Summary
Résumé
Introduction
State and focus
Constraints
Breeding strategies
Strategic breeding schemes
Conclusion
References
G.H. Kiwuwa
Department of Animal Science, Makerere University, Kampala, Uganda
The combined population of small ruminants in sub-Saharan Africa numbering 350 million is approximately 31% of the world total and substantially contributes to human nutrition and the economic requirements of the African people. However, the majority (62%) of the small ruminants (sheep and goats) are raised in arid and semi-arid zones where the nomadic production systems do not render systematic genetic improvement feasible. However, animals found in the arid/semi-arid areas have, for a long time, been naturally selected for the harsh conditions and should be the best genotypes for those areas. A relatively smaller (38%) proportion of small ruminant populations are traditionally kept in humid, semi-humid and highland ecozones. Human populations in these areas practice sedentary agriculture and agropastoral production systems. It is in these zones that strategic small ruminant breeding schemes could be applied. However, under smallholder production systems in the agricultural and agropastoral areas, conventional breeding methods are constrained by communal grazing, small flock sizes, single-sire flocks, low levels of literacy and lack of proper recording practices. Under these circumstances breeding strategies will have to be modelled along stratified schemes based on central nucleus elite breeding and multiplier flocks to generate sires for distribution to smallholder farmers. As recording improves, cooperative breeding schemes based on central nucleus herds without concurrent multiplier herds could also be adopted following on-farm performance evaluation for selection of elite females for comprehensive testing in the on-station nucleus herd. The implications of instituting a successful small ruminant breed improvement complemented by on-farm performance testing are also discussed.
L'ensemble des effectifs ovins et caprins de l'Afrique subsaharienne s'élève à 350 millions de têtes et représente environ 31% du cheptel mondial. Ces animaux, qui permettent aux populations africaines de subvenir à une bonne partie de leurs besoins alimentaires et économiques, sont dans leur grande majorité (62%) élevés dans les zones arides et semi-arides du continent, où les systèmes de production pastorale à caractère nomade font obstacle à la mise en oeuvre de programmes systématiques d'amélioration génétique. En revanche, l'hostilité du milieu a opéré Une sélection naturelle qui a donné lieu à la constitution de génotypes particulièrement adaptés à la région. Une part relativement moins importante (38%) des populations ovines et caprines est élevée traditionnellement dans les zones humide et semi-humide et dans les hauts plateaux de l'Afrique. De type agricole sédentaire ou agropastoral, les systèmes de production en vigueur dans ces régions autorisent la mise en oeuvre de programmes stratégiques de sélection, encore que l'exploitation en commun des pâturages, la taille réduite des troupeaux, l'insuffisance du nombre de géniteurs au sein des troupeaux, le faible taux d'alphabétisation et l'absence de méthodes adaptées de relevés de données qui caractérisent les systèmes de la petite exploitation agricole rencontrés dans ces zones viennent s'opposer à l'utilisation des méthodes classiques de sélection. Dans ces conditions, les stratégies adoptées en matière de sélection devront se modeler sur des schémas stratifiés s'appuyant non seulement sur des troupeaux de sélection composés d'animaux d'élite mais aussi sur des troupeaux de multiplication destinés à produire des mâles reproducteurs aux fins de diffusion auprès des petits exploitants. Avec l'amélioration des contrôles, des schémas de sélection à l'échelle collective, fondés exclusivement sur des noyaux de sélection, pourront être mis en place. L'évaluation des performances au niveau des exploitations aura pour objectif de sélectionner des mères d'élite. Celles-ci seront intègrées au noyau de sélection de la station en vue d'une évaluation exhaustive de leurs performances. Les conditions indispensables à la réussite d'un programme d'amélioration génétique des petits ruminants et de l'évaluation en exploitation sont également analysées.
The goat, a key actor in the small ruminant production theatre of animal agriculture in Africa was once referred to by the colonialists as "a problem child of African agriculture", causing soil erosion and deforestation and being an absolute nuisance in crop-based farming systems. Not until work carried out in the late 1950s at Serere Research Station in Uganda proved these allegations to the contrary, did the scientific world realise that small ruminants are not villains but have a role to play in the overall livestock production systems on the African continent.
