Livestock Research for Rural Development 15 (1) 2003

Citation of this paper

Livestock production and farm animal genetic resources in the Usangu Wetland of the Southern Highlands of Tanzania 

R Trevor Wilson

 Bartridge House, Umberleigh, North Devon EX37 9AS, UK



The Usangu Wetland in the Southern Highlands of Tanzania is, and always has been, an important area of animal production.  Secondary and primary sources of information were used to establish or confirm the types of livestock production systems and animal genetic resources present in the area and the basic productivity parameters of the various domestic livestock species. 


This paper first describes the physical and social environment of the Wetland and then the main production systems.  The farm animal genetic resources of the area comprise cattle, goats, sheep, donkeys, pigs and four species of poultry.  These are briefly characterized and information on production and productivity is provided. 

Keywords: environment, genetic resources, livestock production systems


Tanzania has the third largest cattle population (13.6 million) in Africa after Ethiopia and Sudan.  It has many goats (8.6 million) and sheep (2.7 million), lesser numbers of donkeys and pigs and a very few horses and domestic buffalo.  Poultry comprise domestic fowl (more than 90 per cent of all poultry), Muscovy ducks, Guinea fowl, pigeons, geese and turkeys (Statistics Unit/Bureau of Statistics 1996).  Livestock contribute 30% of agricultural Gross Domestic Product (GDP) which itself is equivalent to 50 per cent of total GDP.


The main preoccupation of most livestock owners in the often harsh environments of the tropics in the past has been and now is to maintain and reinforce the inherent natural survival strategies of their animals.  It has been said that "nature holds sway and elimination of the constitutionally unfit is the rule rather than fostering of the physically desirable" (DVSAH 1929).  Tanzania (and its predecessor the UN Trust Territory of Tanganyika) has long been conscious of the economic and ecological value of its indigenous farm animals.  Thus, in statements well ahead of their time (DVSAH 1926) it was noted that:


"Few people to-day except the natives themselves 'in whose minds there is no dubiety' appear to realise that in Tanganyika the actually convertible capital value of the livestock of the Territory if sold up to-morrow exceeds the sum of the values of the assets of all other industries combined".


"... the, as it were, hidden attributes of the Zebu, such as disease immunity and hardiness are of at least as great value as the revealed ones of physical form and what such form stands for, and that because of this no justification exists for despising such animals because they fail to come up to European standards of excellency".


It would not therefore be surprising if local indigenous domestic animals do not produce large quantities of milk and do not grow rapidly.  This must not be considered synonymous with poor productivity.  Production has been, and continues to be, adapted to the environment.  This fact is not always understood by administrative services who often believe that traditional livestock production is inefficient and wasteful and competes with other sub-sectors of agriculture and the economy as a whole for scarce resources or even deprives them of such resources.  This attitude is particularly prevalent with respect to the Usangu Wetland where livestock are highly visible in the local economy and where most are owned by ethnic groups that are not particularly well represented at the higher political and administrative tiers.  Data on the real contribution of livestock to local production systems and the regional and national economy are essential if sustainable development in financial, socio-economic and environmental terms is to become a reality.  This paper aims to provide some of this data for the construction of a regional development programme and for use by wider administrative, political, development and scientific communities. 


Materials and methods

This study was part of a broader examination mounted to identify the resources of the Usangu Wetland and to make proposals for their sustainable use.  Data for the livestock component were obtained from a combination of secondary and primary sources. 

Secondary data sources included:


Primary sources of data covered:

Usangu is in the Southern Highlands region of Tanzania.  Mbarali District, in which it is located, is bounded by coordinates encompassing 33o 30'-35o 00' East and 7o 30'-9o 00' South.  The vast alluvial wetland or Plains themselves, also known as the Usangu Flats, occupy about half of the District's area of 16 000 km2 and are at an altitude of 1100 to 1220 masl.  The Poroto Mountains and the Kipongoro Range lie to the south but the north is lowland.  Low mountains lie to the east and the Mbeya Range is to the west.  The southern mountains receive rainfall that averages more than 1000 mm in the catchment's more elevated areas.  Several rivers debouch from here onto the Plains before they spread out as large alluvial fans and then converge to join the Great Ruaha River as it flows east through the central Plains.  The northern area of small alluvial fans with extensive clay soils receives less rain than the south.  Rainfall to the east is also low and only one perennial river enters the Plains from this direction.  The network of seasonal and perennial rivers that crosses the Plains mean that much of the area is submerged every year for long periods: this type of land form with black cracking clay soils is known in Kiswahili as 'mbuga'.  To the northeast the Great Ruaha eventually flows over a rock bar that slows its movement and has resulted in the major perennial swamp known as 'ihefu' locally but as the Utengule swamp by others.  Some smaller permanent swampy areas are scattered throughout the region.


