Component research in the farming...
 The role of agroforestry in FSR...
 Farming systems research in ILCA's...
 Extending FSR results in Lesot...

Title: Farming Systems Support Project newsletter
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00071908/00003
 Material Information
Title: Farming Systems Support Project newsletter
Alternate Title: FSSP newsletter
Physical Description: v. : ill. ; 28 cm.
Language: English
Creator: Farming Systems Support Project
University of Florida -- Institute of Food and Agricultural Sciences
Publisher: The Project
Place of Publication: Gainesville Fla
Publication Date: 1983-
Frequency: quarterly
Subject: Agriculture -- Periodicals -- Developing countries   ( lcsh )
Agriculture -- International cooperation -- Periodicals   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
periodical   ( marcgt )
Dates or Sequential Designation: Vol. 1, no. 1 (spring 1983)-
Issuing Body: Issued by: Farming Systems Support Project, which is administered by: Institute of Food and Agricultural Sciences, University of Florida.
General Note: Title from caption.
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00071908
Volume ID: VID00003
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 10387162
lccn - sn 84011294

Table of Contents
    Component research in the farming systems approach - FSSP extending methodology
        Page 1
        Page 2
    The role of agroforestry in FSR & D
        Page 3
        Page 4
        Page 5
        Page 6
    Farming systems research in ILCA's highlands programme
        Page 7
        Page 8
        Page 9
        Page 10
    Extending FSR results in Lesotho
        Page 11
        Page 12
Full Text


Component Research in the Farming Systems Approach
Robert K. Waugh*

One of the principal differences
between traditional research and
farming systems research and extension
(FSR/E) is the approach taken in the
generation and dissemination of
technology. Traditional research has
taken a Component (Commodity and
Discipline) Approach while FSR/E is a
Systems Approach. This seems to mean
to some people that Component
Research and FSR/E are not compatible.
The Component Approach em-
phasizes commodities and disciplines;
it generates technological components
but places very little emphasis on their
integration into farming systems and

thus tends to remain independent of the
systems approach. The Component
Approach is not necessarily the same
thing as Component Research,
however, because Component
Research can be systems-oriented to
facilitate the integration of technologi-
cal components into farming systems,
and which is reason to include it in the
farming systems approach.
Component Research is necessary;
but in order to contribute to agriculture
the results must be integrated into
production systems that are useful to
the farmer. In some cases the producer
himself integrates the results into his

production systems. But in those cases
where the farmer is not in a position to
do the integration, it seems logical that
a systems approach that integrates the
components is more effective than a
component approach. In fact this is one
of the basic premises of FSR/E and a
strong argument for making component
research a part of it.
The criticism of the component
approach does not lie so much with the
attention given to components as with
the inadequate integration of the results
into farming. That is how a systems
approach evolved and what it is
designed to do.m

*Dr. Robert K. Waugh is an international consultant in the area of farming systems, working with the FSSP at the University of Florida.

FSSP Extending Methodology

The Farming Systems Support Project
has been developing a series of slide/tape
presentations that address and explain
farming systems methodology. These
have undergone a development and
refinement process in conjunction with
workshop activities of the project and of
program associates. Valuable feedback
in the form of criticism and suggestions
from workshop coordinators and par-
ticipants alike has not only helped to
crystalize concepts in the FSSP presenta-
tions, but has also pointed the way for
additional slide/tape presentations that
need to be developed. A brief description
of those currently in use for orientation
and training is given below.

Introduction to Farming Systems
Research and Development describes
FSR&D approach to agricultural develop-
ment and offers some insight into its
evolution and purpose. The concept of
limited-resource farmers is explained in
terms of their commonality world-wide
and the importance of including them in
the development process.
Overview of Farming Systems Re-
search and Extension explores the
farming systems approach to technology
research, development and dissemina-
tion for limited-resource, small-scale
family farmers. It defines the role of FSR/E
and uses examplesto show how an FSR/E
program works. This module answers the

following important questions: Who is
expected to benefit from FSR/E? How
does FSR/E work to benefit this group?
Why would one expect FSR/E to work in
their country or area?
Economic Characteristics of Small-
Scale, Limited-Resource Family Farms -
Part 1 discusses the implications for
technology development in terms of the
resources and constraints these farmers
have to work with. It discusses the
concept of "limited-resource" farmers as
a social, cultural and economic environ-
ment of the family.
Part 2 of this module sequence moves
beyond the economic characteristics of
the farmer into the economic considera-

tions of technology inputs. It considers
the resource base and various outcomes
a farmer might expect in the allocation
of those resources. The presentation also
looks at learning curves, or learning to
use a new technology, as a function of
The Small Scale Family Farm as a
System discusses relationships among
and between the household, the crops,
the animals and the market. Three kinds
of systems in Asia are examined: swid-
den, humid uplands and lowland rice
agriculture. A farming system typical of
the Central American Highlands is also
examined. The models representing
these systems reflect the interactions
within the various systems and the
modifications that evolve as population
pressure increases and infrastructure
Land Use in Upper Volta is a case
study of the relationship between family
and farming systems. It describes the
subdivision of household land and some
of the implications of these subdivided
rights to its use. It stresses the importance
of understanding family economic and
social roles, since these can have a
pronounced effect on farming systems.
Defining Recommendation Domains -
uses a case study of Santiago Sac-
atepequez, Guatemala to define its
concept. In farming systems research,
recommendation domains are useful
delineations of geographical boundaries
for groups of farms with similar, farming
systems. The definition and usefulness of
this boundary is the topic of this presen-
Designing Alternative Solutions is a
series of case studies that portray the
farming systems approach to different
kinds of agricultural problems in various
parts of the world. Each slide/tape
presentation describes specified prob-
lems and factors influencing the design
of solutions for limited-resource family
farms. Current case studies in modular
form include: Zapotitan, El Salvador;
Jutiapa, Guatemala; and the North
Florida (USA) FSR/E Project.
Design and Analysis of On-Farm
Agronomic Trials concerns one of the
main tools in the farming systems ap-
proach to the development of technology
for small-scale, limited-resource family

