Previous Page Table of Contents Next Page

Strategies for sustainable development of small ruminants: Burundi - a case study

by R. Branckaert


The majority (75.3 percent) of the mountainous regions of Africa are located in the east, notably in Ethiopia, Kenya and Tanzania. The high plateaux generally has good soil but high population densities. Agriculture is, therefore, intensive and permanent cultivation with little or no fallow is common. Crop and animal production are usually practised on the same farm, but in parallel rather than a real integration of activities (Jahnke, 1984). Despite these somewhat favourable condition, actual levels of subsistence are not much higher than in other tropical zones.

The greatest population densities in the high plateau areas exist in Rwanda and Burundi with an average of between 160–180 inhabitants per km2 but with some areas in excess of 330 per km2. Over 90 percent of these populations live in rural areas.

In Burundi, which will be taken as an example, agriculture is based on small farms (less than two hectares), supplemented by communal grazing land. For various reasons, these grazing lands are rapidly declining, both in area and in value, due to:

In certain regions, population pressure is so intense that grazing land is disappearing. The country's future must inevitably lie in an expansion of agro-pastoral systems or, preferably, efficient agro-sylvo pastoral systems. Several such projects have been tried in recent years with relative success and are based on a synergy of food plants, cash crops and livestock production.

For subsistence, farmers plant various food crops according to altitude and climate and include: sweet potato, bananas, cassava, beans, maize, rice, eleusinia and “colocase” with surplus crops being sold. In addition, limited cash cropping includes: coffee, tea, oilpalm fruit and tobacco. Farmers may also keep a few head of livestock to serve as a marketable reserve and, at the same time, to produce manure for the crops. This latter function is often the primary reason for keeping livestock since breeding animals are rarely capable of producing sufficient milk for human consumption and little meat is required for home use. The manure is indispensable for assuring the survival and growth of plants.

Animal productivity, especially in ruminants, is generally low, due to the genetic quality of local breeds, poor nutrition (due to deteriorating rangelands) and animal health problems. Milk production is generally non-existent, fattening of the animals is very time consuming (also the manure produced is of poor quality) and is rarely undertaken.

Thus a multi-disciplinary approach is necessary to improve the agro-pastoral system. Efforts to increase productivity must be undertaken simultaneously with the provision of necessary inputs: seeds, improved breeding stock, processing and marketing facilities. Erosion control and improved supply of quality manure are also important objectives. When such systems are in action, the farmer will be able to produce sufficient for both subsistence as well as earning a supplementary income to raise their living standards and for farm improvements. Of course, such a system would necessitate thorough training of both extension staff and farmers for whom such systems may introduce new concepts and technologies.

Thus most existing projects aim at setting up efficient agro-pastoral systems based on the multitude of sedentary and diverse small farms. They address smallholders directly and aim at replacing semi-nomadic herds in favour of better managed smaller herds/flocks. The lack of adequate grazing land and intensified cultivation will mean that settling migratory herds, at least in the short-term, will necessitate a greater use of intensive housing. Thus, for animals to provide a sufficient source of income and manure, rapid improvements in genetic make-up, nutrition, reproductive performance and health must occur.

That such a system can lead to a reduction in cattle numbers is observed in Burundi where the cattle population has decreased from approximately 800,000 in 1977–1978 to around 400,000 in 1985–1986. The reduction in cattle numbers was associated with an increase in the number of small ruminants from 850,000 (1977/78) to 1,050,000 (1985/86). In Burundi, goats (700,000 head) are the more prevalent than sheep (350,000 head). Sheep are sometimes ignored for reasons of prejudice but usually from a lack of interest; cattle raising is a preferred activity and under present production conditions the performance of sheep is mediocre. However, they are better suited than goats to the agro-pastoral systems now being established. This is due to their less selective and destructive feeding habits and their better adaptability to intensive management.


Before examining the possibilities for improved small ruminant production through disease control, management, feeding, genetics and marketing, it is essential that the existing small ruminant production systems are fully understood. It would be useless to promote interventions to improve productivity, unless they were adapted to the existing systems and are fully understood by the producers and fit in with their personal expectations.

Problems in developing small ruminant production manifest themselves mainly in the areas of animal health, management, lack of technical skills, feeding, genetics and marketing. Therefore it was in these fields that applied research has to be conducted and solutions sought.

