Previous Page Table of Contents Next Page


R. T. Wilson


Latest estimates of livestock populations in nine East African countries provide a total of about 64 million sheep equivalent to about one half of a sheep per head of the human population in the area. Sheep represent about 8 percent of the total domestic ruminant biomass but, because of their rapid reproduction rate and the early age at which offtake occurs, they provide a disproportionate amount of animal products. This is probably equivalent to about twice the figure represented by their biomass.

With the exception of some commercial enterprises in Kenya whose production goals are either wool or meat all sheep in eastern Africa are managed in traditional systems. Ownership patterns and flock sizes vary, depending on whether livestock is the principal occupation or a subsidiary one of the owners. The end product is almost entirely meat, either for home consumption or to an internal or external market through sales. In parts of Sudan, sheep are also kept to provide milk.


The productivity levels of traditionally managed sheep are higher than has generally been accepted. Long-term studies carried out over periods of several years have shown that the annual reproductive rate (number of young produced per breeding female per year) varies from 1.5to over two lambs. This rate of reproduction results in part from the uncontrolled access of rams to ewes on a permanent basis and in part from the litter size. Some representative annual reproductive rates for a number of African countries are shown in Table 1.

The interacting effects of litter size and the parturition interval are clearly demonstrated by Table 1. “Control” of breeding usually means attempts to induce lambing at what is considered to be the most favourable period of the year. This control normally results in longer parturition intervals than are found when breeding is allowed year round. Control leads to only one lamb crop per year for example in Rwanda, under station conditions and slightly in excess of one in the traditional Masai system in Kenya. In the semi-arid to arid zones of Mali and Sudan, uncontrolled breeding results in shorter parturition intervals enabling approximately three lamb crops to be obtained in two years. Shorter intervals coupled with moderate litter sizes lead to relatively high annual reproductive rates compared with longer intervals even if sheep are highly prolific.

Total lifetime production of young can be increased by encouraging first lambing at early ages. In most traditional societies, first lambing occurs at 15–18 months when ewe weights are 80–85 per cent of mature size. Control of age at first breeding usually means delaying this and may result in first lambing not taking place until 2 years or older.

Small Ruminant and Camel Group, International Livestock Centre for Africa, P.O. Box 5689, Addis Ababa, Ethiopia.

The growth rate is an important factor in livestock productivity. In traditional systems, because of overstocking, genetic potential is rarely expressed. Growth rates vary from as little as 40 g per day in Kenya Masai sheep to as much as 70 g per day in Sudan Desert type from western Sudan. These are averages for all sheep from all birth types over all seasons. Male sheep born as singles in the pre-rains period usually have much higher growth rates to weaning. As an example of the potential for increased growth under improved conditions of nutrition and management, the “Mouton de Case” sheep in West Africa achieves a growth rate of 117 g per day to 40 weeks of age compared with only 60 g for its range-reared contemporaries.

Management practices in many traditional societies are such that the best adapted sheep or those with superior genetic potential are not used as breeding stock. This is because of the cultural or religious requirements for large fat sheep for slaughter at social and sacrificial occasions.

Pre-weaning mortality has been shown to be an extremely important constraint on productivity of sheep. Levels of up to 30 or even 40 per cent losses before weaning are not uncommon. Examples of the rates of death in several areas and some of the environmental variables affecting the death rate are shown in Table 2.

From the three production parameters just discussed, it is possible to construct a series of simple productivity indices. These can be calculated as total liveweight of weaned young produced per breeding female per year (I) or liveweight produced per unit weight (II) or per unit metabolic weight (III) of female per year. Some examples of overall productivuty indices are provided in Table 3. These are affected by the same variables as shown in Table 2.


The standard approach to improving the supposedly unproductive indigenous African sheep types has been to import exotic breeds, usually of European origin. They have then been kept on research stations, “under mosquito nets” as one French colonial administrator once described it, and given all the care and attention needed to force them to survive under climatic, disease and nutritional conditions totally alien to them. There have rarely been successful transfers of these breeds (or crossbreds with high levels of exotic blood in them) to traditional systems. In East Africa, successes have almost entirely been confined to those cases where modern management practices can be assured and high levels of veterinary and nutritional inputs maintained.

It does not appear that this classic method can be expected to produce worthwhile improvements in sheep production across the whole spectrum of pastoral and agro-pastoral systems to be found in East Africa. The methods employed to improve productivity must be geared, especially in the initial stages, to the existing systems of production and their capacities to provide managerial, nutritional and health inputs. “Improved” animals to be used in or released into these systems must also be capable of surviving under prevailing conditions. Table 3 has shown the variation among African indigenous breeds in productivity: it should be perfectly feasible to use these as improvers by making use of their comparative advantages, whether these relate to reproductive performance, capacity for growth, high survival rates or tolerance of mediocre management. Experiments under controlled station conditions should be paralleled with continuous monitoring of traditionally owned and managed flocks.

