Guidelines For Sheep Breeding-2

ICARDA - TOOLS AND GUIDELINES 1
About ICARDA and the CGIAR

Established in 1977, the International Center for Agricultural Research in the Dry
Areas (ICARDA) is one of 15 centers supported by the CGIAR. ICARDA’s mission
is to contribute to the improvement of livelihoods of the resource-poor in dry
areas by enhancing food security and alleviating poverty through research and
partnerships to achieve sustainable increases in agricultural productivity and
income, while ensuring the efficient and more equitable use and conservation of
natural resources.
ICARDA has a global mandate for the improvement of barley, lentil and
faba bean, and serves the non-tropical dry areas for the improvement of on-
farm water use efficiency, rangeland and small-ruminant production. In the
Central and West Asia and North Africa (CWANA) region, ICARDA contributes
to the improvement of bread and durum wheats, kabuli chickpea, pasture and
forage legumes, and associated farming systems. It also works on improved
land management, diversification of production systems, and value-added crop
and livestock products. Social, economic and policy research is an integral
component of ICARDA’s research to better target poverty and to enhance the
uptake and maximize impact of research outputs.
The Consultative Group on International Agricultural Research (CGIAR) is a

strategic alliance of countries, international and regional organizations, and
private foundations supporting 15 international agricultural Centers that work
with national agricultural research systems and civil society organizations
including the private sector. The alliance mobilizes agricultural science to
reduce poverty, foster human well being, promote agricultural growth and
protect the environment. The CGIAR generates global public goods that are
available to all.
The World Bank, the Food and Agriculture Organization of the United
Nations (FAO), the United Nations Development Programme (UNDP), and the
International Fund for Agricultural Development (IFAD) are cosponsors of the
CGIAR. The World Bank provides the CGIAR with a System Office in Washington, Guidelines for Setting up Community-based
DC. A Science Council, with its Secretariat at FAO in Rome, assists the System in
the development of its research program. Sheep Breeding Programs in Ethiopia
Lessons and experiences for sheep breeding in low-input systems
Aynalem Haile, Maria Wurzinger, Joaquín Mueller, Tadele Mirkena, Gemeda Duguma,
Okeyo Mwai, Johann Sölkner and Barbara Rischkowsky
Guidelines for Setting up
Community-based Sheep Breeding
Programs in Ethiopia
Lessons and experiences for sheep breeding
in low-input systems
Aynalem Haile, Maria Wurzinger, Joaquín Mueller, Tadele Mirkena, Gemeda

Duguma, Okeyo Mwai, Johann Sölkner and Barbara Rischkowsky
Acknowledgments
This guideline is an output of the Austrian Development Agency funded community-based sheep breeding
project implemented in four locations representing four breeds and diverse production systems. The project
is a collaborative project between the International Center for Agricultural Research in the Dry Areas (ICARDA),
the International Livestock Research Institute (ILRI), the University of Natural Resources and Life Sciences
(BOKU) and the National Agricultural Research Systems in Ethiopia. We thank the staff of the research centers
of Bako, Bonga, Debre Berhan, and Worer for their support in implementation of the breeding program.
The target-community members are acknowledged for cooperation during the whole phase of design and
implementation of the breeding program.
Edited by: Azage Tegegne
ISBN: 92-9127-255-8
Key words: Community-Based Breeding; Guidelines; Sheep; Ethiopia
International Center for Agricultural Research in the Dry Areas (ICARDA)

P.O. Box 5466, Aleppo, Syria.
Tel: (963-21) 2213433
Fax: (963-21) 2213490
E-mail: [email protected]
Website: www.icarda.org
The views expressed are those of the authors, and not necessarily those of ICARDA. Where trade names are
used, it does not imply endorsement of, or discrimination against, any product by the Center. Maps have been
used to support research data, and are not intended to show political boundaries.
ICARDA encourages fair use, sharing and distribution of this information for non-commercial purposes, with
proper attribution and citation.
Copyright © 2011 ICARDA (International Center for Agricultural Research in the Dry Areas)
All rights reserved.
Citation: Aynalem Haile, Maria Wurzinger, Joaquín Mueller, Tadele Mirkena, Gemeda Duguma, Okeyo Mwai,
Johann Sölkner and Barbara Rischkowsky, 2011. Guidelines for Setting up Community-based Sheep Breeding
Programs in Ethiopia. ICARDA - tools and guidelines No.1. Aleppo, Syria, ICARDA.
iii
Table of contents
Executive Summary v
1 Introduction 1
2 User Guidance 3
2.1 Purpose and objectives of the guidelines 3
2.2 Target groups 3
2.3 Conditions under which the guidelines should be used 3
2.4 Structure of the guidelines 4
2.5 Using the guidelines 4
3 Implementation Guidelines 6
3.1 Selecting target breeds and communities 6
3.1.1 Selection of breeds 6
3.1.2 Selection of community 6
3.2 Characterization of target sites and breeds 7
3.2.1 Description of the production system 7
3.2.2 Breed characterization 8
3.3 Definition of breeding objectives 9
3.3.1 Personal interviews 10
3.3.2 Workshops (focus group discussions) 10
3.3.3 Choice cards experiment 11
3.3.4 Ranking of live animals 11
3.3.5 Comparison of methods 13
3.4 Assessment of alternative breeding plans 15
3.5 Developing adequate breeding structures 17
3.5.1 The breeding program 17
3.5.2 Animal identification 21
3.5.3 Data recording and management 21
3.5.4 Selection of candidate rams 24
3.5.5 Management and use of breeding rams 26
3.5.6 Institutional backup: organizational issues 27
4 Creating an Enabling Environment 28

4.1 Community-implementers’ relationship 28
4.2 Complementary interventions, services and capacity development 28
4.3 Government support 30
4.4 Links with other projects/activities 30
5. Monitoring and Evaluation 31
6. Concluding remarks 34
7. References 35
8. Further readings 36
ICARDA - Guidelines for community-based sheep breeding in Ethiopia

iv
List of tables
Table 1. Characteristics of the four sites 2

Table 2. Advantages and disadvantages of alternative methods
of defining breeding objectives 14
Table 3. Traits recorded in four sheep breed improvement
communities in Ethiopia 23
Table 4. Structure of community-based sheep breeding implementation
team in the ICARDA–ILRI–BOKU sheep project 28
Table 5. Monitoring and evaluation of community-based sheep
breeding program 32
List of figures
Figure 1. Logical sequence of steps to implement community-based
breeding program 5
Figure 2. Example of a ‘choice card’ tool used to describe breeding
ram trait performance 11
Figure 3. Open and closed nucleus schemes 18
Figure 4. Open nucleus scheme: nucleus and base flocks. Example,
Merino Sheep Breeding Program, Argentina 19
Figure 5. Open nucleus scheme: dispersed nucleus. Example, Bonga
and Afar Sheep Breeding Program, Ethiopia 20
List of boxes
Box 1. Group ranking of sheep 12
Box 2. Input parameters required to run ZPLAN 16
Box 3. Open Nucleus schemes for Merino sheep 18
Box 4. Selection of rams 25

v
Executive Summary
Guidelines for setting up community-based sheep breeding programs in Ethiopia
These guidelines are designed for all those involved in planning and implementing sheep breeding activities with
resource-poor farmers in developing countries. This includes research centers, non-governmental organizations
(NGOs), farmers’ associations and livestock development projects, and government extension officials.
The guidelines address the lack of generic direction on designing and implementing community-based breeding.
Community-based breeding programs are proposed as an option for genetic improvement of livestock in developing
countries. This new approach has been tested in a few places with promising results (e.g. with dairy goats in Mexico,
llamas and alpacas in Bolivia and Peru). They draw on practical experiences from implementing community-based
sheep breeding programs in four agro-ecological zones in Ethiopia and provide guidance for continuing and out-
scaling the breeding program in Ethiopia and for planning similar projects elsewhere.
The breeding programs in Ethiopia have achieved important ouputs. For example, negative selection has been
reverted as fast growing lambs are now being retained for breeding instead of ending up in markets. The acute
shortage of breeding rams, observed previously in flocks of participating communities, has also been rectified
as farmers are now fully aware of the importance of breeding males.
Preliminary analysis of the recorded data indicates that the market outlet has increased through more births of
lambs, bigger lambs at birth and weaning, and reduced mortality rates due to the combination of breeding with
improved health care and feeding.
The objective of these guidelines is to:

• Describe the prerequisites and context in which community-based breeding program can be successfully
implemented
• Explain how communities can be best engaged to get actively and sufficiently involved in all the critical
stages (i.e. definition of breeding goals and decisions on the best implementation options and plans) of
the project
• Take the user through the main steps in the design and implementation process leading to operational
breeding programs
• Propose appropriate local institutional arrangements within the communities’ capacity to effectively manage
performance recording, selection, and delivery of improved genetics
• Suggest a system for monitoring and evaluating progress and the impact of the breeding program during
and after the project.
The guidelines have five parts:

1. General background information
2. User guide
3. Implementation modalities
4. Issues related to enabling environments
5. Monitoring and evaluation for community-based sheep breeding programs.
While based on the Ethiopian experience, these guidelines propose generic approaches to community-based
breeding for resource-poor sheep farmers. As community-based breeding is a new approach, the guidelines
will be refined and updated as experience in community-based breeding accumulates and our tools improve.
The project team welcomes inputs and perspectives from interested readers.
Contacts: [email protected]; [email protected]

