Valuation of indigenous farm animal populations and breeds
in comparison with imported exotic breeds with a focus on
sub-Saharan Africa
Henner Simianer (PD Dr)
Applied Genetics Network, Pelargusstr 
3, D-70180 Stuttgart
Tel.: +49-711 6075067
Fax: +49-711 6075066
e-mail: simianer@genetics-network.de
February 2000
Compiled for: Deutsche Gesellschaft für Technische Zusammenarbeit GmbH, Eschborn
Funded through: Project for Promotion of Postgraduate Education in Agriculture in SADC
(Project number 94 2072 001.200)
In co-operation with: BMZ/GTZ Sector project: Managing Agrobiodiversity in Rural Areas
(Project number 98.2020.0)
PD Dr Henner Simianer, applied genetics network, Stuttgart, compiled this study with the
assistance of Mr Alexander K. Kahi (MSc), on leave from Egerton University, Kenya, at
present PhD student at the University of Hohenheim, Germany. The input of Mr Kahi,
especially to sections on poultry, pigs, and unique traits and properties of indigenous breeds
of this report, is gratefully acknowledged. 
Table of contents
List of participants
Executive Summary
Introduction
Valuation of farm animal genetic resources
Definitions
Valuation of biological diversity
Biodiversity, species and breeds
Economic values of farm animal genetic resources
Economic valuation tools
Cost-benefit approaches
Risk-based approaches
Completed and planned research on economic valuation of farm animal genetic
resources
Factors affecting results on the relative economic superiority of breeds or crosses
Unit of comparison
Limiting factors
Genotype × management level interaction
Data editing
Performance comparisons of indigenous and exotic breeds and their crosses
Cattle
Dairy traits
Beef traits
Small ruminants
Sheep
Goats
Pigs
Poultry
Chickens
Guinea fowls
Unique traits and properties of indigenous breeds
Disease resistance or tolerance
Trypanosomosis
Gastro-intestinal parasites
Ectoparasites
Bursal disease
Heat tolerance
Cattle, sheep and goats
Poultry
Camel
Adaptation and nutrition
Other properties
Salt tolerance
Scavenging ability
Reproductive performance
Final remarks
Valuation of farm animal genetic resources--prospects and possibilities
Crossbreeding with exotic breeds as a threat to farm animal biodiversity
Naive crossbreeding schemes
Structured crossbreeding programmes
Developments of synthetic breeds
Decision making
Conclusions
References
Appendix: AnGR valuation method evaluation (Drucker et al. 2001)
List of abbreviations
 List of abbreviations
AnGR (Farm) animal genetic resources
BMZ Bundesministerium für Wirtschaftliche Zusammenarbeit
CABI Commonwealth Agricultural Bureaux International
CBD Convention on biological diversity
Fi i-th crossbred generation
FAO Food and Agriculture Organization of the United Nations
FARM-Africa Food and Agricultural Research Management-Africa
G × E Genotype by environment (interaction)
GTZ Deutsche Gesellschaft für Technische Zusammenarbeit
ILCA International Livestock Centre for Africa
ILRI International Livestock Research Institute
IPR Intellectual property rights
KARI Kenya Agricultural Research Institute
LW Large White breed
Ne effective population size
SACCAR Southern African Centre for Co-operation in Agricultural Research
and Training
SMS Safe minimum standard
SSA Sub-Saharan Africa
WTP Willingness to pay
Executive summary
Executive summary
This study reviews the different concepts and approaches regarding economic valuation of
farm animal genetic resources (AnGR) in the context of programmes for AnGR management
to be implemented in sub-Saharan Africa (SSA). The different types of values and methods to
estimate them are presented and critically discussed.
Decisions related to AnGR management are not exclusively or even primarily of economic in
nature, but other aspects, like socio-economic, livelihood or development implications are
relevant. Nevertheless, all decisions and activities have economic implications, and it is this
aspect the study focuses upon.
The relative value of local breeds is observed especially in crossbreeding programmes with
exotic, non-adapted breeds, which historically often have led to non-sustainable
improvements and complete market failures. The respective studies are reviewed and
discussed for the main farm animal species.
Values of indigenous breeds are mainly due to unique traits and characteristics (like adaptive
capacity, disease tolerance etc.), which are reviewed for the main farm animal species in SSA.
Only few (actually two) studies have valued AnGR and few more are underway in developing
countries, yet the need for such studies is considered and activities are prioritised. We argued
that certain values, which are difficult to assess, have little relevance for conservation
decisions.
Therefore, and due to the relative lack of knowledge and of experience with respect to
valuation methods, we do not recommended to conduct empirical valuation studies using
some of the described methods.
Alternative approaches have been suggested that focus more on risk of loss and cost of
AnGR management, rather than on potential returns. These concepts are extensively
discussed in the present study, since they may be a valid and pragmatic alternative to breed
valuation studies.
Risk-based approaches, like the safe minimum standard approach or the methods Smith
(1985); Weitzman (1992) and Weitzman (1993) suggested require the assessment of risk
functions and the cost-efficiency of AnGR management, i.e. the risk of extinction has to be
quantified for each breed and the return (in terms of risk reduction) of different management
activities has to be assessed. These data are generally lacking for developing countries.
A further feature to be taken into consideration is the resilience of AnGR management.
Respective activities should be designed in a way that even the total failure of some of the
components will not lead to a total failure of the whole project, i.e. to the loss of a breed.
Taking into account the high requirements of time, labour, money, and logistic input to conduct
a sensible set of AnGR valuation studies, we argued that all related activities should be co-
ordinated. Focusing on SSA, International Livestock Research Institute (ILRI) has taken the
lead to initiate a set of studies based on different valuation approaches and targeting the
different species in different regions.
If Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) intends to start major
activities in this field, projects should be designed to fit in the overall incomplete matrix of
methods, regions and species to be built. We argued that such activities should be planned
and carried out as co-ordinated effort in close linkage with other organisations, which are
already active in this field. In SSA, ILRI clearly is the organisation with the highest competence
and experience in the area of economic valuation of AnGR. Since ILRI activities are mainly
based on contingent valuation methods, GTZ studies focusing on risk-based approaches
would be a useful complementary effort.
Since results of valuation studies are virtually non-existent and consequences of different
activities for AnGR conservation are difficult to assess, decision makers could not be able to
make specific recommendations. We argued that decision-making with respect to AnGR
management activities should not be postponed until sound evidence is available. This could
initially be made based on inadequate information by using, e.g. panels of experts to assess
the relative values (or conservation priorities) of different breeds.
Concerning future activities, the report recommends that highest priority be given to a region-
and species-wide framework for decision making and priority setting, which combines
information on values, conservation costs, diversity patterns and risk of extinction. Such a
framework includes creation of public awareness, training and implementation of economic
balancing mechanisms between stakeholders.
Introduction
Terms of reference
Introduction
There are numerous national, regional and global initiatives aiming at the management and
conservation of farm animal genetic resources (AnGR). Respective activities are within the
scope of the Southern African Centre for Co-operation in Agricultural Research and Training
(SACCAR), which has, together with the Bundesministerium für Wirtschaftliche
Zusammenarbeit/Deutsche Gesellschaft für Technische Zusammenarbeit (BMZ/GTZ) project
Managing Agrobiodiversity in Rural Areas, commissioned this study. The latter project aims to
contribute to a sustainable use of agrobiodiversity for food security and rural development,
and to help reduce agrobiodiversity losses.
Terms of reference
Aims:
Documentation of the actual state of economic valuation of indigenous farm animal
populations
Development of a model for relevant options of activities, which can be implemented in
developing countries (with examples).
Working plan:
cf. 1:
Review of existing data and current research projects or proposals concerning animal
genetic resources and economic valuation of indigenous farm animal populations.
Evaluation of competitiveness of indigenous vs. imported exotic breeds and crosses.
Compilation of unique qualitative characteristics of indigenous populations and breeds
(disease tolerance, adaptation etc.) that may determine the value of these breeds.
Review of existing approaches for the economic valuation of biodiversity and their
relevance for farm animal biodiversity and agrobiodiversity
cf 2:
Economic values for some farm animal populations have been determined in case
studies and consequences for the management of these resources are discussed, taking
the risk into account.
Sectoral and global strategies are discussed, taking into account different scenarios
(market failure, loss of unique characteristics).
Based on the framework suggested by Weitzman 1992; Weitzman 1993 a methodological
approach should be developed to assess the consequences of different options for farm
animal biodiversity.
In a world of limited financial resources, it is necessary to develop rules for decision making
based on economic arguments, i.e. compare the relation between investments (e.g. for setting
up a conservation scheme) and returns (e.g. through the exploitation of the conserved gene
pool for some purpose in the future). For objective decisions, it is necessary that both the
costs and the returns be quantified.
Both due to the considerable time scale (conservation thinks in decades or even centuries,
rather than years) and to the vast uncertainty of future changes and events, it is inherently
difficult to derive the respective values. Therefore risk-based analyses have been suggested
for the purpose of decision-making, which may avoid some of the problems associated with
traditional cost-benefit analysis approaches. The different aspects of valuation are described
in the first part of the study.
The value of indigenous populations under tropical conditions is often most clearly expressed
in comparisons with exotic breeds of European origin, both in pure breed comparisons or
crosses. On the other side, crossbreeding is one of the major threats of the exotic gene pool,
in that uncontrolled mixtures of indigenous and exotic genes displaces the exotic genotypes.
In the second part of this report, literature on crossbreeding and breed comparisons in the
main farm animal species under tropical condition is reviewed and interpreted from the point of
view of genetic conservation.
In many cases, indigenous breeds are highly adapted to the conditions they are exposed to,
including climate, feed supply, disease challenge etc. The respective traits are in many cases
unique and constitute a qualitative difference compared to exotic breeds or crosses, which
hardly cope with the respective conditions. Such problems have in many cases led to failures
of upgrading programmes, which aimed at improving productivity, but unintended led to a
reduction of the adaptive capacity of the generated populations. The third part of this study
reviews the main types of unique properties of indigenous breeds across all major farm animal
species.
The information presented and summarised in this study is intended to serve as a basis for the
development of options for effective and efficient activities in AnGR conservation. Knowledge
is far from being satisfactory or complete in different areas.
Different farm animal species are studied with different intensity: in SSA, by far most
information is available on cattle, while information on other important species like pigs,
poultry and small ruminants is relatively scarce.
Different aspects of methodology are applied with different emphasis: as Ruane (1999)
stated molecular distancing seems to be the main activity in AnGR research, while
phenotypic description or economic valuation of breeds is done with much less
emphasis.
While there is a lot of theory about the methodological approaches that can be used to
valuate genetic resources, concrete studies on the economic valuation of AnGR are
scarce. However, activities in this direction are considered and some pilot projects are
started, the state of the discussion and the present and planned activities are discussed
in this report.
AnGR characterisation is mostly descriptive, while little or no conceptual work is done on
how to decide which populations should be maintained, how limited resources should be
allocated, and what measures are most effective to reach the often-not-well-defined goal
of biodiversity conservation.
Methodology development is mainly done for the conditions of the northern world, in
that, e.g. conservation concepts are based on the conservation of breeds in the
European sense. It must be doubted, whether these concepts can or should be used
under the conditions of developing countries, where, e.g. breeds in this sense are often
not implemented. It certainly is necessary to develop AnGR conservation strategies that
are based on the actual conditions in developing countries.
These aspects will be discussed in more detail in the conclusions of this study, taking into
account the available information as summarised in the following chapters.
Valuation of farm animal genetic resources
Valuation of farm animal genetic resources
Definitions
Valuation of biological diversity
Biodiversity, species and breeds
Economic values of farm animal genetic resources
Economic valuation tools
Cost-benefit approaches
Risk-based approaches
Completed and planned research on economic valuation of farm animal genetic
resources
Factors affecting results on the relative economic superiority of breeds or crosses
Unit of comparison
Limiting factors
Genotype × management level interaction
Data editing
Valuation of farm animal genetic resources
Definitions
Biodiversity can be defined as the diversity of biological units. A biological unit may be an
individual or a group of individuals. In the very general sense, diversity1 may focus on very
different attributes, like phenotypes (coat colour, body size), production traits (milk yield,
growth), behaviour, nutritional aspects etc.
Diversity is an attribute of a group. While two units (individuals, breeds etc.) may be different,
a whole group (>2 units) is diverse.
Genetics have revealed that the genetic code, as deposited in the nucleic and mitochondrial
DNA, is the major criterion for the individuality, in other words, the genetic code is the blueprint
for an individual, and differences between the DNA sequences of two individuals will to a
certain extent lead to differences in the criteria mentioned above.2 Therefore, biodiversity
studies prefer to work with genetic data (sequence data, allele frequencies etc.).
Biodiversity in non-agricultural applications mostly operates on species, habitats, or
ecosystems, but usually not on subgroups within species, like breeds, strains, or lines.
Species is a taxonomic term and defines the largest group of individuals that can mate among
each other with fertile offspring. Farm animal species are, e.g. horse, goat or chicken, but
horse and ass are not the same species, since they can mate, but the offspring usually is not
fertile. So, biologists focus their research on between species biodiversity (e.g. between
different species of butterflies etc.).
Agrobiodiversity, both in the context of farm animals and cultivated plants, usually focuses on
within species biodiversity. Farm animal species are subdivided into breeds, which may be
further subdivided in strains, lines, local populations etc.
Breeds are not at all well defined, and historically, farm animal breeds are a European or at
least northern concept in animal breeding, as they have been constituted through breed
societies in the early 19th century, first in Great Britain, later in other European countries and
in North America (Willham 1987). Lloyd-Jones (1915) already made the point that `... a breed
is a group of domestic animals, termed such by common consent of the breeders. There as
such is no sharp and objective definition of a breed, but it is a subjective classification.'
Köhler-Rollefson (1997) suggested the following criteria for the definition of a breed, which
may be applicable not only in the northern hemisphere with its traditional breed concept, but
also on a global scale. A group of animals is considered as a breed, if it
is subject to a common utilisation pattern
shares a common habitat/distribution area
represents largely a closed gene pool
is regarded as `distinct' by their breeders.
The general idea behind within species biodiversity can be demonstrated with the following
example: suppose, a species is made up of three breeds, A, B, and C. The whole gene pool of
the breed is the sum of all the alleles present in either A, B, or C. Typically, some of the alleles
are present in all three breeds, others are represented in a subset of breeds (e.g. only in
breed A and B), while some alleles are typical for a single breed and do not exist in the others.
This unique set of alleles of, e.g. breed B would be lost if breed B was extinct. As a
consequence, the between breed diversity would be reduced by this proportion.
Finally, agrobiodiversity tends to use the term `resource' (e.g. AnGR). This is a somewhat
unsharp terminology, which reflects the unavailability of a clear definition of breeds, strains,
and other units used to substructure species. In an internal proposal, ILRI defines animal
genetic resources as `those animal species and populations with them that are used, or may
be used, for the production of food and agriculture. These populations within each species can
be classified as wild and feral populations, landraces, and primary populations, standardised
breeds or strains, selected lines, and any conserved genetic material.'
 Valuation of biological diversity
Conservation of biological diversity, of which agrobiodiversity and even more specifically farm
animal biodiversity is an important component, has been formally provided significant global
importance and interest in the Convention on Biological Diversity (CBD), signed by 157
countries at the Rio Earth Summit 1992, and put into force in 1993. Among others, the
importance of agrobiodiversity for food security, poverty alleviation, and improvement of
livelihood are stressed. Valuation of specific segments of agrobiodiversity-in the case of AnGR
the most common, though problematic unit, a breed-is an essential component of the whole
complex of description of diversity, which is a prerequisite for any rational conservation
strategy.
However, purely economic considerations are not the only, probably also not the most
important aspect in this context. Other aspects, like the contribution of a breed to food
security, its ecological role, the insurance function of genetic diversity with respect to eventual
future changes etc. may be seen as complementary arguments favouring conservation
programmes. Environmental economists however have tried to set up a complete framework
of different types of values, claiming that in principle all these values of, e.g. a breed can be
quantified in economic terms, so that the whole problem can be discussed in a quantitative
economic context.
There is some dispute about this approach in the scientific community, and even those in
favour of the `economic' approach are forced to admit that not all values can be quantified with
sufficient accuracy (Figure 1) to be useful for decision making.
Source: Adapted from Virchow 1999.
Figure 1. Economic values of genetic resources.
Valuing AnGR was preceded by major attempts to value plant genetic resources (PGR),
especially crops, so that some methodological issues may be adopted from this field. PGR
conservation has certain advantages compared to AnGR conservation, mainly the much less
complex biology of plants, their high fecundity, the availability of wild ancestors, the ease and
relative inexpensiveness of maintaining long-term seed collections etc.
Somewhat simplifying, plant breeding (at least in some of the major crop species) can be seen
as sort of experimental combinatorics, i.e. germplasm of different sources is combined and the
properties of the hybrids are tested to select the best. Transgenic technology has added
efficient tools to introgress single genes across species boundaries. These principal
approaches should allow a relatively straightforward valuation of an additional source of
germplasm. Despite this claimed relative ease, the economic quantification of the benefits of
germplasm to crop development has been described to be still extremely difficult (Evenson et
al. 1998). There are some examples of post festum valuation of single genes transferred
(through traditional non-transgenic breeding techniques) from exotic resources to cultivated
strains. A resistance gene against the yellow dwarf virus extracted from a single Ethiopian
barley plant, for instance, has been found to generate a value of US$ 160 million after being
transferred to cultivated barley strains (Munasinghe 1992). In the context of plant
agrobiodiversity, Swanson (1998) stated that the fundamental nature of the public interest in
genetic conservation is supplying those values that the private sector has little or no incentive
to pursue. Thus, it is mainly the option value (with a strong emphasis on potential pest
resistance) that is seen as the major motivation for plant agrobiodiversity conservation efforts.
Another major field of valuation of genetic resources focuses on ecosystems, habitats and
wildlife. In this context, some studies have focused on the valuation of large mammals (e.g.
elephants) being a major component of safari tourism industry. The usual approach is to use
the `willingness to pay', which empirically can be derived, e.g. from travelling costs, people are
willing to pay to see elephants. Another approach uses survey based contingent valuation
methods, where people are offered, e.g. safaris to parks with or without elephants to derive
the marginal added value of elephants. Both types of studies have been successfully
conducted (Brown and Henry 1989), and correspondingly resulted in a value/elephant per
tourist-contact of US$ 100, leading to an overall economic contribution of elephants to the
Kenyan economy of about US$ 25 million/year. Similar studies have been conducted to value
entire habitats (Tobias and Mendelsohn 1991).
Such approaches especially if they are based on survey techniques offering hypothetical
choices, may suffer from biases emanating from discrepancies between claimed and realised
willingness to pay. This is especially so, if the good to be valued is of high moral or aesthetical
value or if it is thought to be politically correct to choose this good.3
Therefore, methodological approaches are more reliable that are based on empirical data on
realised prices from market surveys, rather than on subjective `willingness to pay' data.
Another lesson to learn from environmental economics aiming at wildlife and habitat valuation
is that species should always be seen and valued as an element of a natural system, and part
of the value of a species is derived from the interaction with the whole habitat. An example for
such a systematic interaction is the Ronalds day sheep breed with its high copper absorption
capacity and salt tolerance that has the ability to feed almost exclusively on seaweed and is
an essential factor to maintain the seaweed habitat (Ponzoni 1997). In this case, the value of
the breed should reflect the value of the whole habitat, which likely would not continue to exist
if the breed was extinct.
Similar interactions may be found between indigenous breeds and, e.g. forage plants that are
more or less exclusively selected and fed by these animals, so that the feed usage maintains
this plant which otherwise would disappear. Of course, there are also close links of indigenous
breeds with human cultures, especially in pastoralist societies, so that a holistic approach has
to be used to assess the value of a breed in such a system.
 Biodiversity, species and breeds
In farm animal biodiversity studies, the general and widely accepted approach is to focus on
breeds as the unit to preserve biodiversity. Under tropical conditions and especially in SSA,
we must be aware that the breed concept is not established to a comparable extent for all farm
animal species. While for Africa there are 120 cattle breeds on file (Simon 1999), the
corresponding number of swine breeds in Africa is only 13 (Hammond and Leitch 1998). Does
this mean that there is less genetic diversity in swine than in cattle? Most likely not, it may
rather be seen as an indication that even in the non-Islamic African countries the breed
concept is less adopted for swine than for cattle.
About 50% of genetic diversity of farm animals is between breeds, while the other half of the
diversity is within breeds of a respective species (Oldenbroek 1999). This, of course, relies on
the assumption that species are stratified into breeds, which not necessarily is the case in less
developed regions. As a consequence, a considerably larger share of the within species
genetic diversity may be allocated within rather than between the existing breeds, if
stratification into breeds is incomplete.
Given a sufficient stratification into breeds, the two principal approaches to conservation of
AnGR are to preserve breeds and to preserve diversity within breeds. This includes
determining which breeds are different enough, so that their maintenance conserves a
significant part of the between breed variation. Eding and Laval (1998) established and
summarised the methodology to assess phenotypic and genetic diversity. Ruane (1998)
suggested criteria for the selection of breeds for conservation. In a critical review, Ruane
(1999) argued that biodiversity research focuses too much on distancing, i.e. on evaluating
genetic distances between breeds, while too little work is done in the phenotypic description of
breeds (including productive and adaptive performance).
If breeds to be maintained are below a critical effective population size, measures have to be
taken to prevent the loss of this breed or the uncontrolled incline of inbreeding in the breed.
Meuwissen (1998) and Toro and Mäki-Tanila (1998) discussed there are multiple options for
this task, ranging from cryoconservation to in vivo, in situ schemes.
There is, however, a lack of theory and rules for decision making, which breeds should be
maintained and when respective activities should start. Using an approach more targeted to
species conservation rather than to breed conservation, Weitzman (1993) presented a formal
decision-making framework based on a unique concept of diversity (Weitzman 1992). Both
Solow et al. (1993) and Thaon d'Arnoldi et al. (1998) discussed a decision-making framework
based on the farm animal context. It must be seen that preservation of biodiversity is different
from preservation of individual breeds or populations. While each population may have a value
as such, the contributions to biodiversity of different units within a species or a taxonomic
class may differ. In a world of limited resources, it has to be decided how to allocate the
available resources so that the defined goal is achieved with highest probability. As was
shown by Weitzman (1993) it may not be the optimum strategy to try and preserve all
populations, but it may in some cases be more rational to concentrate resources on the
preservation of the major contributors to biodiversity.