The climate is tropical semiarid.  A long dry season lasts from April to November.  Rain falls from November to April but there is often a drier spell in mid season.  Long term average annual rainfall in the Plains is 703 mm but there are large between year variations: in the period 1989/1990 to 1997/1998, for example, the variation was from 349 mm to 1046 mm.  The spatial variation of rainfall within the Plains is also very high and local droughts are common.  There is relatively little intra-annual temperature variation but this is highest in October and November at the end of the dry season.  Lowest temperatures occur in May to August.  Relative humidity is highest in February at 70% and lowest in October at 46%.  Increasing temperatures and relative humidity before the rains break in November or early December lead to very uncomfortable conditions for people and animals.  Wind run is high and the strong winds that blow during the dry season carry much dust and add to the general desiccation of the area.

Livestock Production Systems

Livestock -- especially cattle -- are a traditional component of Usangu indigenous production systems.  The Sangu (who were the original inhabitants and from which the Plains take their name) were notable owners of livestock  in the second half of the Nineteenth and the first half of the Twentieth Centuries.  In the early 1890s they are known to have had many cattle.  At least 90% of these were lost in the years preceding the turn of the Twentieth Century as the first devastating rinderpest pandemic swept through the whole of continental Africa's previously unexposed and entirely susceptible cattle population (Mack 1970).  Sangu cattle numbers never completely recovered to their former level.


In the early 1950s cattle totals again began to build up (Table 1) as immigrant herders and especially the Masai from Arusha, Dodoma and Tanga Regions began to arrive in the Usangu Plains.  Immigrant numbers remained low for some years but further influxes of herders, mainly Sukuma from Mwanza, Shinyanga and Tabora Regions, in the 1970s and the 1980s led to a massive increase in livestock numbers.  The animals of smaller numbers of other pastoral groups contributed to the total livestock biomass as they moved into the Usangu Plains in the period 1970-1998.  In 1998 the slightly more than half a million cattle considered by the local authorities to be occupying the area would have been equivalent to 4.0% of the national herd on 1.9 % of the country's land area.  In reality, however, aerial surveys carried out as part of the broader survey of which this study was part indicated a probable total of 301 000 cattle occupying the Wetland itself in the dry season of 2000 (Graham 2001). 


Table 1.  Livestock numbers in the Usangu Wetland, 1926/1927-1997/1998




Livestock species and number




Total TLU#


Source of data












66 865












Mbeya Regional Book, Volume 1




44 924












Mbeya Regional Book, Volume 1




187 244


31 828


24 107






136 840


Mbeya Regional Book, Volume 2




226 269


38 785


27 555






165 156


Mbeya District Annual Report, 1953




541 645


16 047


23 826


1 346




383 811


Mbeya District Livestock Office, File VYMC/2558




437 821


34 204


39 178


2 443




314 034


National Livestock Census, Mbeya Wards




457 759












Mbarali Subdistrict Livestock Office Annual Report




548 291


42 616


43 783


2 393




393 641


Mbarali District Agriculture/ Livestock Extension Report




300 963


94 597


3 202




221 735


Aerial survey dry season 2000


















% change 1948-1998















# Tropical livestock unit of 250 kg - 1 head of cattle = 0.7 TLU, 1 sheep or goat = 0.1 TLU, 1 donkey = 0.5 TLU (author's conversion factors based on estimated livestock demographic structure and mean weights for age)

##           Mbeya District total


There are four agroeconomic zones: a southern crop production zone; a central crop-livestock transitional zone; a pastoral zone to the north of it; and a northern sparsely inhabited zone.  Most cattle, sheep, goats and donkeys occupy the central crop-livestock transitional zone and the pastoral production zone to the north of it.  Here there are seasonal movements to and from the 'mbuga' as the annual inundation proceeds and recedes during and after the rains.  An aerial reconnaissance in which the author participated in early December 1998 showed that, except for relatively small numbers belonging to the mixed farmers of the southern crop production zone, animals were concentrated in large herds on the western and eastern edges of the Utengule swamp.  Livestock in the southern crop production area are in small herds.  Most of the oxen kept for draught and other agricultural work are also in this zone and it is here that most poultry and almost all pigs are found.  The northern zone is virtually permanently devoid of domestic stock due to the problems associated with tsetse-borne trypanosomosis and lack of water during much of the dry season.  There are thus two major livestock production systems in the catchment and a third minor system.


The first major system is the almost pure grazing one of the immigrant herders or pastoral tribes.  There is little integration with crops and the system mainly uses native grassland for its production inputs.  Livestock present little other than a minor burden on the environment in their possible eutrophic effects within and without the Usangu Wetland.