farms. For farming systems research to
properly evaluate the technology, it is
necessary for the trials to be conducted
under the real conditions of the farmers
for whom it is being developed. This
module discusses both the trials and the
importance of farmer participation in the
evaluation process.
In addition to the slide/tape presenta-
tions listed, development is underway on
a continuing and complimentary set of
additional instructional materials. This
includes the following: InitalCharacteri-
zation The Rapid Survey or Sondeo;
Continuing Characterization; Heirarchy
of Constraints; Promotion ofAcceptable
Technology; Enterprise Records; Di-
rected Surveys; Evaluation of Results;
Household; and Nutrition. Additional
case studies are being sought, from
various farming systems research teams,
to broaden the scope of the material that
the FSSP can make available in its
orientation and training efforts. Case
studies are continuously being revised
and updated to introduce new and
targeted information.
At the same time the FSSP has formu-
lated a book of readings to supplement
the background of course and workshop
participants. The readings, Farming
Systems Research and Extension
'Methods, draws from recent and current
experiences in farming systems research
and development efforts throughout the
world. Readings have been selected to
further define the current thinking in
farming systems research, and to cite
examples from various authors on the
application of the methodology. It is
anticipated that the book of readings will
ultimately be published as part of the
FSSP effort to disseminate information
about farming systems.
Training and Orientation Workshops
Since the first FSSP offering of a
Farming Systems Research and Extension
Orientation Short Course at the Univer-
sity of Florida in June there have been six
additional course offerings. These in-
cluded another at the University of
Florida in July, at Michigan State Univer-
sity an' "irginia Polytechnic Institute in
August, and at Colorado State University
in September. In September an FSR/E
Orientation Workshop was also offered
in Spanish in the Dominican Republic,
and in French in Upper Volta.

The FSSP will continue to respond to
requests for training workshops, with a
specific focus on West Africa. Two
one-week farming systems research ana
extension orientation workshops are
planned for the near future. Each work-
shop will involve participants from
several countries, includingAID mission
personnel, AID contract employees,
national researchers and research ad-
ministrators. These workshops, designed
to provide an overview of philosophy
and methodoloy, will initiate participants
in the structure of the FSR approach. The
workshops will be pertinent to an African
setting and discussion of issues and
concerns relevant to representative
national projects and programs will be
Beginning in February, the FSSP will
begin its response to AID mission re-
quests (routed via AID/W/S&T) for
one-week FSR orientation workshops
delivered in-country. In content these
workshops will be very like those just
described, including a specific focus on
the national setting. Participants will be
mainly national researchers, exten-
sionists, and administrators. The FSSP
will use the occasion of these workshops
to further discuss FSR project and prog-
ram training and technical assistance
needs with AID mission officials.
A four-to-six week FSR Methods
course for national researchers and
extensionists will be ready for in-country
delivery by mid-June. The course will be
practical, intensive, and will carry
participants through the several stages of
the FSR process. These will include
diagnostic surveys, design of alternative
technological solutions, design of on-
farm trials and agronomic data analysis,
farm records, and the gathering and
analysis of economic data.
Also to be ready for delivery by
mid-1984 is a one-week program in
Managing and Administering FSR. This
course will be for managers and adminis-
trators of FSR projects and programs. The
course will address not only issues of
day-to-day management, but, since FSR
is being introduced to many settings fc
the first time, will address also the matter
of reckoning the approach and its
requirements with current institutional

Kathama, Machakos District, Kenya Above: Field research assistant checking progress of Leucaena 1., (Peruvian variety)
intercropped with maize (less than one year). Upper right: Women carrying fuelwood. Lower right: Farmer harvesting wood
from grazing land. Photos: R. Vonk.

The Role of Agroforestry in FSR & D

Dianne Rocheleau and Remko B. Vonk*

Agroforestry systems research focuses
on the tree and woody shrub components
of land- use systems, their interactions
with other components, and their end
use by land managers (usually small farm
households). Although such research has
been initiated at the international level
(ICRAF') in some regional centers
(CATIE2) and national institutions (ICAR3
India), some agroforestry has not yet
been integrated into farming systems
programs. There is, however, potential
to improve and strengthen farming
systems research, development and
extension programs through incorpora-

tion of agroforestry components and
methodology (Raintree 1983; Lundgren,
The diagnostic and design (D & D)
methodology developed at ICRAF
incorporates rapid appraisal and
on-farm research methods from FSR &
D. The D & D approach takes the basic
needs of the farm household (food,
water, fuel, shelter, cash and raw
materials) as a point of departure. A
strong emphasis is placed on the
existing and potential role of trees and
shrubs. The agroforestry D & D
methodology has been tested in a wide

range of agroecological zones through-
out the tropics. Two contrasting exam-
ples are given to illustrate the approach
and to demonstrate the importance of
woody perennials for small farm
In Kathama, Northern Division,
Machakos District, Kenya, on-farm
research has been carried out by
Wageningen Agricultural University,
(The Netherlands), and ICRAF since
The Kathama research area is in a
medium potential agricultural region at
1200m above Msl. The average rainfall

1) ICRAF International Council for Research in
2) CATIE Centro Agron6mico Tropical de
Investigaci6n y Ensefianza
3) ICAR Indian Council of Agricultural Research

*Dianne Rocheleau, is a geographer, currently
working for ICRAF on a post doctoral research
fellowship with the Rockefeller Foundation.
Remko, B. Vonk, is a post-graduate student from

Wageningen Agricultural University, the Nether-
lands. He has worked for ICRAF from 1981-1983
on Kathama Agroforestry On-Farm Research
Project, in Machakos District, Kenya.

is 850mm, with biomodal distribution.
The farming system is characterized by
subsistence-oriented food and livestock
production on small holdings averagee
3.5 ha) Table 1 gives the basic needs,
problems identified and their antece-
dent causal factors.