This meeting, today, bears witness to one of the major achievements of the last two decades, regarding animal agriculture in Africa. That achievement has been the fact that greater awareness of the role that small ruminants play in subscribing to the animal protein pool for human consumption has been created. Yet there are still numerous constraints against increased small ruminant productivity. Some of these have been spelt out by Adeniji (1989) and include low genetic potential for milk and meat production, inadequate nutrition and poor management practices, prevalence of diseases and reproductive wastage and limited market outlets for small ruminant products. Secondary limitations cover a range of logistic inadequacies including inappropriate technologies, poor extension services and linkages between research and extension, lack of economic incentives and unavailability of inputs for the utilisation of research findings.
In this address, privilege has been accorded to the speaker to switch on the state of the art and focus on breeding strategies for small ruminants in Africa. In the process, attempt will be made to highlight some of the critical bottlenecks against genetic improvement through known conventional methods and suggestions will be made in relationship to unconventional approaches that could be pursued for various farming systems in the different ecological zones that characterise the environments of sub-Saharan Africa.
Sheep and goat populations in Africa have recently been put at about 191 and 159 million heads, respectively. These estimates represent about 30% and 32% of the world's total population of sheep and goats, respectively. Distribution trends by ecozones indicate that the arid areas take a substantial share of the African small ruminants, accounting for 36% of sheep and 39% of goat populations. The semi-arid, the subhumid and the humid zones account for 22%, 14% and 8% of sheep and 27%, 16% and 9% of the goat populations, respectively.
It would appear therefore that the majority of small ruminants (58% sheep and 66% goats) are kept in arid and semi-arid zones which predominantly practice pastoral-nomadic system of livestock production. Implications from distribution trends are clear. Breeding strategies for nomadic flocks face contradictions between adaptation and absolute productivity. The type of small ruminant breeds for the harsh environment are determined by the ability to survive prolonged droughts, walk and cover long distances in search of water and forages, and utilise coarse forages for growth, reproduction and lactation (Osinowo and Abubakar, 1989).
Breeding strategies for intensive small ruminant production presuppose a sedentary rather than a nomadic system of production. It is under the subhumid and humid ecozones, where communities practice agropastoral and sedentary agricultural farming system, that breed improvement technologies are likely to be successfully adopted. These areas, however, constitute roughly 35% to 40% of the small ruminant population out of which about 10% are on the extreme humid side with the attendant problems arising out of the stressfully humid climate, low nutritive value of fodder and major disease vectors such as ticks, tsetse flies and helminthic parasites. Breeding for heat and disease tolerance would be desirable but could in the process contradict the genetic improvement of the economically productive traits.
However, irrespective of the strategic options, it would be highly desirable that breeding plans match small ruminant breed resources to the available feed, management and socio-economic resources in a given ecozone. The bottom line to future approaches in genetic improvement should be to promote the utilisation of the more promising and adapted indigenous breed resources rather than opting for introduction of temperate goat and sheep germplasm that are often ill-adapted to the tropical conditions. The main goal would be to develop simple breeding procedures involving farmers at the grassroot level and be able to register progress towards predictable and sustainable levels of small ruminant meat and milk productivity in agricultural and agropastoral farming systems around Africa.
In designing breeding strategies for small ruminant improvement in Africa one should not overlook a number of constraints that must be overcome before strategies are put in place. A few that come to my mind include:
· Lack of adequate data on breed/type characterisation and standardisation of breed names within and between the five regions of Africa (North, West and Central, East and South). It is possible that there could be a lot of duplication of breed names by country or region. Determination of genetic distances among some of the populations may therefore by useful. The developments in genome mapping and biochemical genetics may, in future, be able to provide information on DNA sequences that can be used to measure genetic distances and thereafter narrow the present breeds/strains to a smaller number that can be evolved into a dominant type for a given ecozone.· Much as on-farm breeding data recording schemes would promote participatory research, extension and feedback attributes of a strategic breeding plan, flock mobility under the extensive nomadic systems would pose a major problem. In smallholder systems, small flock sizes coupled with communal herding and single sire flocks also creates problems for isolating sire, farm or flock effects from the true genetic effects. These problems immediately suggest that breeding strategies will have to be modelled along group breeding schemes with on-station nucleus breeding units to provide tested sires to participating farmers.
· Small ruminants, particularly goats, tend to perform less satisfactorily under onstation than on-farm environment. Peters (1989) attributes this phenomenon to differences in management practices. On-station production systems follow modern husbandry methods (fencing, routine disease control, restricted breeding, housing etc.). These lead to reduced feed selection possibilities, higher disease risks and restricted mating freedom. Hence, transposing onstation results on breed comparison to field conditions may not necessarily reflect the true life situation on the range or free choice herding systems in villages. On the other hand, on-farm studies deal mainly with non-genetic factors and may not lead to proper assessment of the desired genotypes.