To adopt Weitzman's (1992) concept to farm animal conservation schemes, it would be
necessary to quantify exactly a number of `soft' parameters, e.g. what is the probability p that
a certain breed still exists in, e.g. 50 years, and what are the costs of increasing this
probability from p to p + _?
It should be noted that time horizons for genetic conservation approaches like, e.g. 50 years
(as chosen by Weitzman 1992) correspond to quite different equivalents in generation intervals
in farm animals, which range from less than one year in small but important species like
poultry, rabbits etc. to up to six years in cattle. If, as is often done, the increase in inbreeding
per year is seen as the main criterion for genetic erosion within breed, larger effective
population sizes have to be maintained for farm animal species with shorter generation
intervals.
The biodiversity value of farm animal breeds
The conventional approach to valuation of biodiversity conceptually uses the same principle to
that used to decide, whether a company should continue to produce and sell a certain
product. That is, actual and future costs and benefits are assessed, values for the future are
discounted to corresponding prices of today, and then the decision is made. The major
difference arising from valuating AnGR is that different types of values (Figure 1) have to be
considered that it is difficult to attach monetary prices to these values, and that most of them
will be realised in the (distant) future and with considerable uncertainty. One of the
consequences of discounting is that high economic returns in the distant future have to
compete with relatively low (but guaranteed) returns in the present or the foreseeable future.
The straightforward application of the respective economic theory may be misleading. Clark
(1973) showed that it was economically more profitable, to allow an extinction of blue whales
(by selling `whaling licenses' to Japanese companies) and to invest the short-term profit into
industries with high economic returns, like the computer industry, than to invest in whale
conservation with some expected return in the distant future. While this analysis was
principally correct, the result will not be acceptable for most people, not just environmentalists.
For plants and wildlife, Ehrenfeld (1981); Ehrenfeld (1986) pointed out that certainly not all,
possibly even not the majority of all existing species have a real value, neither at present nor
ever in the future. Ehrenfeld (1981); Ehrenfeld (1986) called species without actual or future
value non-resources and suggests that, e.g. many of the 600 thousand existing and described
species of bugs will belong to this group. The problem, however, is that it is strictly impossible
to know, which of the existing variety of species will have a value at some point in the future
and which ones will be non-resources. Therefore it is not only necessary to discount future
values to today's values, but also to weigh them with the probability that any of the set of
species to be conserved will turn out to be a valuable resource.
It must, however, be seen that some species potentially have an extremely high value. For the
relatively recently discovered Mexican wild grass Zea diploperennis, which is closely related to
corn, but is frost resistant, Fisher and Hahnemann (1984) have estimated (under certain
assumptions) a potential value of US$ 6.82 billion. Similar returns might be expected from
pharmaceuticals derived from natural contents of wild plants.
It certainly is difficult to adopt these concepts directly for the situation in farm animal
conservation, but some of the arguments deserve further consideration in this context.
One major point is that the value is not only attached to any of the breeds in itself, but that
there is an additional value, which results from the contribution of the breed to the diversity of
the species. Weitzman (1993) gave an illustrative example based on crane conservation that
may be translated to the farm animal situation as follows: consider a farm animal species to be
a set of N sufficiently distinct breeds. The between breed diversity is to a large part due to two
breeds, A and B that are quite similar to each other, but different from all other breeds (the
respective phylogenic tree is depicted in Figure 2.
a) Complete set b) A missing c) B missing d) A and B missing
Let DABR be the diversity for the set of breeds A, B, and R standing for all breeds C to G, and
let DÐ BR be the diversity for only B and R etc.
Figure 2. Assumed phylogenic tree for seven breeds, and corresponding Weitzman (1992)
diversity values Dx for the complete set (a) or for reduced sets (b-d).
If, say, breed A is lost for some reason, the reduction of the diversity will be minor, because
the genetically very similar breed B is still there. Thus, the relative contribution of A to the
diversity of the species is minor, and so is the contribution of B under the assumption that A is
still there. If, however, both A and B are lost, there may be a significant reduction of the
diversity. This is shown in Figures 2a to 2d, where, as suggested by Weitzman (1992), a
diversity measure Dx is used that is equal to the total height of the maximum likelihood
phylogenic tree for the set of breeds x.
Let us, for example, assume that DABR = 1.00, DÐ BR = 0.98, DA Ð R = 0.98 and DÐ Ð R = 0.80,
i.e. diversity is almost complete, if either A or B are present, but is severely reduced, if both of
them are lost.
Then the relative value of the contribution of A to the whole diversity is (DABR - DÐ BR)/DABR =
0.02 if B is conserved and is (DA Ð R - DÐ Ð R)/DA Ð R = 0.184 if B is extinct. This means that
the diversity value of breed A varies-based on what happens to the other breeds (in this
example to breed B) in the species. The diversity value of A is almost ten-fold if B is extinct,
compared to the situation when B is conserved.
As a consequence, it is conceptually impossible, to assign values to individual breeds that
reflect their relative impact on within species diversity without considering the genetic structure
of all the other breeds in the species at the same time. This, after all, may be one important
argument justifying the considerable number of genetic distancing studies for farm animal
breeds that have been undertaken in recent years, since reliable distance matrices are an
essential prerequisite for assessing a breed's contribution to within species diversity.
In the present case, a conservation scheme should be aiming at the preservation of at least A
or B. In what Weitzman (1993) called the `crowned-crane paradox' he demonstrated that it
may be a more sensible decision to sacrifice one of the two (actually the more endangered
one) and to spend all the available resources to ensure the conservation of the less
endangered one, rather than to split the money to conserve both with an increased risk of
losing both of them.
Although the conceptual framework is very clear, applications of Weitzman's (1993) concept
are not trivial since a considerable number of parameters have to be specified, like the
probability for any one breed to be extinct at the end of the assumed planning horizon (e.g. 50
years from now), the cost of increasing this probability by a certain increment etc. Weitzman's
(1992, 1993) theory is also tailored for species rather than for breeds, thus it totally neglects
the estimated 50% of the within species diversity that are allocated within breeds, nor does it
consider possibilities like genetic exchange between breeds, merging breeds etc. which are
not relevant if species conservation is considered.
 Economic values of farm animal genetic resources
While actual breeding programmes operate on costs and benefits related to production under
market conditions valid today or in the (relatively) near future (Weller 1994), AnGR economics
have a longer time horizon and are not restricted to production values only.
Gandini and Oldenbroek (1999) gave the following list of possible objectives for AnGR
conservation, which may be considered as a list of actual or future values:
opportunities to meet future market demands
insurance against future changes in production circumstances
socio-economic value
cultural and historic value
ecological value
opportunities for research.
Virchow (1999) brought these components of value in a system of use- and non-use values
with a suggested decline in quantifiability (Figure 1).
Valuation obviously is easier and more reliable, if it refers to actual markets and prices. This
implies that a higher degree of uncertainty is involved if valuation aims at future market
situations. Valuation becomes even more speculative, if future changes in production
circumstances, like environmental changes, changed consumers' wishes or newly arising
disease challenges are considered. It must be questioned, whether there is a possibility to
derive values for these items that are anything but speculative.
A major set of methodologies used for AnGR valuation is based on cost-benefit analyses. In
this context, asymmetric incurrence of costs and benefits is an important issue, i.e. which
group of stakeholders will have to pay for AnGR management, and which group will have the
benefit.
Costs are associated with all in vitro and in vivo, ex situ conservation approaches, but also in
vivo, in situ approaches will have explicit and/or implicit costs. In the context of developing
countries, a cost-related decision affecting conservation matters may not be to replace a local
breed through a (presumably) more productive exotic imported breed. The relative inferiority in
productive efficiency are costs that are effective now and accrue to the farmer, while the
benefit of not loosing the genes of the indigenous population are effective in the long-term and
accrue to the `society', or even the `world as a whole' (Drucker et al. 2001).
This aspect becomes even more relevant, if property issues are considered, with the key
question: who owns AnGR. This question is intensively discussed in the field of plant genetic
resources (Zilberman et al. 1998). According to Drucker et al. 2001, the two general market
mechanisms in this field are intellectual property rights (IPR) and exploration and extraction
contracts. An essential component in both approaches must be an adequate refunding for the
initial provision and maintenance of AnGR to the individuals, societies, or states that are
providing the actual `raw material', i.e. indigenous breeds with their respective adaptive
capacity. Perrings et al. (1992) stated that establishing IPRs for genetic resources could be
seen as a means for generating markets, which may lead to incentive effects beneficial for
AnGR management.
This asymmetry of cost and benefit allocation with respect to time scale (when do they
become effective) and target groups (who pays and/or who takes the benefit) must be seen as
one major driving force of AnGR loss. Other important driving forces on the macro level are
habitat destruction, human population growth, and decision making processes including policy
in general and impact of international markets, subsidies and developmental aid (Perrings et
al. 1992).
 Economic valuation tools
 Cost-benefit approaches
Mendelsohn (1999) summarised the set of methodological tools that might be suited to
measure values of AnGR; most of them have not yet been used for this purpose, though. This
set of methods contains the following approaches.4
The hedonic method is designed to value the observed traits of animals. This approach
is based on empirical data on market sales and the characteristics of the sold animals or
products. From these empirical data, values of new or alternative products or animals of
breeds with specific characteristics can be assessed. Jabbar et al. (1998) reported
applications of this approach for cattle breeds in West and central Africa and Jabbar
(1998) reported the same applications for sheep and goats in southern Nigeria.
Aggregate market analyses are based on information concerning aggregate quantities
both on the input and output side of a system following a top-down approach. This
information can be derived from empirical time series and/or panel data. The approach
takes into account the multiple interdependencies of a production system, so that, in
principle, it should be useful to evaluate, e.g. the consequences of a genetic
improvement of a breed, taking into account all the corresponding changes like
increased food demand, higher management requirements etc. It also reveals to whom
the benefits of a potential genetic improvement will accrue.
Cross-sectional household studies are based on aggregate demand functions derived
from empirical data on consumer's behaviour in different markets. While the aggregate
market analysis is based on empirical data from markets differing over time, cross-
sectional household studies are based on markets differing geographically. These data
may be easier to sample in studies focusing on developing countries. Together with
supply functions, derived, e.g. with cross-sectional farm studies, this approach may
allow welfare analyses on alternative scenarios.
As mentioned before, the cross-sectional farm study allows derivation of supply
functions from empirical data on farmer's decisions on price differences in different
geographical locations.
Contingent valuation methods are based on surveys. Consumers are offered a set of
(real or theoretical) products with specific characteristics and are asked for the price
they would be willing to pay (in currency or equivalent goods, e.g. sheep or cattle), to
rank them or to make pair-wise choices. This approach is especially suited to assess
non-use values (e.g. cultural values) or any hypothetic characteristic that is not yet
available.
Further useful approaches can be to do an economic input-output analysis of breeding
programmes, farms or herds involved in animal production. Both are data-intensive and rely
on a good quantitative knowledge of the genetic and biological processes underlying the
respective systems. It must, however, be seen that examples for economic evaluation of
actual breeding programmes are rare even for developed countries (Weller 1994) so that it will
not be trivial to assess realistic input and output values for the much more volatile biological
processes in developing countries.
It must be stressed that the valuation tools that have been briefly introduced are mainly
designed to assess direct use values. Most of the other value categories can hardly, if at all,
be quantified through these approaches. This is more true the more abstract a characteristic to
be valued becomes, the more distant in the future it might arise, and the more it depends on
external changes.
Some authors stress that there are principal limitations to what can be valued. Norton (1992)
puts serious doubts on the question, whether a value for the `insurance' function of a
conserved species will ever be anything but speculative as long as it is not clear, which
characteristic of the species will be of value under which specific circumstances.
Given these limitations, it should be clear that valuation efforts have to be undertaken to gain
better insight on the economics of AnGR management. Nevertheless, conservation decisions
will always have to be made under considerable uncertainty. Therefore, decision-making
processes have to be designed to be robust towards limited or wrong perception of the
economic background and consequences of the alternative options to chose from.
The cost-benefit analysis tries to assign definitive numeric values to all the components of
costs and benefits, and to probabilities for uncertain future events etc. Especially in the
context of biodiversity management, the cost-benefit analysis has been criticised to be
extremely speculative (Randall 1992) since many of the required quantifications will in most
cases suffer extreme uncertainty.
To give an example, option values of breeds are based on potential values of these breeds
under unforeseeable changes in the future with respect to environmental changes, new
diseases, changes in the markets etc. If, for example, a new infectious disease appears, some
breeds may be genetically resistant or tolerant against this disease, which gives them an
immanent value. However, not all (most likely only a minority) of breeds can be expected to
realise their option value, but since we do not know the future conditions, we have to maintain,
say, ten breeds to exploit the option value of one of them, not being able to identify which one.
 Risk-based approaches
As an alternative, ecological economists have suggested the concept of `safe minimum
standards' (SMS) (Ciriacy-Wantrup 1952), which starts from the assumption that in a certain
case, there is biodiversity which is valuable enough not to be acceptable (for the society) and
that the respective diversity would be lost. The problem then is to find (and quantify) the costs
for the management of this diversity. Of course, it is still necessary to assign values to the
required activities, and the objective is to develop a scenario with minimum costs that
guarantees the maintenance of the diversity with sufficient probability. This scenario will then
be adopted, unless the social costs are unacceptably large (Bishop 1978).
Of course, there are some obvious and inherent problems with the SMS concept: e.g. when
are social costs `unacceptably' high? And can maintenance of diversity ever be guaranteed by
any plan of action? Crowards (1998) pointed out that following the SMS concept does not
imply that a quantification of actual or future returns unnecessary.
However, the SMS approach has some advantages over the cost-benefit approach in that it
poses the null hypothesis on the assumption that diversity (e.g. a set of breeds) should be
maintained and this hypothesis only will be rejected, if the society decides that it is not willing
to pay the minimum costs required to achieve this goal. SMS therefore is much closer to a
decision making tool than cost-benefit analysis, which is sort of vague in its objectives and
priorities.
The prejudiced approach of SMS, which is a priori in favour of conservation, is justified through
the irreversibility of the loss, if diversity is not maintained (Bishop 1979).
SMS has been used empirically for some wildlife species, e.g. the preservation of the
Californian condor or the Rwanda mountain gorillas, leading to the conclusions that required
costs were rather limited for the preservations activities (Bishop 1980).
The concept to reduce the risk in the future by maintaining genetic diversity has been
discussed by Smith (1985) for the context of farm animal breeding programmes. He
introduced an extra discount rate for the remaining uncertainty, which is a function of the
number of preserved selection stocks. While this discount rate drops with an increasing
number of maintained stocks, the steepness and the shape of the declining curve may vary
depending on the considered situation.
For the European situation, Smith (1985) concluded that maintaining a sufficient number of
reserve stocks is not only economically justified, but will be economically beneficial under a
wide range of assumptions, as demonstrated in a sensitivity study.
A generalisation of this approach to the AnGR management problem seems possible even
though the approach should be based on a genetic valuation rather than on an economic
valuation as in the original paper.
This means the `currency' should be `useful genes' under actual or future environmental
conditions. Let us assume that a given population has a set of Na polymorphic genes that are
crucial for fitness and production under the actual environmental and management conditions,
but that there are also Nf genes that are polymorphic and may be of some value in the future.
Since future production conditions are unknown, it is uncertain which of the Na + Nf genes will
be of any value. The objective must therefore be to preserve as large a share of the genetic
diversity of these genes as possible under the given budget restrictions.
Conceptually, it is possible to define samples of the population, like groups of, say, 100
females with the corresponding males at dispersed geographic spots, which are monitored on
a very extensive level. Analogous to the approach of Smith (1985) it then would be necessary
to compute the reduction of the risk of losing genetic variation if additional samples are added
to the existing ones.
To give a simplistic example: if we assume that at a given locus the rarest allele has an allele
frequency of 1%, we expect from quantitative genetic theory (Falconer 1984) that with an
effective population size of Ne = 50 the probability that this allele is lost in one generation due
to genetic drift is 36.6%. The probability that the respective allele is still present after five
generations is only 10.2%, that means the vast majority of rare alleles will be lost due to
random genetic drift within a few generations.
Figure 3 shows how this probability of loss is reduced if more than one sample of size Ne = 50
are maintained. However, the relative benefit of maintaining additional populations is variable
depending on the frequency of rare alleles to be conserved. If the focus is on conservation of a
very rare allele with p = 0.005, the benefit of an additional sample is almost linear, while for
more frequent alleles (e.g. p = 0.02), there are diminishing returns. While the second sample
increases the probability of conservation by almost 50% (49.1-74.1%), adding the fifth sample
to already maintained four samples increases the conservation probability only from 93.3-
96.6%, i.e. by 3.3%.
.Figure 3. Probability that an allele with initial frequency p is conserved over five subsequent
generations, as a function of the number of samples with Ne = 50 in each sample.
Considering the design of AnGR management programmes in developing countries, high
weight should be given to the robustness of the implemented approach, i.e. the prevention of a
total failure of the programme if the underlying assumptions are not met. This also has
economic implications as may be shown with the following example:
It may be decided to establish a conservation scheme for a breed based on the maintenance
of four samples of animals of this breed. However, especially under the given conditions in
developing countries it never is completely certain that a sample will not be lost within a given
time frame. External changes (like environmental or political changes, theft, diseases etc.) and
management failures may lead to partial or complete losses with a certain probability.
Let us assume that in scenario A, all efforts are made to establish a scheme that the
conservation probability over a given time period for any one sample is maximum, which is,
say, 95%.
In scenario B, much less effort is made to make sure that the samples are conserved over the
chosen time period. While this probability for any one sample is only, say, 75%, it is chosen to
start with six rather than four samples to balance this reduced individual probability.
Figure 4 gives the probabilities for different numbers of conserved samples. While in scenario
A the probability that four samples are conserved over the whole period is 81.45%, the
probability that four or more samples are conserved in scenario B is 83.06% (sum of
probabilities that four, five or six samples are conserved in scenario B).
Figure 4. Probabilities for different numbers of conserved samples in the two scenarios.
The expected outcome of the two scenarios thus is very similar, and thus the cost of the
implementation of the two schemes will determine their relative superiority. In general, it must
be assumed that increasing conservation security will be encumbered with over proportionally
increasing cost, i.e. increasing the probability of a sample to survive over a given time period
from 90-95% will be much more expensive than increasing the same probability from, say, 70-
75%. This is so, because in the latter case, it may be sufficient to improve a traditional
management scheme by introducing some health services etc., while in the latter case it may
be required to set up a centralised station with professional management staff etc. Figure 5
depicts this increase in costs (on an arbitrary scale) for a range of survival probabilities.
Figure 5. Cost and expected return per sample as a function of the sample survival probability.
The optimum strategy is to establish a sample survival probability so that the extra costs to
conserve an additional sample are equal to the expected return in terms of conserved
diversity. In the example shown, this is the case for a rather low sample survival probability of
about 65%.
As mentioned earlier, risk is another important aspect to be considered when establishing a
conservation scheme. Let us consider a situation where we would aim at conserving 4.8
samples of a population. This can be achieved either by conserving eight samples with a
sample survival probability of 60% or by conserving six samples with a sample survival
probability of 80%. Although the expected number of conserved samples is identical, the
strategy to conserve more samples allows a better dispersion of samples, e.g. across regions,
environments etc., which reduces the risk of a total failure due to an epidemic, natural
catastrophes, drought etc.
 Completed and planned research on economic valuation of farm animal
genetic resources
To date, the number of empirical studies that can be interpreted as valuations of AnGR is
extremely limited. In fact, only the studies conducted by Jabbar et al. (1998) concerning small
ruminants and cattle in West and central Africa can be counted as such. Even these studies
initially had been conducted with a different intention of understanding market mechanisms in
developing countries. The interpretation from the genetic resources point of view has been
imposed a posteriori. Few other studies aiming at the valuation of AnGR are underway, e.g. a
study valuing the criolo pig genetic resource in Mexico conducted by Drucker and Anderson
(2000, personal communication) that will be finished and published later this year.
Nevertheless, the toolbox of potential valuation methods is well described especially in the
FAO/ILRI workshop proceedings (Rege 1999) and the survey presented by Drucker et al.
(2001). Potential valuation methods including data requirements, data availability, strengths
and weaknesses from this study are summarised in the Appendix of this report.
n February 2000, ILRI invited an expert panel to a planning workshop to co-ordinate future
research activities in this field. The main goal was to conduct a series of valuation studies on
different farm animal species at different locations (with emphasis on SSA) and using different
valuation tools.
In the very intensive discussions, different aspects were seen quite controversial. First of all,
the problem of biodiversity can be seen both from a global (species) perspective and from a
breed perspective.
The species perspective tries to valuate the contribution of a breed (or generally
speaking of a group of animals within a species) to the within species biodiversity. Thus,
this approach mainly focuses on the option value of a breed, i.e. its eventual `insurance'
function against future changes in production environments, market preferences etc.
Weitzman 1992; Weitzman 1993 suggested a theoretical framework for this type of
perspective, which, however, was applied only once to farm animal data (Thaon
d'Arnoldi et al. 1998).
The breed perspective tries to validate the actual value of a breed with respect to
productivity, non-consumptive and indirect use values etc. The approach is strictly
empirical, in that values are derived from actual market data, surveys etc.
Within this context, again two principles of valuation can be identified:
Subjective valuation approaches based on data what stakeholders (animal breeders and
owners, buyers of animals, buyers of products etc.) think a breed, an animal a product,
a special characteristic of an animal or product is worth. These subjective valuations can
be based on surveys or on empirical market data, which both reflect actual subjective
valuations of goods or characteristics.
Objective valuation approaches aiming at the identification of real values of certain
characteristics (e.g. disease resistance) of an animal with respect to its productivity or of
a breed within a production system. These studies are based on system analysis
approaches, using farm or production system simulation to study the consequences of
breed replacement in such complex systems.
Both approaches do have their strengths and limitations.
Subjective approaches may fail because the future developments may not be
adequately taken into account by today's stakeholder's judgements or market decisions.
In Europe, there are quite a few examples where breeds were highly valued by
breeders, but subsequently were almost totally replaced within very short time as a
consequence of changes in the market or the political system.5
Objective approaches based on farm or production system simulations require a very
precise and realistic knowledge of the relevant biological and economic processes
involved. Many of these processes are not very well known, especially in the production
systems of developing countries, so that wrong assumptions may lead to wrong
conclusions.
