director of the Guayape Valley Project have been able to integrate
regional research and extension, including the sharing of planning, vehicles, and field work
activities. While this regional effort is a special case, it points out the advantages of
coordinated FSR efforts.
IV. KEY PROBLEMS IN ORGANIZING AND INTRODUCING FSR
The purpose of this chapter is to highlight key problems encountered in introducing
Problems of Introducing FSR to Personnel of
PNIA and Other Institutions
PNIA program headquarters and the CU had two major tasks for implementing FSR:
(1) promotion of the idea within PNIA and (2) securing internal political and budgetary
support. The accepted role of program headquarters was one of orientation, coordination,
resource procurement and distribution. Regional operations were controlled by regional
directors. After expressing initial reservations, many regional directors tentatively
accepted the philosophy of FSR.
The visits of the CU FSR researchers were received with complete acceptance in La
Esperanza but were opposed in San Pedro Sula. More indifference to on-farm research
was encountered from the more traditional crop specialists, who favored the traditional,
technology package (tech pack) approach. This approach is of limited use for long-term
rural development in Honduras. CU researchers spent considerable time explaining that
the gradual approach of on-farm research was not incompatible with high technology, high
production research methods, but was complementary to it, as FSR addresses the
problems of those farmers who cannot afford traditional tech packs.
A sub-group of SRN researchers opposed FSR from the beginning. Members of this
sub-group attacked the importance of farm research while defending the traditional
emphasis of commodity projects. These researchers also objected to the FSR emphasis on
research conducted under the predominant conditions of the farmers and their correspond-
ing crop systems, and the system of in-service training. They thought the status of the
researcher within the SRN was in jeopardy. A lack of social sensitivity and excessive
disciplinary zeal rounded out the attitudes of those opposed to FSR. Finally, this group
expressed legitimate concern about the extra time required to implement FSR.
The management system of the SRN did not permit the head of PNIA to reward
researchers of merit, or to replace or transfer those who were uncooperative. At times,
some of the better researchers in PNIA were hampered by regionally-imposed administra-
tive and logistic constraints, and PNIA leadership was powerless to counteract this.
However, some research personnel soon adopted the FSR guidelines and suggestions
provided by the CU, and the La Esperanza research group employed innovative FSR from
Another source of internal friction was the emphasis on on-farm research. FS
research was viewed as being directly opposed to improving the physical infrastructure of
existing experiment stations. The CU could have emphasized more that strong commodity
projects are necessary for FSR support, and that FSR was complementary to, not a
competitor of, such projects.
The PROMYF project, funded for three years (1976-78) by the Interamerican
Development Bank (BID) and based in Danli, was a joint research effort backed by
CIMMYT and CIAT to increase production of corn and beans. This project was based on
tech packs using free credit from BID. PROMYF trials were production/demonstration
oriented, with research being limited to the verification of imported technology.
Collaborators were selected on the basis of production potential, not need. Methodology
and problems of execution limited the value of the results of these trials. Very little
cooperation existed between PROMYF and other interested Honduran institutes, including
In 1979, when external funding for the project ended, PROMYF was incorporated
into the DGOA of the SRN and renamed PROMYFSA. PNIA and the CU became involved
in planning experiments carried out by PROMYFSA. Finally, PROMYFSA became the
Basic Grains Project and included rice and sorghum.
Methodologically, the problem of PROMYF is that technology packages are only
useful to a small percentage of Honduran farmers, leaving the majority unaided.
PROMYF would tell these farmers to eliminate various inputs from the tech pack based
on their (tech pack-1) experiments. But, since most small farmers cannot afford to
change more than one or two small practices during a given cropping season, these
growers need the opposite information: what tech pack component, then added to his
existing system, yields the highest marginal return? The (tech pack-1) experiments do not
permit calculating such returns.