The Usangu Wetland is a special location for animal production in the Southern Highlands peopled as it is by "professional" livestock owners.  Herds owned by these groups are well above the average size for the Region as a whole and have a different age and sex structure.  An earlier study (Charnley, 1994) showed that all of 41 Masai households included owned cattle.  In 25 households that gave information on the size of holdings 36% had herds that were under 50 head in size with an average number of cattle owned per person of 1.87.  Some 36% of herds had 51-150 cattle with an average holding of 5.65 cattle per person and 24% of herds were 151-350 head in size with an average holding of 15.80 animals per person.  Only 4% of herds comprised more than 350 cattle with each person having 17.24 cattle.  The overall average herd size was 130 animals equivalent to an availability of 8.07 animals per person.  Other studies have estimated that East African pastoral households that consume meat, milk and blood from their herds need 8.33-10.67 cattle per person for subsistence alone.  In addition to cattle 67% of households owned goats with the average flock size being 17 (range 5-40) animals at an availability of 0.86 head per person.  Some 57% of households owned sheep with an average flock size of 15 (2-40) animals at an availability of 0.82 animals per person.  An average of 4 (1-10) donkeys were owned by 61% of households at an availability of 0.16 persons per person.


In the same study reported for the Masai all 50 Sukuma households that provided data on ownership kept cattle, 86% had goats, 83% had sheep and 51% owned donkeys.  Sukuma cattle herds were larger than Masai.  They averaged 285 (range 4-800) animals, goat flocks averaged 22 (5-50), sheep flocks averaged 23 (5-63) and donkey herds averaged 4 (1-12) animals.  Average availability of animals per person in these Sukuma livestock units was 15.5 cattle, 1.2 goats, 1.1 sheep and 0.2 donkeys.  Within the households holding cattle, 20% owned under 50 animals (3.17 per person), 25% had holdings in the range 51-150 head (6.29 per person), 23% owned 151-350 head (14.69 per person) and 14% owned 351-500 head (17.34 per person).  More than 500 head (32.61 per person) were owned by 18% of households.


The second major system is that of mixed crops and livestock practised by some of the cultivators of the southern part of the area and the higher reaches of the catchment.  In this system nutrient flow is a two way thoroughfare from crops to animals and from animals to crops.  Energy circulates from crops (as residues and by-products for feed) to livestock and back again (as draught power, transport and manure).  A sample census of agriculture in 1994/1995 (Statistics Unit/Bureau of Statistics 1996) showed that the 39.9% of households in Mbeya Region that owned livestock averaged 4.77 persons and kept 8.95 cattle, 5.11 goats, 3.96 sheep and 8.13 "chickens".  (Note that the pastoral herds of the Usangu Plains were not part of the sample frame.)  In the study carried out for the purposes of this paper the average number of cattle per family in 16 mixed farming or agropastoral households in three villages was 17.7 in the range 4-45.


Cattle herd demography derived from the author's sample survey includes 27.2% male and 47.2% female (cows 29.4 per cent, heifers 17.8 per cent) animals older than "calves".  If half of the 25.6% "calves" are male the percentage of males and females in the herd would be 40 and 60.  Among older males 9.5% of the total herd is bulls, 7.1% oxen and 10.6% steers or castrates.  Assuming that half the heifers are "breeding" females the imputed calving rate of 0.67 young per breeding cow per year (25.6% ¸ (29.4% + 8.9%) x 100) is rather high.  The authorities in Mbarali estimate 8089 (1.47 per cent) of the total of 547 691 cattle in the District are work oxen.  This very low percentage of oxen in relation to the regional one can at least in part be accounted for by the large population of pastoral as opposed to "agricultural" cattle.  In the sample of herds in the 16 mixed farming households in the three villages already referred to the breeding females were equivalent to 40% and work oxen and other "non-reproductive" males to 19% of the herd.

The goat population comprises about 30% males and 70% females.  The sheep population has a lower proportion of less than 20% males and a higher one of over 80% females (Table 2).  These data support a hypothesis of early offtake of males for slaughter or sale as even in the class under 1 year old there are more females than males.  They also favour the premise that goats are reproductively more efficient than sheep as there are fewer mature females and more young in the goat than in the sheep flock.


Table 2.  Demography (per cent) of goat and sheep flocks in Mbeya Region









1 year +


< 1 year




1 year +


< 1 year


























The sample census shows that 9.8% of the pig population is mature males, 45.8% is mature females and 44.4% is immatures.  Some 46.5% of the donkey population is male and 53.5% female.  The poultry population is overwhelming composed of domestic fowl which account for 97.3% of all poultry: Guinea fowl are equivalent to 0.1% and ducks 2.6% of the total poultry population.