ILRI plans to implement a research focus entitled developing and testing tools for
economic valuation of AnGR. This initiative has received an external funding of about
US$ 300 thousand to conduct a first phase.
The studies planned so far by ILRI for the first phase of the project mainly focus on the
breed perspective and are mostly based on subjective approaches. This is mainly due to
the fact that results have to be obtained within a rather limited period (until end of 2000).
We, therefore, planned to base the majority of the studies on available data on market
surveys etc. Developing realistic models for the objective approach is quite demanding
and likely not feasible within the given time frame of the first phase.
However, there is a strong group stressing the need to also take the global perspective,
i.e. to assess the contribution of breeds to the within species biodiversity. We decided to
conduct a case study on diversity of African taurine breeds, based on genetic distance
data and data on degree of endangerment of breeds available at ILRI. The case study
will yield two types of results:
A quantification of marginal contributions of the considered breeds to the actual and
future within species biodiversity.
Estimated marginal benefits of investments to AnGR management and optimal
allocation of financial resources to conserve a maximum proportion of within species
biodiversity.
This case study will also help to identify research priorities to generate empirical data required
for the implementation of the global approach.
Based on the results of the case study and the integration of all available data, it will be
attempted to establish a more systematic approach aiming at the development of a decision
making tool for global policy makers.
Both approaches, the more global species perspective and the breed perspective are
complementary, in that the economic valuation of breeds provides essential data to be used
for within species decision making. However, knowing all actual use values of the breeds
forming a species will not allow to make rational and optimal conservation decision, since
clearly additional information, mainly on the genetic structure of the species, the different
degrees of endangerment, and (in the case of limited resources), the respective conservation
costs and efficiencies for different breeds have to be taken into account.
Therefore both approaches have to be pursued simultaneously.
To derive the methodological framework for the global perspective is mostly desk-work and an
informal group of scientists (J. Gibson, E. Rege, O. Hanotte, all ILRI; T. Meuwissen, Lelystad,
E. van Dusen, Davis, C. Wollny, GTZ; H. Simianer, Stuttgart) have already agreed to go
forward in this direction and conduct a case study to develop a project proposal to develop
and implement this framework with the respective operational tools.
Breed valuation studies have to be planned and carried out systematically with respect to
methodological approaches, species and breeds within species considered, and geographical
distribution of studies. The optimum design would be a three-dimensional matrix study with
methodology, species and geographical region as the three main factors. This goal will not be
achieved in the first phase of the ILRI project (which focuses to a large extent on ruminants in
SSA plus one study on pigs in Latin America).
Since experience with different methods for valuation of AnGR is scarce, it remains to be
evaluated if the approaches (especially the survey-based methods) are useful or even feasible
in different socio-economic contexts. A survey approach that works well in one situation may
entirely fail in another context, since, e.g. the ability to abstract is necessary to value
`hypothetical' properties that are present to a different extent in different groups of
interviewees.
Given the amount of resources required for a sound study (realistically about six to twelve
scientist months plus additional personnel, transport etc. for the surveys) it must be doubted
that a totally comprehensive matrix approach can be put on the ground. In any case, this is a
very convincing argument for a concerted action with a joint and co-ordinated effort.
Factors affecting results on the relative economic superiority of
breeds or crosses
When comparing the relative economic value of breeds or breed crosses under specific
conditions, a number of factors have to be taken into account.
Unit of comparison
First of all, the unit of comparison has to be specified. This can be, e.g. an individual (or even
an individual in a given time interval, for instance a lactation period), a herd or a whole
production system. As Kahi et al. (1998) pointed out the relative value of imported animals
with a higher live weight and a higher absolute productivity per animal will be overestimated as
compared to smaller size indigenous breeds if the comparison is made on the individual level.
If, however, the comparison is made on a herd level with the main constraint on available feed,
light animals will appear more competitive because the ratio of produced output (e.g. milk or
meat) to the animal's maintenance requirements are higher for smaller animals.
When comparing different types of crossbreeding systems (inter se matings, rotations,
crisscross etc.) it must be taken into account that some of these schemes require the
continuous availability of purebred males and/or females of the underlying exotic breeds. In
this case, the total economic valuation has to take into account the costs of maintaining these
animals or alternatively the costs of purchasing semen or live animals to be brought into the
system. While in many cases straight F1 animals of a Bos indicus × Bos taurus (e.g. Sahiwal ×
Friesian) cross have the highest expected performance, the low performance of the pure B.
indicus cows that have to be maintained as mothers lowers the productivity of a system based
on continuous replacement of F1-genotypes. This was shown in an analytical comparison of
different crossbreeding schemes (Kahi et al. 2000c) where a three-breed rotation (Sahiwal,
Brown Swiss, Friesian) clearly was the method of choice when the productivity of the entire
system was used as criterion, while on an individual base the highest productivity was
achieved with F1 Friesian × Sahiwal animals.
Limiting factors
Kahi et al. (1998) also argued that the most informative comparison of breeds or crosses is
made, if productivity is set in relation to the main limiting factor of the production system. If, for
example, the availability of concentrates is the limiting factor, the net profit per kg of
concentrates should be considered. If, alternatively, available land for pastures is limited, the
productivity per unit of land should be taken as the criterion for breed comparison. Kahi et al.
(1998) have shown that the application of these different criteria leads to quite different
rankings of the genetic groups compared.
Genotype × management level interaction
Madalena et al. (1990) considered the relative superiority of various grades of B. taurus × B.
indicus crosses (Brazilian Guzera × Holstein Friesian) on high and low management levels
and observed considerable re-rankings of the crosses. In Figure 6, the production criterion `kg
milk per day of calving interval' for the second lactation is given for various proportions (25-
100%) of Holstein genes on two different management levels. In this study, management
levels differed both with respect to feed and pasture management and with respect to
veterinary supervision and parasite control. The high management group animals were
virtually free of ticks, while the low management group animals were not.
Source: Madalena et al. 1990.
Figure 6. Milk yield (second lactation) per calving interval (kg/day) of different crossbred
genotypes under low and high management conditions.
On the high management level, pure Holsteins and ¾ or Holsteins perform best and are
clearly better than the F1, there is a distinct superiority of the crosses under less
advantageous management conditions. Similar significant genotype × management interaction
effects were found for all major production criteria in this study.
For organisational reasons, breed comparisons are often made on research stations or larger
farms with a relative good management and an above average production level. Even the fact
to be included in a respective study or survey may positively reflect on the management and
the production level during the course of the study. As the results of Madalena et al. (1990)
show, the differences found under these conditions do not necessarily apply to lower
management and production levels, which may however be more typical for the production
system studied.
Data editing
Madalena (1989) also pointed to the fact that breed or breed cross comparisons may be
biased due to selective data editing, e.g. by eliminating extremely short lactations. As a
consequence, genotypes that are not able to enter into a normal lactation profile in harsh
environments due to insufficient adaptive capacity will benefit from the exclusion of `abnormal'
lactations. Rege et al. (1994) reported extreme variability of production traits of exotic crosses,
like lactation performances ranging from 5 kg in 9 lactation days to 3059 kg produced over 454
days with a small proportion of cows producing more than 100 kg per lactation in that study.
This is not only a result of the genetic capacity of the breed or cross, but again has an
interaction with the management, since indigenous breeds tend to go dry within a few weeks if
not allowed to suckle.
1. Note that diversity is also relevant for non-biological objects like buildings, languages,
landscapes etc. Thus, a major part of `diversity theory' has been developed in non-
biological contexts.
2. However, the opposite is not necessarily true, as can be seen with monozygotic twins,
which have identical DNA, but still may differ in many ways. This is due to the fact that
on average the larger part of phenotypic variability is due to environmental factors.
However, sustainable diversity, i.e. diversity that will be passed over to subsequent
generations is mainly genetic diversity, and hence, this one is accepted as the major
criterion in many diversity studies.
3. Such discrepancies have frequently been reported in market analyses, where potential
consumers were asked, how much extra price they would be willing to pay, e.g. for
`ecologically produced' goods, which largely overestimated the achievable price in the
market (and subsequently often led to market failures).
4. This list is largely a summary of Mendelsohn's (1999) presentation of the topic.
5. See, e.g. the fate of the `SMR' synthetic cattle breed in the former GDR, which was
almost entirely replaced through Holstein as soon as the market for animals and
concentrates was liberalised.
Performance comparisons of indigenous and exotic breeds
and their crosses
Performance comparisons of indigenous and exotic breeds and their crosses
Cattle
Dairy traits
Beef traits
Small ruminants
Sheep
Goats
Pigs
Poultry
Chickens
Guinea fowls
Performance comparisons of indigenous and exotic breeds
and their crosses
Cattle
With respect to the general performance of exotic cattle (B. taurus type of European origin) in
the tropics, de Vaccaro (1990) summarised that on almost every criterion of survival,
European cattle in the tropics did less well than their crosses with Zebu. The impact of
heterosis is shown by the stayability data summarised in Figure 7, showing that crosses with
50-62.5% exotic genes achieved on average more than twice the number of lactations (6.6)
than purebred European cattle (2.9 lactations).
Figure 7. Average herd life (in lactations) of European purebred cattle and different grades of
crossbreds with zebu.
Another interesting result is that imported purebred exotics produced only 0.74 heifer
replacements, i.e. female offspring alive at first calving, while locally born purebred exotics
produced 0.98 heifer replacements (de Vaccaro 1990). This indicates that in the experiments
studied imported pure-bred exotics could not sustain and continuous importation was
necessary to maintain the populations.
A relevant alternative to continuous crossbreeding schemes is to establish `new' synthetic
breeds with gene contributions from different sources, in most cases mixtures of B. indicus
and B. taurus. The great advantage of such synthetics is that they are genetically more stable
and uniform and that selection may operate on establishing epistatic gene complexes with
desired effects.
Developing such a synthetic, however, is a complex and demanding long-term exercise. Most
of the relevant synthetic strains thus have been developed in the US (e.g. Santa Gertrudis,
Beefmaster, Brangus, Charbray), Latin America (e.g. Indubrasil, Occampo, Jamaica Hope),
Australia (e.g. Droughtmaster, Quasar) and South Africa (Bonsmara).
There are, however, also risks attached to this strategy. Development of a synthetic breed
implies to go through a `genetic bottleneck', which may have consequences for the
performance in certain traits, especially related to fitness and reproduction. For Santa
Gertrudis cattle (made up from Shorthorn and Zebu) there are reports of poor libido of bulls
and long calving intervals in females while Jamaica Hope (½ Jersey, ½ Zebu), suffers from
low reproductive efficiency (McDowell and Jones, 1972).
 Dairy traits
In a recent and comprehensive review of the relevant literature, Rege (1998) performed a
meta-analysis on the results of 80 reports on crossbreeding in the tropics, including Africa,
Asia and Latin America. The least square means for the main traits milk yield, calving interval
and age at first calving are depicted in Figure 8. The trait values are expressed relative to the
trait value of the F1 (50% exotic genes), which was set to 100.
Source: Rege 1998.
Figure 8. Relative values (F1 = 100) of the least square means of results from 80 studies on
crossbred performance for the traits: calving interval, age at first calving and milk yield.
For milk yield, there is a considerable improvement of performance when the percentage of
exotic genes is increased from 0-50%, while the average stays approximately on a constant
level between 50-100% exotic germplasm.
On the individual level, crossbreds with a similar share of exotic genes, but resulting from
different crossbreeding schemes (F1, synthetics, rotation) were found to be not much different
with respect to milk yield (Kahi et al. 2000a).
The situation is similar for the trait age at first calving, where a considerable improvement is
observed when increasing the proportion of exotic genes from 0-50%, while the trait remains
on a comparable level for higher proportions of exotics.
The shortest calving intervals are observed for animals with 50% exotic genes, while calving
intervals are longer both for lower or higher grades of crossbreeding.
There is, however, a considerable variability between exotic breeds used for the crosses. For
the trait milk yield, Brown Swiss crosses outperformed most other crosses over a wide range
of crossbreeding grades, while Jersey crosses appeared to be inferior. Crosses with Holstein
Friesians and Red Danes were intermediate for intermediate grades of crosses but performed
best with high proportions of exotic genes (>).
Lactation length is a crucial trait for the overall performance of exotic crosses under tropical
conditions. Rege (1998) reviewed that across breeds, lactation length is linearly increased by
6 days with a 10% increase of exotic germplasm in the cross. This, again, is depending on the
exotic breeds used. While the increase is similar to the overall picture with Holsteins, Red
Danes and Jersey, the average lactation length stays on a low level (260 to 270 days) for
Brown Swiss crosses over a wide range of crossbreeding grades.
Generally, F1 animals tend to be better than subsequent inter se crosses. While there are
numerous reports on the performance of F2 animals, empirical results on F3+ generations are
limited. In his literature review, Rege (1998) found considerable average heterosis of up to
40% of the pure parent breed average for Brown Swiss × indigenous breed crosses for the
trait milk yield.
Sharma and Pirchner (1991) reported 13-18% heterosis for milk yield traits in Sahiwal ×
Holstein crosses in India, but 30% heterosis for the length of the dry period. Kahi et al. (2000b)
found significant heterotic effects of about 15% of the mid-parent value for milk yield in
Sahiwal × Brown Swiss crosses, but no positive heterosis for Sahiwal × Ayrshire.
Due to recombination loss, this heterotic superiority in the F1 animals is expected to break
down in subsequent generations, which is confirmed by findings of Syrstad (1989), who
reported considerable declines for all relevant dairy traits when comparing the F2 with the F1
generation. In a literature review, Syrstad (1990) found a comparable development of the
relative performance of crossbreds at different grades as Rege (1998) reported (see Figure 9),
however when comparing the F2 (inter se crosses of F1 animals) with the F1, he reported a
reduction by about 25%.
Source: From literature reviews of Syrstad (1990) and Rege (1998).
Figure 9. Relative values (F1 = 100) of the least square means for the trait milk yield; F2
average separately indicated.
For the trait milk yield of a B. indicus × B. taurus cross, Mackinnon et al. (1996) reported a
recombination loss of 679 kg, which exceeds the reported heterosis of 616 kg (20% of the
mid-parent value). Similar results were obtained for other dairy traits, e.g. milk yield per day of
lactation or per day of calving interval.
Both Syrstad (1989); Mackinnon et al. (1996) stated that the superiority of crossbreds is mainly
due to the high heterozygosity in combination with dominance effects. This has consequences
for the relative benefits and sustainability of crossbreeding schemes. McDowell (1985) stated
that `the real challenge is to establish breeding programmes that retain the merits of the first
cross'.
Madalena et al. (1990) concluded that a continuous F1 heifer replacement strategy has the
highest expected benefit, but has also the highest implementation costs. Crisscross schemes
(with 50% or 67% exotic genes) are also competitive and may be organisationally less
demanding, while continuous upgrading or synthetic breeds through repeated inter se crossing
are less advantageous. A continuous production of F1 females through the use of embryo
transfer technology, as considered by Teodoro et al. (1996), may be an option for some more
developed countries in Latin America. A sustainable implementation of such a scheme must
however be doubted in the developing countries in SSA.
 Beef traits
Data on beef production with indigenous breeds vs. exotic crosses under tropical conditions
are very limited compared to those on dairy production, especially since performance
comparisons of crosses with the pure breeds under the same environmental conditions are
scarce (McDowell and Jones, 1972; Mpofu et al. 1998)
McDowell (1985) summarised the results of published studies on performance of crossbreds
under tropical or subtropical conditions conducted mainly in the southern parts of the US,
South America and Northern Australia (Queensland). He reports that superiority of crossbreds
over the parent breed average are observed growth rate to 15 month of age (11% heterosis),
but no heterosis was observed for carcass traits. He also points to the fact that reciprocal
crosses may differ due to the significance of maternal effects. In Queensland, crossbred
mothers showed better maternal abilities than imported exotics (75% vs. 56% live calves for
crossbreds vs. pure exotic--Shorthorn and Hereford--cows).
As a consequence of improved viability, growth rate and reduced cost for tick control,
McCarthy and Hodgson (1970) estimated 11.3% higher returns from European × Zebu
crosses compared to pure Herefords under Queensland conditions.
Chantalakhana (1998) reported experiences with imported cattle and crosses used for beef
production in South-East Asia. He refers to factors like tropical disease and parasite
challenge, poor management, low quality food and especially marketing and economic
problems that pose considerable limitations on the use of upgraded animals. Cattle have multi-
purpose functions, being not only used for meat production, but also for work, utilisation of
farm wastes and natural pastures. Cattle also play a considerable socio-economic role-being
used as long-term savings, and an essential element of ceremonies in some regions (Jabbar
et al. 1998).
Increasing occurrence of theft and cattle smuggling across national borders has the potential
to hinder sound development of profitable (or at least sustainable) beef industry based on
upgraded genotypes both in South-East Asia (Chantalakhana 1998) and in the SADC region
(Wollny 1999, personal communication).
In the semi-arid African subtropics of Botswana, Mpofu et al. (1998) compared performance of
crosses of indigenous cows (Tswana or Tuli) with exotic sires of the breeds Charolais, South
Devon, Sussex and Santa Gertrudis with purebred Brahman animals. For birth weight,
weaning weight and weight at 18 months, all crosses outperformed the Brahman purebreds by
up to 25% (Figure 10). Especially for the later weight traits recorded at weaning and at 18
months of age, the Charolais and South Devon crosses appear to be clearly superior
compared to the other crosses and compared to the Brahman. The authors suggest that it
may be possible that crosses are reared and fattened more intensively and slaughtered at a
younger age (than the usual 36 to 48 months) where the required improved management level
can be implemented.
Source: Mpofu et al. 1998.
Figure 10. Beef performance traits for four different crosses relative to pure bred Brahmans in
semi-arid Africa.
Mpofu et al. (1998) also reported that genotype-environment interaction favoured the smaller,
slow growing genotypes on the poorer agro-ecological locations and in dry years.
For cow live weight in crosses of Sahiwal with Ayrshire, Brown Swiss raised in the tropical
lowlands of Kenya, Kahi et al. (2000b) reported consistently positive, but mostly not significant
heterotic effects. The heterotic effect of about 5% of the mid-parent value for Sahiwal × Brown
Swiss crosses however is exceeded by an even higher recombination loss.
In a comparative study of different indigenous breeds and exotic crosses for beef production
traits in semi-arid Zimbabwe, Moyo et al. (1994) found little evidence for a general superiority
of exotic crosses over highly adapted indigenous breeds and concluded that indigenous cattle
should play a major role in livestock development programmes in the region. Experiences from
Northern Australia show that long-term selection in B. taurus type cattle populations may lead
to significant improvements in beef production performance (+10-20%) and in adaptation
characteristics (tick count reduced by 80%) over a 20 year period (O'Neill and Frisch 1998).
 Small ruminants
For small ruminants, most scientific studies on the effects of crossbreeding have been done
with breeds and in locations of temperate zones. If studies were undertaken under tropical
conditions, management level was typically high. Data on crossbred performance under
smallholder situations are virtually not existent in the scientific literature. In a comprehensive
review of crossbreeding studies with sheep in temperate climates, Young et al. (1986)
concluded that heterosis retention was significant, but differing with respect to the respective
combination of breeds and crossbreeding schemes, so that the optimum scheme has to be
determined specifically for any project based on specific combinations. In general,
crossbreeding and use of `exotic' breeds is seen as a means to improve production efficiency
in temperate zones (Hohenboken 1986). For New Zealand, different economic studies
reported high expected returns from a systematic import of exotic breeds (like the Finnsheep),
with predicted internal rates of return on public investment between 27% (Sorrensen and Scott
1978) and over 80% (Bushnell and Hutton 1982).
Sheep and goat genetic resources account for 62% of the total domesticated ruminant
livestock in SSA (Lebbie and Ramsay 1999). Lebbie et al. (1996) listed 61 sheep and 42 goat
genotypes for which the real genetic distinctiveness, particularly of the indigenous ones, is not
known. In many cases, the breeds are named after ethnic groups (e.g. the Red Maasai sheep)
or regions (e.g. West African Dwarf), which says little about there distinctiveness.
The role of small ruminants in these agricultural and eco-systems thus cannot be
overestimated. Small ruminants both are an indispensable component to environmental and
farming systems and in many cases are already or become a threat (through overstocking
and overgrazing) of these ecosystems in tropical SSA (El Aich and Waterhouse 1999).
 Sheep
Charray et al. (1992) stressed that genotype-environment interaction with respect to adaptive
traits is the main obstacle to successful crossbreeding with exotic sheep breeds. In Ghana, the
Nungua Blackhead, a cross of the local West African Dwarf with Blackhead Persian, is known
to produce heavier lambs at weaning than the pure-bred Dwarf lambs. However, this
advantage is nullified by the fact that litter size of West African Dwarf ewes is larger than that
of Nungua ewes.
An attempt to cross West African Dwarf sheep with Sahelian sheep in Côte d'Ivoire had to be
abandoned in the first year due to high lamb mortality. Similar problems have been
encountered with the same cross in Ghana.
Osuagwuh et al. (1980) reported a 10% incidence of dystocia and 14% perinatal mortality
when West African Dwarf ewes were mated with Permer, Uda or Yankasa rams, while no
such problems occurred when the ewes were mated to rams of the same breed.
Steele (1983) reported an improved growth rate and wool yield in the Sultanate Oman, when
the indigenous Omani breed was crossed with Mediterranean Chios sheep. However,
performance of the crossbreds was only acceptable under good management conditions,
while poor management conditions led to considerable losses. In South Africa, composites
based on Finnsheep × indigenous crosses compared to local Dorper animals were smaller,
more fecund and produced faster growing lambs (Schoeman and van der Merwe 1994). The
overall efficiency of the composites was 84% above the one achieved with purebred
indigenous. However, this and most other results reported in literature refer to performance
under station conditions. These may be far away from the conditions given in smallholder
farming systems, and it must be doubted whether the results can be generalised in this
respect.
In Barbados, crosses of Barbados Blackbelly hair sheep with European wool sheep (Suffolk
and Dorset) showed low fertility in both sexes, while the carcass traits were similar to the
pure-bred indigenous animals (Patterson 1983). In India, about 50% of the variability of wool
quality (fibre diameter) could be attributed to the genetic grade level in a crossbred population
of indigenous Nali sheep with exotic Russian Merino and Corriedale sheep (Amar et al. 1998).