Animal Health

No action had as yet been taken to improve the situation regarding the health of small ruminants in Burundi. Epidemiological studies were non-existent and no prophylaxis programmes have been recommended, except for control of external parasites where dips and spray races already exist.

Where such facilities do exist, producers usually only bring their animals for treatment at irregular intervals. This resulted in considerable losses, especially of young stock, estimated between 15 and 40 percent, although, the causes of morbidity and mortality differ considerably between regions. Moniezia, Oestrus ovis, heartwater and plant poisoning (seasonal) can cause considerable problems at localized sites.

Veterinary services, therefore, must draft, as precisely as possible, pathological profiles for the regions where they operate, to provide the basis for suitable prophylactic programmes. However, some afflictions such as helminthiasis are universal and the advantages of a regular worm-control programme are self-evident and permit a considerable reduction in mortalities at a reasonable cost. However, experience showed that it was difficult to launch a development programme for small ruminant production without first identifying all the local pathological problems. Such activities would greatly increase the chances of success and avoid the discouragement of producers faced with diseases beyond their control.

All too often, the health aspects are neglected in projects when compared with the emphasis given to genetic improvement. It is essential to remember that changing an animal's genotype will not increase productivity unless it is in good health and properly nourished.

Improvement of the Farming System

As indicated in the introduction, this represents an essential factor in the development of small ruminant production in highly populated zones. Most projects have attempted to implement efficient agro-sylvo-pastoral systems which link agriculture, afforestation and animal production. In order to understand this type of study and project, one must consider the following:

Considering these factors, two improved farming systems are presently being implemented to improve small ruminants production:

An Integrated Small Ruminant/Crop Programme. This programme is aimed at establishing various mixed farming systems which include: the introduction of highly productive goats and sheep; the transformation of some fallow areas into forage plots where animals can either be tethered or herded; and the development of small farms for producing limited supplies of breeding females, weaned animals (for fattening) and high-quality manure. The programme foresees a growing number of pilot farms, the owners of which must subscribe to certain conditions, notably:

1 An important factor in improving animal production is livestock housing. Sociological factors make it difficult to have specially built pens for small ruminants that are well situated, ventilated and permit the collection of manure. On the contrary, apprehension about possible theft induces farmers to keep their small stock into their own houses, usually in a poorly ventilated, dark corner-but safe from predators. For the highland producers, the acceptance of either permanent or temporary housing seems to be one of the most difficult components of the programme.

A Programme Combining Sheep Production and Re-afforestation. This programme would have the following goals:


The existing feeding practices depend on farming system and degree of intensification.

Natural Pasture is the basis of the small ruminant diet, although utilization differs between species. Goats constantly seek a varied diet of grasses, shrubs and forbs, while sheep prefer shorter grasses better suited to their labial morphologic structure - although at the same time increasing the danger of parasite infestation.

Natural pasture may be utilized by scavenging, tethering or herding. If well managed, the two latter methods permit a more rational use of the available biomass, combining good nutrition for the animal with adequate regeneration of the plant cover. In tree-crop plantations, natural vegetation can best be used by rotational grazing. In the pilot farms the predominant grasses were species of Hyparrhenia, Eragrostis and Digitaria.

Fallow land and roadsides are an important feed resource and which often differ in agrostologic composition and nutritional value from other natural pastures. The commonly occurring palatable species include: Erlangia spissa, Bidens pilosa, Sorghum vulgare, Monathoxalis orophila, Guizatia scabra and Melinis minutiflora.

Cultivated pasture is occasionally used. A diverse range of fodder grasses and tropical legumes have proven successful in long-term testing. Among the best-known fodder grasses are Brachiaria ruziziesis, B. mutica, Setaria sphacelata, Panicum maximum, Cenchrus ciliaris, Cynodon plectystachyon and Digitaria unfulozi. The better known legumes are Centrosema pubescens, Desmodium uncinatum and Stylosantes guianensis.

However, these pastures are expensive to establish and maintain and require costly fertilizer. In practice, only part of the available biomass is utilized and wastage may be in excess of 50 percent, however, controlled grazing can limit this wastage to 25 percent. Neither the species or the limited composition of cultivated swards make them suitable for goats. To sum up, the establishment of cultivated pastures can seldom be justified for small ruminant production.