1
1. Introduction
Genetic improvement of livestock is often viewed as a complex set of tasks requiring a high level
of organization and technical sophistication. In Europe, animal breeding has been traditionally
supported by the state and implemented by large national breeding programs. Data recording,
channeling of the recorded data towards a data processing center, estimation of ‘breeding values’
with complex statistical methods and central decisions about the use of male breeding animals
are ingredients of such breeding programs. In developing countries, the required supportive
infrastructure is largely unavailable, so attempts to replicate developed-country approaches have
met with little success.
As alternatives, centralized breeding schemes, entirely managed and controlled by governments –

with minimal, if any, participation by farmers – were developed and implemented in many developing
countries through a nucleus breeding unit limited to a central station. These centralized schemes
were usually run by a governmental organization attempting to undertake all or part of the complex
processes and breeding strategy roles (i.e. data recording, genetic evaluation, selection, delivery
of genetic change, and feedback to farmers). Although well intended, these centralized schemes
failed to sustainably provide the desired genetic improvements (continuous provision of a sufficient
number and quality of improved males to smallholders) and also failed to engage the participation
of the end-users in the process.
Another alternative widely followed by many developing countries or individuals is importing

improved commercial breeds in the form of live animals, semen, or embryos. These are crossbred
with the indigenous and ‘less productive’ breeds to upgrade them towards the imports, but in most
cases, it is done without sufficient pre-testing of the appropriateness (suitability and adaptability)
of the breeds and their resulting crosses to local production systems or conditions. Where
indiscriminate crossbreeding with the local populations has been practiced, genetic erosion of the
adapted indigenous populations and breeds has occurred.
A new approach is therefore required. One such approach that has recently stimulated global interest
is a community-based breeding strategy. Programs that adopt this strategy take into account the
farmers’ needs, views, decisions, and active participation, from inception through to implementation,
and their success is based upon proper consideration of farmers’ breeding objectives, infrastructure,
participation, and ownership (Mueller 1991; Sölkner et al. 1998; Wurzinger et al. 2011). Designing a
community-based breeding program is much more than genetic theories and increased productivity.
It is a matter of infrastructure, community development, and an opportunity for improved livelihood of
livestock owners through productive and adapted animals and markets for their products.
Cognizant of this, the International Center for Agricultural Research in the Dry Areas (ICARDA), the
International Livestock Research Institute (ILRI), and the University of Natural Resources and Life
Sciences (BOKU), in partnership with the Ethiopian National Agricultural Research System, have
designed and implemented community-based sheep breeding programs in Ethiopia. Ethiopia was
selected as a case study because sheep play an important role in the livelihoods of resource-poor
farmers/pastoralists there. The current level of productivity of the indigenous Ethiopian sheep breeds
under the smallholder production systems is low. The average annual off-take rate and carcass
weight per slaughtered animal for the years 2000–2007 were estimated at 32.5% and 10.1 kg,
respectively, the lowest even among sub-Saharan African countries (FAO 2009).

2
In parallel, the demand for sheep products has increased due to a growing human population
and urbanization. There is, therefore, an urgent need to improve productivity in order to raise
smallholders’ incomes and meet the demands of the growing human population. Furthermore,
recent assessments of the views of farmers and research and development views in the highlands
of Ethiopia have shown that genetic improvement should receive a similar priority to feeding and
health issues (Edea 2008; Getachew et al. 2010). Therefore, an integrated approach is needed,
taking into consideration genetics, nutrition, health, input supply and services, and markets.
The guidelines presented here are therefore based on the experiences gained from the research
project titled ‘Designing community-based breeding strategies for indigenous sheep breeds of
smallholders in Ethiopia’. The project is funded by the Austrian Development Agency and operates
in four regions representing different agro-ecologies that are the habitats of four indigenous sheep
breeds: Afar, Bonga, Horro, and Menz (Table 1). The project is being implemented with the full
participation of farmers and pastoralists, and to date, about 500 households (120–125 households
per breed) owning about 8,000 sheep have been enrolled in the project.
The guidelines consist of the key requirements of, and implementation modalities for, community-
based small ruminant breeding in low input systems, with specific reference to Ethiopian conditions.
Table 1: Characteristics of the four sites

Breed Habitat Production system Major use
Afar Hot to warm arid plains (565–1542 m.a.s.l.) Pastoral/agro-pastoral Milk, meat
Bonga Wet, humid (1070–3323 m.a.s.l.) Mixed crop–livestock Meat
Horro Wet, humid (1600–2800 m.a.s.l.) Mixed crop–livestock Meat
Menz Tepid, cool highland (1466–3563 m.a.s.l.) Sheep–barley Meat, wool

User Guidance 3
2. User Guidance
2.1 Purpose and objectives of the guidelines
The guidelines are intended to assist users with planning and implementation of community-based
breeding programs for resource-poor sheep and goat farmers. They draw on practical experiences
from implementing a community-based sheep breeding program in four agro-ecological zones in
Ethiopia and provide guidance for continuing and outscaling the breeding program in Ethiopia and
for planning similar projects elsewhere.
More specifically the objectives of the guidelines are to:

• describe the prerequisites and context in which community-based breeding program can be
successfully implemented
• explain how participating communities can be best engaged to get actively and sufficiently
involved in all the critical stages (i.e. definition of breeding goals and decisions on the best
implementation options and plans) of the project
• take the user through the main steps in the design and implementation process leading to
operational breeding programs
• propose appropriate local institutional arrangements within the communities’ capacity to
effectively manage performance recording, selection, and delivery of improved genetics; and
• suggest a system for monitoring and evaluating progress and the impact of the breeding
program during and after the project.
2.2 Target groups

The guidelines are intended for use by all persons and organizations interested and involved in
planning and implementing breed development activities, in particular research institutions, non-
governmental organizations (NGOs), private institutions (e.g. farmers’ associations and livestock
development projects), and government officials. In the case of the Ethiopian sheep breeding
program the target group for the guidelines included livestock keepers, national and regional
research institutions, the extension system, universities, NGOs, and policy makers. As the guidelines
target a diverse group of actors, knowledge of the principles of animal genetics and breeding would
be an advantage but is not essential for using the guidelines.
2.3 Conditions under which the guidelines should be used

The guidelines are designed for users that wish to develop breeding programs in situations where:
• a developed infrastructure for animal genetic improvement under smallholder production system is
not in place, thus precluding direct adaptation of approaches from more developed situations
• systematic processes for identifying and delivering genetically superior breeding stock from the
local populations are lacking; and
• national research and development organizations have limited experience and a limited
number of qualified staff.
Although most of the elements described here are applicable beyond the local Ethiopian situation
– and hence can be easily adjusted to specific cultural, social, economic, or ecological conditions
and similar production systems – the guidelines specifically address the situation of resource-poor
sheep keepers in four different agro-ecological zones in Ethiopia. The guidelines describe the
steps to be taken to develop and implement a straight breeding program but are also applicable to
organized crossbreeding programs in local communities.

4
The guidelines can be used in connection with the guidelines provided in ‘Breeding Strategies for
Sustainable Management of Animal Genetic Resources’ which were developed and tested by FAO
(2010). Once national stakeholders have completed the decision-making process described in the FAO
guidelines and prioritized local breeds for breed improvement programs, the guidelines presented here
can help to plan and implement the breeding strategies with the livestock keeper communities.
2.4 Structure of the guidelines

The main part of the guidelines consists of three sections:
• Implementation of the program (Section 3)
• Creating an enabling environment for a community-based breeding program (Section 4)
• Monitoring and evaluation of the breeding program (Section 5)
Section 3: covers the three most important steps in the implementation process, and hence
forms the core of a breeding program:
1. Identifying the target site and target group(s)
2. Developing the breeding plan, including:
a. Definition of breeding goals and selection criteria, and
b. Assessing alternative breeding plans
3. Building adequate breeding structures
For each step, the activities that need to be undertaken are described and appropriate tools introduced.
Fig. 1 Logical sequence of steps to implement community-based sheep breeding

User Guidance 5
Section 4: explains the support required from different institutions to initiate, implement, and sustain
a breeding program in the long term. It also proposes complementary activities to be undertaken
and tasks to be implemented in order to create enabling environments for the breeding program.
Section 5: describes and discusses the monitoring- and evaluation-related activities required to
continuously assess progress being made and the final success of the breeding program and its
impact. In this section the various actors are mapped and their roles at the different levels described.
2.5 Using the guidelines

Developing any livestock breeding program requires teamwork of a number of actors at community,
regional, and national level, each with different expertise and institutional backgrounds. The
guidelines are intended to provide a practical and technical roadmap for participating teams and
team members, including support for taking the decision whether community-based breeding is an
option (or the appropriate option) under the prevailing conditions.
The guidelines are presented in sections arranged in a logical sequence to help users to follow
the steps in implementation. However, as outlined in Section 4 genetic improvement is only one
component of population breed improvement and development. Strategies aimed at improving
nutrition, marketing, health, housing, and the welfare of the animals, as well as other related services
have to be taken into consideration when developing a breeding program.
Although the guidelines outline the necessary requirements and the implementation process, they
also point out and discuss the limitations. Embarking on the development of a breeding program,
no matter whether community based or centralized, is not simple and therefore should not be
taken lightly. For a breeding program to be successful and sustainable, long-term commitment
of all stakeholders is crucial as success and tangible impact will only be achievable after several
generations and many years of consistent collaboration among the key actors.