· Much as some sheep breeds (Dorper Blackhead Persian, Yankasa, Balami, Uda:Ouda etc.) and goats (West African Dwarf, Sudan Desert, Nubian, Galla, Mubende, Boer, Red Sokoto and Somali) have been reasonably documented, there are other types that have not received sufficient attention to explore their full potential. Low genetic potential among the majority of small ruminants is often assumed and breeding plans to cross the indigenous stocks with exotic germplasm are implemented regardless. This is a constraint in a sense that breeders are pushed by economic forces to adopt germplasm for immediate (short-term) benefits without projecting sustainability consequences. The results have been improvement in productive traits at the expense of fitness traits, which in all, could lead to disaster in the long run.
Strategic planning for small ruminant breeding must, from the start, therefore, put into consideration a number of factors. These include the production system (i.e. nomadic, sedentary, agropastoral, mixed farming etc.), the priority order of economic traits (i.e. meat, milk, fibre skins etc.), the national flock structure and the socio-economic mix, including level of literacy, product consumption habits, market outlets and type of ownership (i.e. smallholder, pure or mixed ranching).
With regard to the production system, the majority of small ruminant populations are located in arid and semi-arid zones of Africa. The owners pursue a nomadic production system which is ecologically imposed but makes full exploitation of arid and semi-arid ranges possible. Nomadism, however, does not render itself to systematic genetic improvement of productive traits other than ensuring and promoting traits for survival.
Strategies for nomadic systems will, for some time, be restricted to disease control and regulation of carrying capacities by creating socially acceptable market cutlets.
When we turn to the prioritisation of traits, and assuming that breeding strategies will be best suited to the agropastoral and smallholder crop production systems, two issues come to the fore:
First, with the exception of the milking Nubian-Riverine and the Abyssinian goats, most of the goat and sheep breeds are kept for meat. Those that produce limited amounts of milk (40-80 kg/year) are mainly located in arid and semi-arid environments (i.e. Sudan Desert/Sahel, Maradi/Red Sokoto, Small East African upland semi-arid, Sudan Desert sheep and Ethiopian fat-tailed sheep).
Having noted already that breeding efforts should focus on sedentary farming and agropastoral production systems, one would be inclined to exclude milk production as a key trait for genetic improvement. The effort to breed for milk production would involve importation of exotic dairy goat breeds for cross-breeding. The associated costs of importation followed by ill-adaptation of the crossbreds to the ecosystems in the sub-Saharan region cannot be justified. Increases in milk yield from 0.5 kg to 2 kg per day are possible, but would still be less than what could be obtained from similar efforts with cattle.
Second, accepting that milk can more effectively be obtained from cattle than from goats or sheep under intensive agro-farming practices, the next consideration is to decide between sheep and goats as meat sources. Sheep can be cheep sources of mutton as they tend to weigh more than goats. However, by the nature of their coat, sheep may not adapt well to the humid climatic zone. Their natural ecosystem is the highlands whose total geographical area is relatively small with extremely high human population densities as compared to the semi-humid and semi-arid areas. Sheep breed, improvement strategies for semi-arid zones may be hampered by the nomadic nature of the production systems in those areas.
Furthermore, due to their close-to-the-ground grazing behaviour, sheep kept together with cattle can easily cause soil erosion, particularly in the hilly and mountainous areas of eastern Africa. In the West African region, breeds of sheep such as the Yankasa, Uda and Balami may have a role to play in the agropastoral production systems of that region. Overall, it would appear that for the humid and semi-humid zones breeders should concentrate on goat improvement for meat and skins which offer decisive advantage as goats fit in well with peasant smallholder production systems in those areas.
The third issue to consider is that modelling breeding strategies along the conventional methods of progeny testing over a wide range of farm conditions, often supported by recording and artificial insemination services will not work for small ruminants. In most of sub-Saharan Africa, record keeping for meat animals has been difficult to institute due to range production systems where animals stay away from homesteads for some period of time and kidding or lambing are asynchronous with seasons. Furthermore, ownership of small ruminants ownership is in small numbers and is scattered far and wide in the area.
Concentrating large numbers of small ruminants for conventional breeding techniques may not be feasible in the near future, as most progressive farmers and agropastoralists would rather invest the high capital outlay into cattle than .goats or sheep. Under such circumstances it would be strategic for breeders to pursue procedures that would enhance superior gene flow from nucleus elite herds to the farmers, supported by performance testing under smallholder production systems.