In sheep, repeated inter se matings with selection to create `synthetic' breeds have been
exercised, e.g. in India (Avikalin breed) or South Africa (Dorper breed). Again, the
development of a new breed or establishing and maintaining a complex mating scheme (such
as crisscross or three breed rotation) requires considerable infrastructure and has a high risk
of failure. Additionally, populations go through a genetic bottleneck in the formation of
synthetics, with a high risk of loss of genes of potential future importance.
Like in other farm animal species, F1 animals are reported to show the highest heterosis,
which is lost in subsequent inter se matings. Boujenane and Chafik (1994) reported
experiences with D'Man × Sardi crosses in Morocco, where an almost 20% heterosis for
number of lambs born in the F1 is reduced to 13% and 5% in the F2 and F3, respectively.
Indigenous local breeds with special characteristics in one or few traits may be seen as a good
resource for improving breeds in the same region through crossbreeding. Mason (1980)
suggested using the prolific West African Dwarf sheep to increase the reproductive rate of
larger, less prolific breeds in West Africa. Such a cross would have the advantage that the
`exotic' breed is adapted to a similar environment and has comparable `hardiness'. In some
regions, like in the tsetse-infested rainforest region of western Africa, indigenous
trypanotolerant breeds cannot be replaced by exotics, but may act as the base for an
upgrading programme with exotic breeds of higher productivity, while the trypanotolerance of
the indigenous breeds has to be maintained (Devendra and McLeroy 1982).
Indigenous sheep breeds seem to be less threatened than breeds from other farm animal
species. In general, sheep use to live in unmodified harsh environments and large-scale
genotype replacement is very slow (Gatenby 1986). This, however, is not necessarily true for
all regions in SSA.
McDowell and Jones (1972) referred to the failure of government breeding programmes in
North Africa, resulting from the fact that nomadic shepherds normally are not willing to accept
rams that have been `artificially reared'. On the other hand, King et al. (1984) reported
crossbreeding with Blackhead Persian and Dorper rams into Maasai flocks, which has been
carried out by the stock owners without external encouragement, e.g. through development
agencies.
 Goats
According to Shelton (1986), the value obtained from goat milk production exceeds the one
derived from other products (meat and fibre) on a world wide scale, although only a small
proportion of the world's goat population is milked. This is due to the fact that milk production
is a continuous process, while meat production involves slaughtering of the animal and fibre
production is done with only a very small proportion of all goats. Most studies in the literature
have been conducted in Asia, especially in India, while reports from Africa are scarce. When
expressed per unit of input (e.g. per kg dry matter of feed), goats may even outperform cattle,
buffaloes and sheep under comparable environmental conditions in the tropics (Knights and
Garcia 1997).
A review study of Sahni and Chawla (1982) compared native breeds and crosses with exotics
(mostly Saanen and Alpine goats). With ½ to exotic crosses, milk production is increased by
100% to almost 150%, with an apparent optimum of ¾ crosses. It should, however, be noted
that these results are based on a limited number of studies (26 in total).
More detailed results on crossbreeding experiments with Indian Beetal goats and exotic
breeds (Saanen and Alpine) are depicted in Figure 11. For milk yield, there is a clear heterotic
effect of more than 20% of the mid-parent value, which erodes if the proportion of exotic
genes is increased to 75%. For lactation length, however, the ¾ exotic genotypes show the
best performance, which is also true for the kidding interval, where the ¾ exotics have the
shortest interval. Devendra and Burns (1970) reported similar results for crosses of Malaysian
indigenous breeds with Anglo Nubians.
Source: Sahni and Chawla 1982.
Figure 11. Performance of pure indigenous (Beetal), pure exotic (Alpine and Saanen) and ½
and ¾ crosses for the traits: lactation milk yield, lactation length and kidding interval,
expressed as per cent of the pure line average.
In SSA, a number of dairy goat crossbreeding schemes are under way or have been
conducted with `improver' breeds like the Toggenburg, Alpines, Saanen and Anglo Nubian.
These projects utilise the genetic adaptation of indigenous breeds in terms of hardiness,
disease tolerance and resistance to exploit the productive superiority of the more disease
susceptible and less adapted exotic breeds (Lebbie and Ramsay 1999). Examples for such
projects are the FARM-Africa Dairy Goat Development Project in eastern Ethiopia (Peacock et
al. 1990), the Ngozi Goat Development Project in Burundi (Rey and Jacob 1991) and the
Small Ruminant Collaborative Research Support Program Dual Purpose Goat Initiative in
Kenya (Semenye et al. 1989).
In other cases, composite or synthetic breeds like the Kenya dual purpose (East African ×
Toggenburg × Galla × Anglo Nubian) or the Tanzania Blended (Boer × East African ×
Kamorai) have been successfully developed.
The majority of goats are exploited only for meat production. In some tropical regions,
however, goat meat is hardly competitive to other sources of meat, due to low dressing
percentages, small and bony pieces of meat etc. The major benefit of goats for meat
production therefore must be seen in their excellent adaptation to poor environments and the
high reproductive efficiency of some breeds at least. Singh and Singar (1981) concluded that
while larger breeds are superior when growth rates or body weights at a given age are
considered, smaller breeds were more efficient if reproduction was taken into account and
meat production per unit of input was measured.
A systematic diallel crossbreeding study using the Tennessee stiff-legged, the Spanish and
the Nubian goat breeds under hot and dry conditions in the Southwest of the United States
revealed no significant improvement of Nubian × Western breed crosses compared to pure-
bred Nubians in growth and carcass traits, while Boer goat × Tennessee stiff-legged and Boer
× Spanish performed well under these conditions (Dzakuma et al. 1998).
There are considerable differences in breeds with respect to the different traits of interest. A
crossbreeding scheme to produce F1 animals from highly fertile females (e.g. Black Bengal)
with large fleshy males (e.g. Boer or Anglo Nubian) should lead to well adapted and
sufficiently productive stocks. However, such a scheme requires considerable infrastructure to
maintain the pure lines, which in many cases will be difficult to establish under smallholder
management schemes.
 Pigs
As already alluded to, SSA is the home of 13 pig genetic resources (Hammond and Leitch
1998). While pig genetic resources have been described for some of the tropics (Alba 1972;
Rigor and Kroeske 1972), there have been few (mostly from Nigeria) detailed and accurate
studies on the performance of the SSA's indigenous pig genetic resources (Ilori et al. 1976;
Chiboka 1981; Somade 1985; Somade and Makinde 1985). Most of the indigenous pig breeds
are kept under village management and are normally for subsistence use. Like other
indigenous livestock species, the production potential of these indigenous pig breeds is low.
These breeds should not, however, be wholly criticised because they have important
characteristics that contribute to their adaptability, which are tolerance to heat and dietary
changes. There is clear evidence to indicate that some indigenous West African pig breeds
are tolerant to porcine trypanosomosis (Onah 1991). Porcine trypanosomosis has been shown
to affect the productivity of exotic pig in Nigeria (Omeke 1989a). While some Chinese breeds
are considered prolific (Bidanel 1990), this has not been confirmed in any of the SSA
indigenous pig breeds. In some production systems, indigenous genotypes or their crosses
with some exotic breeds may in fact be the most suitable. With considerable variability among
indigenous pigs and their crosses with exotic breeds, their potential is highly subjective to
selection procedures to attain higher levels of performance (Adebambo 1995).
In most of SSA, the exotic breeds tend to dominate intensive piggeries because of their
genetic superiority over the indigenous breeds although the adverse environmental factors
may limit their performance. Therefore, there is increasing indiscriminate crossbreeding
between the indigenous and the exotic breeds and even between some of the exotic breeds
(Omeke 1989a). The performance of the crosses is determined by the additive genetic values
of the two breeds and the amount of heterosis. It should, however, be noted that efforts to
improve production at the village level, where the indigenous breeds might continue to be
important, are very rare. Since studies comparing the performance of the SSA's indigenous
pigs breeds, the exotic and their crosses are few, attempts will be made to review few studies
from other tropical areas experiencing climatic conditions similar to those of SSA. However,
because of the possible genotype × environment interactions, results from these areas cannot
be extrapolated to the African conditions but give an insight on what to be expected.
Preweaning traits
Reproductive performance of pigs is influenced by complex interplay of environmental factors
and physiological processes. In the tropics, climate is the most important environmental factor
influencing reproductive performance. Steinbach (1976) discussed the effects of a tropical
climate on the fertility of the boar and sow. Reproductive performance is reflected in litter size
and weight at birth and at weaning. Several studies on the performance of the indigenous pig
breeds, pure-bred exotic breeds and various crosses among them have been made under
tropical conditions. In Sri Lanka, Goonewardene et al. (1984) compared preweaning traits of
the indigenous pigs, pure-bred Large White and indigenous × Large White crosses and found
that the Large White was significantly better than the indigenous in litter sizes, litter weight at
birth, birth weight, weaning weight and average daily gain. However, there were no significant
differences in weaning weight and average daily gain between the indigenous breed and the
Large White-sired crossbred (Table 1). The litter size reported in that study was slightly higher
than that reported by Adebambo (1983) (cited in Goonewardene et al. 1984) for the Nigerian
indigenous pig.
Table 1. Breed cross means for preweaning traits.
Breed cross1 Litter size Litter weight (kg) Birth weight (kg) Weaning weight
(kg)
Average
preweaning daily
gain (kg/day)
I × I 6.38 3.03 0.48 4.50 0.074
I × LW 8.20 8.26 0.97 7.73 0.108
LW × I 9.50 5.47 0.56 4.69 0.074
LW × LW 10.60 14.02 1.30 7.70 0.118
1. I = indigenous pig breed, LW = Large White breed. Breed of sire shown before breed of
dam.
Source: Goonewardene et al. 1984.
In India, the pure-bred Large White Yorkshire breed was superior to it's crossbred with an
indigenous Indian pig breed for reproductive performance (Mohanty and Nayak 1986). In
southern Nigeria Steinbach (1971) reported significant differences between the Large White
and Landrace breeds for birth weights, litter weight at birth and in percentage of stillborn
piglets. The Landrace breed had the heavier piglets, heavier litters but the higher percentage
of stillborn piglets than the Large White.
In a tsetse infected area of Nigeria, Omeke (1989a) showed that the crossbreds between the
Large White and Landrace breeds had the bigger litter size and the lower preweaning mortality
rates than purebreds and recommended crossbreeding of these breeds in such areas to
improve productivity. The performance of exotic breeds is better during the rainy season than
during the dry season (Omeke 1989a). There is a significantly higher farrowing rate when
exotic sows are mated during the cooler months of the year than during the hotter months
(Dan and Summers 1996).
Postweaning traits
Crossbreeding of the indigenous pig with high productive exotic breeds usually results in
heavier weaners and faster growth of fattening stock when compared with their indigenous
parents. Usually indigenous pigs have carcasses that have more fat than the exotics. If
consumer demand for pork from these pigs is to increase then leaner indigenous pigs will have
to be produced. Ilori et al. (1976) compared the performance of indigenous and exotic pigs on
restricted feed intake and found that the exotic pigs had leaner carcasses, faster growth rates
but lower feed efficiency than the indigenous pigs.
In western Samoa, Udo (1982) simulated three different systems of management, namely
village, semi-commercial and commercial levels and compared the postweaning performance
of the Large White breed, indigenous Samoan pig and crosses between them (Table 2). Under
the simulated village management level, crossbreds grew significantly faster and had
significantly higher dressing percentage than Large White and indigenous pigs. In contrast to
the study of Ilori et al. (1976), there were no significant differences in backfat thickness but
carcasses of local pigs were significantly shorter than those of Large White. Based on these
results (Table 2), we concluded that there is no need to introduce pure-bred Large White pigs
into villages except as breeding boars for crossbreeding purposes.
Table 2. Performance data of Large White (LW), LW × Indigenous Samoan crossbred and
indigenous Samoan pig (I) under different management regimens.
 Village Semi-commercial Commercial
 LW LW × I I LW LW × I I LW LW × I I
Initial weight (kg) 10.45 11.59 10.45 11.70 10.71 13.25 10.68 11.36 11.45
End weight (kg) 34.00 44.35 32.04 82.04 80.08 80.80 79.73 81.41 70.41
Fattening period
(days)
168 168 168 153 153 153 103 112 168
Daily gain (g/day) 140 195 129 461 453 442 670 625 351
Dressing
percentage
63 70 63 71 70 68 68 68 69
Carcass length
(cm)
61 70 49 74 72 68 76 75 64
Source: Udo 1982.
For the breeding pigs, one of the most important postweaning trait is the age at puberty. It has
been shown that the indigenous tropical pig breeds attain puberty earlier than their exotic
counterparts. In Nigeria, Chiboka (1981) reported an earlier age at puberty in the indigenous
pig breed when compared to the Large White, Duroc or Ukranian White. In contrast to the
situation in exotic breeds Young and King (1981) reported the litter size, viability of offspring or
normalcy of the parturition process (as measure by frequency of stillborn and milk ejection)
were not affected by age at first service in that study. In Sri Lanka, Goonewardene et al.
(1984) found that the indigenous gilts were sexually mature at an early age of 4-5 months in
spite of the low nutrition and could therefore be bred for the first time when they are four to
five months of age. Early mating is not necessarily harmful to the indigenous pig, and that
sows bearing normal litters after early mating may be utilised for increasing maturity within
these breeds and the exotic through crossbreeding.
Post-farrowing traits
The total number of piglets weaned by each sow per year determines the production cost and
profit margin in a swine enterprise. The number of piglets weaned is in turn determined by the
regularity of the successive reproductive cycles of the sow. The most important post-farrowing
traits include, the interval from weaning to first oestrus, conception rate, gestation length and
farrowing interval. Under conditions in SSA, some of these traits have been estimated in exotic
breeds (Omeke 1989a; Omeke 1989b; Omeke 1990) and in indigenous breeds (Chiboka
1981; Somade 1985; Somade and Makinde 1985). Studies to compare exotic breeds,
indigenous breeds and crosses between them for these traits in SSA are few.
Genotype × environment interactions
Genotype × environment (G × E) interactions simply mean that the effect of the environment
on different breeds or genotypes is not the same; or in other words, there is a change in
ranking of breeds or genotypes when taken from one environment to another environment.
Environment might mean a climatic condition, system of management or feeding level.
Therefore, there is no universally `best' breed or genotype. In this sense, genotypes selected
in developed countries may not be suitable for extensive and semi-intensive production
systems in the developing countries (Pathiraja 1986).
As shown by Udo (1982) for daily gain (Table 2) in West Samoa, there was a change in
ranking of the Large White breed and its cross with the indigenous Samoan pig under village
and commercial management. This is clearly represented in Figure 12. The levels of heterosis
for this trait were higher under village (45%) than under commercial management (22%).
Cunningham (1981) suggested that when there is a substantial difference between the F1 and
the local strain, there would be G × E interaction when heterosis heavily influenced production
in a poor environment and breed additive effects and small heterotic effects largely
determined production in a good environment.
Source: Udo 1982.
Figure 12. An illustration of genotype × environment (management system) interaction in pigs.
 Poultry
Poultry is the most numerous species of farm animals in Africa with more than 80% of it kept
in rural areas (Guèye 1998). Worldwide, poultry out-levels the human population by almost
30% (Horst 1990). Several species namely chickens, guinea fowls, ducks, geese, turkeys,
pigeons and quail represent poultry. The ostrich has also been included as a poultry species.
The small poultry species are generally kept in conjunction with ruminant animals and equines
especially in the agricultural subsystems and in the pastoral system. Despite the substantial
contribution of all small poultry species to the SSA's supply of protein, chickens and guinea
fowls have received more research attention in recent years than the others. There is also
some attention on ostrich under extensive systems especially in the southern part of Africa.
However, its significance to the supply of protein to the rest of SSA is yet to be realised.
 Chickens
Of all the poultry species, the domestic fowl (Gallus gallus domesticus) is widely represented
in all parts of SSA where different strains or breeds have been known to exist. The indigenous
fowls have been variously referred to us the African chickens, bush chickens or runner
chickens. However, distinct local varieties have been reported in Egypt, Cameroon, Burkina
Faso, Morocco and Sudan (Guèye 1998). Indigenous chickens tend to be robust and are well
adapted to harsh environmental conditions such as hot or cold weather, rain and periodic feed
shortages. These birds have many advantages such as good egg and meat flavour, hard egg
shells, high dressing percentages, and especially low cost with little special care required for
production. However, they have some disadvantages arising from the fact that they suffer high
rearing mortality (Trail 1962), have slow growth rates and are poor egg layers and attain
sexual maturity late. Such birds can be improved genetically through selective breeding or by
crossbreeding with exotic stocks depending on the management conditions (Omeje 1986;
Mukherjee 1990; Katule 1991; Katule 1992). This section briefly reviews the performance of
indigenous chickens and their crosses with exotic breeds under SSA conditions.
Growth traits
Fast growing chickens have an advantage in that they reach the market weight early hence
reducing production costs. It has been shown that crossbreeding the indigenous chickens with
the exotic breeds improves the growth traits, which include traits like live weights, daily gains
and feed intake (Omeje and Nwosu 1988; Asiedu and Weever 1993). While this is the case, it
has also been reported that the indigenous chickens possess the potential to grow well at the
early stage of life (Nwosu et al. 1984). Under a good nutritional environment, Katule and
Mgheni (1990) reported similar growth rates of indigenous and exotic breeds to 12 weeks of
age. In that study, the merits of the indigenous chickens were supported by the fact that the
backcross to the indigenous strain grew faster than the exotic backcross. Ihemelandu and
Nwosu (1986) also found that the indigenous chicken showed early faster growth rate than the
exotic breeds and demonstrated that this early faster growth was also true for other organs
(e.g. liver, heart and kidney). This indicates that the growth patterns of some indigenous
chickens especially in Nigeria favour that of broilers. Various ecotypes of indigenous chickens
produce meat that is in composition, consumer appreciation and acceptability similar to meat
from exotic breeds of same age and management (Obanu et al. 1984). Improvement in body
size and growth of indigenous chickens is important from economic considerations bordering
on the need to increase egg size and to improve the post-lay value of the chickens (Ibe 1995).
With well-designed selection programmes, this can easily be achieved in the indigenous
chickens because of the appreciable additive genetic variance observed in these breeds
(Ebangi and Ibe 1994; Olori 1994).
Egg production traits
Egg number and weight are major traits of economic interest in commercial egg production.
Egg size determines to a large extent the price received in any market. Asuquo et al. (1992)
reported that the internal egg quality traits are also important to the consumers especially the
bakery industry. In that study, the exotic Isa-Brown breed was superior in all egg quality traits
except per cent of yolk to the indigenous chickens. Soltan (1993) compared the egg quality
traits of the selected Sinai fowls with those of the Fayoumi and Baladi fowls, which are the
standard Egyptian indigenous breeds and found that the differences in these genotypes for
these traits were not significant. Essien and Ekanem (1990) reported the effects of the age of
layers and the period of oviposition on egg quality produced under humid tropical conditions.
Because of the positive genetic correlation between body size and egg size, crossbreeding of
the small indigenous chickens with an exotic breed also tends to improve the egg size of the
crossbred progeny. In Nigeria, Omeje and Nwosu (1988) compared egg production traits of
the indigenous chickens, the Gold-Link exotic breed and crosses between them. Significant
differences were observed in the egg weights of the different genotypes. The Gold-Link exotic
breed and the Gold-Link-sired backcross laid heavier eggs compared to the first cross, F2 and
backcross populations that were sired by the indigenous chickens (Figure 13). The indigenous
chickens had the smallest egg size. For total egg mass, a similar trend was observed. In
Tanzania, Katule and Mgheni (1990) found that indigenous chickens produced more eggs in
early life but were less persistent and laid smaller eggs. Surprisingly, in that study, the F2
generation laid more eggs than any other genotype. This is contrary to the deterioration in
performance from F1 to F2 in crosses between tropical and temperate cattle for milk
production (Syrstad 1989; Rege 1998). It should be noted that there exists variation in egg
production traits within the indigenous chickens in some parts of Africa. In Egypt, for example,
comparison of the three indigenous breeds (Sinai fowls, Fayoumi and Baladi) showed that the
Sinai fowls produced fewer and smaller eggs than the other two (Soltan and Ahmed 1990).
The indigenous Fayoumi and Baladi breeds have been subjected to several generations of
selection for these traits and hence their superior performance.
Source: Omeje and Nwosu 1988.
Figure 13. Some egg production traits for four different genotypes relative to the indigenous
chicken in Nigeria.
Fertility traits
Fertility traits can be classified into those that are associated with the egg, e.g. fertility and
hatchability of the egg and those directly associated with the chickens, e.g. age at sexual
maturity, which is defined as the age from hatch to the day of first egg. Before the introduction
of the exotic chicken breeds for crossbreeding with the indigenous chickens, Trail (1962)
carried out an experiment in Uganda to compare the fertility and hatchability of eggs from
indigenous and exotic chickens and crosses between them. The fertility and hatchability rates
of indigenous eggs were higher than that from the Rhode Island Red. When indigenous hens
were crossed with cockerels of the Rhode Island Red, Light Sussex, White Leghorn and Black
Australorp breeds, there was little difference between the fertility and hatchability levels of the
four breeds of males used.
In a study that was necessitated by the need to have information on the level of heterosis in
the F1, and that of the residual heterotic effects in the F2, backcross and subsequent
generations before starting a meaningful trait-group selection on crossbred populations,
Omeje and Nwosu (1988) reported that the indigenous chickens and the indigenous-sired
backcross laid the first egg about 10 days earlier than the pure-bred Gold-Link breed and its
backcross. The F1 and F2 generations were intermediate. Katule (1992) reported small and
non-significant differences in age at sexual maturity between the indigenous and the exotic
strain of chickens in Tanzania. In that study the F1 generation matured earlier than any of the
parental indicating existence of heterosis for this trait. Normally breeds that attain sexual
maturity early end up laying lighter eggs than late maturers (Oni et al. 1991).
Genotype × environment interactions
Because of the method of rearing chickens in SSA, it would be expected that most
investigation of G × E interactions would involve differences in management levels or systems.
There are three poultry management systems- intensive, semi-intensive and
extensive/scavenging. The intensive system is normally based on specialised breeds and is
found mainly in urban areas. The extensive/scavenging is based on indigenous chickens and
is mostly found in the villages (Kitalyi 1999). Comparison of breeds or strains of different ability
to utilise feed under conditions where there is a scarcity of high quality feed (such as in most
villages in SSA) would therefore be expected to favour the breed that is better able to utilise
low quality feed.