6
3. Implementation Guidelines
3.1 Selecting target breeds and communities
3.1.1 Selection of breeds

When initiating and implementing community-based breeding programs it is important to pick
the right breeds, populations, and locations to work with. There are a number of criteria to follow
in selecting target breeds and communities. In our sheep project we selected four breeds for the
following reasons:
• The breeds are the most populous in Ethiopia with a wide area coverage, thus interventions
would have far-reaching impact
• The breeds are kept by resource-poor farmers/pastoralists
• The breeds are genetically diverse as evidenced by previous phenotypic and molecular
characterizations
• There is a good potential for genetic improvement of the breeds
• Regional research centers with relevant expertise and interests are available within reasonable
reach of the communities who keep those breeds
• Reasonably good background information is available on the breeds and production systems,
on which the planned and future research and development work could be based
• The areas are comparatively easily accessible.
3.1.2 Selection of community

Selection of the right community has been recognized as key to success of community-based
programs. Some essential factors to consider in selecting target communities for a community-
based breeding program include:
External factors
1. Market access: distance to market, transportation of products, and quality of roads. This is also
critical as the market is the major driving force for improvement and development projects
2. Consider potential or possible negative or positive impacts by other projects: e.g. irrigation
might result in more cropping and less livestock activities. A crossbreeding program could
jeopardize the breeding program as farmers would see impacts in the short term that trigger
their interest and cause them to abandon or disregard the agreed breeding plans
3. Synergies with other projects: it is important to scan for possible involvement of other
stakeholders, if available allow room for their participation; for example a development program,
that could actually provide the enabling environment for the realization of the project ideas
4. Government support: although this applies to the whole sector and not to a specific community,
it should be considered what local developments are occurring in relation to policies and
government priorities; e.g. the development of abattoirs and feed producing plants
5. Support from NGOs
6. Availability of inputs and services (public vs private): existing or potential for development.
These include forage seeds, feeds (roughage and concentrates), veterinary drugs, veterinary
services, drug vendors, extension systems (technical advice), and market information systems.

Implementation Guidelines 7
Community-related factors
1. Willingness/interest of the community to participate in the project
2. The key species of interest, in this case sheep, should be a priority: a substantial portion of
income should be generated from targeted livestock species. Set a minimum percentage for
selection in relation to the importance of the target species at the national level
3. The community should have a sufficiently large (combined) sheep flock (> 500 ewes): the
distribution of the flock within the community should be fairly equal (disparities should be
avoided, e.g. situations where one farmer has 400 ewes and a few farmers have 10 ewes each).
4. Existence of communal/shared resources or institutional arrangements. For example, common
grazing land or watering points, and or common use of breeding rams, herding or marketing
facilities. These indicate that already some common facilities exist that require collective actions;
the existing institutional setup could be therefore used as the starting point for developing the
institutional structures of the breeding program
5. Presence of community leaders (elders) and champion farmers/ pastoralists who are very
important in social and traditional structures in the region. They should be involved as
community-level facilitators to work closely with the project’s team. It is critical to identify such
persons as early as possible with the help of farmers/pastoralists and also extension workers,
researchers, and NGOs that have previously worked in the area or are still working in the area.
Religious leaders could also play important roles.
Suggested steps to follow for selecting the communities

1. Consult with extension people, researchers working in the area, former livestock specialists
who know the area, NGOs, and development projects. It is useful to build an inventory of
stakeholders, and a map of actors including roles and responsibilities. Let these resource
persons suggest the potential or candidate communities to visit
2. Visit the communities, if possible accompanied by people that have already developed trust
(provided they agree with your goals)
3. Organize a participatory workshop – this is a very important key step. The beneficiary community
as well as key stakeholders including public sector (extension, researchers, cooperatives,
microfinance, and administration) and private sector (NGOs, traders, brokers, butchers, export
abattoirs, feed suppliers, and drug vendors) should be carefully identified for participation.
Gender balance also needs to be considered. Such workshops should be organized when
the community is not occupied by farm activities and facilitated by somebody who knows the
culture and language of the community
4. Document the whole process, preferably done by a communications expert.
3.2 Characterization of target sites and breeds
3.2.1 Description of the production system

Assuming that some broad information on the production system is already available from secondary
sources, the characterization of production systems for the purpose of the breeding program should
concentrate on additional issues, including:
• Importance and function of livestock in the system and use of livestock products
• Economic evaluation of production (production costs and returns from sales)
• Current breeding practices (management of males and females, herd structure, gene flows,
including exchange and/or acquisition of new breeding animals)

8
• Marketing channels and opportunities for marketing animals and animal products
• Institutional settings that affect breeding and animal management, including marketing
(decision mechanisms within the community).
This information should be collected by standard methods such as Rapid Rural Appraisal and
farm monitoring, with active participation of farmers to provide the answers to the ‘what’ type of
questions, followed by workshops with focal groups to provides answers to the ‘why’ and ‘how’ type
of questions. This will help in more precise design of surveys and to validate information collected
at household level.
Mixed crop livestock production system in Menz, Ethiopia
3.2.2 Breed characterization

Populations of livestock species in developing regions are traditionally recognized as distinct types
by ethnic group or geographical locations, from where they often derive their names. Preliminary
identification of breeds or populations involves phenotypic characterization of distinct populations
using a combination of stratified and purposive sampling strategies. Qualitative and quantitative
descriptions, including morphometric measurements of animals, are collected through farm-
level surveys to identify and describe the representative samples of animals from the targeted
populations or breeds or breed groups. For this purpose, a comprehensive list of animal descriptors
was developed by FAO (1986) and Ayalew and Rowlands (2004).
Phenotypic characterization
Qualitative and quantitative variables to be observed and recorded include:
• Phenotype: qualitative variables such as coat color, fiber type, face profile, presence of horn,
and tail type, and quantitative characters to be measured are body weight, withers height, body
length, and heart girth
• Phenotypic performance characteristics, such as body weights of adult males, daily milk yield
at onset or peak lactation, and lactation length
• Flock/herd-level reproductive performance data (e.g. ewe fertility, lambing rates, prolificacy, and
pre-weaning survival rates).
Means for each quantitative measurement are calculated to describe each population sampled.
Related indigenous knowledge systems can also be collected at this stage.

Horro ram, Ethiopia
Molecular genetic characterization

Molecular genetic characterization of populations broadly involves assessment of the level of genetic
diversity within and between populations. Genetic characterization tools include biochemical
(protein) polymorphisms and molecular polymorphisms. Polymorphic molecular genetic markers
include microsatellites, single nucleotide polymorphisms, restriction fragment length polymorphisms,
randomly amplified polymorphic DNA, mitochondrial DNA markers, Y-specific alleles, and amplified
fragment length polymorphisms. Microsatellites have become markers of choice for diversity
studies because of their co-dominant nature, ease of amplification, and hyper variability. FAO (2007)
recommends about 31 microsatellites markers for characterizing sheep. Recently, SNPs have been
used extensively.
3.3 Definition of breeding objectives

The breeding objectives and the selection criteria or traits, on which the livestock keepers wish
to improve and base their selection, have to be understood. In this regard the homogeneity and
heterogeneity of breeding objectives and selection criteria among community members, as well as
between neighboring communities, should be assessed. Uniform and consistent views of farmers
facilitate the creation of a common understanding and working towards a clear formulation of
common objectives.
As the market demand is of utmost importance for the long-term economic benefits from keeping
a certain breed or breed combinations, information on current and predicted consumer demands
(e.g. size of carcass and meat quality) from different markets need to be collected. Valuable market-
related information from key market agents such as traders, abattoirs, butchers, food industries,
restaurants, and in some cases also individual consumers (end-users) are important as well and
should be collected and analyzed.

10
A cross-check of the current breeding objectives of the communities and the market demands
allows a validation of the suitability of current objectives. The findings from the market study have to
be presented to the involved communities, and in the case of discrepancies required adjustments
have to be discussed and decisions taken.
Different participatory approaches can be used to describe breeding objectives of communities.

Some of these methods are briefly described below, and advantages and disadvantages of each
method are summarized (Table 2).
3.3.1 Personal interviews

One option, often the starting point in defining breeding objectives is to interview individual farmers
and ask each of them to list and describe the traits that are of economic interest to them, and what
selection criteria they respectively employ toward achieving the stated objectives. This process is
best done with key and knowledgeable, not necessarily all, local villagers. The listing of preferred
traits is then followed by asking or facilitating the producers to independently rank or assign a score
for each of the traits or trait categories. The respondents should not be influenced by a pre-defined
format. If a person does not recall any criteria, the facilitator can help them by suggesting some
points; however, forced answers must be avoided.
3.3.2 Workshops (focus group discussions)

A group of 8–15 persons can be invited to discuss their opinion on breeding objectives and
selection criteria. Such a workshop has to be facilitated by at least two persons: one moderating
the discussion and one recording the information on a flip chart or a board. Where necessary,
an interpreter, should be used to ensure clear and common understanding is maintained. The
information should be clearly presented and made visible, and where necessary diagrams or charts
should be used to illustrate the issues during the whole workshop to all participants. The task of
the moderator is to ensure that each participant can freely express his/her point of view. In such
discussions it is usual that long lists of preferred traits are compiled initially; however, the facilitator
should help the participants to shorten/limit the list to key and most valuable traits only, by pooling
related traits, ranking all the re-listed, agreeing on the ranks, and then finally ranking the final listed
traits ensuring that traits that are valued most are captured.
Farmers’ group in Bonga, Ethiopia, discusses breeding objectives

3.3.3 Choice cards experiment

To design choice cards experiments, valuable information about farmers’ preferences for different
traits should be gathered from the survey questionnaire results. Respondents are presented with a
series of choice sets, each containing five or six alternative traits. For example in our sheep project,
six traits for ewes (body size, coat color, mothering ability, lambing interval, twinning rate, and tail
type) and five traits for rams (body size, coat color, tail type, libido, and presence or absence of
horns) were used, with the exception of one sheep breed (Afar) where one additional trait (milk yield)
was included for ewes. From each choice set, respondents are asked to choose their preferred
alternative. The attributes used are common across all alternatives. Each of the traits is grouped
into contrasting classes (i.e. ‘good’ or ‘bad’). There is also the possibility for opting out. The trait
categories are described to interviewees using drawings of hypothetical types of sheep (Figure
2). For those traits that cannot be described using drawings, trade-offs between the different trait
categories are described verbally. Additional information, often picked up from group discussions
could be added to the choice cards to investigate interactions between such information and the
traits presented in the cards.
Fig. 2 Example of a ‘choice card’ tool used to describe breeding ram trait preferances
3.3.4 Ranking of live animals

Ranking of own animals
The sheep owners are asked to rank their own animals (female and males) from the 1st (best),
2nd (second best), 3rd, and worst, and to indicate reasons for their ranking. After receiving detailed
information from the owners, linear body measurements including body length, chest girth, heart
girth, tail length, tail circumference, body condition, and dentition are taken on each individual ranked
animal. This is to assess the correlation between the farmers’ rankings and the actual quantitative
measurements taken on animals of same age or age groups, and what such relationships/
correlations mean. All family members can participate in this exercise. Often there are not many
rams in the flocks and therefore it is difficult to accomplish this exercise with males.