The firewood situation is critical.
Already more than 50% of the farmers
rely on dwindling off-farm sources to
meet their demands. Detailed studies
have shown that the off-take exceeds
production rates of firewood species.
The farmers, though, do not yet perceive
the problem to be as serious as the
research has demonstrated.

Agroforestry technologies can offer
solutions to some of the observed
problems. Existing agroforestry systems
are taken into account in the design of
interventions, although some new
components and practices are intro-

Food Production

Existing system: Farmers tolerate
Terminalia brownii (broad leaf tree
species) in their fields for its yield-im-
proving effects on marginal cropland.
Tested agroforestry interventions:
Mulch farming hedgerow system,
whereby hedgerows of N-fixing trees
are planted between crops at regular
distances. The hedge provides mulch
that can improve crop growth, reduce
weed growth and provide firewood.
The system, can evolve into a minimum
tillage system. Trees used: Leucaena
leucocephala. Cassia siamea.
Suggested modification: improve the
existing system by interplanting Ter-
minalia brownii.

Feed Production

Existing systems: Trees and shrubs are
important fodder, provide foliage and


Basic needs Problem

Food Seasonal food
shortages of farm
produced staples

Dry season feed
shortage for

Energy Lack of firewood

Shelter Lackoftimber,
fencing materials,
and big logs to
burn bricks
Raw material Lack of poles for
construction pur-


Antecedent Causal Factors

-low soil fertility due to soil mining
- high run-off rates at onset of rains due to
hard baked top soil.
- lack of labour at weeding time
- tillage problems due to limited
availability and poor condition of draft
-overgrazing and subsequent soil degradation
-high land requirements for food crop
- disappearance of browsing species due to
fuelwood and other wood needs.

- depletion of tree resource base due to high
need for fuel, fencing material and
building material.
- Lack of adequate forestry extension service,

- Limited availability and poor quality of
available seedlings.

Lack of big logs
for commercial
brick and charcoal

high protein pods, especially in the dry
season when grass is scarce and alterna-
tive feed sources are low in protein. The
most favoured species are: Acacia
tortilis (used for firewood, fencing,
medicine and termite-resistant con-
struction material); Aspillia morsam-
bicenus (also medicinal); Acacia
mellifera (firewood, fencing and local
medicine) and Securimega virosa (good
fuel, local medicine). Suggested inter-

1. Put upright, prune and protect
favoured species like Acacia tortilis
and A. senegal that are continu-
ously kept low and have no chance
to grow into mature, pod producing
trees. Micro catchments can be
dug around the base of the treesto
increase infiltration, reduce runoff,
trap silt and improve grass growth.
(Acacia tortilis especially has a
positive effect on the grass growth
under its canopy).

Editor's note: The International
Council for Research in Agrofor-
estry (ICRAF) has initiated a
series of three-week training
courses in Agroforestry Research
and Development. The first course
in this series will be held at ICRA F
headquarters in Nairobi, Kenya
from 1-18 November, 1983, and
will be limited to 25 participants
from African countries.
Each course will focus on availa-
ble knowledge in agroforestry
practices and systems, consider-
ing methods for assessing land
use problems and evaluating ag-
roforestry potentials. The ICRA F
courses are intended to enhance
the professional capabilities of
research scientists and develop-
ment planners from developing
countries for initiating and im-
plementing agroforestry research.
Courses are directed toward the
development of systems and
technologies that are both suited
to local conditions atrd adoptable
by farmers.
Information about future course
offerings and applications are
available from: Training Officer,
ICRA F, P.O. Box 30677, Nairobi,

2. Planting of multipurpose fodder
species like Zizyphus mauritiane
and Parkinsonia aculeata as living
fences around compounds, culti-
vated area and grazing area for
fodder, fencing,firewood and (for
Z. mauritiane) fruits.

3. Planting of woodlots for cut and
carry fodder and firewood produc-
tion. Trees that stay green through-
out the dry season are rich in
protein, like Leucaena
leucocephala, can be used.

The basic needs approach can also
improve the quality of research and
extension in more commerical cash-
cropping systems. Emphasis is usually
placed on cash production through
maximizing yields of the main com-
modity. Even where tree crops are the
main commodity the multiple uses of
individual plants or of mixed stands are
often overlooked. Shaded mulitipur-
pose coffee stands on small plots are a
case in point (lagemann and Heuveldop

An example is presented from a D & D
exercise carried out in Costa Rica in
March, 1983 by the Ministry of Agricul-
ture, CATIE and ICRAF. The study area,
Acosta, is located in the central high-
lands at 800-1200m elevation. Annual
rainfall is 2300mm, mean annual
temperature is 210 and the area is
characterized by rough topograhphy
(30-60% avg. slope).

Many of the small farms in Acosta (<10
ha) are based on diverse stands of coffee
intercropped with bananas, plantains,
citrus and leguminous fuelwood/shade
trees. These stands are complemented
by grain crop, small animal and milk
production for on-farm use. The stands
are maintained almost exclusively with
family labor (except harvest) and most
small farm households depend partly
on off-farm employment including
seasonal coffee harvesting on larger
holdings. The results of the farming
systems diagnosis are given in Table II.