The third and minor method of livestock ownership and management in the area has developed in and around the main towns and larger villages and can best be referred to as a peri-urban system.  Stock are mainly kept as an investment or as a cash generating enterprise.  A large part of the urban livestock population is male that is growing out for eventual sale and another important proportion comprises crossbred dairy cattle.  These supply the demand for milk that develops in cash economies at household level and in the innumerable "guest houses" and restaurants that spring up in this situation.  Some potentially high yielding goat breeds are beginning to appear in this system.  Most owners are not "stakeholders" in any real sense in the local natural resource endowment but merely exploit it for short term and entirely personal gain.


The supposedly very large livestock population is widely considered by politicians and administrators to be responsible for reduced water flows in the Great Ruaha River leading to lowered hydroelectric output and destruction of the natural vegetation of the Wetland.  (The District Commissioner informed the author that the "real" number of cattle was twice as high as admitted by owners and was in excess of one million head.)  As, however, livestock do not have access to the major part of the Wetland during the flood season they do little damage to the perennial grass vegetation which largely comprises Echinochloa stagnina (author's observation).  Large areas of rice in the lower catchment and around the edges of the flood plain certainly use much more water than livestock and are also likely to be responsible for more pollution from increasing use of mineral fertilizers. 

Livestock genetic resources of the Usangu Wetland

The FAO data base on domestic animal diversity lists a total of 20 "breeds" of all species combined for Tanzania.  Three of these are listed as at risk of extinction (Scherf 2000) although it is far from clear that they are, in fact breeds.  The Zanzibar Zebu "is a variety of Small East African Zebu with some Indian, Somali and Boran influence" and is classed as critical.  The Chagga is also "a variety of Tanzanian Zebu ...... not yet [..] characterized" and is considered endangered.  The Mpwapwa or Indo-African Zebu "originated from Sahiwal in the 1940s and the remaining animals are now found mainly in research stations": in reality many more types of cattle were used on the foundation Tanganyika Shorthorn Zebu, the animals have hardly ever been off research stations and it has never been a fixed and true-breeding line in the 50 years devoted to its attempted development.


The farm animal genetic resources of Usangu are a microcosm of those of Tanzania as a whole.  They comprise cattle, goats, sheep, donkeys, pigs and poultry.  Goats have a broader ecological range than the other main species (Hornby and van Rensburg 1948) but in general are unable to take advantage of their eclectic dietary tastes as they are herded along with cattle and sheep.


Cattle are the major species in Usangu as indeed they are in Tanzania as a whole.  They are overwhelmingly thoracic-humped Bos indicus Zebu of the Tanganyika Shorthorn Zebu (TSZ) type (Mason and Maule 1960; Mason 1996).  This genotype has always been (DVSAH 1929) the country's most common bovine and accounts for more than 95% of indigenous cattle.  Several local strains were known in the past but one authority considers that they are no longer recognizable (Mason 1996).  In addition to the TSZ there are some cervico-thoracic long lyre-horned Sanga cattle that probably originated from west and southwest of Lake Victoria.  These are almost certainly of Ankole provenance but do not have the typical red colour.  They are kept by the Masai rather than the Sukuma whose own origins are geographically closer to the Ankole cattle homeland.  There are some crossbreds between the two main cattle types but most are bred pure.  The animal husbandry services report already referred to (DVSAH 1929) stresses that Usangu cattle were grazers and "suffer exceedingly" if moved to bush areas.


TSZ cattle are of various and usually mixed colours.  They grow slowly.  Males reach an average maximum of about 300 kg live weight (although 240-270 kg is a more common range) at five years of age (Hutchison 1958; this study).  Females usually average 160-180 kg so that the weighted average mature mass of all adult cattle is 210-240 kg.  They produce little milk and meat.  They are, however, well adapted to the harsh environment and long periods of under-nutrition during the dry season.  The possible innate resistance of Masai cattle to tick infestation (de Castro 1991) may be reinforced by behavioural characteristics such as mutual grooming to reduce tick burdens even further (Norval et al 1988).  The Ankole type have longer legs than the TSZ and are physically bigger at about 300 kg live weight.  They probably have better milk and meat production characteristics but may be less resistant to some local stress factors.


There is a considerable number of crosses of zebu cattle with Bos taurus or European stock in the periurban system.  These are mostly of black and white Holstein-Friesian provenance but many animals show clear indications of Ayrshire and possibly of Guernsey blood (the last two breeds were favoured by European settlers in the Iringa and Mbeya Regions [personal observations during the early 1960s by the author]).  Almost all crossbred Bos taurus-Bos indicus cattle are of very poor conformation with light bones and narrow guts.  This is almost certainly because they do not receive the management, nutrition and health inputs that they require if they are to perform to their genetic potential.