Where outstanding performance in indigenous flocks is noted, the reasons for this should be identified and, where possible, extended to contemporary owners and even research stations. If better performance is shown to be due to individual animals, these should be used to improve further the overall production. There are good reasons to believe, from the evidence of recent studies of traditional systems, that such outstanding animals do exist and may carry single genes which would enable rapid improvement in overall reproductive performance or in tolerance to internal parasites, for example.

A considerable degree of improvement could be achieved, however, in traditional flocks by identifying the environmental variables within the productivity indices which contribute most to the variation in productivity. From these variations, it should be possible to design pathways which, sequentially in order of their importance, would improve in a cost- and resource effective manner overall flock productivity. As examples of the kind of exercise being advocated Table 4 shows ratios of comparative advantages calculated on an index of productivity for unit weight of dam for a number of environmental variables. It should be noted that some of what might be considered the obvious improvements, such as selection for twinning, might not be appropriate in every case. The disadvantage to twins in Table 4 results from the very high death rate of this class of animal coupled with the low growth rate of any surviving twins. Increasing the viability of twins would, of course, then lead to an improved index and a high precedence in the improvement programme.

The greatest advantages are due to flock, which can to a large extent be associated with individual management practices and abilities. Identifying these practices and abilities and extending them to other owners would lead to overall improvement. A plan for improvement of a traditional flock with the minimum of outside and costly interventions is shown in Figure 1. At a later developmental stage, the order of application of improvements might change if for example, as already mentioned, it were possible to reduce mortality of twins, to improve the growth rate of twin-born lambs or to distribute superior stock with high genetic potential for reproductive improvement or disease resistance, yet capable of withstanding the local environmental and management conditions.

TABLE 1. Litter size, parturition interval and annual reproductive rate (ARR) for sheep in some African livestock systems

Country-SystemLitter SizeParturition interval (days) ARR
Sudan:- pastoral1.142751.51
Ethiopia:- pastoral1.033151.20
 - agro-pastoral1.053651.05
Kenya:- pastoral1.053121.23
Rwanda:- station1.543651.54
Mali:- agro-pastoral1.042611.45

ARR = Litter Size × 365/Parturition interval

TABLE 2. Pre-weaning death rate (% ) for different variables at six sites

VariableSudanMaliKenya I1Kenya II1NigeriaEthiopia
Overall LS Mean30.
Sex: male30.927.918.216.715.010.5
Birth type:single23.1a22.9a6.7a9.6a16.012.1
 twin36. 7b33.0b31.4b22.9b14.013.1
Birth season3 :A26.9a32.8a17.2a12.2a-4.6a
 D22.7a27. la22.7c13.2a-21.4b
Parity: 127.6a38.6a28.6a-19.019.5a
 9434.4c26.2b12.2c-13.05.4c la

Notes: 1 Kenya 1 = Elangata Wuas Group Ranch 1978–81; Kenya 11 = Three Kaputeo Section Group Ranches 1981–83

2 Sudan A = sedentary, B = migratory; Mali A = rainfed millet
B = irrigated rice; Kenya 11 A = small flocks, B = large flocks

3 Sudan, Mali, A = cold dry, B = hot dry, C = rains, D = post rains;
Kenya, A = short dry, B = long rains, C = long dry, D = short rains; Ethiopia, A = best month, D = worst month

4 Parity 9 = parities ≥ 4 except Nigeria ≥ 2 and Ethiopia ≤ 3

Within variables means in same column with different superscript differ significantly.

TABLE 3. Productivity indices for some African sheep types under conditions of traditional management

Sheep type
Index 1
Index 11
Index 111
Sudan - Desert22.25981.47
Ethiopia - East-Africa Fat Tail16.95821.45
 - Menz16.17041.64
Kenya - Masai14.14731.18
Mali - Sahel29.18702.24

TABLE 4. Ratios of comparative advantage for sources of variance in an index for unit weight for Kenya sheep

Sources of varianceRatio
Flock: best to worst1.76
Birth season: best to worst1.67
Birth type: singles to twins1.31
Parity: overall mean to first1.29
Sex: female to male1.02

Figure 1 Potential improvement pathways for traditionally managed small ruminant flocks on Masai group ranches

Figure 1

Previous Page Top of Page Next Page