In an experiment to explore the possibility for developing strains of chickens which would
perform reasonably well under low input production conditions (a characteristic of most of
SSA), Katule (1992) presented evidence of existence of the influence of G × E interaction on
both the body size and egg production traits. In that study, the performance of two exotic
breeds of chickens (an egg type and a meat type), indigenous breed and crosses between
them were compared in two experimental phases. Management (feeding and floor space
allowance) was better in the first phase than in the second phase. In the first phase of that
study, both the exotic breeds performed well in traits for which they had been developed while
in the second phase, the meat and egg breeds performed less satisfactorily than their crosses
and backcrosses with respect to body size and egg production traits, respectively. This
indicates that genotypes, which have been developed for high performance in good
environments, are expected to be more sensitive to environmental changes than those that
have not been developed for any specific environment.
 Guinea fowls
In some West African countries, guinea fowls are second only to chicken as a source of
poultry meat and eggs (Ayeni 1983). In West Africa, there are two subspecies of guinea fowls.
The grey breasted, helmeted guinea fowl (Numida meleagris galeata) is the most common
and there are four varieties of this that are domesticated namely Lavender (Ash), Black, Pearl
and White (Ayorinde 1991). Ayorinde et al. (1988); Ayorinde (1989a); Ayorinde (1989b);
Nwagu et al. (1997) described the external and performance characteristics of these four
varieties. The other type is the crested guinea fowl (Guttera edouardi edouardi) and is
restricted in distribution to forest and derived savannah-forest edges (Ayeni 1979, cited in
Ayorinde 1991). The grey breasted, helmeted guinea fowl is widely raised by peasant farmers
extensively. We considered it uneconomical for its low fertility, slow growth and seasonal egg
production. However, in France, the uses of light control programmes and selection have
resulted in making guinea fowl an all year-round breeder (Nwagu and Alawa 1995).
The local species of the guinea fowls is normally characterised with lower performance than
the exotic (Ayorinde et al. 1988). Therefore, there has been introduction of exotic species of
guinea fowl (Numida meleagris galeata), especially from France, for crossbreeding to improve
the overall performance of the indigenous species. Other traits targeted for improvement are
egg production and fertility traits. Nwagu et al. (1997) reported that the crossbred guinea fowls
were considerably fertile and their eggs more viable than those of any of the local breeds.
There was also variation within indigenous species for hatching characteristics. The Pearl
variety produced eggs with the highest percentage of fertility and hatchability. We concluded
that this variety could be a candidate for commercial hatchery production of a day old kites.
There is therefore the need to properly explore the potential of guinea fowls as sources of
protein for the ever-increasing population of SSA. Nwagu and Alawa (1995) listed a series of
interventions that might help to realise this.
Unique traits and properties of indigenous breeds
Unique traits and properties of indigenous breeds
Disease resistance or tolerance
Trypanosomosis
Gastro-intestinal parasites
Ectoparasites
Bursal disease
Heat tolerance
Cattle, sheep and goats
Poultry
Camel
Adaptation and nutrition
Other properties
Salt tolerance
Scavenging ability
Reproductive performance
Unique traits and properties of indigenous breeds
Different hereditary characteristics of breeds and even types within a breed have resulted in
difference in reactions to environmental stimuli. These reactions are intimately associated with
anatomical-physiological characteristics, which have developed as the result of natural
selection. Indigenous breeds, as the name suggests, have been identified with a particular
area or people from time immemorial and are thus considered to be adapted to some of the
environmental stresses. The degree of successful adaptation of such animals is accurately
reflected in their ability to grow, to reproduce regularly and in their production. Therefore, we
can conclude that adaptability and animal's efficiency of production are closely correlated.
SSA is the home of roughly 120 indigenous cattle (Simon 1999), 61 sheep, 42 goat (Lebbie
and Ramsay 1999) and 13 pig (Hammond and Leitch 1998) breeds.1 There is very little
information on the diversity in indigenous African poultry populations, both at phenotypic and
genetic levels. Most small-scale farmers and pastoralists owned indigenous breeds for whom
they are a source of improved nutrition, income and a secure form of investment. The
production systems under which they are produced are normally characterised by low levels of
animal husbandry practises coupled with the normal tropical stresses of disease challenges,
heat stress and poor nutrition. This demands that an animal must have some degree of
resistance or tolerance to these stresses to maintain some degree of production. This section
reviews within species documented examples of unique attributes and properties of
indigenous breeds that are considered to be the most economically important in SSA.
Definitions given were deemed necessary.
Disease resistance or tolerance
Diseases normally have an impact on welfare and production of a production system. The
conventional control measures, such as vaccination and chemotherapy have either been
ineffective, unsustainable or uneconomic. The genetic approaches to disease control have
been suggested. The genetic basis of disease resistance or tolerance is evident from
observations on breed differences and from results of selective breeding experiments and
from laboratory studies (Stear and Wakelin 1998). It is important to differentiate between
disease resistance and tolerance. Resistance can be defined as the initiation and
maintenance of responses provoked in the host to suppress the establishment of parasites
and/or eliminate parasite load while tolerance is the ability of the host to survive and produce
in the face of parasite challenge (Baker 1998).
Trypanosomosis
In 1975, the Food and Agriculture Organization of the United Nations (FAO) launched, as a
project of high priority, a long-term programme for the control of African animal
trypanosomosis (Hoste 1987). This disease is particularly important in Africa and is one of the
major constraints on animal production in areas that have the greatest potential for significant
increases in domestic livestock population and livestock productivity (D'Ieteren et al. 1998).
Cattle
Trypanotolerance in cattle particularly in the N'Dama (a West African Longhorn) is well
documented (Mortelmans and Kageruka 1976; FAO 1980; Starkey 1984). Other
trypanotolerant breeds in SSA include the Baoule (a West African Shorthorn) (FAO 1980) and
the Dahomey breed that is found in Zaire (Mortelmans and Kageruka 1976). There are now
N'Dama herds in nearly all West and central African countries where they are being used as
purebreds or as crossbred with other breeds (FAO 1980). Currently, N'Dama cattle constitute
only 5% of the total cattle population in SSA, with an approximate population doubling time of
16 years at current rates of increase (Teale 1993). The N'Dama and Baoule breeds are
traditionally small animals but their productivity can match that of trypanosusceptible zebu
cattle (B. indicus) in areas where tsetse fly risk is low (ILCA 1979). In high-risk areas most B.
indicus breeds require regular treatment however significant differences in resistance to
trypanosomosis occur also among various B. indicus breeds (Njogu et al. 1985). Maintenance
of exotic breeds even in areas of low tsetse fly risk would require intensive trypanocidal drug
therapy and veterinary care.
Sheep and goats
Studies on the genetic resistance of small ruminants to trypanosomosis are scarce. In West
Africa the Small West African Sheep called Djallonké or Fouta Djallon sheep and the West
African dwarf goats are the only recognised breeds that are trypanotolerant (FAO 1980). In
Kenya, it has been suggested that the Red Maasai and Blackhead Persian sheep and the
Galla and Small East African goats have more resistance than imported breeds (Merino sheep
and Saanen goats) (Gray et al. 1995). Katunguka-Rwakishaya et al. (1997) reported
differences in susceptibility to Trypanosoma congolense infection among three indigenous
goat breeds namely, the Kigezi, Mubende and Small East African goat. The Kigezi goat was
the most susceptible while the Small East African goat was the least susceptible. This
indicates that there is also variation in resistance to trypanosomosis between the indigenous
goats breeds. More studies are required to investigate within and between breed genetic
variation for trypanotolerance in sheep and goats.
Pigs
While a lot of attention has been put on trypanosomosis in cattle, little information is available
for pigs. Pigs, like other domestic livestock, are infected by several trypanosomes and are
therefore hosts for these causative agents that can infect the other domestic species. In West
Africa, the long-snouted West African pig breed is less susceptible to porcine trypanosomosis
than the exotic (Onah 1991). Most studies are required on this indigenous pig breed to
investigate its trypanotolerance status and its possible use both as a purebred and in
crossbreeding programmes in the tsetse infested areas of SSA.
Camels
It is only recently that camels have become the subject of more intensive and systematic
interest in connection with increasing the development potential of drylands in Africa and Asia
(Baumann and Zessin 1992). The camel is generally considered to be disease-tolerant
(Morton 1984). However information on which camel breed is resistant or tolerant to which
disease condition is lacking. There exist a large variety of distinct camel breeds that have
developed along tribal/ethnic lines (Köhler-Rollefson 1993). For example, classification of
camels in Kenya is based on the geographical distribution of the owners and ethnic group. As
a result the terms Somali, Rendille, Gabra or Turkana camel, are widely recognised (Kegode
1990). Using this classification, trypanosomosis was perceived as a major disease affecting
young stock and adult camels (Kaufmann 1998). However, trypanosomosis was ranked
highest in Somali and Rendille breeds. This outcome is not conclusive enough to indicate that
the Gabra breed is more trypanotolerant because it is not yet clear whether the camel
populations named after the camel keeping ethnic groups differ both in phenotype and
genotype. A research programme is underway at Hohenheim University, Germany, in
collaboration with ILRI, the Kenya Agricultural Research Institute (KARI) and the Food and
Agricultural Research Management (FARM)-Africa to characterise camel genetic resources in
dry lands in Africa and Asia. This programme might prove whether these breeds are
genetically homogenous or not.
Gastro-intestinal parasites
Most of the research work on resistance to endoparasites has concentrated on the nematodes
and particularly the Trichostronyles (Haemonchus, Ostertagia, Trichostrongylus, Nematodirus
and Cooperia) species. In most of the tropics, helminths constitute one of the most important
constraints to ruminant production (Fabiyi 1987). The problem is serious, as chemotherapy is
becoming less effective due to multiple resistances of parasites to chemicals and prospects for
vaccination are not encouraging. Therefore, there has been research emphasis on genetic
control of the disease.
Cattle
Surprisingly, evidence for host genetic resistance to gastro-intestinal nematodes even of
African breeds (Frisch and O'Neill 1998) is from Australia. It has generally been observed that
B. indicus cattle are more resistant to gastro-intestinal nematode parasites than B. taurus
(Turner and Short 1972; Frisch and Vercoe 1984). Comparing beef cattle breeds of African,
European and Indian origins, Frisch and O'Neill (1998) showed that the Brahman (Indian
Zebu) was more resistant to gastro-intestinal nematodes than the Boran (African Zebu).
Among the two African breeds represented in that study, i.e. the Boran and the Tuli, the Boran
had higher resistance to worms than the Tuli. The Boran is predominantly B. indicus and
originated from Ethiopia while the Tuli is the southern African Sanga and is sometimes
referred to as the African B. taurus.
Sheep and goats
There is more evidence for both between and within breed genetic variation for resistance to
gastro-intestinal nematode parasites in sheep and goats than in cattle. Breeding for resistance
has attracted considerable research and development attention all over the world. Outside
Africa, several indigenous breeds have been identified in the Caribbean (St Croix and
Barbados Blackbelly) and North America (Florida Native and Navajo) that are more resistant
or tolerant to gastro-intestinal nematodes than Rambouillets or Merinos (Gray et al. 1995),
however within breed genetic variation for resistance to gastro-intestinal have been reported in
the Merino sheep in Australia (Raadsma et al. 1998). In SSA, the indigenous Red Maasai
sheep of East Africa are resistant to naturally acquired and artificial infections with gastro-
intestinal nematode parasites and particularly to Haemonchus contortus (Baker et al. 1999).
The Red Maasai sheep has the ability to maintain good levels of production when under
severe and persistent endoparasite challenge. Based on these results, ILRI has embarked on
a programme to identify genetic markers linked to quantitative trait loci that control resistance
to infections to gastro-intestinal parasites, H. contortus in particular (Baker et al. 1999).
While sheep and goats are considered to share common susceptibility to many diseases, it
has been observed that there were significant differences in disease resistance between the
two small ruminants. In contrast to sheep, there is less information on resistance to gastro-
intestinal nematodes in goats. In East Africa, the Small East African goat has been found to
be more resistant than the exotic and the indigenous Galla goat (Shavulimo et al. 1988;
Waruiru et al. 1994; Baker 1998). The Galla breed has its origin in the semi-arid areas of East
Africa, where gastro-intestinal nematodes are not a major problem. It is therefore probable
that the Galla have had little exposure to nematodes and have therefore had no natural
selection for resistance.
Ectoparasites
The most important ectoparasite that imposes severe economic constraints on livestock
industries in SSA is tick. Ticks are responsible for losses caused through blood loss, damage
to hides, skins and udder, tick worry and the injection of toxins. They also transmit devastating
and often fatal diseases such as theileriosis, babesiosis, anaplasmosis and cowdriosis. This
brings about mortalities, reduced productivity and direct costs associated with preventive and
curative measures (Young et al. 1988). In SSA, there is widespread and indiscriminate
application of acaricide to control ticks that has resulted in emergence of acaricide-resistant
ticks (Nolan 1990). Therefore utilisation of breeds of livestock that are considered to be
resistant to ticks is important to reduce the reliance on chemicals that have negative impacts in
animal products and on pastures. Resistance is manifested by reduction in attachment and
engorgement of ticks (Fivaz and De Waal 1993).
Cattle
Mackinnon et al. (1991); Davis (1993); Frisch and O'Neill (1998) (in Australia) and Lemos et
al. (1985); Garcia et al. (1989); Gomes et al. (1989) (in South America) established that pure-
bred B. indicus cattle and their crosses with B. taurus breeds are more resistant to ticks and
tick-borne diseases than pure-bredB. taurus breeds and indicated that there exists both
between and within breed genetic variation for resistance to the one-host tick Boophilus
microplus. While the indigenous cattle breeds are resistant to ticks, it has been established in
the zebu cattle of southern Uganda that its resistance to some tick species, Rhipicephalus
appendiculatus in particular, is not strong (Kaiser et al. 1982). An evaluation of the
performance of indigenous Nguni, Bonsmara and Hereford in South Africa indicated that the
Nguni was more resistant, the Bonsmara being intermediate (Fivaz and De Waal 1993). The
Nguni breed belongs to the indigenous Sanga group, which evolved along the east coast of
Africa while the Bonsmara is a composite breed developed from crosses between the Sanga
Afrikander and exotic Shorthorn and Hereford (Fivaz and De Waal 1993). In Ethiopia, the
Boran, Barka and Horro indigenous breeds are more resistant to ticks than their crosses with
the Jersey, Friesian and Simmental breeds (Yehulashet et al. 1995). Kiltz and Humke (1986)
reported partial tolerance to theileriosis (East Coast fever) in the Ankole cattle in Burundi and
Paling et al. (1991) reported the same result in the Ankole cattle in Rwanda. This tolerance is
likely to be a result of centuries of natural selection within the Ankole cattle population,
surviving in the East Coast fever endemic areas of central Africa.
Sheep and goats
Although small ruminants are often an important part of the livestock enterprise in tick-endemic
areas, there is little information in this class of livestock on resistance to ticks and tick-borne
disease. There is however some indication that the Red Maasai sheep in East Africa is
resistant to ticks (Ogore 1996). Ogore et al. (1999) reported a simple method of assessing tick
burden in sheep.
Bursal disease
This is an infectious viral disease of poultry caused by the Birnavirus and is a disease of
considerable economic importance, particularly affecting birds between three and six weeks of
age (Bumstead et al. 1993). However resistance to this disease has been reported in the local
chicken breeds in Sudan (Salman et al. 1983) and Nigeria (Okoye et al. 1999). Bumstead et
al. (1993) reported within lines or breeds variation to susceptibility to this disease.
Heat tolerance
SSA is characterised with extreme heat and high humidity in some coastal and inland valleys
and basins. Animals under heat stress dissipate heat by increasing their respiration rate and
sweating and measurements of these parameters plus pulse rate and rectal temperature have
been used to assess the tolerance of ruminants to adverse climatic conditions (Lemerle and
Goddard 1986). Heat tolerance is one of the adaptations that contribute to the performance of
tropically derived breeds and their crosses in warm environments (Turner 1984).
Cattle, sheep and goats
Bonsma (1949) recognised differences in adaptability to the effect of heat between tropical
and temperate breeds in SSA in the first half of the twentieth century. This has been
confirmed in Australia (Turner 1982; Turner 1984), Papua New Guinea (Lemerle and Goddard
1986) and India (Singh and Bhattacharyya 1990). In Nigeria, Amakiri and Funsho (1979)
studied the rectal temperature and respiratory rates of German Brown, Friesian, German
Brown/Friesian crosses, N'Dama, White Fulani and German Brown/N'Dama crosses. The
Friesian and German Brown were found less heat tolerant than the indigenous N'Dama and
White Fulani. Use of the sweating rate as an indicator of heat tolerance indicated that the
White Fulani and N'Dama breeds were similar for heat tolerance (Amakiri and Onwuka 1980).
The superior heat tolerance of the indigenous breeds has been attributed to their coat type
and colour, skin thickness and pigmentation, high sweating capacity (which is a measure of
the density of sweat glands in the skin), low body heat production as a consequence of their
low productivity, body conformation and some physiological aspects (Bonsma 1949; Allen
1962; Allen et al. 1963; Amakiri 1974; Amakiri and Mordi 1975; Turner 1975; Herz and
Steinhauf 1978; Amakiri and Onwuka 1980).
Most indigenous cattle, sheep and goats breeds in SSA have small bodies, low levels of milk
production and slow growth rates and have fat localised in specific areas of the body. For
example, the storage of fat in the hump of zebu cattle and camels and in the tails or rumps of
many indigenous sheep (e.g. the Red Maasai sheep) leads to more heat loss than an equal
amount of fat stored under the skin. The indigenous sheep and goat breeds are smooth-
coated, which is in itself an adaptive mechanism because they lose more moisture from the
skin as a result of evaporation (Bonsma 1949).
Poultry
High ambient temperature is regarded as the most important inhibiting factor for poultry
production in hot climates (Horst 1990). Under these conditions, chickens cannot dissipate
heat produced following meals rapidly enough, leading to reduced feed intake and lower
weight gain or egg production (Cahaner and Leenstra 1992). This has lead to systematic
comparisons of naked neck and normally feathered chickens at various ambient temperatures
(Horst 1988; Mathur and Horst 1990; Cahaner et al. 1993; Ibe 1993). Several breeds exhibiting
the naked neck trait have been described, including the `Cou Nu du Forez' from France, the
`Malay Game' from Malaysia, the `Shingangadi' from Zaire, the `Trandylvanian naked neck'
from Hungary and Romania and the `Peel-Neck' from Belize and Guatemala (Merat 1986;
Mallia 1999).
Some major genes have been found relevant to the hot tropical production environment.
Among these major genes are the feather distribution gene, naked neck (Na) and the feather
structure gene, frizzle (F) (Horst 1988). These genes cause a reduction in tropical heat stress
by improving the breed's ability for convection, resulting in improved feed conversion and
better performance. The Na and F gene confer superiority in some production characters in
the tropics (Horst 1988). Horst (1988) and Mathur and Horst (1990) showed that individuals
with F and Na genes both singly and in combination were superior to those individuals with
normal feathering for egg number, egg mass/weight and forty-week body weight in tropical
environments. The Na and F individuals have been shown to mature earlier than individual
with normal feathering (Ibe 1992). Cahaner et al. (1993) showed that the heterozygous Na
broilers were superior to normal broilers at normal temperatures and were even superior at
higher temperatures. In Nigeria, quantitative genetic principles are being employed to explore
use of major genes and that of polygenes in bringing about improvement in production
characters in synthetic populations of indigenous and exotic strains (Ibe 1990; Ibe 1992).
Camel
As already alluded to, camels are the most drought-tolerant and efficient converters of feed
and water to meat and milk than any indigenous livestock species (Morton 1984). Their unique
anatomical and physiological characteristics enable them to exist, reproduce, and produce
meat and milk during periods of drought, poor grazing and low management. Camels have
begun to receive attention in recent years as far as describing their unique characteristics,
physiology and estimating their production capability as a livestock species is concerned
(Dolan et al. 1983; Schwartz et al. 1983; Morton 1984; Yagil 1985; Chowdhary 1986; Wilson
1986; Wilson 1988; Khanna et al. 1990; Schwartz 1992; Kaufmann 1998; Hülsebusch 1999).
However studies to compare different camel breeds for measure of heat tolerance are lacking.
Adaptation and nutrition
Another important attribute for surviving and producing in most of the tropics is the ability to
utilise lower quality feeds. While studies on the comparison of the indigenous livestock breeds
to their exotic counterparts are lacking in SSA, it has been indicated that indigenous zebu (B.
indicus) is better able to utilise lower quality feeds than the temperate breeds (Ashton 1962;
Moore et al. 1975; Dunkel 1981; Hunter and Siebert 1985a). However, B. taurus cattle have
higher intakes than B. indicus of good and medium quality forage diets (Hunter and Siebert
1985b). It has been suggested that the superiority of the B. indicus on low quality feeds could
be due to their superior ability to recycle endogenous nitrogen in the rumen (Hunter and
Siebert 1985b). Zebus also have higher true digestibility, more extensive ruminal digestion,
more efficient protein synthesis and lower metabolic faecal nitrogen excretion than most
temperate breeds (Ashton 1962; Moran and Vercoe 1972; Vercoe and Frisch 1972; Kennedy
1982; Hunter and Siebert 1985a; Hunter and Siebert 1985b).
The tropical Bedouin goats can maintain a high level of high fibrous feeds compared with the
temperate Saanen breed (Silanikove 1986a; Silanikove 1986b; Silanikove and Brosh 1989).
This has been attributed to their ability to maintain higher microbial activity in the particulate
matter and hence higher total ruminal fermentation rate, which are reflected in the extent and
site of absorption of digested end-products (Silanikove et al. 1993). The Bedouin goats have a
small body size and are found to thrive in extreme deserts where, with the exception of the
camel, no other domestic ruminant lives (Silanikove 1986b). There are reports from Côte
d'Ivoire of the indigenous poultry breeds having the ability to use high fibre feeds (Diambra
1990, cited in Sonaiya et al. 1999).
Other properties
Salt tolerance
There might be some relationship between adaptation to native feeds and heat with tolerance
to salts. Although not published, reports indicate that the Galla goat found in the arid and
semi-arid areas of Kenya has the ability to consume plants that contain high amounts of salts
without getting dehydrated. Normally goats have a lower water turnover than sheep and will
respond to dehydration by excreting a smaller volume of urine with higher osmolalities
(Aganga 1992). Generally in Kenya it has been observed that for grilling, consumers preferred
meat from salt tolerant goats because of its characteristic salty taste.