12
Horro, Ethiopia. Farmers do ranking of their breeding ewes
Ranking of animals not known to farmers

Another option is to ask farmers to rank animals which are unknown or unrelated to them (i.e. other
peoples’ animals). These animals could originate from a research station or from other, distant
farmers. For this method, the focus is first on the phenotypic appearance of the animals. The test
persons can then be provided with additional facts on production and reproduction on each animal
to further inform his/her decision. It is important that each test person is provided with identical
information on a given animal (Box 1) .
Box 1. Group ranking of sheep

In the ICARDA–ILRI–BOKU sheep breeding project, 15 ewes and 15 rams were randomly selected from
the communities’ flocks at each study site, marked and randomly assigned into five sub-groups and
then penned together. A total of 30 sheep owners from each site were moved to the other site (each
location has two sites) so that farmers were ranking animals with which they were not familiar. Each
interviewee was asked by an enumerator to rank the animals within each pen according to his/her own
preferences and give the reasons why s/he had chosen the animals as 1st, 2nd, and 3rd. Then they
were provided with a life history of the animals, including information on productive and reproductive
traits to determine whether they would change their rankings.

Menz, Ethiopia. Farmers rank sheep not known to them
3.3.5 Comparison of methods

The different methods described for defining breeding objectives have advantages and shortfalls
(Table 2). One has to analyze the practical situation on the ground before decding on the methods
to be used.
In general, to make sure that no selection traits are overlooked, a combination of at least two
methods is recommended. Selection traits should have the following features:
• Relate either directly or indirectly to the breeding objectives
• Easy to measure under field conditions
• Heritable
• Not too many (not more than three under smallholders situations)
• Relationships between selection traits should be understood, as antagonistic relationships (i.e.
traits which are negatively genetically correlated) between two traits mean improvement in one
trait will result in deterioration of the other trait.
In the Ethiopian community-based sheep breeding programs, we followed the different methods
described and defined breeding objectives and selection traits of the communities (Table 3).
One difficulty is how to include adaptive traits of local breeds to various stress factors of the
environment (e.g. diseases, internal and external parasites, water scarcity, and walking ability).
These traits are often difficult to record under field conditions. However, given that these are
extremely important traits for small ruminant production in tropical and other harsh environments,
mechanisms should be sought to include them in selection decisions. For example, resistance to
internal parasites could be measured by fecal egg counts (easy to measure) and has been shown
to be heritable (Baker 1998). Because all animals are equally exposed to the same stress factors,
often to similar magnitudes, one can assume that the best performing animals under the given
production environment must be the best adapted to the prevailing conditions. For example, under
conditions where animals all graze on poor quality pastures the fastest growing animals, among
those grazing the same pastures, must be those able to cope best with such forages, and
hence should be selected as the future sires and dams.

14
Table 2: Advantages and disadvantages of alternative methods of defining breeding objectives
Properties Personal Workshops Choice cards Ranking of live animals

interviews Own animals Unknown to farmers
Advantages • A large • Information • Large sample • Relatively • Easily done by

number of from size easy to farmers
persons different • Enumerator handle • Closer to reality
can be persons introduced • Closer to than choice cards:
interviewed collected bias likely reality than seeing a live animal
• Possible to at once to be lower choice cards: is better than a
verify the • Differences than in Seeing a live picture
consistency can be interviews animal is
of the directly • Price can be better than a
responses discussed included as a picture
• Additional characteristic • Information
information from different
can be family
gathered at members can
the same be considered
time
Dis • Language • Some • Limited • There may not • Large ‘pool’ of
advantages barrier people (e.g. number of be enough animals often not
• Enumerator- with higher animal profile animals of readily available
introduced social choices can the same Hypothetical life
bias may be status) be made per category history provided
high might person available in with a given
• Important dominate • Visual small herds animal may not be
traits may the illustration compatible with the
not be discussion of some visual appearance
mentioned traits can be according to
complicated farmers’ experience
or impossible • Limit to the number
of additional
traits that can be
included
General • Access to farmers is often difficult
comments • Clear sampling strategy needed
• Unclear results, if farmers have no common breeding goal

3.4 Assessment of alternative breeding plans

There is no single most suitable method for designing breeding plans that fits all possible
circumstances. Thus, one option is to evaluate the results of alternative designs through modeling
in order to choose the best under the given circumstances. There are two basic approaches for
modeling and evaluating breeding programs: deterministic and stochastic models.
Stochastic simulation is the easiest of the approaches. Its advantage is that one can mimic the
true breeding program in detail with more precision because the individual animal is simulated.
Its disadvantages relate to time/computer power requirement and the user does not gain much
insight compared to the deterministic approach. Simulation of a large number of replicates of a
large breeding scheme may take from several hours to days, making the approach less suited as
an operational tool to quickly evaluate alternative schemes. Since stochastic simulation does not
explicitly model mechanisms (e.g. accuracy and generation interval) the user may not be able to
appreciate the relationship between the determinants. Hence it is difficult to extend results to other
breeding schemes that have not been simulated. Examples of the stochastic computer programs
include ADAM, EVA, and SixS.
The deterministic method does not mimic the breeding program on the individual animal level but
use deterministic equations and population parameters to predict gain and inbreeding. Hence it
requires more insight into quantitative population genetics than stochastic simulation. Advantages
of deterministic methods are short computation time (many alternatives can be computed within a
limited time) and it gives a lot of insight into gain and inbreeding within breeding programs because
the mechanisms are modeled explicitly. Software packages available for deterministic calculation
include ZPLAN, ZPLAN+, and SelAction.
In this guideline we limit ourselves to the most widely used deterministic model, ZPLAN, which was
developed in 1980s at the University of Hohenheim, Germany. It was designed to optimize livestock
breeding strategies by deterministic calculations. It evaluates both the genetic and economic
efficiency of breeding programs considering one selection cycle. ZPLAN is written in FORTRAN
and allows flexible modification of existing subroutines to model desired breeding scenarios as
realistically as possible. The optimization of a particular breeding program in ZPLAN is based on
three functional core areas: selection index procedure, gene flow method, and economic modeling.
Important outcomes of ZPLAN include annual monetary genetic gain for the aggregate genotype,
annual genetic gain for each single trait, and discounted return and discounted profit for a given
investment period.
Depending on the particular situation, in the design and evaluation of a potential breeding program,
the following basic decisions must be taken and selected in ZPLAN:
• Defining of tiers in the breeding plan
• Defining of sexes in the selection group
• Indicate paths of gene transfer from one group to the other.
Then input parameters (input files) such as population, biological, and economic cost parameters
are defined by the users. Furthermore, phenotypic and genetic constants are required for modeling
alternative breeding plans (Box 2).

16
Once a breeding program is operational, the realized genetic improvement over the period of
interest can be compared with the predicted values. Reasons for the observed differences, if any,
can be examined and new strategies developed to rectify or accelerate progress towards the
desired outcomes.
Box 2. Input parameters required to run ZPLAN

Input parameters: input parameters (input files) are defined by the users and can be subdivided into:
1. Population parameters
• Population size (females)
• Number of proven males/year
• Proportion of male and female animals in different tiers
2. Biological parameters
• Duration of breeding females’ and males’ use (time unit)
• Mean age of females and males at birth of first offspring (time unit)
• Mean time between subsequent lambing/kidding/calving (time unit)
• Mean number of offspring per litter (e.g. litter size in sheep and goat)
• Mean number of offspring per female per time unit
• Survival to weaning/yearling
3. Economic/cost parameters for a given investment period
• Fixed and variable costs/breeding female: increased cost per unit should be discounted when
calculating relative economic values. These values may vary from breed to breed or from region to
region within the same breed. Only additional feed or labor costs spent over the normal husbandry
practices have to be included during simulation.
• Interest rates of return and costs: have to be based on real rates of interest/cost (i.e. bank account
interest rates of specific region or country). It is commonly recommended to use slightly higher
discount rates for returns than for costs, because returns are realized later than costs.
• Investment period is defined in time unit. For instance, for cattle and sheep one time unit is one year
and for pigs it is six months. It has been quite common in animal breeding studies to define the
investment period as three or four times the mean generation interval of the particular species under
consideration.
Phenotypic and genetic constants:

• Phenotypic and genetic standard deviations for goal traits
• Phenotypic and genetic correlations between each pair of goal traits
• Heritability estimates (heritable fraction of the variance in each trait)
Estimates for phenotypic and genetic constants are lacking for most indigenous breeds in low input
systems. In that case, literature-based information estimated from breeds found in similar production
systems or production environments should be used.