Fodder Crops represent a further opportunity for feeding sheep and goats. The productivity of fodder crops, combined with the limited dry matter intake of small ruminants, allows satisfactory results to be obtained from a small cultivated area.

Intensive fodder plots permit either year-round feeding using zero-grazing or tethering, or to assure sufficient feed during critical periods of the year. However, in densely populated regions with high crop intensities, available land is extremely limited for either cultivated pasture or fodder production, particularly during the first planting season.

Thus the following procedures have been recommended:

Non Conventional Feeds (NFC) are supplementary feeds which should not be overlooked and can represent a significant percentage of small ruminant rations at certain periods of the year. Principle NFCs in the pilot farms include:

Genetic Improvement

Genetic improvement may be achieved through either selection from within local breeds or by introducing exotic breeds for crossbreeding. However, even in those countries which have followed a policy of importation, priority is still given to studying of the potential of indigenous breeds which have show excellent adaptation to the local ecological conditions and which represent an indispensable genetic resource.

Therefore, before ambitious crossbreeding programmes are undertaken, a careful selection of the best local breeding animals, principally male, would be advisable.

Internal Selection within Local Breeds: Such a selection could include two aspects: culling and establishment of a reservoir of quality breeding animals, principally males.

Small ruminant flocks appear to lack good breeding males. Often producers rely on the free-ranging males from neighbouring farms to service their females. Moreover, they usually sell their young males quickly to avoid the risk of theft or conflicts resulting when females in heat are pursued over planted fields. The consequence is that negative selection takes places with the weakest and slowest growing males surviving and often only immature males being available to serve on-heat females. Even when a producer acquires a high-quality male it is usually sold immediately after the females become pregnant.

Therefore, the introduction of service stations is desirable. Such stations would be responsible for acquiring good quality bucks and rams and making them available on demand. Selection criteria should be based on phenotypic appearance plus whatever performance information is available eg. weaning weights.

Such service stations could later be responsible for taking some of the resulting offspring to initiate progeny testing. After a few years, this would identify selected local animals as the possible basis of a line-breeding programme. In order to facilitate collection, analysis and use of data, service stations would need to be associated with and academic or scientific institution.

Introduction of Foreign Breeds Through Cross-breeding. The choice of cross-breeding scheme depends on the objectives of the breeding programme:

However, not all individual animals are suitable for crossbreeding and success is best assured when starting with a stock of local animals of recognized quality based on performance registration.

A number of cross-breeds have been tested in the densely populated highland regions with varying degrees of success. Considering the reduction in cattle numbers and milk being a traditional component of the diet, local goats were initially crossed with milk goats including the Alpine, Anglo-Nubian and Saanen. These crossbreeds were not however popular and their lean conformation was not appreciated by local goat-raisers. Furthermore, the market for goat milk is limited principally to the urban areas. The Boer breed, better suited for meat production, has been more favourably received. However, it remains difficult to obtain a sufficient supply of these animals for breeding purposes.

A few crossbreeding attempts have been undertaken with sheep. The Romney-Marsh was introduced to cross with indigenous breeds but was not a success and the programme has been discontinued. One reason is that the situation in Burundi is totally different to the highlands of neighbouring Kenya and Ethiopia were numerous exotic sheep breeds have been tested with some success, notably: the Merino, Corriedale, Hampshire, Romney-Marsh, Awassi and Dorper.


Very little is known regarding the marketing of sheep and goats and systematic studies are required to determine the following factors:

Extension Service

A general complaint of producers concerns the lack of technically qualified extension staff to assist in small ruminant production. Such services are practically non-existent at present and without them, farmers cannot see the potential for improvement.

Not only do the extension services provide insufficient advice and training in the small ruminant production they also lack logistical support, infrastructure and even the simplest support materials.

It is interesting to note that those African countries that are able a policy to expand their small ruminant production (Kenya, Ethiopia, and the Ivory Coast) have also progressively developed their extension services. In Burundi significant efforts have been made in the area of training, both of extension staff and farmers.