A commodity-oriented approach to
such farms results in replacement of
Traditional and improved varieties
under shade by monocropping of the
caturra var. without shade. Disadvan-
tages of this technology package

Acosta-Puriscal, Costa Rica Above: Fuelwood harvest from a coffee stand. The
MAG-CATIE-ICARF D&D Team found this farmer selling part of his annual fuelwood
harvest. Below: Multi-story, multi-purpose coffee stand. The farmer is showing his
plot to the D&D Team. Photos: F. Torres.
4bIW- r ._ lw

include high agrochemical, labor and
management requirements, and elimi-
nation of the multipurpose functions of
the existing system (staple food,
firewood4', citrus and timber produc-
tion; erosion prevention; and risk
spreading).' Moreover, such
technologies may be adopted in part,
as was the case with some farmers who
later lost their crop due to lack of
fertilizers (Espinoza 1983).

Given the multiple objectives of the
farm household the following agrofores-
try interventions were suggested for
on-farm trials. To improve cash, food
and fuel production from coffee stands
research is focused on:

1. improvement of shade tree type
and management, especially with
regard to N-fixing and soil cover
properties and fuelwood produc-
tion; and



Food & Water

Low productivity of*
coffee stands

Lack of employ- *
ment opportunities

Difficulty in mar- *

Erosion in coffee

Lack of milk, *
meat for home

Low yeilds of beans *
in cut-over fallow

Low productivity *
of corn

Erosion in corn field**

High cost of*
(egg production)

Shortage of wood**
over medium
and long term

Shelter Shortage of timber**
over medium
and long term

Savings/ Unableto maintain *
Investment cattle

Basic needs Problems

*Problems perceived by farmers
**Problems perceived by D & D team, not farmers

2. increasing the density and yield of
fruit trees in coffee stands5'

Suggested innovations to increase food
and cash production in the other plots

1. cut and carry fodder production for
small ruminants (goats) integrated

2. hedgerow intercropping of maize
with N-fixing tree species and,

3. improved fallow for bean crop with
sowing of fast-growing N-fixing
woody legume.

While tree and shrub components do
not play the same role in all farming
systems they do play an important part
in meeting basic needs of most small
farm households. Fuelwood and dry
season fodder are especially important.
These examples from Kenya and Costa
Rica demonstrate how agroforestry
systems research can help FSR & D to
improve and develop tree and shrub
components on small holder farms.m

4'A family needs 3 ha of coffee, or 2 ha
shaded coffee to meet fuelwood needs.
(lerncekert and Campos, Inform. Tecnico No.
18 CATIE 1981)

"Linked to a feasibility study for development
of a fruit processing industry in Acosta.


Espinoza, Leonardo, 1982. Estructura General de
Cafetales de Pequenos Agricultores en Acosta-
Puriscal, CATIE, GTZ. Agroforestry Project,
CATIE Turrialba mimeo 21pp.
ICRAF, 1983b. Draft Guidelines for Agroforestry
Diagnosis and Design. ICRAF. Nairobi.
Kathama Agroforestry Project, H. Gielen, E.
Flierveet, J.v. Wijngaarden, E. Boer. O.
Nijssen, R.B. Vonk, A.v.d. Hoek, Wag-
neningen Agricultural University. ICRAF.
Lagemann, J. and J. Heuveldop, 1982. Charac-
terization and Evaluation of Agroforestry
Systems, the Case of Acosta-Puriscal, Costa
Rica. CATIE/GTZ. Turrialba, mimeo 19pp.
Lundgren, B. 1982. Introduction. Agroforestry
Systems 1 3-6.
Raintree, J.B. 1983. A Diagnostic Approach to
Agroforestry Design Paper submitted to the
International Symposium on Strategies and
Designs for Afforestation, Reforestration and
Tree Planting, Hinkeloord, Wageningen
(Holland) Sept. 19-23, 1983.

Antecedent Causes

Low soil fertility, high cost of fertilizers.
Inadequate management of shade

Lack of mktg processing facilities forexisting
and potential food products

Inadequate local mktg and processing

Inadequate ground cover and/or erosion
control structures

Shortage of grazing land, fodder

Low soil fertility, seasonal labor shortage for

Low soil fertility, heavy runoff

Planting on steep slopes; poor ground cover,
no erosion control

High cost feed

Partial dependence on off-farm sources,
deforestation, removal of second growth.

Dependence oi off-farm sources, deforestation,
removal of second growth.

Insufficient land and other inputs to maintain

The farming systems approach to research is a comparatively recent development in the study of smallholder
mixed crop/livestock agriculture. In its Highlands Programme, ILCA has acquired considerable experience with
the application of the approach focussing on the livestock enterprise. The impact of new technologies on
competition for lands, labour and capital between the crop and livestock enterprises has been a major part
of the work. A key lesson in the work to date has been the crucial importance of regular close contacts with
the farmer clients so that their production problems and opportunities are clearly defined, and the subleties
of the links between the dynamics of the various components of the farming systems are sufficiently well
understood. Without knowledge of these factors, the FSR approach would degenerate to a disciplinary research
programme, thereby losing the system's perspective which must apply if research is to provide the basis for
sustained and stable increases in agricultural production from smallholder farms.