3.5 Developing adequate breeding structures
3.5.1 The breeding program

In Sections 3.1 and 3.2 the aspects critical for the target community, farmer group, and breed were
indicated and discussed. Also discussed were how to target the animal population to be genetically
improved. This section on the breeding program considers the identified animal population in terms
of its biological characteristics, together with husbandry practices under which it is raised, the
prevailing and anticipated infrastructure, as well as the constraints and opportunities – all of which,
if appropriately considered, enable the design of a program which maximizes both genetic gain
and profit for the community.
The simplest and most straightforward design is one in which the best males and females are
selected as replacements (i.e. future parents of the next generation) from the whole population.
This means that all the herds/flocks would be monitored and be involved in screening for these
‘best’ individual animals. The problem with such a design is that each member of the participating
community would have to be involved, somehow, in the selection process. Each farmer therefore
will undertake performance recording, pedigree recording, and rearing of male candidates; the
latter, at least up to some defined age, which may not always be practical.
Performance recording may be kept at a minimum or performed in stages. Rearing of male

candidates, at least post-weaning, may be centralized at a test station or may be entrusted to a few
selected members of the community.
An alternative design would be to have some farmers with ‘best’ animals and often ‘best or average
practices’ to breed males for use by the whole population. Such designs with structured populations
are called ‘nucleus systems’. Nucleus farmers concentrate on maximizing genetic gains while the
remaining ‘base’ farmers can concentrate on production. In this case best males and females are
mated in the nucleus, in order to produce the ‘best’ next generation of young animals, thus increasing
the probability of having better gene combinations in the progeny at the nucleus compared to the
rest of the population.
For the designs above to deliver, the nucleus must be functional, that is nucleus farmers not only
have to make genetic progress but also have to consistently produce and disseminate appropriate
number of genetically superior males to nucleus and base populations (farmers’ flocks/herds). Thus,
the size of the nucleus, or proportion of females to the total community herd/flock which should
be in the nucleus, depends on the number of males needed by the entire system – taking into
consideration a desired selection pressure or intensity. A minimum effective population size is also
required to avoid inbreeding at the nucleus. For example if we desire an annual rate of inbreeding
to be < 0.5% in a nucleus with average generation length of 3 years, effective population size must
be > 33 (for example 9 males and 100 females).
The nucleus can be either closed or open. A closed nucleus means no upward (from base to
nucleus) gene flow is allowed, while an open nucleus allows the best sheep to enter the nucleus
from the base population (Figure 3).

18
Fig. 3 Open and closed nucleus schemes
Open nucleus systems (Figure 4) require base farmers to do some selection, usually on the females.
Usually females for the nucleus are supplied in exchange of males, but other arrangements are
possible (e.g. cash, in kind, or percentage of sales). A very important feature of open nucleus
systems is that adaptation traits and other breeding objective preferences in the base population
are secured in the males produced since these are born from ‘best’ base females (Box 3).
Box 3. Open Nucleus schemes for Merino sheep

In Argentina open nucleus systems are common for Merino breeding communities. Nucleus flocks are
established with best females and half of their replacements are selected from the base flocks. Males
are selected on measured performance and visual inspection. Rams performing above average and that
are visually acceptable get a special identification from a breeder’s society. From these the best remain
in the nucleus and the next best are used in the base flocks.
The rate of inbreeding can be reduced and genetic progress increased in the so called ‘open
nucleus’ systems. In such systems the nucleus is ‘open’ to best animals detected in the base
population. Typically very good females born in the base replace worst females in the nucleus.
Typical designs have 5–15% of the total female population in the nucleus and have about half
of the nucleus replacements come from the base. For example a community with a total of 700
breeding females would need about 70 breeding females in the nucleus. If 20 nucleus female
replacements are needed each year, 10 should be selected from nucleus progeny and the other
10 from base progeny. The proportion of females in the nucleus and the proportion of base females
going to the nucleus can be smaller if selection is more accurate in the nucleus or reproduction rate
is high or the female to male ratio is high.

Fig. 4 Open nucleus scheme: nucleus and base flocks. Example, Merino Sheep Breeding
Program, Argentina
We assumed that all animals of nucleus farmers breed males since all their females are supposed
to be ‘best’. This is not necessarily the case, as a nucleus farmer may also have inferior females. In
that case the nucleus farmer may identify his ‘best’ females and mate only these with ‘best’ males,
or he may mate all his females with ‘best’ males but consider for selection only the progeny of ‘best’
females. There will be several nucleus farmers with only some of their females qualifying as nucleus
animals. Such a system requires controlled mating at the nucleus farm or early castration of male
progeny from non-nucleus females.
For practical reasons, it is difficult for the individual farmer to raise male candidates from birth until
final selection. Variations in level of husbandry between farms can create serious confounding,
making clean separation between genetic and environmental superiority rather difficult. To get
around such a problem, young candidate males, usually at weaning, are gathered and placed in one
common station or farm; here their performances are jointly monitored under the same conditions
for a fixed period of time. This process is known as ‘performance testing’ and the common station
as the performance-testing station. Such a station may belong to the community itself or may be
facilitated by an external organization (Figure 5).

20
Individual farmer
Sire pool
Base nucleus
Test station
Any farmer can produce male Candidates are evaluated

candidates and therefore at a central performance
has potential nucleus females testing station
Fig. 5 Open nucleus scheme: dispersed nucleus. Example, Bonga and Afar Sheep Breeding
Program, Ethiopia
In cases where nuclei are run and managed by several farmers, but all following similar management
procedures and selection processes, then the nucleus is referred to as ‘dispersed’. There are also
programs with a single nucleus-farmer producing males for a group of farmers. Such systems are
also called ‘group breeding systems’. The principles of open nucleus systems apply; however,
considering that 5–15% of the female population should be in the nucleus, group breeding
schemes are appropriate for farmers with typically > 100 breeding females each or for a situation
with a nucleus farmer is individually running a large flock.
For smallholders, large individual flocks are uncommon, but there are examples with community
single-nucleus flocks or cooperative-nucleus flocks (Mueller et al. 2002). Sometimes a single
‘central’ nucleus is run by an external organization such as a university or a governmental body.
In this case the design essentials regarding nucleus size and gene flow still apply but the control
of the community over its breeding program may be low. Nevertheless there are also examples
of community breeding systems which started with a ‘centralized’ nucleus and developed into a
dispersed nucleus system. Such dispersed nuclei in turn may develop into reference sire schemes
and eventually into population-wide evaluation schemes (Abad et al. 2002).
In conclusion, there are many alternative breeding structures and tools for its optimization. The tools
are useful for a strategic optimization of a breeding program. In practice, however, many variables
are fixed and a program can rarely start with an optimum structure in terms of layer size and gene
flows. Thus, practical situations need tactical optimization, which means finding the best solution at
each step of the program while having a target structure in mind.

3.5.2 Animal identification

Animal identification is crucial in genetic improvement programs. Animals should be uniquely
identified, so as to accurately trace their respective pedigrees and link the performance of individual
animals to her/his progeny and relatives through known genetic relationships. Combining
performance and pedigree records enables more accurate computations of the genetic worth or
breeding values of the animals to be estimated or predicted, and used for selection.
The identification methods employed can vary between regions and communities. Ear tags, collars,
tattoo, branding and ear notches can be used. Ear tags are the most commonly used identification
methods because they are relatively cheap, easy to apply, and are less stressful to animals. However,
in some situations, where for some reasons ear tags are not acceptable or practical, for example
because of cultural taboo or shape of the ear, other alternatives should be sought.
Unique numbering should be embraced, such that any two animals in the breeding program should
never have the same identity, both in time and space. In our project we used a unique identification/
numbering system (five digits) per community. Plain plastic ear tags were procured, identification
numbers were hand-written using indelible markers and all sheep belonging to community member
households were ear-tagged.
The identification of the base population was done by the research team. Thereafter, identification of
newly born lambs was undertaken by village enumerators. Ultimately, community members should
be trained to handle animal identification by themselves.
Animal identification using ear tags
3.5.3 Data recording and management

Development and use of a simple, flexible, and cost-effective performance recording and evaluation
system is essential for the breeding program. The recording formats should be kept as simple and
as practical as possible for easy use and adoption.
In accordance with the agreed selection traits, three recording formats were developed for each
location in our sheep project: two for ewes and one for lambs. The ewe format contained information
such as lambing date, parity, and litter size. One of the ewe data formats had detailed information

22
about the ewe (Annex 1) and was kept with the enumerator. The second ewe data format, however,
had less information to be recorded by the household and was kept by each household. The lamb
data format had information about lamb identity and performance. These formats were developed
after a thorough discussion with the community.
Major traits considered for all breeds were weight (at birth, weaning, six months, and yearling) and
number of lambs weaned. In addition, milk yield for Afar, number of lambs born (twinning) for Bonga
and Horro, and wool yield for Menz were included.
Weighing of animals
In a community-based breeding set up, the ultimate goal should be to ensure that the community
can handle and indeed does handle all activities required for the program to be functional at
community level. At the initial stages, however, assistance is needed. In our project, an enumerator
was employed for each community to assist households in measurement and record keeping and
to continuously train them on how to do it. A recording book was prepared for each household for
day to day follow up and one recording book, kept with enumerators, was provided per community.
All necessary equipment (e.g. weighing scales, ear tags, and markers) should be made available for
effective recording and follow up until the community masters the major activities.