The extension service should be responsible for addressing the following main constraints associated with sheep and goat development:

Training programmes need to be conducted at all levels:


Integrated Small Ruminant/Crop Programme

This project was well accepted by farmers, who were enthusiastic and the number of volunteers exceeds the project's capacity to provide fodder and animals. At the outset, farmers fulfilled all conditions required for the loan of 5 ewes and/or does, although, thereafter, various features and impediments were identified.

Apprehension regarding theft of animals induced farmers not to use pens but to bring their animals back to the family house. A number are continuing to keep sheep and goats outside in pens but will usually guard them at night. In both cases the production of a large amount of well-rotten manure is definitely compromised.

Fodder Crops. This component was not satisfactory and supplementary feeding is irregular. Harvesting of fodder was not carried out properly or at the right time. Increase of fodder production did not match increases in flock size. Furthermore, most of the fodder crops are multi-purpose and are also used for mulch and thatch - these alternative uses require an advanced stage of lignification. It is therefore important to make an inventory of all the alternative uses of fodder crops at the farm level and outline a chronological schedule that takes into account seasonal priorities and the required vegetative stage in order to fulfil both family and animal requirements.

With planning it should be possible to develop cropping schedules that allow for the maximum production of fodder, using conservation techniques as appropriate, as well as satisfying the thatch, stake and mulch requirements of the farm.

Non Conventional Feeds. Available NCF are not fully utilized correctly. A survey was conducted in 1989–1990 in three similar projects to assess the quantity and quality of NCF at farm level, seasonal availability and possibilities of storage. Data collected is being analyzed and a report is in preparation.

Special attention has been given to Acantus spp., a natural thorny shrub widespread throughout the whole country and growing mainly on short-time follows and road sides. It has good nutritive value (±25% Crude Protein) and is well accepted by goats, but not by sheep. Farmers hesitate to use it because it is supposed to favour the expansion of echthyma.

Genetic Improvement. The programme is constrained by the lack of young males and negative selection where the fastest growing males are sold at a young age for slaughter and not kept for breeding. A need for good males, therefore has been identified and a programme has been started to purchase, breed and exchange young bucks and rams in Selection Centres located within the different projects involved in the National Network for Research and Development of small Ruminants. All performance data collected by the projects are collated in a data bank and analyzed by the Department of Animal Science (University of Burundi) which co-ordinates the National Network.

Elite males, identified in one experimental site, are exchanged with other projects to avoid possible in-breeding.

Sustainability: Effective sustainability is a long term process and should assure the effective integration of the various components of the system. After four years experience with the National Network the following positive features can be highlighted:

Outputs of the programme were mostly positive, however, a negative impact was identified concerning women's rights. Within the traditional system, only small numbers of goats and sheep were kept and some could be owned by women and children. Subsequently, when the flocks increased and became an important part of the family income, men no longer accepted such sharing and claimed total ownership of all animals.

Another negative aspect is the lack of motivation amongst extension staff. The majority of the extension staff express little interest in small ruminant production and are, in any case, unskilled in the area. It would probably take a long time to build up an effective extension service at the smallholder level.

Combined Sheep Production/Reafforestation Programme

This programme has been conducted at three different sites:

All three projects were implemented in state-owned forests which had over-grazed common pastures. Since the first project has been abandoned, this paper will only discuss the last two projects which are funded by the World Bank. Research programmes examined potential management systems in order to maximise both outputs (timber and sheep) without detrimental effect on environment.

The results could be briefly summarized as follows:

Fertility (percent)96.097.6
Prolificacy (percent)111.0130.0
Lambing interval (months)8.88.0
Age at first lambing (months)15.015.0
Mortality rate 0–12 months (percent)15.06.9
Birth weight (kgs)  

The main objective of the project was to establish an experimental farm which, in addition, will also demonstrate to local sheep producers the benefits of combining sheep in reafforestation areas. It will also allow producers to exploit progressively the state forest. To achieve this it is necessary to gather individual flocks together in communal flocks to facilitate and to simplify shepherding. Due to the individualistic nature of most local farmers it is extremely difficult to put this into practice.

Under-tree grazing is not sufficient to satisfy the feeding needs of sheep, especially during the dry season (May to October). Therefore, it is important to grow fodder crops, which can be established along the fire breaks, to supplement the diet. The main fodder crops/trees used are: Tripsacum laxum, Setaria splendida and Leucaena diversifolia.

Previous Page Top of Page Next Page