Scavenging ability
This is an important attribute of the indigenous poultry breeds and is the ability to find feed
from the surrounding environment. The land area available for scavenging and the distance a
flock can travel to scavenge will depend on many factors such as flock size, feed availability,
population density, agricultural activities and predators (Gunaratne 1999). Studies have shown
that indigenous birds have better scavenging ability compared to others (e.g. exotic or
crossbreds). In Kenya, for example, one of the main reasons for the failure of the National
Cockerel Exchange Program was the progeny of the exotic cockerel tended to restrict their
scavenging area close to the household where, with no supplementation, is an area with less
feed. Therefore, the productivity of the crossbred progeny was low and the farmers refused to
give up their indigenous cockerels in favour of the exotics. Roberts (1995) described a simple
model for production systems of scavenging village chickens in which the biomass of the
village flock is maximised at the capacity of the scavenging feed resource base.
Reproductive performance
A limited number of indigenous breeds exhibit exceptional reproductive ability. While there is
less information on such breeds in SSA, in China for example, the Meishan breeds of pig are
very prolific but are characterised by very poor production performance (Bidanel 1990). It has
been reported that the Mossi dwarf goat (a local caprine breed reared by smallholders in
central and eastern Burkina Faso, West Africa Sudano-Sahelian region) has excellent
reproductive performances (Tamboura et al. 1997). Unpublished reports indicate that the
twining rate in the Small East African goat is high.
1 . Note the difficulty to define breeds in developing countries as discussed in the section on
Valuation of biological diversities.
Final remarks
Final remarks
Valuation of farm animal genetic resources--prospects and possibilities
Crossbreeding with exotic breeds as a threat to farm animal biodiversity
Naive crossbreeding schemes
Structured crossbreeding programmes
Developments of synthetic breeds
Decision making
Final remarks
Valuation of farm animal genetic resources--prospects and
possibilities
Economic valuation of indigenous breeds or populations in tropical countries, especially in
SSA, has been done to a very limited extent, so that to date only few results are reported in
the literature.
There are numerous methodological approaches that can be used and different types of
values to be determined, as outlined earlier in this study.
For the purpose of decision making with respect to AnGR management, it is necessary that
comparable values are determined for different breeds. If, say, breed A has a specific indirect
use value in that it fulfils an ecological function, and this breed may be replaced through
another breed B that has a higher production value, but presumably is not able to fulfil the
ecological function to a comparable extent, the costs and benefits have to be balanced on the
same scale. This will be difficult, if, e.g. the ecological value of breed A is determined for the
country or region and the production value of breed B is expressed on a farm level.
As mentioned earlier, there are serious doubts that an overall economic valuation of breeds
will be possible with sufficient accuracy, especially, if indirect and potential future values are
considered, the latter being often used as the main argument justifying conservation activities.
A comparable, if not greater, lack of knowledge exists with respect to costs of conservation
strategies. Breed conservation is not absolute, but is related to extinction probabilities. To give
an example: at a given point of time, the probability that a certain breed will be extinct in, say,
10 years, may be 50%. Conservation activities aim at reducing these probabilities. However, it
will hardly be possible to reduce it to 0% (i.e. the conservation of the breed is guaranteed), but
the extinction probability may be reduced to, say, 30% or 20%. Different activities will result in
different extinction probabilities, but will also have different costs.
Rational conservation strategies require knowledge of costs per, say, 10% reduction of
extinction probability.
Since relevant approaches like the `safe minimum standard' concept are based more on
quantifications of costs rather than benefits, determination of the costs and the efficiency of
conservation strategies is assumed to be more important than valuation of the populations to
be conserved.
Crossbreeding with exotic breeds as a threat to farm animal
biodiversity
Crossbreeding appears to be an attractive option if productivity (in terms of output units) is to
be increased in short-term. Under tropical conditions, productivity (like milk yield per day or
lactation, daily gain, eggs per year etc.) is only one factor among many others determining the
overall economic efficiency of livestock and certainly is of much lesser impact than it is in
temperate regions. It has been shown in many studies that the price to be paid for increased
productivity is reduced adaptation and hardiness, health and fertility problems, increased
treatment costs, losses etc. Fair economic valuations of crossbreeding schemes are not
available, since in many cases real costs are masked through free provision of exotic genetic
material (sperm or live animals) and crossbred animals initially often are kept under station
conditions financed by donor organisations.
When judging crossbreeding schemes, one has to differentiate between three principal
approaches:
Naive crossbreeding schemes
These schemes are characterised through a more or less unplanned import of exotic genetic
material, which is used for one or few generations to `upgrade' local indigenous breeds.
Further characteristics are that the exotic genotypes usually are not selected to fit well to the
local populations they are merged with. Often, the costs of the import of these exotic genetics
are taken over by donors, while the consequences have to be taken by the local farmers.
A good example is the frequently practised insemination of indigenous cows with imported
Holstein semen, which often is of a quality that cannot be sold in the temperate countries of
origin.
In the starting phase of such a programme, results are promising due to heterotic superiority
of F1-animals. However, recombination loss leads to a breakdown of this superiority in
subsequent generations, so that the productive advantage is reduced, while the problems with
insufficient adaptation remain.
Typically, such schemes lead to continuous indiscriminate and uncontrolled crossbreeding, i.e.
after the termination of the project, crossbred animals remain in the country and are
disseminated and used without any plan or control. Their different phenotype and (seemingly)
higher productivity makes them attractive to the farmers, so that on the long run exotic
genotypes replace a significant proportion of the indigenous genotypes. However, the resulting
animals will have a lower adaptive capacity and increased disease susceptibility and
management and feeding conditions are in many cases not sufficient to allow them to express
their eventual productive superiority, so that an overall economic analysis in many cases
would not favour such a scheme. On the other hand, this type of indiscriminate dissemination
of exotic genes, resulting in a complex mixture of genotypes, is a severe threat to indigenous
AnGR.
Structured crossbreeding programmes
It has been shown that complex crossbreeding schemes like continuous production of F1-
animals, crisscrossing, two- or three-breed rotational crosses etc. have the capacity to
produce animals with improved productivity and sufficient adaptive capacity to be profitable
under the given environmental conditions. Such schemes imply a continuous maintenance of
the required infrastructure and logistics, i.e. pure lines have to be maintained or exotic
material has to be continuously imported (as semen, embryos or live animals).
Establishing and maintaining such schemes requires considerable infrastructure and stable
financial and marketing conditions. Therefore, this sort of sustainable crossbreeding schemes
will in many cases be limited to special situations, like semi-intensive milk production in peri-
urban regions. In this case, the threatening of the biodiversity of the local breed is limited. If,
however, such schemes are used to produce `improved' crossbred products to be
disseminated as breeding animals to smallholders to be mated to local females, the risk of
indiscriminate use and therefore considerable adverse effects on biodiversity are to be
expected.
Developments of synthetic breeds
Synthetic breeds development from a systematic cross of local and exotic breeds has been
shown to be a promising strategy to combine desired properties from both sources. If the
products are selected under typical environmental conditions, the optimal rate of productivity
under the given constraints can be achieved in combination with sufficient robustness and
hardiness. Typically, such composite breeds go through a genetic bottleneck since they are
derived from a limited number of founder animals. Therefore, within breed diversity is limited
and a relatively high level of inbreeding is programmed.
Development of synthetics is a multi-generation project, which requires substantial input over
a long time period. It is likely for that reason that most existing synthetics in the tropics have
been developed in regions like tropical northern Australia, South Africa or Latin America,
where both the required infrastructure and the funding are easily accessible than in most of
the less developed countries of SSA.
Once synthetics are released, they compete on the market with indigenous breeds and may
replace them to a certain extent. It must however be realised that these animals have
relatively high prices and only perform well under good management and feeding conditions,
so the risk of a complete replacement of indigenous genotypes is limited.
Crossbreeding certainly is one elementary technique to improve productivity and has its
justification to improve the farm animal productivity under tropical conditions. Situations must
be avoided, where crossbreeding does not achieve its goal (i.e. improvement of productivity
under the given conditions) but still is executed with harmful effects on farm animal
biodiversity. This is usually the case, if exotic crosses are made accessible to the farmers
without sufficient training, control and monitoring. Already the fact that these animals look
different, may be larger, or may perform well on station, makes them attractive, although they
or their offspring may not be able to perform well or even to survive under the smallholder
conditions.
Biodiversity is not static. Breeds have always been newly developed and others have
disappeared. It is a matter of the speed of such processes, and it is a matter of rationality,
especially it must be avoided that well adapted genotypes are replaced by genotypes that are
not competitive under the given conditions. However, it must also be accepted that adapted
breeds are improved through well-planned crossing and upgrading. There is a certain trade-off
between the goals of conservation and improvement, but many examples have shown that
sustainable improvement of breeds is hardly possible without conservation of the most
important characteristics of indigenous breeds.
Decision making
Decisions with respect to AnGR management and conservation have to take into account a
variety of implications, e.g. development in the region, effects on the livelihood of
stakeholders, socio-economic impacts etc. It was clearly beyond the scope of this study to
analyse and discuss all the implications of AnGR management in the different areas, but it is
obvious that economic considerations are only a part of the whole framework under which
AnGR management has to be discussed. However, all decisions have an immanent economic
component, in that present or future costs are involved and/or actual or future values are
preserved or sacrificed.
It is difficult to assess these economic aspects (in the narrow sense) with sufficient accuracy,
so that rational decision making on the base of objective data and economic valuation results
under the present conditions in SSA1 must be seen as a major challenge.
The problems arising in this context may be illustrated with the following examples:
There is no obvious and clear definition of the goals to be achieved through AnGR
management activities.
The scope of decision making is limited. It is difficult to imagine a negative decision with
respect to the conservation of a breed, i.e. to decide that the extinction of a certain
breed will be accepted. The real alternative rather is either to become active with
respect to certain breeds, or to do nothing about these breeds, which may have the
consequence of extinction or replacement for them.
Most of the theoretical work on AnGR management assumes well-established breeds
and breeding infrastructure. These conditions are not given for most farm animal species
in SSA. It must be doubted that this will be changed in the foreseeable future.
Scientists questioned the important components of the total economic value, e.g. the
option value reflecting the `insurance' function allowing adaptation to future changes,
can ever be quantified with sufficient accuracy. Therefore, total economic value
comparisons will always be affected by a considerable amount of uncertainty.
Apart from the approach suggested by Weitzman (1993), which is aimed at the problem
of conservation of biodiversity between species, there is hardly any conceptual work on
how to decide which animal groups (breeds, strains etc.) should be preserved.
Taking these problems and methodological shortcomings into account, approaches gain
attractiveness which are suited to decision making under high levels of uncertainty.
One such approach may be to adopt the concept of `safe minimum standards', which mainly
requires decision making with respect to finding and implementing the most cost-effective
strategy to prevent total extinction of a certain breed or population.
This is based on the assumption that a priori every population should be conserved, if
possible, since it may have a future value which, however, is largely unknown by definition. In
such a case, decisions are required with respect to, e.g.
the exact unit to be conserved (definition of the breed or strain, exclusion of crosses
etc.);
the conservation scheme to be implemented (in situ or ex situ, number of animals,
monitoring system etc.);
the possibility of combination of this activity with other activities (i.e. improvement of the
breed, health services etc.).
Of course, there will be costs for the implementation of such schemes, and, given that
financial resources are limited, it will in most cases be necessary to set up a priority list of
activities and breeds to be conserved. This, again, is a decision-making process that should,
in principle, be based on objective data and information. However a fast and inexpensive (but
may be suboptimal) decision will in many cases be adequate and tolerable. It may, e.g. in
many cases be sufficient to ask experts with good knowledge of the populations in a region to
set up a ranked list of breeds with highest priority to be conserved.
It also must be doubted, whether a scientifically `sound' description and valuation of the actual
biodiversity is always required, which is both expensive and demanding with respect to time
and manpower to be invested. Economic resources are limited, and there is a competition
between funds to be used either for description or for conservation activities. A considerable
proportion of biodiversity may already be lost at the time the respective populations are
described and valuated accurately.
1 . In fact, the situation with respect to sufficient knowledge of population structures, breed
values and conservation costs is not much better in more developed regions of the world.
Conclusions
Conclusions
Farm animal genetic resources are subject to conservation efforts, because we assumed that
they have a value. This value may be of very different nature (productive, ecological, aesthetic
etc.) and may be realised at different times (past, present and future).
Management of AnGR has to be based on a rational decision-making framework, part of
which is the definition of conservation priorities and development of tools for the optimal
allocation of resources. For this, it is essential to assess economic values of genetic
resources.
Environmental economies have suggested different types of values for species or breeds. Use
values are related to actual productive and consumptive functions and non-consumptive
aspects like aesthetic, cultural or ecological functions. Option values are use values to be
realised in the future, e.g. the potential of a breed to satisfy market requirements that may
arise in the future, caused, e.g. by climatic changes etc. Non-use values are related, e.g. to
the contribution of exotic breeds to the understanding of biological mechanisms.
Actual use values are most relevant for the role a breed plays and whether it becomes
threatened to be extinct (e.g. if its productivity is clearly inferior to other breeds, which are
comparably viable in the given environmental conditions otherwise). Option values are
relevant for conservation priorities, since it is most sensible to maintain breeds that will have
highest use values in the future.
Different types of values can be quantified with different accuracy. It is easier to quantify
values that are realised in the present than those values that may be realised in the future,
and it is easier to assess values for traits or animals that are traded on markets compared to
traits (e.g. ecological functions) for which no market exists.
aluation methodologies can be subdivided into empirical approaches and system analysis
approaches. Empirical approaches are based either on the analysis of real market data (paid
prices) or survey based, trying to assess `willingness to pay' for animals, traits or functions.
The latter approach, usually called `contingent valuation' allows assessing values for
`hypothetical' traits or for functions, for which no market exists.
System analysis approaches try to model a complex system (farm, household, sector, country,
market etc.) by defining detailed relations and functions between all relevant elements. This
allows to assess the effect of specific changes (e.g. the replacement of one breed through
another breed) on the productivity of the studied system.
Decision making can be made in a cost-benefit framework, trying to assess all the costs and
benefits arising from certain decisions, and maximising the expected economic return. Since it
is difficult to quantify all values (and costs) with sufficient accuracy and since especially option
values underlie considerable immanent uncertainties, alternative concepts like the safe
minimum standard concept have been suggested to avoid these problems.
The safe minimum standard approach is based on the assumption that a loss of a breed
should be avoided and tries to find a strategy that guarantees conservation with minimum
costs. Genetic approaches try to set up least cost conservation schemes that limit the loss of
genetic diversity to an acceptable rate. An important and necessary feature of such schemes
is robustness, i.e. that they still should prevent a total loss even if the assumptions they are
built on are not entirely met.
Decision making and priority setting for conservation activities requires well defined objectives.
In the AnGR context, these objectives are neither specified nor discussed with sufficient
rigidity. There is no uniformly accepted measure of diversity, nor is there any formal concept
how to value diversity and how to combine its value with, e.g. production and option values.
In environmental economics, Weitzman (1992); Weitzman (1993) suggested a general
concept of diversity, which integrates information on structural diversity (in form of a distance
matrix), extinction probabilities and cost efficiencies of different conservation activities to an
overall operational framework for decision making and priority setting. Adopting this
framework for the farm animal situation would require both methodological developments (e.g.
Weitzman's approach is designed for species rather than for breeds) and empirical derivation
of a large variety of values and costs. However, this approach is the only available formal
framework that appears suitable for the desired implementation.
There are only very few studies available that have actually derived values of AnGR. ILRI has
now initiated a major project to conduct a series of such studies (with focus on SSA) to
evaluate the usefulness of different valuation techniques in the farm animal context, since
most of these techniques have been developed and used for valuation of non-agricultural
species or habitats.
These activities will reveal mainly quantifications of actual use values on the basis of empirical
studies. This may be useful for situations where, e.g. breeds are threatened due to market
distortions and externalities (e.g. due to highly subsidised import of semen of exotic cattle
breeds), but it will be of limited use for the assessment of the diversity value of a breed.
Diversity is a feature of a set of breeds, not of a single breed (a single breed has a
quantifiable contribution to the species' diversity, though). Therefore, decision making at the
local level (e.g. on conservation activities for a breed on a national level) has to take into
account implications related to the global level (e.g. conservation of the diversity within a
species with breeds distributed over a whole continent or worldwide). To achieve this, different
aspects have to be considered:
A clear framework for priority setting and decision making has to be agreed upon,
including the definition of objectives.
Information has to be provided and shared to allow the operational implementation of
such a framework.
Rights and duties of all relevant stakeholders have to be defined, including intellectual
property rights (IPR) and ownership issues in relation to animal genetic resources.
Balancing mechanisms have to be implemented for situations, where costs and benefits
arise asymmetrically, e.g. a national conservation scheme conserves with high cost an
actually unproductive breed since this breed contributes highly to the global within
species diversity.
Education and training is essential especially on the level of decision makers to motivate
them to co-operate and to take the `global' view rather than the `local' one.
Weitzman (1992), Weitzman (1993) suggested the most promising (and, actually, also
the only available) conceptual framework, which, however has to be adapted to the
specific conditions given in the farm animal situation. To implement this concept, the
following input data are required:
Quantification of between and within breed genetic variability and pair-wise breed
distances (for African cattle breeds mostly available from ILRI studies, for other species
and other regions to be determined).
Quantification of the relative risk of extinction of different breeds and development of
functional relationship, e.g. to effective population size or inbreeding level.
Derivation of costs related to efficiency (in terms of loss of genetic diversity) of different
conservation activities under the conditions in SSA.
Empirical assessment of values of different breeds and breed characteristics, especially
those `unique' characteristics concerning adaptive capacity, disease tolerance or
resistance etc., which form a major component of the option values of indigenous
breeds.
Valuation techniques are available and their usefulness in the farm animal context will be
determined. Given the urgency of starting efficient conservation activities now,1 and given the
limited resources in this area, research priorities have to be considered and discussed very
pragmatically. In many cases, rough estimates that are available today will be more useful
than exact scientific figures available in two years time.
The main conclusions of this study can be summarised in the following theses:
Economic valuation of indigenous breeds is an important part of the description of
AnGR.
Important components of these values, especially non-use and option values, are
extremely difficult to measure and likely cannot be determined with sufficient accuracy.
Many of the factors determining the value of indigenous populations are of a qualitative
nature and are biologically not well understood, so that it is difficult to quantify their
economic impact.
The contribution of a breed to the biodiversity within its species (its biodiversity value)
may be very different from its actual economic value.
Therefore, decision making should not be exclusively based on analytical cost-benefit
considerations.
Alternative approaches are based on the assumption that as much AnGR as possible
should be conserved and focus on safe conservation with minimum cost.
Conservation activities should be designed to be robust, i.e. to prevent total loss of a
breed even if the assumed conditions the design is based upon are not met.
Rational decision making requires empirical knowledge of cost and efficiency of different
conservation and management strategies under the conditions in SSA, which is not
available to date.
Priority should be given to conservation of the most endangered breeds with highest
biodiversity value. However, in many cases immediate activity will be more important
than accurate quantification of all values, genetic distances etc.
Establishing IPR for genetic resources can be a means for generating markets, which
may lead to incentive effects beneficial for AnGR management.
AnGR management should be an integrated objective of all activities aiming at, or
dealing with livestock-related structures in the region.
Crossbreeding can be an effective means to improve productivity while maintaining
adaptive capacity under certain conditions. However, it poses a considerable threat on
indigenous breed biodiversity. Therefore, consequences for AnGR should always be one
aspect to be considered in a project introducing exotic genetic material into the region.
 1 . According to FAO (Scherf 2000) about one farm animal breed is lost per week on the
global scale!
References
References
Adebambo O.A. 1995. Selection response for higher body weight in Nigeria's indigenous
crossbred pigs. Nigerian Journal of Animal Production 22:21-27.
Aganga A.A. 1992. Water utilisation by sheep and goats in northern Nigeria. World Animal
Review 73:9-14.
Alba J. 1972. Productivity of native and exotic pig breeds in Latin America. World Animal
Review 1-4:25-28.
Allen T.E. 1962. Responses of Zebu, Jersey and Zebu × Jersey crossbred heifers to rising
temperature with particular reference to sweating. Australian Journal of Agricultural Research
13:165-179.
Allen T.E., Pan Y.S. and Hayman R.H. 1963. The effect of feeding on evaporative heat loss
and body temperature in Zebu and Jersey heifers. Australian Journal of Agricultural Research
14:580-593.
Amakiri S.F. 1974. Sweat gland measurements in some tropical and temperate breeds of
cattle in Nigeria. Animal Production 18:285-291.
Amakiri S.F. and Funsho O.N. 1979. Studies of rectal temperature, respiratory rates and heat
tolerance in cattle in the humid tropics. Animal Production 28:329-335.
Amakiri S.F. and Mordi R. 1975. The rate of cutaneous evaporation in some tropical and
temperate breeds of cattle in Nigeria. Animal Production 20:63-68.
Amakiri S.F. and Onwuka S.K. 1980. Quantitative studies of sweating rate in some cattle
breeds in a humid tropical environment. Animal Production 30:383-388.
Amar P.S., Singh B. and Poonia J.S. 1998. Genetic divergence among different grades of
sheep. Proceedings of the 6th World Congress on Genetics Applied to Livestock Production
24:67-70.
Ashton G.C. 1962. Comparative nitrogen digestibility in Brahman, Brahman × Shorthorn,
Africander × Hereford and Hereford steers. Journal of Agricultural Science 58:333-342.
Asiedu F.H.K. and Weever W. 1993. Growth rate and egg production of Creole and Rhode
Island Red × Creole fowls. Tropical Animal Health and Production 25:111-117.
Asuquo B.O., Okon B. and Ekong A.O. 1992. Quality parameters of Isa-Brown and Nigerian
local chicken egg. Nigerian Journal of Animal Production 19:1-5.
Ayeni J.S.O. 1983. Studies of grey breasted Helmet Guinea fowl (Numida meleagris galeata
Pallas) in Nigeria. World Poultry Science Journal 59:143-151.
Ayorinde K.L. 1989a. Carcass yield and chemical composition of four indigenous Guinea fowl
varieties at different areas. Bulletin of Animal Health and Production in Africa 37:361-366.
Ayorinde K.L. 1989b. External characterisation of four indigenous helmeted Guinea fowl
varieties. Nigerian Journal of Animal Production 16:47-54.
Ayorinde K.L. 1991. Guinea fowl (Numida meleagris) as a protein supplement in Nigeria.