Guido Gryseels and Frank Anderson*

The International Livestock Centre for Africa (ILCA)
is one of 13 international agricultural research centres
supported by the Consultative Group on International
Agricultural Research (CGIAR). ILCA has field pro-
grammes in the major ecological zones of sub-Saharan
Africa, namely the arid and semi-arid, sub-humid and
humid and highland zones. This article reports on the
Farming Systems Research activities of ILCA's High-
lands Programme.
To date, research in this programme has focused on
the Ethiopian highlands. Ethiopia has the largest high-
land area of the continent, accounting for almost half
of the total African highland zone. It is an ecologically
diverse country with the agricultural sector contributing
the major share of gross national product and almost
all export earnings. Nearly 80% of the population is
directly dependent upon agriculture and per capital
income is estimated to be less than US$140 per annum,
classifying it as one of the least developed countries in
the world.
Of the total area of Ethiopia of 1.22 million km2, half
is highlands above 1500 m altitude. Some 70% of the
total human population of about 32 million live in the
highlands. Ethiopia has the largest livestock population
in sub-Saharan Africa with some 26 million head of
cattle, 24 million sheep, 12 million goats, 7 million
equines and 1 million camels. The majority of the
cattle, sheep and equines are in highland areas. Agricul-
tural conditions vary widely throughout the country,
according to topography, climate, and soils. However,
the highlands are generally temperate and compara-
tively favourable for both crop and livestock produc-
tion. Ethiopia's rugged terrain is a major constraint to
economic development because of related transport
and communication problems. For example, it is often
claimed that three-quarters of the population live more
than one day's round trip walk from an all-weather road.
Since the mid-seventies, rural development has been
organised in a socialist framework. However, collective

farming accounts for less than 5% of the total area
cultivated. The bulk of agricultural output is still
produced by individual subsistence small holders who
have 'farming rights' over the land they till.1
ILCA's field research in the Ethiopian highlands has
concentrated on the cental highlands where mixed
smallholder farming is the dominant mode of produc-
tion. Rainfall here averages between 600 and 1200 mm
per annum, of which about 70% falls in the main rainy
season between July and September. Average farm
sizes range from .5 to 5 hectares, and around 80% of
all farm produce is used for subsistence consumption.
About two thirds of cultivated land is sown to cereals.
The majority of the remainder is sown to pulses. The
fraction of fallow land in different parts of the highlands
is quite variable. In some locales almost no land is
fallowed while elsewhere fallow periods of up to 12
years after 3 years of crop are common. The main crops
are teff, wheat, barley, maize, sorghum, horse beans,
chick peas and field peas. Grain yields average between
500 and 1000 kg per hectare sown. Access to modern
inputs such as chemical fertilizers of improved seeds
is limited. Fewer than 10% of farmers regularly use
either input. Most farmers own livestock and a typical
inventory is two oxen, a cow, a few sheep and a
donkey. As most livestock manure is used as household
fuel, only little is returned as fertilizer to the fields.

*Agricultural Economist and Team Leader,
respectively. ILCA, Highlands Programme, P.O.
Box 5689, Addis, Abada, Ethiopia.
In Ethiopia, land is not individually owned, but is allocated annually
by the Peasants' Association (PA) of which each farmer necessarily is a
member. The size of individual allocations, comprising several plots,
depends on family size, local population density and the policies of the
local PA. Land distribution takes account of the different fertility levels of
the soils within the land area of the PA. Each PA has on average 250
individual members and controls around 800 ha of land.

Cattle are kept mainly as a source of draught power
and manure production. Milk, meat and hides are
relatively less important by-products. Livestock are
privately owned, so are an important form of investment
and financial security. Usually only oxen are used for
cultivation. Productivity is low for all livestock reflect-
ing an underexploited genetic resource and generally
inadequate nutrition, particularly during the extended
dry season of up to seven months each year. For
example, milk offtake from indigenous Zebu cows kept
under traditional management, rarely exceeds 400 kg
for lactations of seven months. Calving intervals average
two years. Sheep also are comparatively unproductive
as, along with cattle, they are subject to heavy endo-
parasite burdens and to extended periods of nutritional
It is against this background of a traditional subsis-
tence agriculture, where productivity gains must be
achieved to ensure survival in the coming years, that
ILCAestablished a research programme in the Ethiopian
Overview of the research programme
The basic objectives of the research are to study ways
and means of improving the overall productivity of
mixed smallholder farms by increasing the technical
and economic efficiency of livestock enterprises.
Particular emphasis is given to enhancing the com-
plementarity of the livestock and crop components.
Results and experiences of the research in Ethiopia will
in many cases have direct relevance to other highland
smallholder situations in sub-Saharan Africa where the
same immediate concerns to increase agricultural
production apply.
The Programme has adopted the farming systems
approach to research. This stresses on-farm technol-
ogy testing and appraisal, backstopped by relevant
station research on components where greater experi-
mental control is advantageous. In addition to re-
search undertaken at its headquarters in Addis Ababa,
field work is undertaken in two study areas: at Debre
Zeit located 50 km south of Addis Ababa atan altitude
of 1800 m, and at Debre Berhan 120 km north east
of Addis Ababa, at an altitude of 2800 m. The area
around Debre Zeit is intensively cultivated with
virutally no arable land kept fallow. Teff (Eragostis
teft) is the principal cereal grown. Debre Berhan
represents the high altitude zones. Frosts, hail and a
shorter growing season, in addition to low soil fertility
levels cause the area to be less productive than Debre
Zeit. Most of the land cultivated there is sown with
barley. Research stations have been established by
ILCA at both Debre Zeit and Debre Berhan.
The station research focuses on key elements of the
traditional farming systems: soil fertility, forage produc-
tion, various aspects of draught animal utilization, and
small scale dairy processing technologies. Innovative
livestock enterprise combinations considered to be
relevant future options for small holders are also under
test on station in research farms.

Baseline surveys were made in both study areas
before beginning experimental work. These have been
followed up by ongoing studies with a random sample
of individual farmers in the areas surrounding the ILCA
sites to provide time series data on the dynamics of
these agricultural systems and to serve as a control
group against which the impact of innovations can be
evaluated. The programme concentrated initially on
research associated with improved dairy husbandry at
the smallholder level. Improvements to these enter-
prises were considered to be the most rapid means of
achieving substantial increases in farm incomes. The
focus of ILCA's research on the topic was on farm-level
adoption problems.