FSR, as instituted by PNIA, was obstructed as often as assisted by some internation-
al institutes. Many external problems stemmed from the bureaucratic tendency of these
institutes to defend their privileges and terms of reference so as to remain in favor with
their donors. In the late 1970's, these institutes were more interested in using national
programs as replicates for progeny testing trials than they were in assisting them in their
The Honduran commodity programs generally have too little manpower to handle all
of the requests from the International Agricultural Research Centers. For example,
CIMMYT requested Honduran corn researchers to carry out International Progency
Testing Trials (IPTT) and Experimental Variety Trials (EVT). In 1977, 9 IPTT's, each with
512 observation plots, were planted in the Guaymas station in Region 3. Such trials
compete with the primary task of the few trained breeders to adapt improved varieties
and technologies to Honduran farming conditions. IPTT's are far too complex for
Honduras to handle. International centers should assess the capabilities of a given
national program or project and not encourage or permit such overloads on trained
During 1977 and 1978, some international center's outreach teams bypassed the head
of PINA to work directly with commodity project heads. Fortunately, this situation was
resolved by 1981. The economics unit of CIMMYT and the new CIMMYT outreach team
support both the FSR philosophy and the training effort. CATIE encouraged FSR and
shared its experiences with the CU. As a regional center, in contrast to an international
one, CATIE attempted to assist PNIA researchers adapt FSR to the physical, biological,
economic and political realities of Honduras. While CATIE does have its own farming
systems methodology, an effort was always made to consider Honduran priorities first.
Other outreach programs have much to learn from this outreach philosophy. While some
of the positive interaction between PNIA and CATIE was based on personalities, the
absence of a patronizing attitude was highly valued and contributed toward getting on
ICTA was perhaps the most relevant source of FSR information for PNIA, and
frequent trips were made to Guatemala. In some cases, ICTA personnel came to Honduras
to participate in CU training activities.
Problems of Linking-Up with Other Units
of the Ministry and Outside Institutions
Extension and Research Linkages
A gap exists between the Honduran extension service and experiment station
research. Honduran extension agents have less formal training, more institutional
mobility, and face the requests and recriminations of farmers more directly, than
researchers do. Rarely does research develop technology directly useful to extension and
farmers. Extension has not participated in research planning, so extension agents have, in
general, ignored past research activities.
Since extension methodology was also in transition during 1978 and 1979, this
provided an opportunity to develop extension and on-farm research linkages. Extension
began by characterizing their zones to permit the agents to become better acquainted
with farmers and their problems. Using this information, each agent submitted a work
plan to address the pressing problems of his zone. This information also provided a
quantitative base for forming research priorities. In 1979, zonal characterizations were
made collectively by regional personnel from both extension and research. One regional
director participated personally to show his support for the approach. By 1979, some good
informal working relationships had developed between on-farm research teams and
extension agents. Implementation and acceptance of FSR was invariably faster when
extension was involved from the beginning.
The CU and PNIA Headquarters
In time, communication between the CU and PNIA headquarters became more
sporadic, often the result of emergencies. Added activities at both levels prevented
PNIA and Sector Planning
Sector planning of SRN is responsible for reconciling the Annual Work Plan
with the SRN program budget (for presentation to the Ministry of Finance), and for
evaluating regional performances. However, Sector Planning has often proposed crop
production and diversification projects in the past with little or no attention to
Recently, working relations between PNIA and Sector Planning have improved,
culminating with the joint preparation of the Annual Work Plan and Budget of 1980. The
work plan was developed upward from the local extension offices through the central
headquarters of all programs. This process was so effective that budgetary discussions
among Sector Planning, the DGOA and the Ministry of Finance, normally lasting three
days, required only a morning meeting.
PNIA and SAID
Project funds from USAID were to have been disbursed directly from PNIA,
but the Honduran Government required such disbursement via the Ministry of Finance.
Funds were disbursed when PNIA submitted vouchers against a revolving fund in the
Ministry of Finance. Initial disbursement was slow, and PNIA was held responsible for any
irregularities. Purchases of equipment had to satisfy both Honduran and USAID
For example, the working relationship between the FSR team and the extension
agent in San Jeronimo was excellent. There was no extension agent in the El Rosario
zone, and those in extension in La Paz showed little interest in FSR. In Olancho,
excellent working relationships have developed between research and extension, largely
due to the efforts of the research coordinator and the regional director.
2For example, the castor bean project was introduced with a large publicity
campaign, but without basic agronomic information and no improved seed. Initial
plantings of imported seed were eliminated by a disease which was normally of little
importance on the scattered castor plants indigenous to Honduras. This project had to be
cancelled after two years because planning was not based on feasibility studies.
An AID project manager with sufficient agricultural experience to reconcile
field operations with administrative issues would be of great assistance to FSR implemen-
tation. Strict adherance to the 3-bid procedure for local procurement, for example, is
absurd during certain FSR field operations.
PNIA and Other External Institutions
Viable working relationships between PNIA and other institutions (e.g., CATIE,
IDRC, IICA, CIMMYT, CIAT, CIP, and the Peace Corps) were achieved with varying
degrees of difficulty, or not at all. In cases where initial reluctance was encountered,
such an attitude was stemmed from a lack of respect for Honduras' past research effort.