Performance records used in the ICARDA–ILRI–BOKU sheep project
Appropriate training is crucial for success and should be organized for both enumerators and the
community and offered in easily digestible components (i.e. not rushed and offered all at once). The
type of data and frequency of collection need to be decided upon in close consultation with the
community and must be based on the agreed breeding objectives and selection traits. The simpler
it is the better and the higher is the probability of its sustainability. First, focus should be on the key
few (3–4) traits only, with additional traits added to as and when necessary as the groups mature
and become more sophisticated.
One should also discuss with the individual households to identify and agree which family member
could be in charge of the data recording and handling. It is sometimes useful to engage school
children, as adults might be illiterate. In this case, supervision by an adult person ensures that data
are recorded on time and kept in a safe place. In our sheep project the type of data collected is
summarized in (Table 3).
Table 3: Traits recorded in four sheep breed improvement communities in Ethiopia
Breeding Traits recorded Afar Bonga Horro Menz

objectives
Body size • Birth weight √ √ √ √
• Three-months weight √ √ √ √
• Six-months weight √ √ √ √
• Yearling weight √ √ √ √
Lamb survival Proportion of lamb weaned/ewe √ √ √ √
Twinning rate Number of lambs born/ewe/lambing √ √
Wool yield Greasy fleece weight (yearling) √
Milk yield Test day milk yield (once a month) √

24
Support in data entry and processing should be provided by the local/regional research centers.
The local or partnering research institution can play this role, but once the database is developed
and in place, the database can be updated on a near real-time basis through innovative use of aids
such as cell phones, and verified via the same devices. Centralized data management tools would
facilitate easy data capture, analysis, and reporting. Simple indices based on the set selection
criteria for each breed should be developed and the overall merit values computed and shared with
the communities and farmers as part of the feedback and for use to effect selection.
3.5.4 Selection of candidate rams

Young rams should be selected based on recorded data (own and maternal performance) for the
set of agreed selection traits. Selection can be undertaken at different stages. For example, the first
stage could involve culling of animals with undesirable phenotypic characteristics (e.g. tail type, coat
color, horns, conformation, and general appearance) and clearly observable and genetic defects
(e.g. testicle deformation and undershot or overshot palates). The retained individuals are then
further judged based on morphometric and body weight. The stages at which the selection process
takes place depend on both the existing traditional practices of ram selection and use, as well as
on scientific requirements. If the selection decision can be made in line with the traditional practice,
it will improve the probability of acceptance by the community.
Holding yard at Horro, Ethiopia
For good reasons, it is important to cull undesirable males before they reach puberty (i.e. before they
can serve). Depending on the breeds, this can be as early as 6–8 months of age. Where communal
grazing is practiced, synchrony and agreement on when to cull is important as flocks can meet in
common pastureland or watering points, when the undesired entire males can breed, and hence
reduce the selection impact. It is also important that the selected young males are effectively used
for breeding before they are sold off in order to avoid negative selection.

Candidate rams for breeding in their holding yards at Horro, Ethiopia
When the breeding program is fully functional the best rams should be identified by their breeding
value computed from recorded data and based on their pedigree. Animal models can be used to
rank rams. If breeding values cannot be computed for whatever reason, rams can be selected based
on simple index values that are computed from the available recorded data from the site population.
The community has to be actively involved in the selection process so that the ram ranking closely
match their (own valuations) goals and desires. This helps to build trust and confidence, buy-in, and
a sense of belonging among the beneficiary community that increases both their confidence in the
selected rams and ownership of the process (Box 4).
Box 4. Selection of rams

In the ICARDA–ILRI–BOKU project two stages of selection were applied: initial screening at the age
when first sales of young rams occur (4–6 months) and final selection for admission for breeding at
12 months of age. All young rams are collected at one central place in each community on an agreed
screening date. Selection is then carried out based on the data analyzed. A breeding ram selection
committee composed of about 3–5 members elected by the community are involved in the selection.
In here, if for example 15 rams were to be selected from a total 100 candidates, say 20 would be
preselected based on their breeding value and the culling of the last five and the ranking of the
selected rams would be made by the committee. The joint selection process strengthens the linkages
between farmers and researchers.
Animal exhibitions or shows could be linked with the ram selection events. During such shows animals
of different sexes and age categories could be ranked and the best ones awarded prices along
with the best young rams. Animal shows are important as they can create awareness at the entire
community of the higher relative worth of selected breeding animals compared to the unselected
ones. Individual livestock keepers who manage their flocks and records better could also be awarded
during such shows, thus creating healthy competition among the community members. This also has
a social dimension in that winners are recognized in public and creates some level of social status,
with pride to the winners and respect from the community. During selections and shows, judging
should be done through a participatory process, and preferably by a panel of committee members,

26
who are formed/nominated from among the local site farmers/pastoralist, by the community members.
This creates some form of ownership and transparency.
Awarding of rosettes to the best breeding ram
3.5.5 Management and use of breeding rams

In cases where the individual flocks are quite small, the flocks should be treated as one flock.
Selection would then be undertaken at community-flock level, with the selected best rams shared
among the community members thereafter. This is often the case in smallholder mixed crop–
livestock systems. In some communities such tradition of ram sharing may already exist; however,
in other communities members may be reluctant to share rams outside their established social
networks. Repeated consultations should be made with the community to arrive at an agreed
modality as there is no single arrangement that applies to all situations. In our project, the modalities
for ram exchange that were discussed with the communities included:
• Sharing rams based on friendship and trust among members of the breeding group
• Exchanging rams based on a written agreement
• Exchanging rams based on purchase between different breeding groups when rams complete
the defined service period in a given flock
• Advancing some seed money from the project or from members’ contributions to purchase the
first round of breeding rams, use these, then sell them to generate a revolving fund to purchase
the next and subsequent rounds of breeding rams.
It should be noted that one or a combination of the above arrangements may be adopted or used
depending on the prevailing circumstances. In our project, the last option was chosen. It is expected
that creating a revolving fund will sustain the program in the long term. This also helps to prevent the
negative selection of rams that is a common phenomenon in the communities. Negative selection
arises from faster growing males being sold off early before they are of breeding age, leaving the
slower growing males as the breeding males in the community flocks. To avert negative selection,
the best young rams are purchased by the project and are owned and used by the community. After
two years of service (period to be agreed with the community), such rams are to be fattened and
sold to support the purchase of the next group of selected rams for the community. Mechanisms
of how to use the revolving funds and how benefits will be shared have to be agreed upon and the
by-laws and an administrative procedure established.
The best way to use rams communally is by forming ‘ram-user-groups’ and this can be based on
criteria such as number of breeding ewes, settlement pattern, and use of communal grazing areas.

Traditional ram use groups are often based on social networks and perhaps resource availability
and thus these should be considered where and when applicable. In order to minimize inbreeding
a strategy for ram rotation among the ram groups has to be established through a consultative
process. Our project was planned such that a ram is used in flocks for one year after which it is
rotated to another ram-group within the community. The ram rotation records must be diligently kept
to avoid inbreeding. The management of selected breeding rams to be used by the community
should be based on prior-agreed modalities. Some of the options include:
• Manage the ram in rotation
• Keep the ram in one agreed household and those who use the ram pay an agreed amount
for the service
• Keep the ram in one agreed household and other community members contribute in kind (e.g.
feed and veterinary drugs) to keeping the ram.
A critical issue that needs to be thought through is how to manage the unselected rams. It should
be recognized that the unselected rams, especially those young rams that fulfilled the initial
requirements but got lower ranks than the selected rams, are better than the rams in neighboring
communities where no selection program exists. Therefore, mechanisms should be designed to
sell these rams. Value addition in terms of fattening could be organized for the unselected rams
and linked to markets. If the animals could be pooled together for targeted markets then their value
would be much higher than when they are individually sold.
3.5.6 Institutional backup: organizational issues

Community-based breeding programs need to be initially supported by a committed team of
researchers, extension personnel, the NGO community, and project staff. The institutional backup
needed to implement such a project can vary depending on expertise and resource availability. For
illustration, the structure that we used in our sheep project is presented (Table 4), but note that not
all community-based breeding programs necessarily need to take such a form. A sufficient period
(at least four years) is required for local communities and the national research institutions, before
they can take over full responsibilities, ensures sustainability and success.
Table 4: Structure of community-based sheep breeding implementation team in the ICARDA–

ILRI–BOKU sheep project
Level Team membership Role/TOR
Project level Project Coordinator (PC) • Overall project leadership/admin
(plus quantitative geneticist for • Liaison with project partners
technical backstopping) • Prepare progress reports
Site level • Senior researcher (Team Leader) • Oversee activities on the site
• Other scientists
• District office of Agriculture
Community • Community leader • Community-level leadership
level/Team • Community representatives • Provide links between project team and the
(elected focal point) such as community
committee/council members, • Reporting community-level developments
elders, women, youth, • Assisting project logistics at community level
government representative • Feedback on progress of project activities
Graduate fellows • Look at critical aspects of the project

28
4. Creating an Enabling Environment

4.1 Community-implementers’ relationship
It is crucial that the program provides the basis for farmers to effectively interact with researchers and
extension staff and to openly discuss their fears, doubts, and ideas about the program. Workshops
should be organized regularly to discuss every step of program design and implementation.
Informal consultations with community elders would help to get information about the program. It is
also vital to provide regular feedback to the farmers. Record sheets of individual flock productivity as
deviations from population averages would help the community members to realize performance of
their sheep compared to their contemporaries. Even preliminary results have to be presented and
discussed with the farmers, thus giving them the chance to comment and share their opinion with
the researchers. Such feedback also reduces the risk of misinterpretation of results, which could
lead to wrong decisions. The relationships between researchers, extension agents, and farmers/
pastoralists should be based on trust, transparency, and respect, which is expected to evolve
through working together closely. An important precondition is to consider and account for the
cultural, religious, and ethical values of the community.
4.2 Complementary interventions, services and capacity development

Ideally a breeding program should be part of a broader livestock improvement program. Genetic
improvement should therefore be complemented by other interventions, notably, access to improved
and affordable health services, market information and market services, improved infrastructure, as
well as supportive policies. Development of, and ensuring adequate quality, feed resources and
supply all year round will ensure that improved genetics are expressed to their optimal genetic
potential.
Legume and grass fields for livestock feeding
The benefits and effects of the complementary interventions are realized within a relatively short
period of time, long before the real effects of genetic improvement become apparent or visible.
Projects, research and extension departments, and NGOs can assist the program in various ways
until the community understands the benefits and start to invest themselves.
The farmers may form cooperatives or farmers associations to ensure better access to markets and
stronger negotiation power, and this also means that a larger number of animals and/or quantity of
livestock products are regularly marketed. Cooperatives also have easier access to credit and can