In short, and according to Olayiwole and Adu (1989), since conventional selection methods are not feasible for small flocks where recording is inadequate or hard to institute, breeding programmes for such a situation should be carefully planned, easy to operate, sufficiently flexible and should incorporate the utilisation of available genetic resources. Within a given breeding scheme emphasis on meat should supersede consideration for other secondary traits such as milk and fibre (skins being a by-product of meat production improvement efforts).
Strategic small ruminant breed improvement could then be approached from three angles:
a) screening and selection within the locally adapted breeds for prolificity, faster growth rate, larger body size and conformation. The programme should be carried out in elite nucleus flocks to generate superior sires for distribution to smallholder farmers.b) crossbreeding or inter-breeding between two most promising local breeds that developed in nearly similar ecozones but that can complement each other's attributes.
c) blending several local breeds irrespective of their original environment with the hope of evolving a synthetic breed for a given production system.
Three possible approaches are hereby proposed with the following assumptions:
a) Meat is the primary target for small ruminant improvement.b) Body size at maturity is generally categorised into three groups: small (20-30 kg), medium (30-45 kg) and large (45-60 kg).
c) Heritability estimates for growth characters are reasonably medium to high (0.30-0.45) and litter size, though it may register low heritability estimate, responds somehow to selection.
d) The majority (over 90%) of small ruminant production is still under traditional systems of production but can change gradually to modern systems with improved extension services.
Open nucleus elite breeder and multiplier flocks
The scheme involves assembling animals through purchases to a central station. The number of animals (does or ewes) should be sufficiently large (300-500) with about one-third to act as control group. High selection pressure is thereafter imposed on the offspring and the dams according to productivity indices for male and female offspring on the one hand and selection of dams to stay in the herd, on the other. The progeny selection index takes into account the reproductive rate of the dam and the offspring's growth rate and weight at weaning, or better still up to one year to avoid maternal effects at weaning stage. Dam selection index is on the basis of fertility and is calculated at the end of each breeding season. It considers the dam's present and previous records in relationship to those of its contemporaries. These indices have been described by Osinowo and Abubakar (1989).
The selection pressure would allow the top 10% of the males to be retained in the elite herd for further breeding while the next top 30-40% would be transferred to form the flock sire multiplication scheme after performance testing and the 50 60% would be culled. On the side of the dams, about one-third would be culled remaining during each cycle and replaced by the progenies of the top (10%) performing offspring. The next one-third of the females would be moved to the flock sire multiplier station. The flock sire multiplier station would ensure an increase of the number of males available for distribution to farmers after further screening. The system would ensure improved males from the elite breeder flock entering multiplier flocks from which greater numbers of improved males are generated for distribution to smallholder village flocks.
Open nucleus elite breeder and participatory farmer flocks
The procedure under this scheme would be essentially similar to the first except for the starting and ending stages of animal selection. It also does away with the intermediate siremultiplier flock stage and requires full cooperation between participating farmers and the Elite Nucleus Breeders. The scheme requires some minimum level of literacy on the part of the beneficiaries (farmers) and outright goodwill to release and receive animals between the station and the farm flocks.
In essence the proposed scheme is a private cooperative breeders' organisation pivoted around a public government institution that funds the central nucleus breeding flock. The procedure assumes that there has been preliminary on-farm performance appraisals of animals that would enter the nucleus flock.
The scheme involves organising groups of interested participants on a provincial or state basis to discuss the concepts, aims and benefits, including the legal and genetic aspects of the scheme.
Each member of the cooperative group, assisted by the field staff, contributes the topperforming females to the elite nucleus central station. This is the concept of preliminary screening of the population for elite females based on weight for age and histories on prolificity.
The next step would be to decide on the exchange rate for the top females contributed to the central nucleus flock, in return for the highly selected sires to pass back to the cooperative farmers. Perhaps a ratio of five females contributed to one tested sire and two super dams received back from the station could be a useful starting point.
Within the central nucleus flock, the top performing females will acquire an elite status as more performance data accumulate. These females must then be bred to the top (1-5%) sires for use within the nucleus flock.
Under this scheme faster genetic progress comes from the relatively high selection pressure made possible in the initial on-farm screening operation followed by equally intensive selection and the technology transfer and extension service between the central nucleus station and the farmer-contributior of superior foundation animals.
On-farm performance testing
Much as intensive selection programmes under centralised elite nucleus flocks generate superior germplasm for distribution to farmers, the final benefits can only be gauged through on-farm performance evaluation. Performance testing identifies and quantifies the nongenetic constraints in order to improve management, disease control and feeding practices. Since we cannot breed beyond the environment the improved animal's genetic abilities should periodically be tested against the prevailing environmental conditions on the farmers' farms.