Such skepticism was not without some merit. It was particularly difficult to establish
better links with CIMMYT, CIAT, and IICA, because of their traditional interactions with
PNIA. Once the goals and objectives of FSR were better defined by PNIA researchers,
collaboration generally improved. One exception was the working relationship between
PNIA and IICA. As PNIA matures and internal regional priorities become more clearly
defined, external agencies will have to tailor their support more toward nationally-defined
needs and goals.
PNIA and Regional Directors
The PNIA/Regional Director relationships depended on the personal attitude of
the regional director towards the new research focus. FSR received regional approval in
Regions 4, 5 and 6, was ignored in Region 2, and continually faced serious problems in
Regions 1 and 3.
The relations of the DGOA, and PNIA in particular, with the seven regional
directors have improved greatly since 1977. Initially, relations were strained as the
normative role of DGOA/PNIA and the operative roles of the regional directors were not
well defined in 1977. By 1978 and 1979, the role of National Programs were better
defined, resulting in improved relationships between PNIA and the regional directors. A
major step in improving this relationship was completion of joint work plans in 1979 and
1980 by PNIA and regional PNIA staff. While redistribution of earmarked research funds
and unapproved personnel changes still occur regionally, integration of work plans and
budgets has reduced these problems.
Problems of Acquiring Sufficient Resources to Implement FSR
While high-level administrators in the SRN approved of FSR, the task of implement-
ing it and selling the philosophy to the rest of the researchers was left to PNIA and the
Regionalization was not begun seriously until 1975 in Honduras.
CU. However, research was not a line item in the annual budget, so PNIA Headquarters
had to locate funds on its own to initiate FSR. The only viable sources of FSR funding
were external from 1977 to 1980.
Disbursement of funds from AID began in mid-1979, eighteen months after FSR
began in Comayagua. Disbursement of AID funds was slow because of administrative
procedures, but this funding became a stable and vigorous source of support to FSR. A
good working relationship was established between PNIA and the local AID mission.
Several CU members were paid with AID funds, and work vehicles and other equipment
were purchased. Sporadic financial support was also obtained from other institutions
(CIAT, FAO, CIMMYT, Peace Corps, and IICA).
The Honduran Government contribution to PNIA was limited to cost of living
increases, with no attempt being made to increase the proportion of budget going toward
research (USAID, 1981). Honduran research funds were administered by regional directors
who did not always adhere to Program budget allocations. However, in some regions,
budgets from other Programs were used to support research activities.
In summary, there was no formal SRN budgetary support for FSR. Funding for FSR
came from external agencies and from whatever sources PNIA Headquarters could
convince to contribute toward FSR.
Separation of the unit implementing FSR from headquarters resulted in a loss
of communication. Even though Comayagua was selected for locating the CU partly
because of its proximity to Tegucigalpa, there were still serious lags in communication.
The head of research could not monitor the FSR team and implementation problems as
closely as he could have if the field unit and headquarters had been located together.
Support Facilities, Supplies and Vehicles
While Comayagua had the physical facilities to accommodate the FSR team,
offices were actually in laboratory or storage areas. While the station had electricity,
there was no running water and no public facilities. Those working out of Tegucigalpa
later in the program had trouble locating either office space or desks.
Little flexibility existed for purchase of inputs and repair parts, with the
exception of the IDRC petty cash fund. FSR depends on local farmer surveys to
determine research priorities and experimental design, trial numbers and necessary inputs.
This sequence of events takes place immediately prior to planting. Because of this
timing, inputs cannot be anticipated and accounted for in a previous budget, but must be
purchased as their need is determined in the field. Most national programs and external
donors do not allow for the budgetary flexibility necessary to account for this change
from traditional research methodology.
Transportation was a major problem. The entire FSR effort began with a
worn-out, 4-year-old Toyota Landcruiser. Using only this vehicle in all three work zones,
such vital steps as the zonal reconnaissance, farmer questionnaires, farm trial planning,
planting and initial farm trial observations were carried out. Work vehicles promised to
some of the expatriate CU researchers in January, 1978 were finally in use in the program
in July, 1979. FSR depends heavily on adequate field mobility for timely trial operations
and observations. Mistimed operations (especially planting) can render trial results
worthless. One vehicle per work zone is absolutely essential to FSR.
The gasoline shortage problem of the SRN is legendary. The regions normally
spend their annual gasoline budget months before the end of the fiscal year.