Goats are all of the generalized Small East African type (Mason and Maule 1960; Wilson 1991; Mason 1996).  These have short backward curving horns in both sexes.  Toggles or tassels are present in 15-20% of both males and females.  Body colours are very variable but most animals have a coat of at least two colours and that of many is of several shades.  Withers height is usually in the range 55-60 cm.  Mature weights of males are in the range 28-35 kg and those of females range from 25 kg to 30 kg.  These animals are good scavengers.  They are also somewhat prolific breeders and produce enough milk for the twins to which they often give birth to grow at a reasonable rate and be ready for slaughter at 8-12 months of age.  Local goats are tolerant of the harsh conditions and of some diseases (Baker et al 1998) and obtain their nutrients from a variety of grasses, herbs and shrubs.


A few exotic goats, mainly Toggenburg but including some Saanen, have been introduced by development projects.  A main condition of introduction of these animals is total confinement and zero grazing, supposedly in response to environmental concerns.  There is thus as yet little to no evidence of this genetic group in local goat populations. 

Sheep in the Usangu Plains are of the widespread East African Long-fat-tailed type (French 1938a; Mason and Maule 1960: Wilson 1991; Mason 1996).  It has been said that this sheep is stoutly built with a good spring of rib (DVSAH 1938).  In good condition, in addition to the fat tail, there is a band of fat on the poll and fat pads on the sides of the muzzle.  The dewlap is prominent.  Tassels and vestigial ears are not infrequent.  Colours vary and include red, white, black, mixed colours and white body/black head.  Shoulder height is 55-60 cm.  Mature live weights are heavier than those of goats but growth rates are slow (French 1938b, 1942; this study) and carcass quality is not very good (French 1944).  Most males surplus to breeding needs are slaughtered before one year of age.  Local sheep are not as hardy as goats nor as prolific and single births are the rule.  Resistance to helminth parasites may be present in Masai-owned sheep as indicated by owners in this study and supported by experimental evidence in the type in East Africa (Preston and Allonby 1979; Gray et al 1995; Baker 1997).


Donkeys are of the common African grey type.  The population is mixed, however, with some animals tall at the withers for this type of donkey and some lighter coloured ones.  They are hardy, resistant to or tolerant of many common equine diseases and thrive on a poor quality diet obtained by scavenging on local feed resources.


A report of the late 1930s (DVSAH 1937) remarked that the author did "not know whence come the pigs found in native kraals in a few parts, particularly the southern parts, of the Territory, but think they are descendants of animals recently introduced by missionaries.  They are of miserable appearance, miserably kept and infested by Cysticerci cellulosae."  It is clear that the few pigs kept around Usangu are mostly of exotic provenance.  The English Large White or Yorkshire breed is the most likely progenitor.  There is some evidence of other breeds (including possibly the French Piétain) or of an aboriginal stock.  Most are very degenerate in type.


Domestic fowl ("chicken"), Guinea fowl, Muscovy duck and pigeon are the species of poultry kept in Usangu.  There is no distinct type of local fowl but there is much genetic variation evidenced by body size, feather colour and comb type.  Mature males weigh 1.2-1.4 kg and females 0.9-1.1 kg.  The bare-neck gene is common in domestic fowl and is an adaptation to the heat of the local environment.  Guinea fowl are the common grey type that weigh about 1.5 kg at maturity.  Muscovy ducks show considerable phenotypic variation in colour with males weighing 2.0-2.5 kg and females 1.3-1.7 kg at maturity.  Pigeons are of the common barred wing type.  

Production and productivity

Native or traditional livestock breeds are commonly considered unproductive.  What is generally meant, however, is that such animals have low levels of output.  Productivity implies some relationship between inputs and outputs whereas production is merely an output function.  The great advantages of the indigenous livestock of Usangu is their adaptation to the stressful local nutritional, health and management environments and their ability to produce any outputs at all.  In all discussions of "production" or "productivity" this essential adaptation must receive full consideration.


Physical outputs of the livestock of Usangu can be considered as end and intermediate products.  End products include meat, milk, eggs, hides and skins and live export from the area.  Professional range ecologists and livestock specialists have long considered the Usangu Wetland to have great potential for the production of meat for the country as a whole (UNDP/FAO 1967).  Intermediate products include draught power and manure.