World's Poultry Science Journal 47:21-26.
Ayorinde K.L., Oluyemi J.A. and Ayeni J.S.O. 1988. Growth performance of four indigenous
Helmeted Guinea fowl varieties (Numida meleagris galeata Pallas) in Nigeria. Bulletin of
Animal Health and Production in Africa 36:356-360. FAO (Food and Agricuture Organization
of the United Nations), Rome, Italy.
Baker R.L.1998. Genetic resistance to endoparasites in sheep and goats. 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 2413-30.
Baker R.L., Mwamachi D.M., Aduda E.O. and Thorpe W. 1999. Genetic resistance to gastro-
intestinal nematode parasite in Red Maasai, Dorper and Red Maasai × Dorper ewes in the
sub-humid tropics. Animal Science 69:335-344.
Baumann M.P.O. and Zessin K.H. 1992. Productivity and health of camels (Camelus
dromedarius) in Somalia: Associations with Trypanosomosis and Brucellosis. Tropical Animal
Health and Production 24:145-156.
Bidanel J.P. 1990. Potential use of prolific Chinese breeds in maternal lines of pigs.
Proceedings of the 4th World Congress on Genetics Applied to Livestock Production 15:481-
484.
Bishop R.C. 1978. Endangered species and uncertainty: The economics of a safe minimum
standard. American Journal of Agricultural Economics 60:10-18.
Bishop R.C. 1979. Endangered species, irreversibilities, and uncertainty: A reply. American
Journal of Agricultural Economics 61:376-379.
Bishop R.C. 1980. Endangered species: An economic perspective. Transactions of the North
American Wildlife and Nature Conference 45:208-218
Bonsma J.C. 1949. Breeding cattle for increased adaptability to tropical and subtropical
environments. Journal of Agricultural Science 39:204-212.
Boujenane I. and Chafik A. 1994. Heterosis retention in advanced generations of crosses
among D'man and Sardi Sheep. Proceedings of the 5th World Congress on Genetics Applied
to Livestock Production 18:75-78.
Brown G. and Henry W. 1989. The economic value of elephants. London, Environmental
Economics Center, Paper 89-12. London, UK.
Bumstead N., Reece R.L. and Cook J.K.A. 1993. Genetic difference in susceptibility of
chicken lines to infection with infectious Bursal disease virus. Poultry Science 72:403-410.
Bushnell P. and Hutton J. 1982. Better use of pasture with Finnish Landrace cross.
Agricultural Economics 3:7-9.
Cahaner A. and Leenstra F. 1992. Effects of high temperature on growth and efficiency of
male and female broilers from lines selected for high weight gain, favourable feed conversion,
and high or low fat content. Poultry Science 71:1237-1250.
Cahaner A., Deeb N. and Gutman M. 1993. Effects of the plumage-reducing naked neck (Na)
gene on the performance of fast-growing broilers at normal and high ambient temperatures.
Poultry Science 72:767-775.
Chantalakhana C. 1998. Role of exotic breeds in dairy and beef improvement in Asia.
Proceedings of the 6th World Congress on Genetics Applied to Livestock Production 25:213-
222.
Charray J., Humbert J.M. and Levif J. 1992. Manual of sheep production in the humid tropics
of Africa. CAB (Commonwealth Agricultural Bureaux) International, Wallingford, UK. 187 pp.
Chiboka O. 1981. The effect of age at first mating on litter characteristics in the native Nigerian
pig. Livestock Production Science 8:155-159.
Chowdhary B. 1986. Some important biological and production characters of the Bikaneri
camel. Indian Journal of Animal Production and Management 2:145-151.
Ciriacy-Wantrup S.V. 1952. Resource : economies and policies. University of California Press,
Berkeley, USA.
Clark C.W. 1973. Profit maximization and the extinction of animal species. Journal of Political
Economy 81:950-961.
Crowards T.M. 1998. Safe minimum standards: Costs and opportunities. Ecological
Economics 25:303-314.
Cunningham E.P. 1981. Selection and crossbreeding strategies in adverse environments. In:
Animal Genetic Resources Conservation and Management. FAO Animal Production and
Health Paper 24. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy.
pp. 279-288.
Dan T.T. and Summers P.M. 1996. Reproductive performance of sows in the tropics. Tropical
Animal Health Production 28:247-256.
Davis G.P. 1993. Genetic parameters for tropical beef cattle in Northern Australia: A review.
Australian Journal of Agricultural Research 44:179-198.
Devendra C. and Burns M. 1970. Goat production in the tropics. Technical Communication 19.
Commonwealth Bureau of Animal Breeding and Genetics, Farnham, UK. 184 pp.
Devendra C. and McLeroy G.B. 1982. Goat and sheep production in the tropics. Longman,
Harlow, Essex, UK. 271 pp.
D'Ieteren G.D.M., Authie E., Wissocq N. and Murry M. 1998. Trypanotolerance, an option for
sustainable livestock production in areas at risk from trypanosomosis. Revue Scientifique et
technique (Office international des épizooties) 17:154-175.
Dolan R., Wilson A.J., Newson R.M. and Field C.R. 1983. Camel production in Kenya and it's
constraints. II. Tick infestation. Tropical Animal Health and Production 15:179-185.
Drucker A.G., Gomez V. and Anderson S. 2001. The economic valuation of farm animal
genetic resources: A survey of available methods. Ecological Economics. 36:1-18
Dunkel R. 1981. Animal production using nutrient-deficient fodders in the tropics and
subtropics. Animal Research and Development 13:32-39.
Dzakuma J.M., Risch E., Johnson P.M. and Johnson B.M. 1998. Meat goat breeding research
at the international dairy goat research center, revisited. Proceedings of the 6th World
Congress on Genetics Applied to Livestock Production 24:125-128.
Ebangi A.L. and Ibe S.N. 1994. Heritabilities of and genetic correlations between some growth
traits in Nigerian local chickens. Nigerian Journal of Animal Production 21:19-24.
Eding J.H. and Laval G. 1998. Measuring the genetic uniqueness in livestock. In: Oldenbroek
J.K. (ed), Genebanks and the conservation of farm animal genetic resources. ID-DLO,
Lelystad, The Netherlands. pp. 33-58.
Ehrenfeld D. 1981. The arrogance of humanism. Oxford University Press, New York, USA.
286 pp.
Ehrenfeld D. 1986. Thirty million cheers for diversity. New Scientist 110:38-43.
El Aich A. and Waterhouse A. 1999. Small ruminants in environmental conservation. Small
Ruminant Research 34:271-287.
Essien A.I. and Ekanem E.O. 1990. Effects of the age of layers, rate of egg production, and
time of oviposition on some shell quality indices in a humid tropical environment. Beiträge Zur
Tropischen Landwirtschaft und Veterinärmedizin 28:457-462.
Evenson R., Gollin D. and Santianello V. (eds). 1998. Agricultural values of plant genetic
resources. CAB (Commonwealth Agricultural Bureaus) International, Wallingford, UK. pp. 285.
Fabiyi J.P. 1987. Production losses and control of helminths in ruminants of tropical regions.
International Journal of Parasitology 17:435-442.
Falconer D.S. 1984. Einfuehrung in die Quantitative Genetik. Stuttgart, Ulmer. pp. 472
FAO (Food and Agriculture Organization of the United Nations). 1980. Trypanotolerant
livestock in West and Central Africa. Vol. 1-General study. FAO Animal Production and Health
Paper 20/1. FAO, Rome, Italy.
Fisher A. and Hahnemann M. 1984. Option values and the extinction of species. Working
paper No. 269. Giannini Foundation of Agricultural Economics, Berkeley, California, USA. 39
pp.
Fivaz B.H. and De Waal D.T. 1993. An evaluation of strategic and short interval tick control in
indigenous exotic and crossbred cattle. Tropical Animal Health and Production 25:19-28.
Frisch J.E. and Vercoe J.E. 1984. An analysis of growth of different cattle genotypes reared in
different environments. Journal of Agricultural Science (Cambridge) 103:137-153.
Frisch J.E. and O'Neill C.J. 1998. Comparative evaluation of beef cattle breed of African,
European and Indian origins. 2. Resistance to cattle ticks and gastrointestinal nematodes.
Animal Science 67:39-48.
Gandini G.C. and Oldenbroek J.K. 1999. Choosing the conservation strategy. In: Oldenbroek
J.K. (ed), Genebanks and the conservation of farm animal genetic resources. ID-DLO,
Lelystad, The Netherlands. pp. 11-32.
Garcia M., Huanca W., and Chavez A. 1989. Body weight, tick burden (Boophilus microplus),
physiological parameters and reproductive efficiency of crossbred zebu cattle. Acta. vet.
scand. 30:347-353.
Gatenby R.M. 1986. Sheep production in the tropics and sub-tropics. Longman, London, UK.
Gomes A., Honer M.R., Schenk M.A.M. and Curvo J.B.E. 1989. Populations of the cattle tick
(Boophilus microplus) on purbred Nellore, Ibage, and Nellore x European crossbreds in the
Brazilian savanna. Tropical Animal Health and Production 21:20-24.
Goonewardene L.A., Sahaayaruban P., Rajamahendran R. and Rajaguru A.S.B. 1984. A study
of some production traits among indigenous pigs in Sri Lanka and its crosses with improved
white breeds. World Review of Animal Production 20:45-48.
Gray G.D., Woolaston R.R. and Eaton B.T. 1995. Breeding for resistance to infectious
diseases of small ruminants. Australian Centre for International Agricultural Research,
Monograph No. 34, Canberra, Australia, 322 pp.
Guèye H.F. 1998. Village egg and fowl meat production in Africa. World's Poultry Science
Journal 54:73-86.
Gunaratne S.P. 1999. Feeding and nutrition of scavenging village chickens. First INFPD/FAO
Electronic conference on family poultry; free communication 2, 9 pp.
Hammond K. and Leitch H.W. 1998. Genetic resources and the global programme for their
management. In: Rothschild M.F. and Ruvinsky A. (eds), The genetics of the pig. CAB
(Commonwealth Agricultural Bureaux) International, Oxon, Wallingford, UK. pp. 405-426.
Herz A. and Steinhauf D. 1978. The reaction of domestic animals to heat stress. Animal
Research and Development 7:7-38.
Hohenboken W.D. 1986. Costs and benefits of breed utilization strategies in sheep.
Proceedings of the 3rd World Congress on Genetics Applied to Livestock Production 9:509-
522.
Horst P. 1988. Native fowl as reservoir for genomes and major genes with direct and indirect
effects on productive adaptability. Proceedings of 28th World's Poultry Congress. pp. 99-
105.World's Poultry Science Association, Nagayo, Japan.
Horst P. 1990. Problems and perspectives of small scale poultry development. Entwicklung +
laendlicher Raum. 4/90:7-10.
Hoste C. 1987. Trypanotolerant livestock and African animal trypanosomiasis. World Animal
Review 62:41-50.
Hülsebusch C. 1999. Immunoglobulin G status of Camels during 6 months post natum. Ph.D.
dissertation, Hohenheim University, Germany.
Hunter R.A. and Siebert B.D. 1985a. Utilisation of low-quality roughage by Bos taurus and
Bos indicus cattle. 1. Rumen digestion. British Journal of Nutrition 53:637-648.
Hunter R.A. and Siebert B.D. 1985b. Utilisation of low-quality roughage by Bos taurus and
Bos indicus cattle. 2. The effect of rumen-degradable nitrogen and sulphur on voluntary food
intake and rumen characteristics. British Journal of Nutrition 53:649-656.
Ibe S.N. 1990. Utilising local poultry gene resources in Nigeria. Proceedings of the 4th World
Congress on Genetics Applied to Livestock Production 14:51-53.
Ibe S.N. 1992. Incorporating adaptability genes in poultry breeding programmes in Nigeria.
Proceedings of the 19th World's Poultry Congress 2:64.
Ibe S.N. 1993. Growth performance of normal, frizzle and naked-neck chicken in a tropical
environment. Nigerian Journal of Animal Production 20:25-31.
Ibe S.N. 1995. Repeatability of growth traits in Nigerian local chickens using early records.
Nigerian Journal of Animal Production 22:5-7.
Ihemelandu E.C. and Nwosu C.C. 1986. Comparison of allometric growth of different organs in
Nigerian native chicken and an exotic. World Review of Animal Production 22:9-12.
ILCA (International Livestock Centre for Africa). 1979. Trypanotolerant livestock in West and
Central Africa. Volume 1. General Study. Monograph 2. ILCA, Addis Ababa, Ethiopia. 148 pp.
Ilori J.O., Adegbola A.A., Adeyanju S.A., Omole T.A. and Akpomudje G.O. 1976. The
performance of local and exotic pigs on restricted feed intake in the tropics. Nutrition Reports
International 14:133-137.
Jabbar M.A. 1998. Buyer preferences for sheep and goats in southern Nigeria: A hedonic
price analysis. Agricultural Economics 18:21-30.
Jabbar M.A., Swallow B.M., d'Ieteren G.D.M., Busari A.A. 1998. Farmer preferences and
market values of cattle breeds of West and Central Africa. Journal of Sustainable Agriculture
12:21-47.
Kahi A.K., Kosgey I.S., Cardoso V.L. and van Arendonk J.A.M. 1998. Influence of production
circumstances and economic evaluation criteria on economic comparisons of breeds and
breed crosses. Journal of Dairy Science 81:2271-2279.
Kahi A.K., Thorpe W., Nitter G., van Arendonk J.A.M. and Gall C.F. 2000a. Economic
evaluation of crossbreeding for dairy production in a pasture based production system in
Kenya. Livestock Production Science 65:167-184.
Kahi A.K., Thorpe W., Nitter G. and Baker R.L. 2000b. Crossbreeding for dairy production in
the lowland tropics of Kenya. I. Estimation of individual crossbreeding effects on milk
production and reproductive traits and on cow live weight. Livestock Production Science
63:39-54.
Kahi A.K., Nitter G., Thorpe W. and Gall C.F. 2000c. Crossbreeding for dairy production in the
lowland tropics of Kenya. II. Prediction of performance of alternative crossbreeding strategies.
Livestock Production Science 63:55-63.
Kaiser M.N., Sutherst R.W. and Bourne A.S. 1982. Relationship between ticks and Zebu cattle
in southern Uganda. Tropical Animal Health and Production 14:63-74.
Katule A.M. 1991. Poultry production: Genetic improvement of fowls for small scale producers.
Entwicklung + laendlicher raum 5:26-29.
Katule A.M. 1992. Crossbreeding as a tool for genetic improvement of chickens in sub-optimal
environments. Bulletin of Animal Health and Production in Africa 41:71-80.
Katule A.M. and Mgheni M. 1990. Performance of crosses between exotic and local
Tanzanian chicken. Proceedings of the 4th World Congress on Genetics Applied to Livestock
Production 15:62-64.
Katunguka-Rwakishaya E., Murray M. and Holmes P.H. 1997. Susceptibility of three breeds of
Ugandan goats to experimental infection with Trypanosoma congolense. Tropical Animal
Health and Production 29:7-14.
Kaufmann B. 1998. Analysis of pastoral camel husbandry in northern Kenya. PhD Diss.
Hohenhein University, Germany.
Kegode J.M. 1990. Preweaning performance and mortality of camel calves on two commercial
ranches in Kenya. MSc thesis. University of Nairobi, Kenya.
Kennedy P.M. 1982. Ruminal and intestinal digestion in Brahman crossbred and Hereford
cattle fed alfalfa or tropical pasture hay. Journal of Animal Science 55:1190-1199.
Khanna N.D., Tandon S.N. and Rai A.K. 1990. Breeding parameters of Indian camels. Indian
Journal of Animal Sciences 60:1347-1354.
Kiltz H.H. and Humke R. 1986. Bovine Theileriosis in Burundi: Chemotherapy with
halofuginone lactate. Tropical Animal Health and Production 18:139-145.
King J.M., Sayers A.R., Peacock C.P. and Kontrohr E. 1984. Maasai herd and flock structures
in relation to livestock wealth, climate and development. Agricultural Systems 13:21-56.
Kitalyi A.J. 1999. Family poultry management systems in Africa. First INFPD/FAO electronic
conference on family poultry. INFBD (International Network for Family Poultry Development).
Knights M. and Garcia G.W. 1997. The status and characteristics of the goat (Capra hircus)
and its potential role as a significant milk producer in the tropics: A review. Small Ruminant
Research 26:203-215.
Köhler-Rollefson I. 1993. About camel breeds: A reevaluation of current classification
systems. Journal of Animal Breeding and Genetics 110:66-73.
Köhler-Rollefson I. 1997. Indigenous practices of animal genetic resource management and
their relevance for the conservation of domestic animal diversity in developing countries.
Journal of Animal Breeding and Genetics 114:231-238.
Lebbie S.H.B. and Ramsay K. 1999. A perspective on conservation and management of small
ruminant genetic resources in the sub-Saharan Africa. Small Ruminant Research 34:231-247.
Lebbie S.H.B., Yapie-Gnaore C.V., Rege J.E.O. and Baker R.L. 1996. Current developments
in the management of small ruminant genetic resources in sub-Saharan Africa. In: Devendra
C. (ed), Conservation and use of animal genetic resources in Asia and the Pacific. Proc.
IGA/FAO round table on glob. managem. of small rum. genet. res. Beijing, China. pp. 33-45.
Lemerle C. and Goddard M.E. 1986. Assessment of heat stress in dairy cattle in Papua New
Guinea. Tropical Animal Health and Production 18:232-242.
Lemos A.M., Teodoro R.L., Oliveira G.P. and Madalena F.E. 1985. Comparative performance
of six Holstein-Friesian × Guzera grades in Brazil. 3. Burdens of Boophilus microplus under
field conditions. Animal Production 41:187-191.
Lloyd-Jones O. 1915. What is a breed? Journal of Heredity 6:531.
Mackinnon M.J., Meyer K. and Hetzel D.J.S. 1991. Genetic variation and covariation for
growth, parasite resistance and heat tolerance in tropical cattle. Livestock Productioin Science
27:105-122.
Mackinnon M.J., Thorpe W., Baker R.L. 1996. Sources of genetic variation for milk production
in a crossbred herd in the tropics. Animal Science 62:5-16.
Madalena F.E. 1989. Cattle breed resource utilization for dairy production in Brazil. Proc.
International Symposium on Utilization of Animal Genetic Resources in Latin-America, Revista
Brasileira de Genetica 12 (Suppl.): 183. Sociedade Brasileira de Genética, Ribeirão Preto SP-
Brazil
Madalena F.E., Lemos A.M., Teodoro R.L., Barbosa R.T. and Monteiro J.B.N. 1990. Dairy
production and reproduction in Holstein-Friesian and Guzera crosses. Journal of Dairy
Science 73:1872-1886.
Mallia J.G. 1999. The peel-neck chicken of Belize and Guatemala, central America. Animal
Genetic Resources Information Bulletin. 25:71-76. FAO (Food and Agriculture Organization of
the United Nations), Rome, Italy.
Mason I.L. 1980. Sheep and goat production in the drought polygon of Northeast Brazil. World
Animal Review 34:23-28.
Mathur P.K. and Horst P. 1990. Single and combined effects of tropically relevant major genes
on performance of layers. Proceedings of the 4th World Congress on Genetics Applied to
Livestock Production 16:65-68.
McCarthy W.O. and Hodgson N.P. 1970. Comparative performance of beef breeds in Central
Queensland. II. Financial aspects. Tropical Agriculture 47:51-63.
McDowell R.E. 1985. Crossbreeding in tropical areas with emphasis on milk, health and
fitness. Journal of Dairy Science 68:2418-2435.
McDowell R.E. and Jones R.G 1972. Improvement of livestock production in warm climates.
Freeman & Company, San Francisco, USA. pp. 711.
Mendelsohn R. 1999. Economic valuation of animal genetic resources: A synthesis report of
the planning workshop. In: Rege J.E.O. (ed), Economic valuation of animal genetic resources.
Proceedings of an FAO/ILRI workshop held at FAO headquarters, Rome, Italy, 15-17 March
1999. ILRI (International Livestock Research Institute), Nairobi, Kenya. pp. 3-13.
Merat P. 1986. Potential usefulness of the Na (Naked Neck) gene in poultry production. World
Poultry Science Journal 42:124-142.
Meuwissen T.H.E. 1998. Operation of conservation schemes. In: Oldenbroek J.K. (ed),
Genebanks and the conservation of farm animal genetic resources. ID-DLO Lelystad, The
Netherlands. pp. 91-112.
Mohanty S. and Nayak J.B. 1986. Reproductive performance of Large White Yorkshire pigs
and their crosses with indigenous pigs in hot-humid climate. Indian Journal of Animal
Production and Management 2:134-137.
Moore R.L., Essig H.W. and Smithson L.J. 1975. Influence of breeds of beef cattle on ration
utilisation. Journal of Animal Science 41:203-207.
Moran J.B. and Vercoe J.E. 1972. Some factors affecting apparent nitrogen digestibility of
roughage diets fed to cattle. Journal of Agricultural Science (Cambridge) 78:173-177.
Mortelmans J. and Kageruka P. 1976. Trypanotolerant cattle breeds in Zaire. World Animal
Review 19:14-17.
Morton R.H. 1984. Camels for meat and milk production in sub-Saharan Africa. Journal of
Dairy Science 76:1548-1553.
Moyo S., Rege J.E.O. and Swanepoel F.J.C. 1994. Evaluation of indigenous, exotic and
crossbred cattle for beef production in a semi arid environment. Proceedings of the 5th World
Cogress on Genetics Applied to Livestock Production 20:344-347.
Mpofu N., Mareko M.H.D., Setshwaelo L.L. and Makore J. 1998. The growth performance of
crossbred progeny of foreign sire breeds in semi-arid Africa. Proceedings of the 6th World
Cogress on Genetics Applied to Livestock Production 25:227-230.
Mukherjee T.K. 1990. Breeding, selection and biotechnological developments for improvement
of poultry in the tropics-current progress and future perspective. Proceedings of the 4th World
Cogress on Genetics Applied to Livestock Production 15:337-348.
Munasinghe M. 1992. Biodiversity protection policy: Environmental valuation and sdistribution
issues. Ambio 21:227-236.
Njogu A.R., Dolan R.B., Wilson A.J. and Sayer P.D. 1985. Trypanotolerance in East African
Orma Boran cattle. Veterinary Record 117:632-636.
Nolan J. 1990. Acaricide resistance in single and multi-host ticks and strategies for its control.