Creating an enabling environment 29
negotiate with service providers (feed industry and veterinarians) a better price for particular services
or products, which can be ordered in larger amounts.
Local, regional, national, and international market-information – for breed, type, sex, body weight-for-
age, and price – is key for market participation and market-orientation of farmers and pastoralists.
All in all, the interventions listed above should be accompanied by capacity development for
the different actors involved in the program. The capacity development programs should start
by mapping of the actors, assessing their respective strengths, and identifying the main gaps in
knowledge, organization, and institutional weaknesses at the site level. This should then be followed
by listing of priority topics that should be covered as part of capacity development. Tailor-made
training programs would then be developed and offered to the different actors. Below are some of
these possible actors and the possible areas that can be targeted for capacity development:
Livestock keepers: could get technical support in order to successfully implement the new
technologies. Different forms of training should be envisaged, such as a few one-day workshops
and trainings on improved husbandry, especially health care, animal housing, feeding, and the
essentials of group dynamics, particularly on effective group management (e.g. meetings, recording,
and conflict resolutions). Such training is best supported by practical demonstrations in the form
of farmer field schools or reciprocal farm visits to neighboring communities to stimulate healthy
competition among groups and farmers.
Staff members of extension services: also need to get refreshed and exposed to new related
technologies, and nave their practical applications presented to them, as these persons are
important key informants for farmers.
Researchers: may need specific training in participatory research methods or to have their
knowledge refreshed on various aspects of a breeding program. They can also be trained on data
recording, analysis, and effective reporting.
Private sector: this can be such people as drug vendors, veterinary service providers, feed suppliers,
traders, brokers, butchers, and export abattoirs.
Financing and insurance institutions (government and private sector): these are essential bodies
in accessing credit and also in establishing community-based insurance systems for livestock to
deal with issues of death of animals.
It is important to note that more often than not, is the lack of soft skills among community members
and the technical support staff that lead to failure of community-based livestock improvement
programs. For success, focus on improving the soft skills of the actors, while simplifying the technical
aspects is recommended.
4.3 Government support

For breeding programs to be sustainable, long-term commitment by the local and central
government is essential. Initially the costs of performance recording and animal identification
cannot be shouldered by the farmers, so the government should pick up such costs. In addition,
supportive policies should be developed and their implementations facilitated by the government.
Adequate funds should be allocated for technical personnel (researchers and extension staff) and

30
infrastructure. Breeding programs require continuous technical and intellectual backstopping from
well-trained technicians and researchers.
It is common to find that most of the smallholder and resource-poor livestock keepers have no
access to affordable financial services, notably credit. In addition, national budgets for livestock
development and research are always limited. Government should therefore facilitate access to
credit, land, and other resources (e.g. watering points, rural access roads, livestock auction yards,
and market information on livestock and livestock products). Therefore, there is need for better
coordination among the various government departments and agencies, scientists, and other
development agents involved in such schemes. Such close coordination and networking should
be maintained sufficiently long to allow the breeding program to incubate and reach a sustainable
stage.
4.4 Links with other projects/activities

The problems facing communities are complex and inter-twined. Piecemeal approaches to
development interventions are undesirable and are usually more costly in the long run. Thus, it is of
paramount importance to follow a holistic development approach for sustainable development of
communities, necessitating integration and coordination.
As much as possible, it is recommended to try and link breeding programs to other ongoing
development initiatives and activities in the local area. This will exploit the potential synergies,
reduce on redundancies, and overall strengthen the projects. One example could be that a NGO
wants to offer training courses to farmers. Members of the breeding program could participate in
these courses. Such linkages would also help to get additional expertise and funding to support
the breeding program.

Monitoring and evaluation 31
5. Monitoring and Evaluation

An integral component of a functional community-based breeding program is monitoring technical
and management issues related to the implementation of the breeding program; whether
outputs, outcomes and impacts are achieved or achievable; and whether mechanisms to ensure
sustainability of the breeding program are in place.
Ultimately, a breeding program should be evaluated by the genetic improvements obtained in all
important traits and the effects on total output of products and outputs per unit of measurement,
e.g. per animal and the economic impacts at both household and community levels. Ideally outputs
should be related to inputs and the status of the natural resources utilized. These change with
time and must be revised accordingly. By regularly monitoring the breeding program, corrective
measures can be taken to improve the program. Showing the impact of the breeding program will
be essential for ensuring continuous support of the program.
For effective monitoring and evaluation, the breeding program team should define key indicators to
measure the progress in achieving the main outputs of the breeding program, as well as indicators
to assess whether or not the program outputs are contributing or will eventually contribute to the
desired outcomes and impacts at individual flock, household, and community levels. Appropriate
tools and procedures for monitoring these indicators have to be devised, which should also include
clearly defined timelines for each indicator. Although the details have to be specified individually for
each breeding program, it is expected that the monitoring and evaluation system for the Ethiopian
sheep breeding program proposed in (Table 5) is relevant and applicable for other small ruminant
breeding programs.

32
Table 5: Monitoring and evaluation of community-based sheep breeding program
Parameters Indicators How to monitor When to

monitor
Technical issues related
to implementation of
the breeding program:
• Proper animal • Percentage of ear-tagged • Follow-up/control in each Continuous
identification animals in participating participating household
households through enumerators
• Proper data collection, • Number of farmers involved/ • Follow-up/control in each Continuous
analysis and use dropped out in recording participating household
• Estimated breeding values through enumerators At each
of rams available at time of selection
selection event
• Selection and • Accurate selection of best • Organization and At each
management of rams rams with the community at documentation of selection selection
agreed intervals events through NARS event
• Number of active rams • Detailed accounts kept
included/culled by community of how
• Community uses revolving revolving funds are being
funds to buy rams managed and used
• Ram sharing • Selected rams shared and • Follow-up and Regular
used as planned documentation by intervals
• By-laws for ram purchase and community committee
ram sharing developed and • Documentation by NARS
implemented of whether agreed modality
is followed or, if there are
changes occurring, the
reasons for the changes
• Establishing breeders • Number and participants of • Documentation by NARS After three
associations formal or informal breeders years
associations
• Complementary • Number of fattened rams sold • Monitoring of quality and Regular
interventions: and cost–benefit of fattening continuous supply of feed intervals
› Value addition for • Increased quality feed supply for flocks and for fattening
unselected rams • Productivity losses and by enumerators and NARS
› Feed supply mortality reduced • Flock records and follow-up
› Disease prevention • Market prices achieved for of health of flocks and
and treatment animals sold by participating shared rams by animal
measures households health workers and vets
› Proper market • Recording of animal
linkages sales by households and
additional rapid surveys
with traders

Monitoring and evaluation 33
Table 5: (Continued)
Parameters Indicators How to monitor When to
monitor
Program outputs
• Level of engagement of • Active community breeder • Detailed studies on the After 2–3
the actors in program committees at each site behavior of different actors years and
activities • Attitudinal change among (comparison against repeated in
different actors in the livestock the base year when the subsequent
development program started) years
• Superiority of breeding • Appearance and performance • Evaluation of visual At each ram
animals of the selected animals is better observation and selection
than their contemporaries performance of the
• Demand from neighbors, etc. for selected versus non-
breeding stock produced by the selected rams by
breeding program community and NARS
• Higher price paid for breeding • Recording of sales and
animals originating from the market prices for selected
program rams by owners
• Breeding progress • Genetic and economic gain • Detailed analysis of genetic Annually
achieved by the program and economic gain
• Estimation of breeding
values and breeding
progress by NARS
Outcomes and impacts
• Involvement of the • Regular feedback from • Meetings with individuals Regular
community and beneficiaries received and and the community intervals
acceptability of the documented • Documentation by Continuous
scheme • Dropouts of participants community committee and
NARS
• Livelihood improvement • Productivity gain at animal and • Analysis of flock records After 3–4
flock level years
• Changes in the livelihoods • Analysis of incomes as After five
(income, food availability, and well as consumption years
work sharing) of participating patterns of households
households against baseline
information collected at
program initiation
• Sustainability of the • Economic gains from the • Project-related financial Every 3–5
breeding program breeding programs need to be expenditures need to be years
evaluated monitored and reports
• Program being implemented prepared according to
with little ‘external’ support over plans
long period of time • Assessment of whether
• Feasibility of the program in there is any external
terms of economic, social, and support to the program
natural resources dimensions • Survey
• Impact modeling

34
6. Concluding Remarks
Community based breeding programs are proposed as an option for genetic improvement of
livestock in developing countries. This new approach has been tested with promissing results in
few places (e.g. with dairy goats in Mexico and with llamas and alpacas in Bolivia and Peru).
In Ethiopia, we designed and implemented community-based breeding programs for four sheep
breeds representing different agro-ecologies and production systems. Although it will take more
time until changes in genetic gains can be fully evaluated, the breeding programs have already
achieved important ouputs:
• Negative selection has been reverted as fast growing lambs are now being retained for
breeding instead of ending up in markets
• The acute shortage of breeding rams observed previously in flocks of participating communities
has also been rectified as farmers are now fully aware of the importance of breeding males
• Preliminary data analysis indicates that the market outlet has increased through more births
of lambs, bigger lambs at birth and weaning and reduced mortality due to the combination of
breeding with improved health care and feeding.
These guidelines aim to address the lack of generic guidance on designing and implementing
community-based breeding programs. Though based on Ethiopian experience, the modalities
highlighted in the guidelines are largely applicable to similar situations where community-based
breeding is an option. As community-based breeding program is a novel approach, the present
guidelines will have to be refined and updated as experience in community based breeding
accumulates and our tools improve.