Livestock production is considered by the local authorities as secondary to crops in the area's economy.  Most calculations of this type consider, however, only the end products of the contribution of stock to human welfare.  It is likely that if the full value of intermediate products were capitalized the economic contribution of livestock would surpass that of crop production.  Even if the (probably low) official estimate of 8090 draught oxen or 4045 oxen pairs in Mbarali District is accepted,  their contribution to local Gross Domestic Product (GDP) is not inconsiderable.  The cost of hiring a tractor for ploughing, including fuel, is TShs 15 000 per acre (TShs 37 000/ha).  If it is assumed that each oxen pair ploughs 4 ha/year (this may be a conservative estimate as local people with work oxen considered that a pair ploughed 15 acres or 6 ha at the start of the rains) the imputed total value of ploughing is TShs 66 750 000 (just over US$ 800 000 at November 1998 exchange rates).  Other work  in seeding, harrowing and transport may be equivalent to this amount.  The value of manure is also considerable as N, P and K returned to the soil plus improvements in soil structure and water holding capacity.


Livestock have many important non-physical and even non-financial outputs including their functions as repositories of wealth and media of exchange.  Livestock also have roles in traditional religious and other customs and as dowry payments.


Tanzania's native cattle produce little output as individual animals.  In the country as a whole mature weights average well under 200 kg for cows but bulls and especially draught oxen can weigh considerably more.  Cows produce -- or can produce if forced -- 1-2 litres of milk per day in a 6-7 month lactation in excess of that taken by the calf.  In traditional management systems in general high mortality rates of 5-10% in young and adult stock and of as much as 25% in calves coupled to an annual calving rate of 40% (de Leeuw and Wilson 1987) limit potential herd growth and, more importantly, commercial offtake.  These data are supported for Usangu by information from owners who also indicated a lifetime production of about six calves implying culling or death at 12-15 years.  Cattle do, however, provide almost 90% of all Tanzania's meat output.  As an example, the weights at which most TSZ cattle are slaughtered can be imputed from those of 61 829 cattle trekked from northern Tanzania for slaughter at Athi River in Kenya in 1942.  These averaged 215 kg (472 lb) in the range 205-220 kg (450-485 lb) depending on the district from which they had originated (DVSAH 1942).  Some 20 years later 66 977 cattle slaughtered at the export abattoir in Dar es Salaam averaged 237 kg (521 lb) live weight and 22 476 slaughtered at Arusha weighed 250 kg (551 lb) (DVSAH 1961).


The District livestock services in the Usangu area estimate output at 150 kg meat per head and 2 litres of milk per day (MAC 1998).  These are almost certainly grossly over optimistic but they would, in any case, apply only to animals that were actually slaughtered (or exported from the area) or lactating.


Goats and sheep contribute about 12% to national meat supplies.  A rather small proportion of this enters the commercial chain and most is consumed within the family.  The proportion of goat and sheep meat to total meat production in the Usangu Wetland is probably less than the national average in view of their small numbers.  Carcass weights of both species at maturity are probably 10-12 kg (Wilson 1980, 1984; Wilson et al 1982) but many male animals are slaughtered at low weights and provide carcasses of 7-10 kg.  Goats provide some milk -- possibly 0.25 litres per day per lactating doe -- for household consumption but sheep are not usually milked.  Reproductive performance is slightly better in goats than in sheep such that females of the former species produce about 11 kids per lifetime whereas ewes produce nine lambs.


Skins of both small ruminant species but especially of goats are potentially of high value.  In spite, however, of more than 70 years of attempts to encourage -- or force -- owners to improve processing and preservation to increase financial returns very few skins enter the commercial chain.  They are rarely wasted, however, and are used for a variety of purposes at home including as clothes, mats, containers for liquids and in saddlery.


Donkeys provide transport in carts and via pack operations.  Local use of a team of donkeys in two spans of two results in their inefficient use in carts.  They are rather more efficient as pack animals but are used mainly by the pastoral tribes for carrying grain to and from the mill and only in a limited way for other pack activities.


Pig meat is not important in the national diet but Mbeya Region has a large proportion of the country's total swine herd.  Local pigs -- as are more than 90% of those in Tanzania -- are kept mainly under a free range scavenging system with some occasional provision of household waste, crop by-products and overripe fruit.  Total output of meat arises from a litter that averages about six piglets with each sow probably producing 1.5 litters per year.


About 80% of all domestic fowl in Tanzania are managed in small scavenging flocks.  Medium and larger scale commercial flocks are only found close to the main urban centres.  In the Usangu Wetland all poultry are kept in scavenging flocks.  Local informants were unanimous in stating that mortality is very high from both egg predation and in the early post hatching stages.  Start of laying by the local hen is at about one year.  Owners indicated she then lays about 30 eggs per year in three or four clutches but possibly only 50-60% of these result in a "product".  The average of a sample of eggs weighed in local flocks and at the market was 32.5 g.  Most clutches are incubated naturally with chicks left to survive and grow as best they can until they are sold or slaughtered for household use.  These parameters are similar to those recorded for indigenous fowl in scavenging flocks elsewhere in Africa (Wilson 1986).  Large differences between strains in age at laying of first egg and in annual egg production have been noted (DVSAH 1938) but under greatly improved management it appears that up to 150 eggs per year are possible.