The size of the PNIA staff increased from 45 in 1977 to 90 in 1981. Movement of
personnel was frequent during these years, due to resignations, transfers and leave for
advanced study. FSR added a logistic and philosophical burden to a small national staff.
The lack of Honduran professionals with FSR training meant that the new approach was
implemented largely by expatriate CU personnel. Hondurans returning from foreign study
to continue agricultural research generally were unresponsive to the FSR philosophy,
going so far as to oppose it at times. In order to increase the short-run research benefits
to a particular region, some regional directors would assist returning researchers to obtain
assignment to the regional experiment station. Such an assignment would mean that the
researcher could ignore FSR and concentrate on experiment station research.
Counterparts were not formally assigned to expatriates because of the general lack
of trained Honduran manpower. Counterparts could have reduced some of the excessive
visibility of the expatriates at policy-making levels. A lack of counterparts meant that
the CU researchers encountered difficulty in leaving primary duties to attend to
secondary duties in other regions. Expatriates involved in FSR should have assigned
V. CONCLUSIONS AND RECOMMENDATIONS
Conclusions from the Honduran FSR Experience
The FSR approach to agricultural research began in Honduras with an official
attempt to reorganize agricultural research in PNIA. While all of the goals set out
initially were never realized, a good deal was accomplished in modifying research
philosophy and methodology.
FSR is currently at a crossroad in Honduras. Hondurans involved in FSR CU training
had to reevaluate the relationship between the farm, the farmer and agricultural
research. The broader picture acquired by these young agronomists because of their close
work with farmers as a part of a multidisciplinary field research team is an experience
traditional research cannot hope to duplicate. Researchers trained in FSR are much more
sensitive to total farmer constraints.
There is a distinct difference between a nominal and a real FSR multidisciplinary
team. The former is composed of technicians of various disciplines with common goals
working in a common area. The latter is a cohesive unit whose very strength is derived
from an open interchange between, and mutual support of, disciplines. It is an evolving
entity which requires time to reach the necessary level of frankness and interpersonal
trust for truly multidisciplinary interactions.
At times, research within the CU and PNIA was hampered by many obstacles. This
factor, together with a lack of trained Honduran personnel, meant CU researchers
consistently had more to do than could be well-managed. Requests for their time were so
frequent that national research programs suffered. In addition, research programs have
both limited agronomic and administrative capacities. External donors and international
institutions have to be careful to avoid overloading the administrative and technical
capacities of national programs. For various reasons, several of those trained in FSR may
soon leave the program. Moreover, training has been eliminated. On the other hand, if an
effort is made now by the SRN and PNIA to reinforce FSR (through contract renewals and
reinstatement of training), the philosophy may yet be strengthened and be extended to
other regions of the country.
Recommendations for FSR in Other Countries
General Political and Administrative Issues
To integrate FSR into an agricultural research program, FSR leadership must have:
(1) an explicit mandate and assistance from policy makers; (2) the power to hire, promote,
transfer and fire research personnel; and (3) the ability, time and motivation to manage
the research personnel. The integration of FSR into established commodity-oriented
research programs will never be free of friction. It is important to minimize such friction
by involving established researchers from the beginning and using their knowledge and
expertise to help orient the FSR group to national crop research programs, possibilities,
and areas of collaboration. FSR cannot function effectively without strong commodity
research teams, so they should be strengthened as well as reoriented.
International centers, working through national research leadership, should assess
the willingness and ability of national research systems to carry out meaningful
collaborative research. Many Third World countries (e.g., India and Egypt) have more than
adequate technical staff to collaborate with international centers, while others (e.g.,
Central American countries in general) lack sufficient trained manpower. Sensitivity to
the absorptive capacity of a national agricultural research program should be accorded
high priority by international centers.
Specific FSR Integration Issues
(1) Homogeneous Zone Surveys:
A well-conducted, formally-structured reconnaissance survey can replace both
the informal (sondeo) survey and the follow-up formal diagnostic survey (see Table 2).
Less manpower, vehicles and per diem will be used to conduct a single formal
reconnaissance survey than if an informal reconnaissance and a formal diagnostic survey
are employed. A quick group summary of formal field notes should provide enough
information to select homogeneous zones and assist in design of the first year's farm
There is some debate on the types of surveys which are needed to select
experimental zones for on-farm research. On the basis of our FSR work in Honduras and
the experience of ICTA in Guatemala, we recommend the multidisciplinary reconnais-
sance survey approach.