35
7. References
Abad, M., J. Arrigo, A. Gibbons, M.R. Lanari, G. Morris and H.R. Taddeo. 2002. Breeding scheme
for Angora goat production in North Patagonia. Communication 12-14. 7th World Congress on
Genetics Applied to Livestock Production, August 19–23, Montpellier, France.
Ayalew, W., and J. Rowlands (eds). 2004. Design, execution and analysis of the livestock breed
survey in Oromiya Regional State, Ethiopia. OADB (Oromiya Agricultural Development Bureau),
Addis Ababa, Ethiopia and ILRI (International Livestock Research Institute), Nairobi, Kenya.
Baker, R.L. 1998. A review of genetic resistance to gastrointestinal nematode parasites in sheep
and goats in the tropics and evidence for resistance in some sheep and goat breeds in sub-humid
coastal Kenya. Animal Genetic Resources Information Bulletin 24: 13–30.
Edea, Z. 2008. Characterization of Bonga and Horro indigenous sheep breeds of smallholders for
designing community based breeding strategies in Ethiopia. MSc thesis, Haramaya University, Ethiopia.
FAO. 1986. Animal genetic resources conservation by management, data banks and training.
Animal Production and Health Paper 44/1. FAO, Rome, Italy.
FAO. 2007. Livestock report. Animal Production and Health Division. FAO. Rome, Italy.
FAO. 2009. The state of food and agriculture 2009: Livestock in the balance. FAO, Rome, Italy.
FAO. 2010. Breeding Strategies for Sustainable Management of Animal Genetic Resources. FAO
Animal Production and Health Guidelines. No 3. Rome.
Getachew, T., A. Haile, M. Tibbo, A.K. Sharma, J. Sölkner and M. Wurzinger. 2010. Herd management
and breeding practices of sheep owners in a mixed crop-livestock and a pastoral system of Ethiopia.
Journal of Agricultural Sciences 5(8): 685–691.
Mueller, J.P. 1991. Transferencia de tecnología a pequeños productores de caprinos en la Argentina.

II Reunión de la Red de Rumiantes Menores. Santiago de Chile, 5 al 8 de noviembre. Comunicación
Técnica INTA Bariloche Nro. PA 184.
Mueller, J.P, E.R. Flores and G. Gutierrez. 2002. Experiences with a large-scale sheep genetic
improvement project in the Peruvian highlands. Communication 25-12. 7th World Congress on
Genetics Applied to Livestock Production, August 19–23, 2002, Montpellier, France.
Sölkner, J., H. Nakimbugwe, A. Valle-Zárate. 1998. Analysis of determinants for success and failure
of village breeding programmes. Proceedings of the 6th World Congress on Genetics Applied to
Livestock Production, Armidale, Australia, 12–16 January 1998.
Wurzinger, M., J. Sölkner and L. Iñiguez. 2011. Important aspects and limitations in considering
community-based breeding programs for low-input smallholder livestock systems. Small Ruminant
Research doi:10.1016/j.smallrumres.2011.03.035.

36
8. Further readings
Duguma, G., Mirkena, T., Haile, A., Iñiguez, L., Okeyo, A. M., Tibbo, M., Rischkowsky, B., Sölkner, J.,
Wurzinger, M. 2010. Participatory approaches to investigate breeding objectives of livestock
keepers. Livestock Research for Rural Development. Volume 22, Article #64. Retrieved April 5, 2010,
from http://www.lrrd.org/lrrd22/4/dugu22064.htm
Duguma, G., Mirkena, T., Haile, A., Okeyo, A.M., Tibbo, M., Rischkowsky, B., Sölkner, J., Wurzinger, M.
2011. Identification of smallholder farmers and pastoralists’ preferences for sheep breeding traits in
Ethiopia: Choice model approach. Animal, doi:10.1017/S1751731111001029.
Edea, Z., Haile, A., Tibbo, M., Sharma, A.K., Sölkner, J., Wurzinger, M. 2009. Morphological
Characterization of Bonga and Horro Indigenous Sheep Breeds under Smallholder conditions in
Ethiopia. Ethiopian Journal of Animal Production, 9(1): 117-133.
Getachew, T., Haile, A., Tibbo, M., Sharma, A.K., Kifle, A., Terefe, E., Wurzinger, M., Sölkner, J. 2009.
Morphological characters and body weight of Menz and Afar sheep within their production system.
Ethiopian Journal of Animal Production, 9(1): 99-115.
Mirkena, T., Duguma, G., Willam, A., Wurzinger, M., Haile, A., Tibbo, M., Okeyo, A.M., Sölkner, J., 2011.
Community-based alternative breeding plans for indigenous sheep breeds in four agro-ecological
zones of Ethiopia, Journal of Animal Breeding and Genetics (2011) 1–10. doi:10.1111/j.1439-
0388.2011.00970.x
Mirkena, T., Duguma, G., Haile, A., Tibbo, M., Okeyo, A.M., Wurzinger, M., Sölkner, J. 2010. Genetics of
adaptation in farm animals: A review, Livestock Sciences.132: 1-12.

Lamb Format
Farmer Lamb Dam Sire Litter Birth Sex Birth Coat Weight Date Weight Date Yearling Date GFW Tail
format
ID ID ID size date weight Color at 3 at 6 weight
months months
Ewe Format
Farmer Ewe ID Name/ Coat Dam Sire Mating Ram Lambing Parity Litter Lamb Lamb Litter PP WWt
Special Color ID ID date ID date size ID ID weight Wt
identifier
Ewe Card (to be kept in each household)

Ewe ID №/Name: Dam’s ID/Name:
Coat color: Sire’s ID/Name:
Birth Date: Owner’s name:
Parity Number
1 2 3 4 5 6 7 8 9

Mating date
Sire’s ID №/Name
Lambing date
Litter size
Lamb’s ID & sex (Lamb 1)
Lamb’s ID & sex (Lamb 2)
Litter weight at birth
Ewe post-partum weight
Litter weight at weaning
37
Ewe weight at weaning of lambs

Annex 1. Community-based small holder sheep breed improvement in Ethiopia – data recording
Number of lambs weaned

ICARDA - TOOLS AND GUIDELINES 1
About ICARDA and the CGIAR
Established in 1977, the International Center for Agricultural Research in the Dry
Areas (ICARDA) is one of 15 centers supported by the CGIAR. ICARDA’s mission
is to contribute to the improvement of livelihoods of the resource-poor in dry
areas by enhancing food security and alleviating poverty through research and
partnerships to achieve sustainable increases in agricultural productivity and
income, while ensuring the efficient and more equitable use and conservation of
natural resources.
ICARDA has a global mandate for the improvement of barley, lentil and
faba bean, and serves the non-tropical dry areas for the improvement of on-
farm water use efficiency, rangeland and small-ruminant production. In the
Central and West Asia and North Africa (CWANA) region, ICARDA contributes
to the improvement of bread and durum wheats, kabuli chickpea, pasture and
forage legumes, and associated farming systems. It also works on improved
land management, diversification of production systems, and value-added crop
and livestock products. Social, economic and policy research is an integral
component of ICARDA’s research to better target poverty and to enhance the
uptake and maximize impact of research outputs.
The Consultative Group on International Agricultural Research (CGIAR) is a

strategic alliance of countries, international and regional organizations, and
private foundations supporting 15 international agricultural Centers that work
with national agricultural research systems and civil society organizations
including the private sector. The alliance mobilizes agricultural science to
reduce poverty, foster human well being, promote agricultural growth and
protect the environment. The CGIAR generates global public goods that are
available to all.
The World Bank, the Food and Agriculture Organization of the United
Nations (FAO), the United Nations Development Programme (UNDP), and the
International Fund for Agricultural Development (IFAD) are cosponsors of the
CGIAR. The World Bank provides the CGIAR with a System Office in Washington, Guidelines for Setting up Community-based
DC. A Science Council, with its Secretariat at FAO in Rome, assists the System in
the development of its research program. Sheep Breeding Programs in Ethiopia
Lessons and experiences for sheep breeding in low-input systems
Aynalem Haile, Maria Wurzinger, Joaquín Mueller, Tadele Mirkena, Gemeda Duguma,
Okeyo Mwai, Johann Sölkner and Barbara Rischkowsky

Guidelines For Sheep Breeding-2

Uploaded by

Copyright:

Available Formats

Guidelines For Sheep Breeding-2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Guidelines For Sheep Breeding-2

Uploaded by

Copyright:

Available Formats

ICARDA - TOOLS AND GUIDELINES 1

About ICARDA and the CGIAR

The Consultative Group on International Agricultural Research (CGIAR) is a

Lessons and experiences for sheep breeding in low-input systems

Aynalem Haile, Maria Wurzinger, Joaquín Mueller, Tadele Mirkena, Gemeda

Edited by: Azage Tegegne

Key words: Community-Based Breeding; Guidelines; Sheep; Ethiopia

International Center for Agricultural Research in the Dry Areas (ICARDA)

4 Creating an Enabling Environment 28

5. Monitoring and Evaluation 31

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Table 1. Characteristics of the four sites 2

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

The objective of these guidelines is to:

The guidelines have five parts:

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

As alternatives, centralized breeding schemes, entirely managed and controlled by governments –

Another alternative widely followed by many developing countries or individuals is importing

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Table 1: Characteristics of the four sites

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

More specifically the objectives of the guidelines are to:

2.2 Target groups

2.3 Conditions under which the guidelines should be used

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

2.4 Structure of the guidelines

Fig. 1 Logical sequence of steps to implement community-based sheep breeding

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

2.5 Using the guidelines

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

3.1.1 Selection of breeds

3.1.2 Selection of community

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Suggested steps to follow for selecting the communities

3.2 Characterization of target sites and breeds

3.2.1 Description of the production system

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Mixed crop livestock production system in Menz, Ethiopia

3.2.2 Breed characterization

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Horro ram, Ethiopia

Molecular genetic characterization

3.3 Definition of breeding objectives

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Different participatory approaches can be used to describe breeding objectives of communities.

3.3.1 Personal interviews

3.3.2 Workshops (focus group discussions)

Farmers’ group in Bonga, Ethiopia, discusses breeding objectives

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

3.3.3 Choice cards experiment

3.3.4 Ranking of live animals

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Horro, Ethiopia. Farmers do ranking of their breeding ewes

Ranking of animals not known to farmers

Box 1. Group ranking of sheep

ICARDA - Guidelines for community-based sheep breeding in Ethiopia

Menz, Ethiopia. Farmers rank sheep not known to them

3.3.5 Comparison of methods

ICARDA - Guidelines for community-based sheep breeding in Ethiopia