The numbers of eggs produced by Guinea fowl and Muscovy ducks is estimated to be similar to that of local fowl although eggs of the former weigh 35-40 g and of the latter 40-50 g.  On the basis of information supplied by owners and data from Sudan (Wilson 1979), local pigeons probably produce about 16-20 eggs per year in clutches of two eggs. 


Discussion and conclusions

Domestic livestock have always been an important component of the agricultural economy of the Usangu Wetland.  Although generally considered to be second to crops in the local economy the value of output calculated as the sum of end and intermediate products may well exceed that of crop production.  They are a major source of income for the District Council (now expected by central Government to be largely self-financing) through head and other taxes.  In spite of these facts, however, livestock are subject to much prejudice at both District and wider levels.  They are blamed, for example, for degradation of pasture and other vegetative resources and the drying up of rivers and swamps.


There is no empirical evidence to support these postulates.  It is unlikely that severe environmental damage is occurring due to the traditional systems of livestock production practised in the area.  This is especially so on the 'mbuga' – the main source of feed -- which is protected from use during its vegetative growth and reproductive periods by the annual inundation at these times.  The "carrying capacity" [which, according to modern theories expounded, for example by Behnke et al (1993) and Scoones (1995), is a distinctly tenuous concept in traditional management systems] has been estimated at 526 000 TLU (King 2000) whereas the "stocking rate" appears to be considerably less than half of this at under 220 000 TLU (see Table 1).  Preconceptions will persist, however, until detailed studies identify the real causes of any degradation that is occurring.  Whatever the situation the best ecological and economic use of the grasslands are, and will remain, their conversion to animal products for consumption and use by people.  Their contribution to food security, equity and sustainable production in this context is indisputable.


Similar misconceptions exist over the "productivity" of native livestock.  The adaptive qualities acquired over hundreds of generations must be conserved in parallel with the need to raise output and improve productivity.  In particular it should be noted that the modern inputs needed to modify the total environment -- including health and nutrition -- of supposedly "improved" (i.e. exotic) breeds and their crosses if they are to produce to their genetic potential are unlikely to be provided in the short term.  Conversely, only minor improvements are needed to generate productivity increases in native and adapted livestock. 



Information for this paper was obtained while the author was employed under contract to Hunting Technical Services Limited (HTS), who in turn were contracted to the British Government's Department for International Development (DFID) on a project entitled "Sustainable Management of the Usangu Wetlands and their Catchment".  Neither HTS nor DFID can be held responsible for any errors of fact or opinions expressed in this paper as these remain those of the author alone.  The author wishes to express his gratitude to the local livestock owners of the Usangu ecological system and to District, Regional and National level officials for their varied assistance. 



Baker R L 1997 Résistance génétique des petits ruminants aux helminthes en Afrique. INRA Production Animale 10, 99-110.


Baker R L, Mwamachi D M, Audho J O, Aduda EO and Thorpe W 1998 Resistance of Galla and Small East African goats in the sub-humid tropics to gastrointestinal nematode infections and the peri-parturient rise in faecal egg counts. Veterinary Parasitology 79, 53-64.


Behnke R H, Scoones I and Kerven C (Eds) 1995 Range ecology at disequilibrium: New models of natural variability and pastoral adaptation in African savannas.  Overseas Development Institute, London.


Charnley S 1994 Cattle, Commons and Culture: The Political Ecology of Environmental Change on Tanzanian Rangelands. Unpublished PhD Thesis. Stanford University, Stanford.


de Castro J J 1991 Resistance to ixodid ticks in cattle with an assessment of its role in tick control in Africa. In: Breeding for disease resistance in farm animals. Eds. Owen, J.B. and Axford, R.F.E CAB International, Wallingford. pp. 244-262.


De Leeuw P N and Wilson R T 1987 Comparative productivity of indigenous cattle under traditional management in subsaharan Africa. Quarterly Journal of International Agriculture 26, 377-390.


DVSAH 1926 Annual Report, Department of Veterinary Science and Animal Husbandry. Government Printer, Dar es Salaam.


DVSAH 1929 Annual Report, Department of Veterinary Science and Animal Husbandry. Government Printer, Dar es Salaam.


DVSAH 1938 Annual Report, Department of Veterinary Science and Animal Husbandry. Government Printer, Dar es Salaam.


DVSAH 1942 Annual Report, Department of Veterinary Science and Animal Husbandry. Government Printer, Dar es Salaam.


DVSAH 1961 Annual Report, Department of Veterinary Science and Animal Husbandry. Government Printer, Dar es Salaam.