Some initial results of on-farm studies
The baseline study made in 1978 in the Debre Zeit
area covered 151 farmers in 25 PA's. On-farm research
began soon after with 18 traditional farmers testing a
dairy husbandry package. Each farmer bought a
crossbred dairy cow, cultivated 2.5 ha of arable land
and had access to a 30 ha communal pasture. The
Friesian x Boran2 cow was fed mainly on a special-pur-
pose forage mixture of oats and vetch (Vicia dasycarpa).
The productivity of the farming system was to be further
improved by the use of improved seeds and chemical
fertilizer on subsistence food crops, and generally
improved farm management.
These 18 'test' farmers adopted this programme at
their own risk and expense and decide whether or not
to accept ILCA's technology and management recom-
mendations. Extension inputs were provided by ILCA
until early 1981, and then gradually reduced to corres-
pond to conditions which would normally apply in a
development setting. The performance of these test
farmers was closely monitored and compared with the
other farmers not involved in the programme who
served as a control group.
The package proved to be a feasible way of achieving
significant welfare gains at the smallholder level. Over
a four year period the yields of subsistence food crops
of test farmers were on average more than 40% higher
than those of the control group, while adjusted annual
milk yields from the crossbred cows averaged 2300 kg,
more than seven times those achieved from cows of
local breeds. In consequence, cash income per farm
more than tripled. Where most traditional farms have
a gross cash income of no more than US$200, the test
farmers had an average annual gross cash per family
of around US$800. Net farm incomes increased almost
three times. A major portion of the extra income has
been used to improve their quality of life as, for example,
through upgrading the quality of their housing and
extending the period of education of their children. The
extra cash income generated by the dairy enterprise
had a positive spillover effect on the productivity of the
crop enterprises. The dairy enterprise accounted for the
major fraction of the increase in net income.
2The Boran is an indigenous Ethiopian Zebu breed.

Although this rapid increase in farm productivity can
be sustained under smallholder conditions, the research
revealed and allowed quantification of a number of
specific factors with implications for any broad-based
introduction of the basic system elsewhere in Ethiopia.
Some of the main factors are summarized below:
- Farmers replace those crops with the lowest gross
margins when they produce special-purpose
forages. In many cases, the pulse crops (field peas
and chick peas) which make a valuable contribu-
tion to soil fertility,, were replaced by the oats and
vetch forage mixture. Also, labour is allocated to
forage crops only after the requirement for cereal
crops are satisfied. Competition for labour during

Cross-bred cows have milk yields on average
seven times those of local breeds.
the land preparation time means that less care is
given to forages. Futures forage screening will
therefore concentrate on 'easy care' forages
including the opportunity for establishing im-
proved perennial pasture swards which would
require lower labour inputs.
Fertilizer prices in Ethiopia have increased rapidly
in recent years. With present fertilizer/grain price
ratios fertilizer can be applied only to high value
cereal crops. Research should therefore concen-
trate on more economic ways of restoring soil
fertility, on evaluation of alternative crop rotations
where crop varieties are more productive with low
fertilizer dressings.

- Cereal/milk price ratios have changed substantially
from 1976 to 1982 during the study period. Since
1976 the producer price for milk has remained the
same while cereal prices have more than doubled.
As a result, test farmers allocated a smaller area of
land to forage crops than originally anticipated.
Nevertheless, the financial returns to farmers from
the dairy enterprise were sufficiently favourable for
them to continue with the activity, indicating its
economic robustness for a wide range of cereal/
milk ratios.
- Milk marketing in Ethiopia is a severe problem
during certain periods of year indicating the need
to evaluate the options available to smallholders
of processing milk into a range of storable, trans-
portable and high value products.3
ILCA's research in the Debre Berhan area began in
1979. Less information was available on the traditional
agricultural system in the area than at Debre Zeit, and
therefore the opportunities for an immediate start with
on-farm trials of different technologies were limited.
Cereal production in the area is risky due to the inci-
dence of frost immediately at the end of the compara-
tively short growing season. Therefore, the first years
of the research concentrated on experimental crops and
forage agronomy on the station work and on studies in
the traditional farming system.
The baseline survey of agriculture in the area included
226 farms belongingto 24 different PA's. Routine data
collections have continued subsequent to that original
survey with a sample of 60 farmers from the PA's in
the area surrounding the ILCA station. A strong data
base has therefore been developed on the traditional
farming system. The technical feasibility of the
technologies needed to support a smallholder dairy
enterprise was first tested on-station. After this phase,
on-farm testing of the package began in 1982 with 20
traditional farmers. Results to date are most encouraging
with milk yields per cow in first lactation of around
1500 litres. The adoption of the technology is proceed-
ing with a minimum of difficulties.
The fraction of the total land area cultivated by
farmers in the Debre Berhan area is much lower than
at Debre Zeit, reflecting the generally lower soil fertility
at Debre Berhan. Thus, a central question to be addres-
sed in the analysis of this on-farm trial will be the extent
of competition between crop and livestock enterprises
for land and labour resources. The difference between
Debre Zeit and Debre Berhan, as regards the intensity
of land use for cropping, also offer the opportunity for
a comparative study of the impact at the smallholder
level of the same enterprise on farm production,
resource allocation and farm income.
The relative remoteness of most farmers in the Debre
Berhan area from a major urban centre makes milk and
milk product marketing a more crucial issue than at
Debre Zeit. For this reason, a research programme for
'The majority of people in the study areas are followers of the Ethiopian
Orthodox Church which prohibits the consumption of food stuffs containing
animal protein, except fish for a total of around 150 days per year.

smallholder-level dairy processing technologies has
been initiated in 1983. If the dairy enterprise and
associated milk processing technologies prove success-
ful in the Debre Berhan area, the results will have
relevance to a large part of rural Ethiopia where
crossbred cows can be kept, but the sale offluid milk
is not possible.