(2) Formal Farmer Questionnaires:
Unless a FSR team has a large number of researchers and/or extension agents,
as well as excellent mobility, plenty of time, and a good mechanism for analysis of
questionnaires, this step should be eliminated.
(3) Distribution of FSR Manpower and Responsibilities:
During the first year of FSR implementation, distribute available manpower
rationally and as equally as possible among zones. If a zone lacks sufficient manpower or
vehicular coverage, it should not be included in the research effort until sufficient
resources become available. FSR teams should staunchly resist the tendency to spread
themselves too thin to be effective. They should take on additional duties and
responsibilities slowly and after careful deliberation.
(4) Farm Trial Design and Installation:
Between the survey of farmers and the final design of farm trials, the FSR
team should carry out more dialogue with the farmers. During these visits, systems
identified by the survey or questionnaire should be discussed in greater detail. Planting
times, methods, spacing of rows and hills, and other details need to be confirmed and/or
refined. In addition, the research group should share with the farmers their plans for farm
trials, asking for reactions. Perhaps some have already had experience with some of the
proposed treatments. Either positive or negative results would be extremely useful to the
FSR group at this time to refine farm trials to account for regional realities. These
farmer discussions may be group sessions to save time. Follow-up visits should allow
design of check plot treatments for the proposed trials, using information from local
farmers about the average practices per cropping system.
Ideally, the collaborator's farm size should be near the average for the zone.
Researchers should not request more than 10 percent of the land of a farmer's holdings.
Farmers with the largest and smallest holdings should not be asked to collaborate, to
avoid bias of scale.
Since the importance of weeding and diseases are often underestimated by
farmers while the importance of problems of varieties and insects are often overestimat-
ed, we recommend using clusters of simple two-level factorial trials- during the first
season in order to gain information from farm trials about potentially limiting agronomic
factors. These trials may involve some or all of the following: an improved cultivar
versus the farmer's variety, a hypothesized optimum level of fertilizer against the level
used by the farmer (often zero), a general contact (or soil) insecticide against the farmer's
practice (often no insecticide), a general foliar fungicide versus the farmer's disease
protection method (often none), and an extra hand weeding or application of an
herbicicide versus the typical farmer practice (usually a cultivation plus zero, one or two
Enough farm trials need to be placed in a homogeneous zone to represent its
average agricultural potential. FSR teams should make sure that they conduct enough
trials to avoid failure of acceptance when the recommendations are extended to the rest
of the zone.
A two-level factorial trial consists of all combinations of two levels of all inputs
considered variable in a given trial. A trial with (1) seed variety, (2) fertilizer,
(3) insecticide, and (4) planting density, represented as 2 consists of 16 distinct
treatments. All other variables (e.g., weed control, disease control) would be held
constant in the trial.
(5) Harvesting Farm Trials:
Several weeks before the anticipated harvest, the FSR team should arrange
with each farmer a harvest time to avoid prematurely harvested trials. Many seasonal
observations become meaningless if they cannot be correlated with parcel yields.
(6) Analysis of Trials:
Trials should be designed for joint statistical and economic analysis. The
ANOVA and economic analyses should be completed together soon after the trials are
harvested, to allow enough time between the analysis for planning next season's trials
using all available trial information. Interpretation of the results by the researchers and
in-service trainees should be an integral part of any such analysis.
(7) Farm Record Keeping:
If manpower is insufficient to guarantee proper record keeping, this step
should be eliminated from FSR. If farm records are kept, frequent contact between the
FSR socioeconomist and the record-keeping assistant in each zone is necessary, especially
at the beginning of each season. Many unanticipated questions always arise, and decisions
on how to handle them consistently have to be made on the spot in the field. Misplaced
and misinterpreted data entries also have to be corrected at this time.
Formal training courses should not interfere or compete with practical farm
work, but should complement it. Instructors should show their fallibility and willingness
to learn from mistakes. This attitude will affect the trainees, instilling them with more
respect for their collaborators and the farmers with whom they work. Those involved in
training must foster a correct climate for team work to occur. This climate includes the
open interchange of diverse ideas and flexibility of FSR field implementation.
Stress on good field planning and farm trial execution and proper timing is
essential. We recommend at least weekly field contact between supervisors and the
training group. During trial design, development, implementation and location of
collaborating farmers, such contact should occur on a daily basis.
Ideally, training should include both future extension agents and researchers,
and be administered jointly by both programs. The benefits to a given group of trainees of
working in zones where the local extension agents) are competent and have excellent
working relations with the farmers are many, including joint research and extension
The economic analysis can be patterned after the CIMMYT economics manual
(Perrin, et al., 1979), and modified to fit the local conditions of the country.
(9) Personnel Issues:
If a country has sufficient manpower, counterparts should be assigned to work
with, and learn from, expatriates. More disciplinary expertise and farm-oriented
philosophy will remain in the FSR program if counterparts are assigned from the
beginning. In addition, counterparts should be aware that they will be expected to carry
out the duties and responsibilities of the expatriate when he leaves the program.
Countries with severe manpower limitations must place maximum stress upon the FSR in-
service training phase and hope that natural disciplinary leaders emerge from the
nationals in the program.
(a) Between Research and Extension. Extension agents need to be relieved of
some of their mundane administrative duties to be effective co-participants in FSR.
Research should not exploit the extension agent only for his contacts and the extra muscle
he provides, but rather should accept him as a valuable addition to the FSR team effort.
Finally, the relationship between research and extension should be formalized at the
(b) Between FSR Headquarters and FSR Field Unit. The field team should not
be physically separated from headquarters, at least until the program is functioning
(c) Between FSR Leadership and External Donors. FSR leadership and
external donors must agree on the goals and procedures of both institutions, and FSR
leadership must have the political support to insure donor compliance.
(d) Between FSR and Regional Programs. Care must be given in developing
relationships between FSR and existing regional programs, especially if regional autonomy
is highly regarded, and regional manipulation of the national research budget is a political
reality. When regional leadership has control over research personnel, it is essential that
regional leadership be convinced of the value, goals and objectives of FSR, not just at a
national level, but also at each regional level.
APPENDIX: VIVID FSR EXPERIENCES
The original CU FSR researchers were each asked to contribute their most vivid
recollections of their experience. This appendix presents these unedited reflections in
1. Lack of professional advancement opportunities. The time required to com-
plete the task was underestimated and the group members saw little chance to progress in
the immediate future.
2. Administrative inefficiencies. The continuous frustrations of working under a
bureaucratic public organization with its unexpected problems, delays, and faulty logic
made the effort very difficult. This was also true for foreign personnel working under the
3. Lack of definition within the organization. The Ministry suffered from a
generalized lack of clear definition as to the roles played by the different levels and
groups. Implementing FSR became a series of accommodations and approximations
between the central and regional groups. This contributed to the creation of undue wear
and frustration, making it impossible to reach completion of the task within the time
framework considered initially.
1. More than in other activities, on-farm research places the scientist in a
continuous learning situation. On the one hand, the experience is very stimulating; on the
other, one is overwhelmed at times by the magnitude and diversity of the farm problems.
The necessity for integration into a team is quite evident. The learning experience comes
about equally from direct observations, analysis and reiteration of the field problems, and
from conversations with the farmers.
2. It is impressive to confirm the natural intelligence and refined sense of
observation possessed by countless illiterate farmers. I'll always remember what one of
them said: "Peasant farmers are illiterate, but not ignorant." To learn to respect and
appreciate this peasant farmer culture is a substantial part of the on-farm experience.
3. Based on this respect, dialogue with farmers becomes open and fruitful. I never
encountered the stock phrase, resistancey to change," in the majority of these farmers. I
encountered more resistance to change among extension and research personnel.
4. These farmers have a proper, traditional culture that integrates them into their
environment. The trained agronomists have lost this tradition, and do not always propose
a substitute culture which also gives the farmer security. I believe that this feeling of
insecurity adversely affects the behavior of the agronomist to accept changes. In-service
training contributes, in great measure, to the development of self-confidence in the
agronomist, and to the integration of the traditional and the universal cultural technology
which had previously been perceived of as being divorced and incompatible by the
5. The integration of a multidisciplinary team is a long and difficult process. The
high degree of specialization and individualism in traditional research contributes to this
difficulty, as do the existing methods employed in the evaluation of research work. I
believe that sometimes we worked as a team, learning from one another, and the total
work accomplished was greater than the sum of each individual's effort. Perhaps we only
lacked time to form a real multidisciplinary team.
6. I was also very impressed by being involved in and following the process of a
very positive change in some of the young agronomists who went through the in-service
training program. Their progress and enthusiasm was one of the most gratifying of all the
More stress should be given, from the beginning, to true integration during FSR
implementation. This is especially critical for non-traditional disciplines, such as
agricultural economics, rural sociology and anthropology.
Instability in agricultural research organizations makes research continuity problem-
atic, but renders FSR even more difficult because it requires a longer political time-
frame than traditional research does. Unless the rapid turnover of PNIA personnel is
slowed, little continuity can be expected. Without such staff continuity, increasing farm
productivity via FSR will be very difficult if not impossible.
All peasant farmers perform field operations only after they determine their own
logical priorities. The researcher must be made sensitive enough to assimilate and
analyze these individual priorities to arrive at the key technology changes which will also
be acceptable to the farmers of a homogeneous zone. The technical, political and
interpersonal interactions of multidisciplinary on-farm research are always impressively
Many administrative issues interferred with our work during the CU FSR experience.
Frustrations over many well-intentioned national and international laws (e.g., licensing of
work vehicles, tied aid funds) put numerous stumbling blocks in the path of FSR work.
Perhaps my most vivid experience in Honduras was my first day planting with Frank
Peairs and Franklin Rosales in El Rosario. The field we were given to work with was at 60
percent slope and we wanted to lay out a corn variety trial and an insecticide trial using
Aldrin. Three hours later and totally exhausted, we completed our task.
At that point, we were only dealing with the physical difficulties of FSR but it has
always stuck in my mind how much more demanding this type of work is than the
classical, on-station research. In addition to being physically harder to do, FSR is often
less satisfying scientifically as so many factors are uncontrolled and most of the data is
generated through casual observation and ad hoc measurements.
Notwithstanding these two personal hurdles to involvement in FSR, there are many
advantages. Team research is more fun and usually synergistic since each member brings
to the group a different expertise. Many farmers are excellent collaborators, both on the
technical and human level and this helps to stimulate the research. Finally, and most
importantly, FSR is the most logical research approach to promoting rural development
1. Marking a trial to be planted in Los Empates, El Rosario with Posner and
Rosales is one experience that I'll never forget. Three Ph.D.'s, all who did field problems
in graduate school, went to mark three 10 X 10 meter experiments. This took us three
hours on a farmer's hillside. Not only is FSR time consuming and strenuous, but more
importantly, conventional agronomy training does little to prepare you for FSR fieldwork.
In-house training is a must.
2. During a visit from representatives of the Consortium for International
Development, Rosales and I were asked to give a brief presentation of our FSR project.
Although it seemed almost every day to us, it took us about 6 hours to explain ourselves to
a group of three that probably had about 100 years of agricultural research experience
among them. This points out just how different FSR is from traditional agricultural
3. During a discussion with a representative of CIMMYT about our operations in
Comayagua, he told me that the Comayagua Valley was not a Zonaa maicera" (corn-
growing area), and that we shouldn't bother with corn research for that area. I had always
thought Comayagua was a corn-growing area. As it turned out, his Zonaa maicera" was an
area with a high per unit area production potential, while mine was one in which a high
percentage of the farmers grew corn. Since all of the Comayagua farmers grew corn, I
thought corn research to be essential for the area. FSR teaches one to look at an
agricultural area in terms of development--improving the farmer's lot in life by solving
his actual problems. The more traditional approach teaches one to look at the area in
terms of production potential, with little consideration for the farmer's actual needs and
The activities called for in the functioning of FSR are many and complicated. The
magnitude of the challenge to the participating technician is increased and considerably
more difficult than in traditional research. The responsibility of a researcher goes from a
narrow, unilateral focus (e.g., breeding for stalk rot resistance in corn) to a multidisciplin-
ary focus with various possible outcomes.
It is much more stimulating to understand, know and assist in the solution of a
common problem that affects the socioeconomics and agriculture of many farmers in the
country than it is to solve a genetic problem of a cultivar, as the results of the genetic
solution may or may not improve the level of life, production and productivity of the
majority of farmers.
Multidisciplinary participation is a must in FSR, and is the most stimulating and
interesting concept of FSR. Much is shared and many things learned which are not taught
at the university. However, this necessitates a desire to participate, and a flexibility to
adapt oneself to the different personalities, situations, etc., which this focus offers.
In-service training to prepare researchers or technicians in FSR has to include the
concept of direct and immediate service to farmers and to the agricultural problems of
highest governmental priority. Training, trainees and the feelings of the latter, have to
be distinguished from that present with traditional research. FSR cannot be a substitute
but, instead, must be a complementary ingredient in agricultural research. The FSR
technician is a hybrid between a researcher and an extension agent, as fieldwork and daily
contact with peasant farmers and their problems requires it. A great deal of personal
effort, logistic assistance and adequate mobility, among other requisites, are necessary to
succeed in implementing FSR successfully.