French M H 1938a Development of livestock in Tanganyika Territory. II. Tanganyika breeds of sheep, and mutton qualities of the long-tailed type and its crosses with Black-headed Persian sheep. Empire Journal of Experimental Agriculture 6, 112-124.


French M H 1938b Development of livestock in Tanganyika Territory. III. Skeletal measurements of local tailed and grade Black-headed Persian sheep. Empire Journal of Experimental Agriculture 6, 112-124.


French M H 1942 The growth-rates of local and cross-bred sheep. East African Agricultural Journal 8, 24-25


French M H 1944 Development of livestock in Tanganyika Territory. VI. The development of mutton qualities in Masai and grade Black-headed Persian sheep. Empire Journal of Experimental Agriculture 12, 207-228.


Graham A 2001 Aerial surveys of livestock. In: Sustainable Management of the Usangu Wetland and its Catchment (unpublished report). Hunting Technical Services Ltd, Hemel Hempstead.


Gray G D, Woolaston R R and Eaton B T 1995 Breeding for resistance to infectious diseases in small ruminants. Australian Centre for International Agricultural Research, Canberra.


Hornby H E and van Rensburg H J 1948 The place of goats in Tanganyika farming systems - (1) in deciduous bushland formation. East African Agricultural Journal 14, 94-98.


Hutchison H 1958 Variations of liveweight of cattle on farm and ranch in Tanganyika.  Typescript prepared for submission to the East African Agricultural and Forestry Journal.


King A 2000 Carrying capacity of the Usangu grazing system (Technical Report of the SMUWC Project).  Hunting Technical Services Ltd, Hemel Hempstead.


MAC 1998 The Annual Agriculture and Livestock Extension Report for 1997/98 Mbarali District. Ministry of Agriculture and Cooperatives, Rujewa.


Mack R 1970 The great African cattle plague epidemic of the 1890s. Tropical Animal Health and Production 4, 210-219.


Mason I L 1996 A world dictionary of livestock breeds, types and varieties (4th ed.). CAB International, Wallingford.


Mason I L and Maule J P 1960 The indigenous livestock of Eastern and Southern Africa (Technical Communication No 14). Commonwealth Agricultural Bureaux, Farnham Royal.


Norva R A I, Floyd R B and Kerr J D 1988 Ability of adults of Amblyomma hebraeum (Acarina: Ixodidae) to feed repeatedly on cattle and sheep. Veterinary Parasitology 29, 351-355.


Preston M and Allonby E W 1979 The influence of breed on the susceptibility of sheep to Haemonchus contortus infection in Kenya. Research in Veterinary Science 26, 134-139.


Scoones I (Ed) 1995 Living with uncertainty: New directions in pastoral development in Africa.  Intermediate Technology Publications Ltd, London.


Scherf B (Ed) 2000 World Watch List for Domestic Animal Diversity (3rd ed.). Food and Agriculture Organization, Rome.


Statistics Unit/Bureau of Statistics 1996 National Sample Census of Agriculture 1993/94-1994/95 - Tanzania Mainland. Volume I - Technical and Operational Report; Volume II - Holding Characteristics, Livestock, Poultry, Agricultural Practices, Extension Services and Census Evaluation; Volume III - Planted Area, Crop Production, Yield Estimates, Agricultural Inputs and Related Characteristics. Statistics Unit, Ministry of Agriculture/Bureau of Statistics, Planning Commission, Dar es Salaam.


UNDP/FAO 1967 East African Livestock Survey (3 volumes) (FAO/SF: 21 REG). United Nations Development Programme/Food and Agriculture Organization, Rome.


Wilson R T 1979 Studies on the livestock of Southern Darfur, Sudan. VII. Production of poultry under simulated traditional conditions. Tropical Animal Health and Production 11, 143-150.


Wilson R T 1980 Population and production parameters of sheep under traditional management in semi-arid areas of Africa. Tropical Animal Health and Production 12, 243-250.


Wilson R T 1984 Indigenous goats: Productivity in traditional livestock systems of semi-arid Africa. International Goat and Sheep Research 2, 243-251.


Wilson R T 1986 Poultry production in Sub-Saharan Africa. Outlook on Agriculture 15, 121-127.


Wilson R T 1991 Small ruminant production and the small ruminant genetic resource in tropical Africa (Animal Production and Health Paper N° 88). Food and Agriculture Organization, Rome.


Wilson R T, Peacock C and Sayers A R 1982 Livestock production on Masai Group Ranches. 2. Growth and live­weight in goats and sheep at Elangata Wuas and the factors influencing them. Beiträge für Tropischen Landwirtschaft und Veterinärmedizin 21, 191-198.



Received 7 October 2002, Accepted 30 November 2002


Go to top