Performance evaluation is a logistical complement to genetic improvement schemes. It diagnoses constraints pertaining to a given production system, provides on-spot checks and analysis of husbandry problems, and generates feedback information to breeders and farmers alike. Very often, the results obtained from central station tests are of little relevance to traditional production systems and may not contribute much towards understanding of the specific adaptation ability of animals to farmers' conditions.
One further attraction of on-farm performance testing, is that it provides information on location-specific production conditions that could lead to breed improvement options that are appropriate to the system. Individual animal appraisals and breed comparisons, however, can best be carried out under controlled on-station conditions.
Successful implementation of on-farm performance evaluation is, however, likely to be constrained by a number of problems. Peters (1989) gave a comprehensive outline on the subject and highlighted the following aspects:
a) on-farm recording must be based on a large sample of flocks carefully stratified following a baseline survey.b) comprehensive recording of farm covariates is required to explain systematic differences and major reasons for variation in performance.
c) where small ruminants are reared on an extensive scale, flock mobility is often a major problem to performance testing.
d) in smallholder systems, small flock sizes and communal herding also create problems for isolating farm or flock effects.
e) seasonality of production would facilitate accurate recording of reproductive data, but asynchronous production is most common in African systems and demands regular visits and data recording in herds participating in the on-farm performance schemes.
f) in order to delineate covariates such as season, disease, age of dam, parity, litter size and sex effects and adjust the data accordingly, one would require a data base spanning over a period not less than two years.
g) individual animal identification may face difficulties when herds have mixed ownership and continuity may be disrupted by decisions of the owner to sell or slaughter the animal.
However, there are a number of complementarities between on-farm performance evaluation and central nucleus herd stations. According to Peters (1989), genetic improvement schemes on the latter require a field base if they are to be successful. Furthermore, where nucleus breeding units are integrated with on-farm performance evaluation, immediate improvement through selection of superior foundation animals (see strategy 2) to attain faster and more effective progress can be achieved.
A successful on-farm performance evaluation scheme should be concerned with the whole farm environment; cover all components of the farm environment; identify factors limiting production within the system; initiate integrated research to find solutions based on onstation controlled procedures; test breeding or production interventions (on-farm); evaluate the consequences of the interventions; and be instrumental in the design of small ruminant programmes (Peters, 1989).
Breeding strategies for small ruminant improvement in Africa should focus on meat production characteristics, with skins as a by-product. Multiple births phenomenon should be effectively exploited to provide the maximum meat output per unit of input. Under smallholder production systems of humid and subhumid zones of Africa, conventional breeding methods are constrained by communal grazing, small flock sizes, single-sire flocks, low levels of literacy and poor recording schemes. A partial solution has been suggested to institute elite nucleus breeding stations, screen and test animals (sires) for distribution to farmers. These schemes, however, in the long run should be complemented by simple but systematic on-farm performance evaluation programmes to reinforce and monitor continual genetic improvement.
Adeniji K O. 1989. Statement on the improvement of small ruminants in West and central Africa. In: Adeniji K O (ed), Improvement of small ruminants. Proceedings of the Workshop on the Improvement of small ruminants in West and central Africa held in Ibadan, Nigeria, 21-25 November 1988. OAU (Organization of African Unity), Nairobi, Kenya. pp. 43-59.
Olayiwole M B and Adu I F. 1989. Past and present research on sheep and goat breeding in Nigeria. In: Adeniji K O (ed), Improvement of small ruminants Proceedings of the Workshop on the improvement of Small Ruminants in West and Central Africa held in Ibadan, Nigeria, 21-25 November 1988. OAU (Organization of African Unity), Nairobi, Kenya. pp. 61 69.
Osinowo O A and Abubakar B Y. 1989. Appropriate breeding strategies for Small Ruminant production in West and Central Africa. In: Adeniji K O (ed), Improvement of small ruminants. Proceedings of the Workshop on the Improvement of Small Ruminants in West and Central Africa held in Ibadan, Nigeria, 21-25 November 1988. OAU (Organization of African Unity), Nairobi, Kenya. pp. 71 84.
Peters K J. 1989. Trends in on-farm performance testing of small ruminants in sub-Saharan Africa. In: Wilson R T and Azeb Melaku (eds), African small ruminant research and development. Proceedings of a conference held at Bamenda, Cameroon, 1825 January 1989. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. 439-469.