Parasitologia 32:145-153.
Norton B. 1992. Waren, Annehmlichkeiten und Moral. Die Grenzen der Quantifizierung bei der
Bewertung biologischer Vielfalt. In: Wilson E.O. (ed), Ende der Biologischen Vielfalt?
Heidelberg, Berlin, New York, Spektrum Akad. Verlag. pp. 222-228.
Nwagu B.I. and Alawa C.B.I. 1995. Guinea fowl production in Nigeria. World's Poultry Science
Journal 51:261-270.
Nwagu B.I., Fulayi B.A. and Nwagu F. 1997. Hatchability of Guinea fowl eggs in Nigeria.
Tropical Animal Health and Production 29:63-64.
Nwosu C.C., Obioha F.C., Omeje S.S., Gowen F., Bellonwu C.T. and Onuora G.I. 1984.
Growth performance of local and Shaver Starcross chickens under deep litter system of
management. World Review of Animal Production 20:17-26.
Obanu Z.A., Obioha F.C., Nwosu C.C. and Nwafor W.E. 1984. Evaluation of the organoleptic
and chemical characteristics of meat from the Nigerian native chicken and an exotic strain.
World Review of Animal Production 20:53-58.
Ogore P.B. 1996. Genetic and environmental variation in infestations with ticks and infections
with gastrointestinal nematodes in semi-arid and semi-humid zones of Kenya. MSc thesis.
Egerton University, Kenya.
Ogore P.B., Baker R.L., Kenyanjui M. and Thorpe W. 1999. Assessment of natural Ixodid tick
infestations in sheep. Small Ruminant Research 33:103-107.
Okoye J.O.A., Aba-Adulugba E.P., Ezeokonkwo R.C., Udem S.C. and Orajaka L.J.E. 1999.
Susceptibility of local Nigerian and exotic chickens to infectious bursal disease by contact
exposure. Tropical Animal Health and Production 31:75-81.
Olori V.E. 1994. Quantitative variation in the Nigerian indigenous chicken. Juvenile growth
characteristics. Proceedings of the 5th World Cogress on Genetics Applied to Livestock
Production 20:417-420.
Oldenbroek J.K. (ed). 1999Genebanks and the conservation of farm animal genetic resources.
ID-DLO, Lelystad, The Netherlands. pp 119
Omeje S.S. 1986. Suitable methods for the improvement and conservation of local chickens in
the developing countries. Proceedings of the 3rd World Congress on Genetics Applied to
Livestock Production 12:428-534.
Omeje S.S.I. and Nwosu C.C. 1988. Utilisation of the Nigerian chicken in poultry breeding:
Assessment of heterosis in growth and egg production. Journal of Animal Breeding and
Genetics 105:417-425.
Omeke B.C.O. 1989a. Field performance of exotic pigs and their crosses in tsetse infested
area of Nigeria. Bulletin of Animal Health and Production in Africa 37:79-83.
Omeke B.C.O. 1989b. A comparison of seasonal effects on fertility of Landrace and Large
White sows in a subhumid tropical environment. British Veterinary Journal 145:462-466.
Omeke B.C.O. 1990. Seasonal breeding performance in the field of Landrace and Large
White pigs under humid environment. World Review of Animal Production 25:33-36.
Onah D.N. 1991. Porcine trypanosomosis in Nigeria: Infections in local and exotic pigs in the
Nsukka area of Anambra state. Tropical Animal Health and Production 23:141-146.
O'Neill C.J. and Frisch J.E. 1998. Consequences of selection in two Bos taurus breeds in the
tropics. Proceedings of the 6th World Cogress on Genetics Applied to Livestock Production
25:123-226.
Oni O.O., Abubakar B.Y. and Ogundipe S.O. 1991. Genetic and phenotypic associations of
juvenile body weights and egg production traits in two strains of Rhode Island chickens.
Nigerian Journal of Animal Production 18:66-70.
Osuagwuh A.I.A., Taiwo B.B.A. and Ngere L.O. 1980. Crossbreeding in tropical sheep:
Incidence of dystocia and parturition losses. Tropical Animal Health and Production 12:85-89.
Paling R.W., Mpangala C. Luttikhuizen B. and Sibomana G. 1991. Exposure of Ankole and
crossbred cattle to Theileriosis in Rwanda. Tropical Animal Health and Production 23:203-214.
Pathiraja N. 1986. Improvement of pig-meat production in developing countries. I. Exploitation
of hybrid vigour (heterosis). World Review of Animal Production 60:18-24.
Patterson H.C. 1983. Barbados Blackbelly and crossbred sheep performance in an
experimental flock in Barbados. In: Fitzhugh H.A. and Bradford G.E. (eds), Hair sheep of
Western Africa and the Americas. A genetic resource for the tropics. Westview Press, Boulder,
Colorado, USA. pp. 151-162.
Peacock C.P., Alemayehu Z., Benjaw K. and Forssido T. 1990. The dairy goat development
programme in Ethiopia. In: Proceedings of the 8th SR-CRSP (Small Ruminant Collaborative
Research Support Program) Scientific Workshop, Nairobi, Kenya. Nairobi, Kenya. pp. 254-
267.
Perrings C., Folke C. and Mäler K.G. 1992. The ecology and economics of biodiversity loss:
The research agenda. Ambio 21:201-211.
Ponzoni R. 1997. Genetic resources and conservation. In: Piper L. and Ruvinsky A. (eds), The
genetics of sheep. CAB (Commonwealth Agricultural Bureaux) International, Oxon,
Wallingford, UK. pp. 437-469.
Raadsma H.W., Gray G.D. and Woolaston R.R. 1998. Breeding for disease resistance in
Merino sheep in Australia. Revue Scientifique et technique (Office international des épizooties)
17:315-328.
Randall A. 1992. Was sagen die Wirtschaftswissenschaftler _ber den Wert der biologischen
Vielfalt? In: Wilson E.O. (ed), Ende der Biologischen Vielfalt? Heidelberg, Berlin, New York,
Spektrum Akad. Verlag. pp. 240-247.
Rege J.E.O. 1998. Utilization of exotic germplasm for milk production in the tropics.
Proceedings of the 6th World Congress on Genetics Applied to Livestock Production 25:193-
200.
Rege J.E.O. (ed). 1999. Economic valuation of animal genetic resources. Proceedings of an
FAO/ILRI workshop held at FAO headquarters, Rome, Italy, 15-17 March 1999. ILRI
(International Livestock Research Institute), Nairobi, Kenya. 72 pp.
Rege J.E.O., Aboagye G.S., Akah S. and Ahunu B.K. 1994. Crossbreeding Jersey with Ghana
shorthorn and Sokoto Gudali cattle in a tropical environment: Additive and heterotic effects for
milk production, reproduction and calf growth traits. Animal Production 59:21-29.
Rey B. and Jacob U. 1991. Farm structure and participation in the Ngozi goat crossbreeding
project in Burundi. In: Proceedings of the 9th SR-CRSP (Small Ruminant Collaborative
Research Support Program) scientific workshop, ILRAD, Nairobi, Kenya. Nairobi, Kenya. pp.
141-148.
Rigor E. M. and Kroeske D. 1972. Indigenous and exotic pig breeds in the tropics. World
Animal Review 1-4:20-24. Roberts J.A. 1995. Assessing the scavenging feed resource base
for sustainable smallholder poultry development. Proceedings of ANPRD (African Network for
Rural Poultry Development) workshop, Addis Ababa, Ethiopia, 13-16 June 1995. ANRPD, Ile-
Ife, Nigeria. pp. 40-51.
Ruane J. 1998. Selecting breeds for conservation. In: Oldenbroek J.K. (ed), Genebanks and
the conservation of farm animal genetic resources. ID-DLO, Lelystad, Netherlands. pp. 59-74
Ruane J. 1999. A critical review of the value of genetic distance studies in conservation of
animal genetic resources. Journal of Animal Breeding and Genetics 116:317-324.
Salman A., Shuaib M.A., Suleiman T.H. and Ginawi M. 1983. Infectious Bursal disease in the
Sudan. Tropical Animal Health and Production 15:219-220.
Sahni K.L., Chawla D.S. 1982. Cross breeding of dairy goats for milk production. Proceedings
of international conference for goat reproduction. Scottsdale, Arizona, USA. pp. 575-583.
Schoeman S.J. and van der Merwe C.A. 1994. Improved efficiency in crossbreeding with
Finnsheep × subtropical ewe composites. Proceedings of the 5th World Congress on Genetics
Applied to Livestock Production 18:91-94.
Schwartz H.J.D. 1992. The one-humped camel in Eastern Africa. Verlag Josef Margraf.,
Weikersheim, Germany.
Schwartz H.J., Dolan R. and Wilson A.J. 1983. Camel production in Kenya and it's constraints.
I. Productivity. Tropical Animal Health and Production 15:169-178.
Semenye P.P., Onim J.F.M., Conelly W.T. and Fitzhugh H.A. 1989. On-farm evaluation of
dual-purpose goats in Kenya. Journal of Animal Science 67:3096-3102.
Sharma B.S. and Pirchner F. 1991. Heterosis in Friesian × Sahiwal crosses. Journal of Animal
Breeding and Genetics 108:241-252.
Shavulimo R.S., Rurangirwa F., Ruvuma F., James A.D., Ellis P.R. and McGuire T. 1988.
Genetic resistance to gastrointestinal nematodes, with special reference to Haemonchus, in
three breeds of goats in Kenya. Bulletin of Animal Health and Production in Africa 36:233-241.
Shelton M. 1986. Breed use and crossbreeding in goat production. Proceedings of the 3rd
World Congress on Genetics Applied to Livestock Production 9:523-532.
Silanikove N. 1986a. Feed utilization, energy and nitrogen balance in the desert black Bedouin
goat. World Review of Animal Production 22:93-96.
Silanikove N. 1986b. Interrelationships between feed quality, digestibility, feed consumption,
and energy requirements in desert (Bedouin) and temperate (Saanen) goats. Journal of Dairy
Science 69:2157-2162.
Silanikove N. and Brosh A. 1989. Lignocellulose degradation and subsequent metabolism of
lignin fermentation products by the desert black Bedouin goat fed on wheat straw as a single-
component diet. British Journal of Nutrition 62:509-520.
Silanikove N., Tagari H. and Shkolnik A. 1993. Comparison of rate of passage, fermentation
rate and efficiency of high fiber diet in desert Bedouin goats compared to Swiss Saanen goats.
Small Ruminant Research 12:45-60.
Simon D.L. 1999. Genetic resources and conservation. In: Fries R. and Ruvinsky A. (eds), The
genetics of cattle. CAB (Commonwealth Agricultural Bureaux) International, Oxon, Wallingford,
UK. pp. 475-496.
Singh K. and Bhattacharyya N.K. 1990. Cardio-respiratory activity in Zebu and their F1
crosses with European breeds of dairy cattle at different ambient temperatures. Livestock
Production Science 24:119-128.
Singh S.N. and Singar O.P.S. 1981. Final Technical Report. Department of Animal Husbandry
and Darying. Raja Balwant Singh College. Uttar Pradesch, India. Research Projrct A7-Ah-89.
Smith C. 1985. Scope for selecting many breeding stocks of possible economic value in the
future. Animal Production 41:403-412.
Solow A., Polasky S. and Broadus J. 1993. On the measurement of biological diversity.
Journal of Environmental Economics Management 24:60-68.
Soltan M.E. 1993. Performance of selected-Sinai fowl in comparison with Fayoumi and Baladi
fowls as standard Egyptian local breeds. II. Egg quality. World Review of Animal Production
28:9-23.
Soltan M.E. and Ahmed B.M. 1990. Performance of selected-Sinai fowl in comparison with
Fayoumi and Baladi Fowls as standard Egyptian local breeds. I. Egg production. World
Review of Animal Production 25:17-26.
Somade B. 1985. The influence of season, parity, and duration of lactation on some
performance traits of sows indigenous to Nigeria. Beiträge Zur Tropischen Landwirtschaft und
Veterinärmedizin 23:81-87.
Somade B. and Makinde A.O. 1985. The influence of season on sow weaning to estrus
interval and subsequent reproductive performance following estrus synchronization by batch
weaning. Beiträge Zur Tropischen Landwirtschaft und Veterinärmedizin 23:339-343.
Sonaiya E.B., Branckaert R.D.S. and Guèye E.F. 1999. Research and development options
for family poultry. First INFPD (International Network for Family Poultry Development)/FAO
Electronic Conference on Family Poultry.
Sorrensen W.J. and Scott M.E. 1978. Economic evaluations of exotic sheep crosses in New
Zealand. Proceedings of the New Zealand Society of Animal Production 38:71-78.
Starkey P.H. 1984. N'Dama cattle--A productive trypanotolerant breed. World Animal Review
50:3-15.
Stear M.J. and Wakelin D. 1998. Genetic resistance to parasitic infection. Revue Scientifique
et technique (Office international des épizooties) 17:143-154.
Steele M. 1983. Kabura Development Project Sultanate of Oman. Initial results of sheep
cross-breeding programme. Chios × Omani. Prel. Reports IB(4). Centre for Overseas
Research and Development, University of Durham, UK.
Steinbach J. 1971. Effects of season and breed on sow performance in the seasonal-
equatorial climate of Southern Nigeria. Journal of Agricultural Science (Cambridge) 77:331-
336.
Steinbach J. 1976. Reproductive performance of high producing pigs under tropical conditions.
World Animal Review 19:43-46.
Swanson T. 1998. The source of genetic resource values and the reasons for their
management. In: Evenson R., Gollin D. and Santianello V. (eds), Agricultural values of plant
genetic resources. CAB (Commonwealth Agricultural Bureaux) International, Oxon,
Wallingford, UK. pp. 67-79.
Syrstad O. 1989. Dairy cattle crossbreeding in the tropics: Performance of secondary
crossbred generations. Livestock Production Science 23:97-106.
Syrstad O. 1990. A genetic interpretation of results obtained in Bos indicus × Bos taurus
crossbreeding for milk production. Proceedings of the 4th World Congress on Genetics
Applied to Livestock Production 14:195-198.
Tamboura H.H., Sawadogo L.L., Tahiri C., Aidara D. and Bogore A.D.E. 1997. Estrus
synchronisation and post-partum management of Mossi local breed of goat in Burkina Faso.
Tropicultura 15:190-194.
Teale A. 1993. Improving control of livestock disease. Bioscience 43:475-483.
Teodoro R.L., Madalena F.E. and Smith C. 1996. The value of F1 dairy Bos taurus-Bos indicus
embryos for milk production in poor environments. Journal of Animal Breeding and Genetics
113:471-481.
Thaon d'Arnoldi C., Foulley J.L. and Ollivier L. 1998. An overview of the Weitzman approach
to diversity. Genetique Selection Evolution 30:149-161.
Tobias D. and Mendelsohn R. 1991. Valuing ecotourism in a tropical rainforest reserve. Ambio
20:91-93.
Toro. M. and Mäki-Tanila A. 1998. Establishing a conservation scheme. In: Oldenbroek J.K.
(ed), Genebanks and the conservation of farm animal genetic resources. ID-DLO, Lelystad,
The Netherlands. pp. 75-90]
Trail J.C.M. 1962. The indigenous poultry of Uganda. The fertility, hatchability and chick
mortality compared with European breeds. Poultry Science 41:1271-1276.
Turner H.G. 1975. The tropical adaptation of beef cattle. World Animal Review 13:16-21.
Turner H.G. 1982. Genetic variation of rectal temperature in cows and it's relationship to
fertility. Animal Production 35:401-412.
Turner H.G. 1984. Variation in rectal temperature of cattle in a tropical environment and its
relation to growth rate. Animal Production 38:417-427.
Turner H.G. and Short A.J. 1972. Effects of field infestations of gastrointestinal helminths and
of the cattle tick (Boophilus microplus) on growth of three breeds of cattle. Australian Journal
of Agricultural Research 23:177-193.
Udo H.M.J. 1982. Fattening of Large White, Local Samoan and crossbred pigs. Tropical
Animal Health and Production 14:224-228.
de Vaccaro L.P. 1990. Survival of European dairy breeds and their crosses with zebus in the
tropics. Animal Breeding Abstracts 58:475-494.
Vercoe J.E. and Frisch J.E. 1972. Apparent digestibility, nitrogen utilisation, water metabolism
and heat tolerance of Brahman cross, Africander cross and Shorthorn × Hereford. Journal of
Agricultural Science (Cambridge) 79:71-74.
Virchow D. 1999. Economic value of genetic resources: An agenda for research. Ag Biotech
Net 1: ABN 009.
Waruiru R.M., Karanu F.N., Ruvuna F., Gichanga E.J., Taylor J.F., Jasmer D., Rurangirwa F.
and McGuire T.C. 1994. Differential resistance among goats to primary and secondary
challenges of Haemonchus contortus. Bulletin of Animal Health and Production in Africa
42:193-197.
Weitzman M.L. 1992. On diversity. Quarterly Journal of Economics 107:363-405.
Weitzman M.L. 1993. What to preserve? An application of diversity theory to crane
conservation. Quarterly Journal of Economics 108:157-183.
Weller J.I. 1994. Economic aspects of animal breeding. Chapman & Hall, London, UK.
Willham R.L. 1987. Taking stock. Iowa State University, Ames, USA.
Wilson R.T. 1986. Reproductive performance and survival of young one-humped camels on
Kenya commercial ranches. Animal Production 42:375-380.
Wilson R.T. 1988. The camel. Longman, Singapore.
Yagil R. 1985. The desert camel--Comparative physiological adaptation. Karger. Basel,
Switzerland. pp 163.
Yehulashet T., Gebreab F., Wakjira A. and Tsega T. 1995. Preliminary observation on ticks:
Seasonal dynamics and resistance of three indigenous and three crossbreed-bred cattle in
Ethiopia. Bulletin of Animal Health and Production in Africa 43:105-114.
Young A.S., Groocock C.M. and Karuiki D.P. 1988. Integrated control of ticks and tick-borne
disease of Africa. Parasitology 96:403-432.
Young L.D., Dickerson G.E., Ch'ang T.S. and Evans R. 1986. Heterosis retention in sheep
crossbreeding. Proceedings of the 3rd World Congress on Genetics Applied to Livestock
Production 9:497-508.
Young L.G. and King G.J. 1981. Reproductive performance of gilts bred on first versus third
estrus. Journal of Animal Science 53:19-25.
Zilberman D., Yarkin C. and Heiman A. 1998. Institutional change and biotechnology in
agriculture: Implications for developing countries. In: Smale M. (ed), Farmers, gene banks and
crop breeding. Economic analyses of diversity in wheat, maize, and rice. Kluwer Academic
Publishers, London, UK. pp. 193-203.
Appendix: Farm animal genetic resources (AnGR) valuation method evaluation
 Appendix: Farm animal genetic resources (AnGR) valuation
method evaluation
Valuation
methodology
Purpose,
objective or
strength
Role in
conservation
Type of data
required
Data availability Conceptual
weaknesses or
difficulties
Methodologies for determining the appropriateness of AnGR conservation programme costs
Contingency
valuation
Identify society's
WTP for the
conservation of
AnGR
Define upper
bound to
economically
justified
conservation
programme
costs
Society
preferences
expressed in
terms of WTP
Not normally
available.
Requires survey
Response difficulties
when used for `non-
charismatic' species
and/or chronic genetic
erosion
Production
loss averted
Indicate
magnitude of
potential
production
losses in the
absence of
AnGR
conservation
Justify
conservation
programme
costs of at least
this magnitude
Estimate of
potential
production
losses (e.g.
percentage of
herd and
market value of
animals)
Animal market
values available
for commercial
breeds. Potential
herd loss must
be estimated.
Not a
consumer/producer
surplus measure of
value. Ignores
substitution effects
Opportunity
cost
Identify cost of
maintaining
AnGR diversity
Define
opportunity cost
of AnGR
conservation
programme
Household
costs of
production and
net income
Not normally
available.
Requires survey
 
Least cost Identify cost-
efficient
programme for
the conservation
of AnGR
Define minimum
cost of
conservation
programme
Household
costs of
production and
profitability
Not normally
available.
Requires survey
 
Methodologies for determining the actual economic importance of the breed
Aggregate
demand and
supply
Identify value of
breed to society
Value of
potential losses
associated with
AnGR loss
Intertemporal or
farm-level data
Available for
commercial
breeds. Not
normally
available for
others.
Requires
survey
Requires shadow
pricing of domestic
labour and forage
Cross-
sectional farm
and
household
Identify value of
breed to society
Value of
potential losses
associated with
potential AnGR
loss
Consumer and
producer price
differences by
location
Not normally
available.
Requires
survey
Requires shadow
pricing of domestic
labour and forage
Market share Indication of
current market
value of a given
Justify economic
importance of a
given breed
Market value of
animal products
by breed
Generally
available but
not always by
Not a
consumer/producer
surplus measure of
breed breed value. Ignores
substitution effects
IPR and
contracts
Market creation
and support for
`fair and
equitable'
sharing of AnGR
benefits
Generate funds
and incentives
for AnGR
conservation
Royalty
payments or
terms of contract
Usually
available when
such
arrangements
exist although
can be
commercial
secret
Limited duration of
contracts
Methodologies for priority setting in AnGR breeding programmes
Evaluation of
breeding
programme
Identify net
economic
benefits of stock
improvements
Maximise
economic
benefits of
conserved
AnGR
Yield effects
and input costs
Available for
commercial
breeds. Not
normally
available for
others. Requires
survey/research
Difficulty in isolating
the contribution of
genetic resources
from other costs of
programme
Genetic
production
function
Identify net
economic
benefits of stock
improvements
Maximise
expected
economic
benefits of
conserved
AnGR
Yield effects
and input costs
Available for
commercial
breeds. Not
normally
available for
others. Requires
survey/research
 
Hedonic Identify trait
values
Value potential
losses
associated with
AnGR loss.
Understand
breed
references.
Characteristics
of animals and
market prices
Available for
commercial
breeds. Not
normally
available for
others. Requires
survey/research
Not a
consumer/producer
surplus measure of
value. Ignores
substitution effects
Farm
simulation
model
Model improved
animal
characteristics
on farm
economics
Maximise
economic
benefits of
conserved
AnGR
Inputs and
outputs.
Technical
coefficients of
all main
activities
Available for
commercial
breeds. Not
normally
available for
others. Requires
survey
Correct definition of
farm objective
function. Aggregation
for estimating
consumer surplus can
also be problematic
Source: Drucker et al. (2001).