Examples of other research
The primary contribution of cattle to agricultrual
production in the Ethiopian highlands is as a source of
draught power. ILCA surveys showed that animal
power used for crop-related work averaged more than
1000 hours per farm per annum. Most of this power was
supplied by oxen, but other cattle were sometimes
employed for threshing. The Programme has allocated
substantial resources to various animal traction-related
studies topics. Work in progress includes the evaluation
of different cultivation systems using oxen of local and
crossbreeds, the technical and economic efficiencies
of using crossbred cows for draught purposes as well
as milk production, the use of oxen worked as singles
rather than as pairs as they are traditionally used in the
Ethiopian setting, and the working efficiency of oxen
subject to nutritional stress. Different harnesses and
yokes are also being developed.
Oxen are now used in the highlands only for cultiva-
tion of crop lands and for threshing. New applications
of animal power are being evaluated for their potential
contribution to the traditional system. These inter-
mediate technology applications are only new in the
sense of their introduction to Ethiopia. They include
the use of oxen to draw metal scoops for earth excava-
tion for pond and dam construction and the use of a
winged narrow tine to rip permanent pasture areas to
raise their productivity through better water penetration
and combined aeration and mineralization effects. In
large areas of the Ethiopian highlands farmers depend
heavily upon permanent pastures for direct grazing of
their livestock and as a source of home-fed hay.
Increased production from these areas would help
reduce the competition between crops and livestock
for land resources.
The link between continuing studies with smallhol-
ders and the on-station and on-farm research is well
illustrated by the Programme's work with oxen worked
as singles rather than as pairs. More than half the
Ethiopian farmers do not own two oxen and are depen-
dent on expensive and complex rental arrangements to
obtain the oxen required for cultivation. In many cases
this leads to late crop sowing and reduced yields. In
addition, analysis of data collected by ILCA have
highlighted the impact of oxen ownership on area
cultivated and crop combination sown. At Debre Zeit
for example, substantial differences were observed
among the farmers in ILCA's control group of outside
farmers in the 1980 crop year, as summarized in the
following table:

Farmers grouped Average area Average sown Average sown
byoxen ownership cropped per to cereals to pulses
farmer (ha) (%) (%)
Owning no oxen 1.2 54 46
Owning one ox 1.9 44 56
Owning two oxen 2.7 67 33
Owning three or
more oxen 3.6 92 8
Cereal crops require more intensive land preparation
than pulse crops. Furthermore, the market value of
cereal crops is substanitally higher than pulses. To-
gether these factors mean that farmers with less draught
power at their command sow less land in total and also
have lower incomes than their counterparts with an
adequate supply of draught power.
Given these factors, ILCA undertook on-station tests
with the local oxen yoked singly and drawing a mod-
ified version of the local wooden plough ('maresha').


The single ox with the modified 'Maresha.'
No particular technical difficulties were encountered
and the work output4 of the singles was considered to
be adequate to warrant initiation of on-farm trials in
the current crop season. At this time (June 1983), some
40 farmers have purchased or constructed the modified
'maresha' and harness and yoke and are using it for
crop cultivation. The use of the variant of traditional
practice is being closely monitored. If this proves to be
a successful innovation and is widely adopted in the
coming years it will make an important contribution to
the efficiency of the livestock component of the Ethio-
pian highlands smallholder system and could contribute
to lessening the competition between crop and livestock
production. ILCA is increasing the research work on
animals worked on singles, including crossbred oxen
which are produced by the crossbred as dairy cow
enterprise. On-farm trials have also begun with
crossbred cows being used as draught animals in addi-
tion to their primary role as milk producers..
4Preliminary trials indicate that the single ox can cultivate at approximately
70% of the rate of paired oxen.


David Youmans and David Holland*

The farming systems approach to
research and extension has now been
well described by several authors
(Byerlee et al. and many others).
Among the claimed advantages of this
approach is that researchers become
oriented to the same problems the
farmers are concerned about, and the
emphasis upon "on-farm trials" pro-
vides a ready-made device for further
extension of research results. However,
the FSR approach can run into "special
problems" unless care is taken to avoid
them. The purpose of this article is to
briefly review some of those "special
problems" and to discuss an approach
to deal with them.
Washington State University's FSR
team has been involved in Lesotho
since 1979 in a USAID-funded joint
effort with the Research Division of the
p inistry of Agriculture and Marketing.
Advantageous to the relative success of
the Farming Systems Research Project
in this southern African kingdom is the
complete integration of the project,
including its nine expatriate team
members, into the mainstream of the
Agricultural Research Division's total
agricultural research program. The
Division has adopted a policy of FSR as
its central philosophy and program
rationale. Expatriate research support
has been sustained in Agronomy,
Animal Science, Extension, Farm
Management, Marketing, Range Ecol-
ogy (and Rural Sociology for the initial
two years).
Most experts agree on a general FSR
procedure which is relatively consistent
although nomenclature may differ and

* David Youmans is Associate Extension
Professor at Washington State University and
Extension/Staff Development Specialist with
the Farming Systems Research Project in

Dr. David Holland is Associate Professor,
Department of Agricultural Econommics,
Washington State University, presently
working as Farm Management Economist in
the WSU's Project in Lesotho.