FSR should not be considered as being better than traditional research, because it is
not a substitute methodology, but is, instead, an extension of it which incorporates a basic
change in the mode of thinking about research method priorities. FSR is the system which
most likely will assist in resolving the problem of the link between research and extension.
CONSUPLANE. 1978. Plan Nacional de Desarrollo Agropecuario, 1979-83. Tegucigalpa,
Contreras, M.R., D.L. Gait, S. Muchena, K. Nor, F.B. Peairs, and M. Rodriguez. 1977.
"An Interdisciplinary Approach to International Agricultural Training: The Cornell-
CIMMYT Graduate Student Team Report." Cornell International Mimeograph 59.
Diaz, Alvaro. 1978. Una Experiencia de Reorganizacion del Programa de Investigacion
Agricola de la Secretaria de Recursos Naturales. Secretaria de Recursos Naturales,
Tegucigalpa, DC, Honduras.
1979. Una Ano de Trabajo en Ensayos de Finca y una Experiencia de
Capacitacion en Servicio en el Programa de Investigacion Agropecuaria. Secretaria
de Recursos Naturales, Tegucigalpa, DC, Honduras.
IADS. 1978. Agricultural Research in Honduras. Secretaria de Recursos Naturales,
Tegucigalpa, DC, Honduras.
Norman, David W. 1980. The Farming Systems Approach: Relevancy for the Small
Farmer. MSU Rural Development Paper No. 5. East Lansing, MI.
Nunez, Mario y Alvaro Diaz. 1980. Informe Anual de Capacitacion en Servicios.
Secretaria de Recursos Naturales, Tegucigalpa, DC, Honduras.
Perrin, R., et al. 1976. From Agronomic Data to Farmer Recommendations: An
Economics Training Manual. CIMMYT, El Batan, Mexico.
SRN/PNIA/Unidad Central. 1978. Guia Metodologica para Investigacion en Finca.
Tegucigalpa, DC, Honduras.
1979a. Analisis y Resultados de las Encuestas sobre Preparacion de Suelos en
La Paz, y Conservacion de Suelos en El Rosario. Comayagus, Honduras.
1979b. Trabajos y Ensayos de Finca: 1978. Comayagua, Honduras.
1980a. Funcionamiento del Programa Nacional de Investigacion Agropecuaria
y su Integracion en su Sistema Tecnologico. Tegucigalpa, DC, Honduras.
1980b. Manual de Actividades de Capacitacion en Servicio. Comayagua,
1980c. Proyecto de Capacitacion en Servicio del PNIA para 1981.
Tegucigalpa, DC, Honduras.
USAID. 1978. Agricultural Sector Assessment for Honduras. Annex K. Washington, DC.
1981. Evaluation of USAID Honduras Agricultural Research Project No. 522-
0139 with the National Agricultural Research Program. Office of Rural Develop-
ment, Development Support Bureau, USAID, Washington, DC.
Whyte, William F. 1981. Participating Approaches to Agricultural Research and
Development: A State of the Art Paper. ARE No. 1. Ithaca, New York: Rural
Development Center, Cornell University.
MSU INTERNATIONAL DEVELOPMENT PAPERS
IDP 1 Carl K. Eicher and Doyle C. Baker, "Research on $8.00
Agricultural Development in Sub-Saharan Africa:
A Critical Survey," 1982 (346 pp.).
IDP 2 Eric W. Crawford, "A Simulation Study of Con- $5.00
straints on Traditional Farming Systems in
Northern Nigeria," 1982 (136 pp.).
IDP 3 M.P. Collinson, "Farming Systems Research on $4.00
Eastern Africa: The Experience of CIMMYT and
Some National Agricultural Research Services
1976-81," 1982 (67 pp.).
IDP 4 Vincent Barrett, Gregory Lassiter, David Wilcock, $5.00
Doyle Baker and Eric Crawford, "Animal Traction
in Eastern Upper Volta: A Technical, Economic
and Institutional Analysis," 1982 (132 pp.).
MSU INTERNATIONAL DEVELOPMENT WORKING PAPERS
WP 1 Daniel Galt, Alvaro Diaz, Mario Contreras, Frank $4.00
Peairs, Joshua Posner, and Franklin Rosales,
"Farming Systems Research (FSR) in Honduras, 1977-
81: A Case Study," 1982 (48 pp.).
Copies may be obtained from: MSU International Development Papers,
Department of Agricultural Economics, Agriculture Hall, Michigan State
University, East Lansing, Michigan 48824-1039, U.S.A. Individuals
and institutions in Third World countries may receive copies free of