certain systems may require some
modification in design. Largely, FSR
projects require 1) target and research
area selection, 2) problem identification
and development, 3) planning on-farm
research, 4) on-farm research and
analysis, and 5) extension of results.
Washington State University's work in
Lesotho has basically followed this
Farmer audiences were targeted and
prototype research areas delineated as
early as 1979. The program emphasis
has been on collaboration with farmers
at both moderate and low resource
levels and the research sites selected
were representative of Lesotho's low-
lands, foothills and mountain zones.
While WSU's baseline survey was
troubled by computer-output delays,
problem identification and planning
both on-station and on-farm research
went forward in a timely fashion based
on considerable local information as
well as cursory observation and assess-
The project has three years of on-farm
research results to consider at this
writing. These have been the product
of both enterprise combination testing
with selected cooperating farmers and
on-farm trials involving specific items
of research interest. Since there is vast
diversity in farms, farming practices,
and farmers' resources within any given
prototype area, and there are so many
unpredictable variables which impact
the lives of Lesotho farm families, the
neatly packaged FSR "recipe for suc-
cess" remains a somewhat elusive
ideal. There are, however, enough
research findings at this stage in the
crop/livestock realm that farmers in
every prototype area can choose to
change various aspects of their farming
enterprise with considerable confi-
dence that the changes will prove
Extension activities began early in the
project. Since extendable FSR results
were few at the outset, extension
personnel began to build the informa-
tion capability of the Research Division.
Soon technical circulars, bulletins and

reports for extension workers, scientists
and government officials were printed.
Radio programs in Lesotho language
(appropriate for a farm audience) came
on line and continue along with the
publications as an on-going activity. In
the prototype areas, FSR Extension
workers organized farmer leaders into
Village Agricultural Committees
(V.A.C.'s) and, with research officers,
trained them during three consecutive
years in community leadership and
agricultural subject matter.
The Research Division through its
Extension Sector staged training events
for District Extension Officers, Subject
Matter Specialists, Area Extension
Officers and Extension Assistants. In
support of the total effort Village Chiefs
received training in FSR philosophy and
agricultural subject matter.
Increased agricultural production in
Lesotho is fundamentally constrained
by a paucity of young, able-bodied
men. Much of the mature male popula-
tion spends most of each year at the
mines in the Republic of South Africa.
The fact that so many farms in Lesotho
are operated by families in which the
men are migrant workers for much of
the year presents special problems in
the development, testing and diffusion
of improved agricultural technology.
The male members of the family are
constantly aware of the alternative of
selling their labor power to earn
income. Since the wages are relatively
high compared to existing returns to
labor from agriculture, agricultural
activities are assigned a low priority
for the men's time. Thus, any technical
innovation in agricultural production
that requries an increase in male labor
may not be adopted unless it is ex-
tremely productive.
The women of the family also present
a problem when it comes to innovation
and diffusion of technology. While they
are left with the responsibility for
carrying out the farming operation they

are often left without authority for
changing the operation. If a wife learns
of an innovation she would like to try,
she may be unable to implement it
without her husband's permission,
especially if the innovation involves a
cash expenditure.
Likewise, there may be understanding
of the results of "on-farm" trials but lack
of knowledge regarding how "new
practices" could be applied on the
farmers' own fields. Farmers may feel
that trials were conducted with farmers
who are known to have large endow-
ments of farming capital and therefore
have relevance only for similarly
endowed individuals.
The FSR approach in Lesotho con-
templates widespread adoption of the
technology that has been demonstrated
in the "on-farm" trials. For this ap-
proach to be considered truly success-
ful, a large proportion of the farming
community should adopt at least some
aspects) of "improved" farming prac-
tices. This, over a period of years, must
involve a community beyond the limited
number of farmers with which their
researchers have worked, and beyond
the committee members which the
extension workers have organized and
In order to minimize problems and
maximize the impact of "on-farm
research special attention must be
given to appropriate strategies for
extending research results.
The research/extension framework in
the Lesotho project emphasizes the
following components: (1) integration

of research and extension personnel at
all levels into the planning and execu-
tion of all "on-farm" trials; (2) training
of members of the Village Agricultural
Committee (V.A.C.) in technical ag-
riculture so that they have an information
base adequate to understand the
implication of results from "on-farm"
trials and (3) matching farmers with a
given resource endowment with the
"on-farm" trial most relevant to their
particular farming problem.
By incorporating extension officers into
the planning and execution of field
trials, understanding the results is
maximized. The extension field staff, as
a result of being thoroughly informed
of the trials, is in a position to suggest
possible modifications, potential
locations for particular trials, etc. In
addition to understanding the intent of
the trial the field worker is able to
integrate the trials into plans for future
extension activities such as field days
which then becomethe basis of a solid
extension program. A direct consequ-
ence of such planning is that the
extension field worker becomes in-
terested and concerned about the
outcome of the trial and active in the
execution and monitoring of the trials
in a given area.
The advantage of establishing a sound
technical base of agricultural know-
ledge among the Village Agricultural
Committee is that such individuals are
then in a position to accurately interpret
the results of the trials to others in their
village. In this way, members of the
V.A.C.'s are able to complement

formal extension efforts in their normal
day-to-day contacts. Members of the
V.A.C.'s can be especially important in
clarifying possible misunderstanding of
the trials and in providing advice on
how a particular innovation could
actually work in farmers' fields.
The emphasis on matching the "on-
farm" trial to the specific resource
endowments of participating farmers
means that a broad range of resource
endowments will be represented
among participating farmers. As a result
the villagers know that notjust the best
farmers have been selected for the
trials, and there is greater willingness to
accept the idea that the outcome for a
given trial may have relevance for their
own farming operations.
Although the "special problems" cited
above may be somewhat unique to
Lesotho, certainly the issue of promot-
ing widespread technological change
will have to be faced by all FSR projects.
It is believed that the approach de-
scribed here provides the basis for a
strong program and a high probability
of success for extensive adoption of
improved methods.m
Byerlee, D., L. Harrington and D.L. Winkelman.
1982. Farming Systems Research: Issues in
Research Strategy and Technology Design.
American Journal of Agricultural Economics, 64
(No. 5) 897-904.

This public document was promulgated
at a cost of $866.88, or 34.7 cents per
copy, to inform the public of activities in
the farming systems support project.

3028 McCarty Hall
Gainesville, Florida 32611

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs