Country Pasture/Forage Resource Profiles


Len Reynolds



     2.1  Climate
     2.2  Landforms, agro-ecological zones and soils

     3.1  Smallholder sector
     3.2  Estate sector

     4.1  Natural vegetation 
     4.2  Cultivated legumes and fodders
     4.3  Crop residues
     4.4  Constraints to pasture and fodder production and improvement


     6.1  Institutional structure
     6.2  Personnel




Malawi is a landlocked tropical country with Zambia to the west, Mozambique to the south and east, and Tanzania to the north and east, lying between 9o 45' and 17o 5' S and 30 to 36o E (see Figure 1). The national borders encompass Lakes Malawi and Chilwa which cover 2,440 km2 leaving a land area of 9,408 km2. The Northern region covers 2,690 km2, the Central region 3,559 km2 and the Southern region 3,176 km2.

malawimap.gif (10662 bytes)

Figure 1. Map of Malawi

40 % of the total land area is suitable for agriculture (Table 1).

Table 1: Land use in Malawi (ha x103 )

Forest and Woodland 3,700
Permanent Pasture 1,850
Permanent crops 125
Arable 1,875
Total Agricultural area 3,850
Non-arable and Non-permanent crops  7,408
Lakes Malawi and Chilwa 2,440
Total land area 9,408
Total Area 11, 848

Source: NSO, 2000

The year 2000 agricultural population is estimated to be 11 million people, with 85% of the total population living in rural areas. Malawi is one of the most densely populated countries in sub Saharan Africa. Population density rises from 46 persons/km2 in the Northern region to 144/km2 in the South with the most populated districts such as the Shire Highlands in the south containing over 265 persons/km2. Population grew at a rate of 3.7% p.a. between 1977 to 1987 but the rate of increase declined to 1.9% pa (Table 2) in the following ten years.

Table 2: Agricultural population in Malawi, 1988-2000

Million persons 1988 1990 1992 1994 1996 1998 2000
Agric pop. 7.59 8.17 8.79 9.37 9.88 10.42 10.99

Source: NSO, 2000

Most of the arable land is under traditional/customary tenure system. Cultivation rights, rather than ownership is granted by the chief through the village headman. Matrilineage is common in the centre and south while patrilineage is common in the north. In the matrilineal system, where the husband leaves his home to live with the wife, cultivation rights are inherited by the wife. By the late 1980s over 56% of households were on holdings of less than 1 ha, and a further 20% on 1.0-1.5 ha. Because of pressure on land there is little opportunity for fallow and rotation to restore soil fertility, and smallholders have expanded their cultivation to marginal, less fertile soils often on hill slopes which are not suitable for intensive cultivation, leading to woodland depletion, soil degradation and erosion. Rainfed agriculture predominates, dependant on a single rainy season between November and April. The growing season varies in length from less than 120 days to over 210 days (Figure 2). Only 10,000 ha of land is currently irrigated, 5% of the potential irrigated area, largely on sugar estates. Other irrigated crops include rice and vegetables. 

MALAWIfig2.gif (5876 bytes)

Figure 2: Length of growing period in Malawi

(Source: Moriniere and Chimwaza,1996)

At the same time per capita food production has been falling, moving from an index of 100 in 1978-81 to 75 by 1991, reaching only about 87% of the recommended minimum daily calorie consumption. Maize is the main staple of the Malawian diet, covering 76% of smallholder farmland. In the late 1960s, intercropping was found on 94% of the maize area, but this had fallen sharply by 1980. Other food crops include rice, sorghum and millet. Legumes, beans, pigeon pea and groundnuts are traditionally grown by smallholders (Table 3). Pigeon pea in particular is important in the densely populated Blantyre district, providing food, animal forage and being beneficial to soil fertility. Cassava, potato and sweet potato are more locally based. The area planted to cassava, often intercropped with maize, has increased sharply over the past 10 years particluarly in densely populated southern areas. Tobacco is the dominant cash crop, providing 71% of export earning from both large scale and small scale farming sectors. Other cash crops include cotton, sugar, tea and coffee. 

Table 3: Crop area and production in Malawi, 1995

Area (ha 103), production (t) Karonga Mzuzu Kasungu Salima Lilongwe Machinga Blantyre Shire Valley
Cropped land 124 374 351 80 522 454 521 389


















G/nuts 4.7
















Beans 5.5
















Pigeon pea 0.9
















Cowpea 0.3





( 178)











Soya 0.4
















Bamabra nuts 0.4
















* Maize production data estimated for 1993/94, Heisey and Smale, 1995

n/a not available

Source: Moriniere and Chimwaza, 1996



The climate in Malawi changes from semi-arid in the Lower Shire Valley , semi-arid to sub-humid on the plateaux and sub-humid in the highlands. Most of the country receives between 763-1,143 mm rainfall p.a. There are three main areas with precipitation over 1524 mm: Mulanje, Nkhata Bay and the northern end of Lake Malawi (Figure 3). Almost 90% of rainfall occurs between December to March, with no rain at all between May to October over most of the country.

Table 4: Mean annual rainfall levels, 1975/76 –90/91 



Karonga Mzuzu Kasungu Salima Lilongwe,






Shire Valley, Ngabu
1975-76 1184 1476 379 1894 1014 915 729 847
1980-81 887 894 1031 1252 905 728 893 745
1985-86 1026 1566 1086 1479 956 1248 1028 911
1990-91 825 1048 771 1073 656 736 833 800

Source: NSO,   2000

MALAWIfig3.gif (11156 bytes)

Figure 3: Mean Annual Rainfall in Malawi

(Source: Moriniere and Chimwaza,1996)

Mean annual temperatures vary with altitude, ranging from 25oC in the Lower Shire Valley to 13oC on the Nyika Plateau. Frost may occasionally occur in lower lying land on the plateaux, but is not a significant limiting factor in pasture production.


There are 5 main landform areas, the Highlands, Escarpments, Plateaux, Lakeshore and Upper Shire Valley, and the Lower Shire Valley:

The Highlands: These consist of isolated mountains between 1,320-3,000 masl. Extensive highland plateaux are found in the Nyika, Viphya and Mulanje, while Dedza and Zomba are more isolated. Slopes can become precipitous, and soils are predominantly leached latosols.

The Escarpments: These are associated with major fault lines along the edge of the Rift Valley, running from Karonga in the north to Nsanje in the south. They are also found around the highland plateaux and mountains. Soils are predominantly thin latosols.

The Plateaux: Three quarters of Malawi consists of plateaux at elevations of 750-1300 masl. The topography is flat to rolling, with scattered rock inselbergs. The soil is deep well drained latosols on higher parts of the catena, with poorly drained sand and clay in the hollows. Poorly drained hollows are locally called dambos, (equivalent to vleis in Zimbabwe and South Africa).

The Lakeshore and Upper Shire Valley: Lakeshore plains occupy 8% of the total land area, at 465-600 masl. The land is flat to gently undulating, with deep calcimorphic soils in the hollows. The upper Shire River flows through a broad flat valley from the south of the lake towards the south of the country. Soils are similar to those along the lakeshore. Mopanosols are found in some areas along the river.

The Lower Shire Valley: The lower Shire extends from Kapachira falls to Nsanje at the bottom of the country, mostly at less than 180 masl. The river flows through two marshes with extensive areas of hydromorphic soils. To the east of the river, up to the Thyolo escarpment, soils are medium to coarse textured alluvial and colluvial. To the west there is a broad plain with vertisols and grey brown earths, rising towards the western escarpment. Some areas of saline soils are found.


Livestock accounts for around 7% of agricultural GDP in Malawi. Current animal population is low with 710,000 Malawi Zebu cattle, 12,000 dairy cattle- Freisian and crosses, 110,000 sheep and 1,260,000 goats. It is worth noting that published animal numbers are nearly always extrapolations, and hence different sources will arrive at different figures depending upon the assumptions made. A picture emerges of relatively static cattle numbers but increasing numbers of smallstock (Table 5). Only 4% of households have cattle pens, 15% have goat pens and 55% own poultry (free range and layer/broilers). This indicates a marked decrease in cattle ownership since the early 1990s when 13% of farmers were reported to own cattle (Munthali et al, 1993). Livestock are most numerous in the Northern region (Table 6). The three northern Agricultural Development Districts (ADD) - Karonga, Mzuzu and Kasunga -   contain almost 80% of the national total of oxen. Over the past 15 years there has been a marked decline in the number of work oxen in the Central and Southern Regions, with increased reliance on hand labour (Kumwenda, 1988; FAO, 2000). Few animals are stall fed; less than 600 were recorded in a recent census for the entire country (reported by NRI, 1998). However, a larger number may be provided with locally produced crop residues to supplement grazing. Nevertheless, the supply of manure from kraals is very limited in relation to the area of cropped land. The maximum cattle density is in the far north, 1.4 cattle/ha cropped land, while over most of the country the density is less than 0.25 cattle/ha cropped land (Figure 4). It can be estimated that over most of Malawi only 1 tonne of manure is available for every 10-25 ha cropped land (NRI, 1998).

MALAWIfig4.gif (9336 bytes)

Figure 4: Cattle area per hectare of cropped land

(Source: Moriniere and Chimwaza, 1996)

Meat production and per capita supply are shown in Tables 7 and 8. There are estimated to be 12,000 dairy cattle in the country, mainly around the urban centres of Blantyre, Lilongwe and Mzuzu (Mpofu, 1998). Consumption of animal protein is very low in Malawi, even compared to other sub-Saharan African countries, a reflection of the low number of households owning animals, and limited purchasing power of consumers, even in urban areas. 

Table 5: Livestock numbers in Malawi, 1990-99

Ruminant livestock


  1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

































Source: FAO Database, 2000

Table 6: Distribution of livestock across Malawi in 1998

Livestock (head) Karonga Mzuzu Kasungu Salima Lilongwe Machinga Blantyre Shire Valley Total
Cattle 86,687 130,502 91,501 44,965 73,128 3,876 42,547 81,129 589,175
Work Oxen 12,217 29,480 17,147 2,137 9,333 1,977 852 1,584 74,997
Goats 23,700 130,345 264,822 149,930 349,680 217,001 211,666 21,937 1,566,514
Sheep 4,521 14,580 11,913 13,499 9,544 38,391 2,482 2,987 97,916
Poultry 46,340 94,660 106,325 121,395 172,694 271,117 241,631 61,837 1,115,999

Source: NRI, 1998

NB. Note the difference in totals between the NRI and FAO data

Table 7: Meat production in Malawi, 1994-1999

Meat Production (Mt)








Beef and veal







Goat meat







Mutton and lamb







Source: FAO Database 2000

Table 8: Meat and milk supply per capita per annum in Malawi, 1990-98

Supply/cap/a. (kg)











Bovine meat










Goat and sheep meat










Whole milk 










Source: FAO Database 2000

3.1  Smallholder sector:

East African Zebu (Malawi zebu), a small slow growing animal, is the dominant cattle breed. In the 1980s Canadian Holstein cattle were imported to stimulate the establishment of a dairy industry. Small East African goats are most numerous with small numbers of Boer and their crosses, and a very small number of crossbreeds, derived from European dairy breeds and indigenous goats. 

Almost all ruminant livestock graze on traditional communal lands. During the rainy season, grazing in the Central and some parts of the Southern Region is restricted to dambos and roadsides, or hillside grazing if most of the land is planted to crops. Cattle are herded while goats are tethered by the roadside or around dambos. In the less populated north, cattle are let loose to graze on natural pasture in the uplands areas during the rainy season, and move into dambos during the dry season. When crops have been harvested animals are allowed free access to residues. During the dry season, natural pasture grazing is supplemented with crop residues. There are an estimated 2.7 million ha of available natural grazing in Malawi. Crop residues, from 385,000 ha of cultivated land are grazed in situ, so that some excreta are deposited directly on cropland. Local Zebu cattle may also be offered maize bran as a supplement, but in the absence of cattle, maize bran will be fed to poultry rather than to small ruminants. Stall-fed dairy animals are offered cut-and-carry roadside grass, supplemented with maize bran and other locally available residues. Around Blantyre these include maize stover, groundnut tops, sweet potato tops, sugar cane tops, rice straw, and banana pseudostems. The area available for grazing increases from south to north in Malawi, reflecting the changes in human population density.

3.2  Estate sector:

The estate sector is oriented towards production of tobacco and other cash crops. On tobacco estates a cropping cycle with maize and a grass ley is used to prevent the build-up of nematodes in the soil, and to maintain soil nitrogen at a level appropriate for tobacco. Rhodes grass was often planted on tobacco estates, grazed by herds of beef cattle, which had access to maize stover grazing in the dry season. 

What commercial dairy farming that exists is found in the estate sector. Holstein and holstein crosses are grazed and supplemented with concentrate feed. The system is labour intensive, relying on hand milkers. Breeding uses AI, backed up with natural service.


4.1  Natural vegetation:

Figure 5, which shows vegetation patterns, and much of the material for this section is taken from Rattray (1960) and Moyo et al (1993). DM production of natural grassland is estimated to be 1.0t/ha/a for the country as a whole (Hodges, 1983). In the north grazing is mainly in forest or regrowth after cultivation. The productivity of edible herbage under bush regrowth is low.

MALAWIfig5.gif (7954 bytes)

Figure 5: Vegetation resources in Malawi

(Source: Moyo et al, 1993)

Montane forests, scrubs and grasslands: These areas are found above 1500 masl, comprising forest relics in valleys and in isolated stands, with rolling grasslands and scrubland between. The pattern seems to be controlled by regular grass fires, and by the moisture of lower lying areas maintaining a greener and more fire-resistant margin to the remaining forest. Rainfall is above 1500mm, and is partly composed of a mist formed in the cold dry season, which provides moisture outside the main rainy period. The forests vary in composition and include communities dominated by Widdringtonia whytei (mulanje cedar) and Juniperus procera (African juniper). 

Between 1500 and 2100 masl a variable grassland occurs of short tufted to densely tangled grasses of low ground cover. Species include Themeda triandra, Exotheca abyssinica, Monocymbium ceresiiforme, Elionurus argenteus, Brachiaria serrata, Andropogon schirensis, Hyparrhenia lecomtei and Loudetiasimplex. On wetter slopes and better soils, Hyparrhenia cymbaria, Rhynchelytrum stolzii, R. nerviglume, R. stuposum and Melinis maitlandii are also found. Fire tolerant Protea species invade this type of grassland and many woody shrubs including Tephrosia aequilata and Humularia descampsii.

Above 2000 masl the grassland is mainly comprised of Exotheca species, and produces a short tufted apparently dense ground cover. Species reported in the Malawi montane grasslands include Exotheca abyssinica, Loudetia simplex, Trachypogon spicatus, Helictotrichon elongatum, Agrostis spp, Monocymbium ceresiiforme, and Elionurus argenteus. Festuca schimperiana is common in the wetter areas, with Danthonia davyi in exposed situations. M. ceresiiforme occupies only shallow soils in the montane area, whereas a lower altitudes it is confined to swampy grasslands. The nutritional value of montane grassland declines in the cold season.

Brachystegia woodlands (miombo): Brachystegia woodland, also known as miombo or savannah woodland, is almost ubiquitous, characteristically containing one or more species of Brachystegia with Julbernadia globiflora. The grass layer is depressed by the relatively light crowned trees, which have the ability to coppice freely after cutting. The woodland varies in density from tall fairly open woodland to dense scrub according to treatment. Miombo is generally found between 600-1500 masl with an annual rainfall of 510 to 1530 mm. Grass species include Hyparrhenia filipendula, Themeda triandra, Andropogon schirensis, Bewsia biflora and Andropogon amplectens. On shallow soils, especially in the Northern region Anthephora acuminata, Tristachya inamoena, Sacciolepis transbarbuta, Rhynchelytrum nyassanum and Homozeugos eylesii are also found. 

Grasses vary according to habitat, but are generally of medium height with low ground cover. When trees are cleared the grasses become more vigorous and a dense vegetative cover results until the scrub regrows. The grazing value is low with marked seasonal variation, and the sparse cover has a low carrying capacity. Although the grasses respond to burning and produce a flush of fresh growth after fires, the main reason for managed burning is scrub control.

Shallow, seasonally waterlogged depressions at or near the head of a drainage network are called Dambos in Malawi. Dambos are easily recognised by the sharp contrast between dry typical miombo woodland, or cleared agricultural land , and the open herbaceous vegetation on the dambo itself. Dambos are nearly treeless areas dominated by grasses or sedges with a build up of organic matter with a hydromorphic, sometimes peaty upper soil horizon. Botanical composition of plant cover varies from the edges to the central, more waterlogged zone in dambos (Table 9). While grasses predominate at the margins, sedges herbs, rushes and ferns are more common nearer the central wetter area. A dambo catchment also acts as a hydrological store, holding water and releasing it as base flow to its headwater stream during the dry season. Because they retain water, dambos support vigorous growth of grass when other forms of grazing are in short supply. Dambos are particularly important in Central and Northern Malawi, and provide a valuable dry season feed resource for ruminant livestock. Dambo margins are also used for gardens providing a more reliable crop output than rainfed dryland farms, and spread labour demand more evenly throughout the year. Gardens are fenced to prevent damage by livestock which compete for space in more densely populated areas. Limited grazing in dambos prevent tall grasses such as Hyparrhenia from producing seedheads, ensuring continuous growth through most of the dry season (Roberts, 1988). On better soils, such as at Chitedze Research station DM production from dambo grasses averaged 5.9 t/year. However, grass DM output can fall as low as 0.2 t/year where poor management has resulted in a dense coppice of Uapaca kirkiana such as in West Mzimba. The overall mean DM production in dambos is estimated to be 3.2 t/ha/a (Hodges, 1983). 

Table 9: Typical non-arboreal plant species in southern African dambos

Marginal wash zone Central seepage zone
Alloteropsis semialata (g) Andropogon eucomus (g)
Aristida sp (g) Arundinella nepalensis (g)
Brachiaria filifolia (g) Eragrostis namaquensis (g)
Hyparrhenia rufa (g) Eragrostis stapfii (g)
Loudetia simplex (g) Hyparrhenia bracteata (g)
Monocymbium ceresiiforme g) Sporobolus subtilis (g)
Setaria sphacelata (g) Cladium mariscus (s)
Cassia mimosoides (h) Scleria hirtella (s)
Emilia integrifolia (h) Dissotis canescens (h)
Impatiens sp (h) Gladiolus sp (h)
  Typha latifolia (r)

g = grass; s = sedge; h = herb; r = rush; f = fern

Source: Roberts 1988

Broad-leaved deciduous woodland: On red clay loamy soils of the Central region plain, and on colluvial slopes and plains elsewhere in Malawi at altitudes of 600-1200 masl with rainfall of 510 to 1530 mm a tall grass mixture is found with a broad leaved deciduous woodland, known as chipeta. Varying density of Acacia polyacantha, Piliostigma thonningii and Combretum molle occurs. Selective tree felling has produced communities with a single dominant tree species, while cultivation and overgrazing have produced scrub and thickets. The grass mixture includes Hyparrhenia variabilis, H.filipendula,H. gazensis, H. nyassae, Setaria sphacelata, S. longiseta, Digitaria setivalva, D. diagonalis, Panicum maximum, and Themeda triandra. These areas are subject to fierce annual burns, and fire tolerance is a typical characteristic of the woody species found. The grazing is of low carrying capacity. Overgrazing of these areas and of the associated swamp grasslands leads to colonisation by Urochloa pullulans, a palatable but poor ground cover plant, and then by Sporobolus pyramidalis, which produces a tough, unpalatable tufted ground cover. 

Rift Valley escarpment and the foothills: These areas, at altitudes of 900 to 1500 masl and a wide range of rainfall (350-1500 mm) are characterised by lowland woodlands with Brachystegia manga, Pterocarpus angolensis, and on the lower slopes Sterculia quinqueloba. On the lower foothills the baobab (Adansonia digitata) with an understory of bamboo (Oxytenanthera abyssinica) are frequent. The grasses comprise Hyparrhenia gracilescens, H filipendula, Themeda triandra,Andropogon amplectens, Schmidtia bulbosa, Euclasta condylotricha, Eustachys paspaloides, Eragrostis superba, and Thyrsia undulatifolia. Grasses form a low ground cover on broken, stony soils. These areas have limited grazing value because of the difficult terrain and lack of dry season pasturage. 

Low altitude woodland and parkland: Low altitude woodland and parkland are found below 600 masl and include areas in the Shire Valley and along the lakeshore. Soils are influenced by drift and colluviation, and rainfall may vary sharply within short distances (350mm to 2500 mm in one area alone). As a result a wide floristic range can be found. Tall grasses are associated with low altitude woodland, including Hyparrhenia gazensis, H. variabilis, H. dichroa. Andropogon gayanus, Setaria palustris, and Panicum maximum. In densely settled and cultivated locations, tall reedy grasses are replaced by Urochloa pullulans and U. mosambicensis. Woodlands are characterised by Sterculia africana, Colophospermum mopane, Acacia tortilis and Faidherbia albida according to locality. Acacia woodland provides valuable grazing from pods to supplement grasses in the dry season. Mature trees may stand within a dense understorey, which includes Commiphora spp, Bauhinia tomentosa, and Popowia obovata. The understorey is likely to be man-induced since lone-standing mature trees are found elsewhere in open areas of cultivated land, and in some cases trees are selectively retained by farmers to maturity (eg Faidherbia albida). In some areas, woodlands and thickets may be dominated by Mimosaceae. Terminalia sericea woodlands form on sandy soils, with Pterocarpus antunesii, Fagara spp and Grewia spp woodland and thickets around the Lower Shire river. Base rich soils support Euphorbia ingens and Commiphora thicket, whilst Hyphaene ventricosa, H. crinita and Borassus aethiopium palms occur where the water table is high. 

Becker and Lohrmann (1992) studied small East African goats at a Salima lakeshore site. Green vegetation covered 79% of the site in the wet season, with 40% cover from dry material in the dry season. In the wet season equal time was spent feeding on grass and browse, whereas in the dry season 93% of time was spent feeding on dry browse. Andropogon schirensis was the most common grass species, with Markhamia acuminata, Bauhinia petersiana, Combretum fragans and Friesodielsia obovata the most common browse. Grass palatability decreased markedly from wet to dry season, but browse palatability remained unchanged. Thorny acacias (A. polyacantha, A. nilotica and A. nigrescens) were very palatable. The leaves of Friesodielsia obovata, and Combretum apiculatum remained palatable, while the blossoms of Cordyla africana and Lonchocarpus bussei and fruits of A. polyacantha were highly accepted. On wetter areas of vertisols under natural pasture, a vigorous growth of Ischaemum brachyatherum holds surface soil together with a dense mat of rhizomes, slowing and diffusing the run-off of water (Mitchell, 1987).

Swampland: Edges of swamps and lakes are fringed with the sedges and grasses including Cyperus papyrus, Echinochloa pyramidalis, Typha australis, Vossia cuspidata and Pennisetum purpureum

4.2  Cultivated legumes and fodders:

Introduction and evaluation of germplasm in the 1960s and 1970s was undertaken mainly to seek alternatives to Rhodes grass (Chloris gayana),  the established commercial species widely used on tobacco estates to suppress nematode population as well as providing fodder for beef cattle. From these screenings, a number of species and cultivars from the genera Panicum, Chloris, Cynodon and Cenchrus were shown to have good forage potential. Over the same period the increasing cost of inorganic N fertilisers promoted a search for a cheap source of biologically fixed N from herbaceous legumes, including the genera Stylosanthes, Macroptilium, Macrotyloma, Neonotonia, Centrosema, and Desmodium (Thomas, 1976). At that time, there was a dual focus on identification of forages for stallfeeding operations as well as for grazing. Pennisetum purpureum v. Gold Coast and Panicum maximum v. Ntchisi Panic were recommended for cut and carry systems which were being promoted around Lilongwe Agricultural Development District in the Central region (Dzowela, 1985; Munthali and Dzowela, 1987).

Buffel grass (Cenchrus ciliaris) had been shown to be more productive than Rhodes grass (Anon, 1975) in terms of DM and CP productivity. Furthermore, buffel grass produced viable seed unlike alternative species such as Panicum coloratum, cv Bushmans mine and Cynodon nlemfuensis. Studies were made of the suitability of different varieties of buffel grass, and buffel/ legume combinations at contrasting sites in different agro-ecological zones, with Chitala representative of a Lakeshore/Shire Valley environment, and Chitedze of the medium altitude plateaux, and appropriate varieties were identified.

Hodges (1983) in the Pasture Handbook for Malawi recommended different forages and combinations of forages for grazing, hay or cut-and-carry feeding in different environments across the country. 

Many grass and legume varieties that showed promise in Malawi had already been tested and introduced in Zimbabwe. Work in the 1970s for a grass that would persist under grazing identified Cynodon nlemfuensis cv Henderson No 2, which was able to support grazing through wet and dry seasons better than Rhodes grass (Anon, 1975). Chloris gayana and Desmodium uncinatum cv Silverleaf were shown to be compatible, and that translated into higher animal weight gains in comparison to Star grass pastures, even though the latter supported increased grazing days (Dzowela, 1985). 

Various attempts were made to introduce legumes into natural pasture. A UNDP/FAO project in the mid-1970s demonstrated that S. guanensis cv Cook improved dry season liveweight gains for grazing cattle at West Mzimba. Similar benefits where shown by the inclusion of stylo in natural pasture at Dzalanyama Ranch in the Central Region. Seed was sown in strips ploughed through natural pasture, after which dispersal was obtained by the passage of ingested seed through grazing animals. However, the benefits proved transitory as it was impossible to maintain an adequate proportion of legume in the pasture under communal grazing management.

During the 1990s increased attention was paid to tree legumes, focusing largely on Leucaena leucocephala and Faidherbia albida, mainly for soil fertility and crop production (Saka et al, 1991). Despite a considerable number of smallscale research and development projects, few have led to farmer adoption of agroforestry practices. The most promising appears to be the use of F. albida as well spaced trees in cropland, a traditional practice with naturally occurring specimens in Salima District. In addition to the benefits accruing from leaf fall to soil fertility and crops yields (and hence crop residues) under the tree canopy, livestock graze on fallen seeds which are highly nutritious. Leucaena had previous been promoted in the mid-1970s by a cattle stall fattening and dairy project (Addy and Thomas, 1976; Savory and Breen, 1979), but interest faded after the end of the project. The merits and demerits of agroforestry for improved livestock production were commented upon by Munthali (1991). Smallholder dairy development work during the late 1980s and 1990s in the milk shed areas of Mzuzu, Lilongwe and Blantyre stimulated further interest in cut and carry forage. Farmers, however, still made greater use of natural pasture from dambos and roadside grasses, in preference to cultivated forages. 

4.3  Crop residues :

A total of 1.2 -1.5 million t of maize stover is available in each dry season to supplement natural grazing. Smallholder farmers form stooks of cut maize to finish drying in the field, before removing the cobs later. This produces a number of stover heaps around the field. Virtually all of the stover is grazed in situ, so that possibly half is trampled and soiled. The contribution of maize stover varies markedly around the country. The maize area/head of cattle varies from <0.3 ha/head in Karonga District in the far north to 45 ha/head in Machinga District. Legume residues, especially groundnut haulms, may be collected and brought back to the household for storage and more efficient use by cattle. Few households own cattle, and little attention is paid to the provision of supplementary feed for small ruminants. Any DM remaining in the field (weeds, crop residue trash) is burnt prior to land preparation for the next season.

In the southern region, plantings of Cajanus cajan and Manihot spp. have increased over recent years. Although there are relatively few cattle, the presence of a long season crop in the field requires protection from free grazing animals. Benefits to soil fertility of incorporation of crop residues depends on C:N ratio in the residue. In the short term, the incorporation of low N material adversely affects fertility through utilisation of free N in soil for bacterial growth. However, high N residue is the most valuable for animal feed. For most smallholder farmers in Malawi, soil fertility (and hence crop yield) is more important than animal feed. This will only change when a market develops for the sale of animal feed, which will probably be dependent upon consumer demand for meat and milk and their ability to pay farmers an attractive price for livestock products.

4.4  Constraints to pasture and fodder production and improvement:

By the end of the 1980s, over 56% of farmers had less than 1.0ha with a further 20% between 1.0 and 1.5 ha. The small size of farms constrains farmers from growing pasture crops when they have difficulty meeting household food needs. It is estimated in an average year that some 60% of households run out of food produced from their own farms three months before the next harvest (MALD, 1995). Forage cultivation has been promoted for smallholder fattening and dairy projects, with mixed success. 

Free access to harvested fields in the dry season limits the opportunity for undersowing or the use of palatable agroforestry species.

Pasture/fodder crop seeds are difficult for smallholder farmers to obtain, even if they have the money to buy.

Although extension staff receive training in both crop and livestock production, the emphasis in extension is placed on food crops. Staff have very limited knowledge of pasture/forage crops.

Technical problems, such as lack of information and seed, are only part of the problem. Smallholder investment in livestock has been depressed by poverty, low demand for livestock products, and the possibility of animal theft. The incidence of theft has become more prevalent with the advent of a freer political climate. Inadequate nutrition has contributed to poor health of rural workers, exacerbating the difficulties faced by households suffering from HIV/AIDS. Education and information dissemination is currently and will be increasingly hampered by staffing problems. A World Bank assessment of AIDS in Malawi has estimated that 40% of employees in specific sectors, including education and health, would die from AIDS by the year 2005 (Cohen, 1999). Labour and capital saving technologies for smallholder farmers, taking into account the specific needs of HIV/AIDS affected households, are likely to be viewed as high priority in future. It may be in the future that cut-and-carry feeding of a few animals, rather than herding has a role to play here, with a resultant demand for cultivated forages.


Pasture seed production has not developed in Malawi, partly as a result of lack of demand. Nevertheless, studies in the 1970s demonstrated the environments where grass and legume production was satisfactory under rainfed conditions (Hodges, 1983). Seed harvesting techniques, suitable for village conditions, were explained, and seed application rates were also shown. Despite this, the provision of forage seed has usually depended on inputs from externally funded projects. 


6. 1. Institutional Structure

Agricultural research in Malawi is directed towards applied or production oriented research to generate information and technologies which can be directly utilised by smallholders to solve technical production problems, with goals of improving incomes and distribution of income, diversification of production and stabilising or improving natural resource levels.

The Agricultural Research Department (DAR) of MoA is mandated to undertake research on a wide variety of crops and Livestock. Research on high value export commodities such as tea, tobacco and sugar is carried out by other specialised institutions. DAR has research stations in each district of the country, with 11 research stations, and 20 sub-stations. Chitedze Research Station, near Lilongwe is the centre for pasture and forage work. MoA has mandated the University of Malawi to conduct specific disciplinary research. Bunda College of Agriculture carries out research into beans, rabbits, pigs and socio-economics. Chancellor College works with DAR on cassava pests, soil pests and cassava detoxification. The Department of Animal Health and Industry carries out research on poultry and goat breeding. Contract research funding is also provided by GM, to other organisations for research activities where DAR does not have sufficient resources. External funding for research may go through GoM or direct to the implementing agency. Details of research intentions are laid out in Agricultural Research Masterplan and Action Plans (MALD, 1993).

International research centres are represented by ICRISAT, based on Chitedze Research Station, and ICRAF at Makoka Research Station. Other autonomous agricultural research and development projects include Malawi Agroforestry Extension Project funded by USAID, and Promotion of Soil Conservation and Rural Production funded by the EU. There is currently no externally funded work focused around livestock, fodder, pastures or herbaceous legumes.

NGO projects are focused on crop production and means to stabilise and improve food security. Other areas of priority for NGO attention are education and health.



Mr A Kumwenda,

Shire Highlands Milk Producers Association,

PO Box 30603, Blantyre

Tel: (+265) 631 691

Dr J Banda, Bunda

College of Agriculture,

PO Box 219, Lilongwe, Malawi

Dr J Mtimuni, Bunda

College of Agriculture,

PO Box 219, Lilongwe, Malawi

Pasture and forages:

Dr H Msiska,

Chitedze Research Station,

PO Box 158 Lilongwe

Prof G Kamyana-Phiri, Bunda

College of Agriculture,

PO Box 219, Lilongwe

Tel: (+265) 277 222


Dr W T Bunderson,

Malawi Agroforestry Extension Project (MAFEP),


D A Saka,

Chitedze Research Station,

PO Box 158 Lilongwe

Mr J Kamangira,

Promotion of soil conservation and rural production (PROSARP),

TEPTC Building,

PO Box 1481, Lilongwe, Malawi

Tel: (+265) 740 704

Crops and Soil:

Dr S Snapp,


Chitedze Research Station,

PO Box 158, Lilongwe

J Kumwenda,

Chitedze Research Station,

P O Box 158, Lilongwe

Tel: (+265) 767 222

Dr W Sakala,

Chitedze Research Station,

PO Box 158, Lilongwe

Tel: (+265) 767 222


Dr S Chimwaza,

FEWS Country Representative Room 216,

Agro-economic survey, MAI,

PO Box 30455, Lilongwe 3

Tel/fax: (+265) 744 083,


Addy B L and Thomas D. 1976. The utilisation of crop residues , madeya and leucaena for winter feeding in Lilongwe District. Research Bulletin No 2/76, MoA, Lilongwe, Malawi.

Anon, 1975. Report on pasture/cattle research project. ODMA research Project R2417 for 1975. MoA Lilongwe, Malawi

Becker K. and Lohrmann J. 1992. Feed selection by goats on tropical semi-humid rangeland. Small Ruminant Research 8: 285 – 298.

Cohen D. 1999. Sustainable development and the HIV epidemic in Africa. In: G Mutangadura, h Jackson and D Mukurazita (eds). AIDS and African smallholder agriculture. Southern Africa AIDS Information Dissemination Service, Harare, Zimbabwe. p 2-7.

Dwozela B H. 1985. Highlights of pasture research in Malawi: 1975-84. In: J A Kategile (ed) Pasture improvement research in Eastern and Southern Africa. IDRC, Ottawa, Canada. p 59-76.

FAO, 2000. Database (website

Heisey P. W. and Smale M, 1995. Maize technology in Malawi; a green revolution in the making. CIMMYT Research Report No 4. CIMMYT, Mexico.

Hodges E. M. 1983. Pasture Handbook for Malawi (3rd Edition). Ministry of Agriculture. Blantyre, Malawi.

Kumwenda M. S. L. 1988. Malawi country report. In: Zerbini E, Shapiro B I and Chirgwin J C (eds). Technology transfer: multi-purpose cows for milk, meat and traction in smallholder farming systems. ILRI, Nairobi, Kenya. p 59-60.

Ministry of Agriculture and Livestock Development, 1993. Agricultural Research Masterplan. MALD, Lilongwe, Malawi.

Ministry of Agriculture and Livestock Development, 1995. The agricultural and livestock development strategy and action plan. MALD, Lilongwe, Malawi.

Mitchell A. J. B. 1987. Management problems of cotton on vertisols in the lower Shire Valley of Malawi. In: M Latham and P Ahn (eds). Management of vertisols under semi-arid conditions. IBSRAM proceedings No 6. IBSRAM, Bangkok, Thailand. p 221-229

Moriniere L. and Chimwaza S. 1996. Malawi vulnerability assesment and mapping baseline 1996. USAID/FEWS. MoAI, Lilongwe, Malawi.

Moyo S, O’Keefe P and Sill M. 1993. The Southern African Environment. Earthscan. London
Mpofu Z. 1998. Sourcing of Dairy Cattle in Southern Africa. ILRI, Nairobi, Kenya

Munthali J T. 1991. The merits and demerits of agroforestry for improved livestock production in Malawi. In: A R Saka, W T Bunderson and J A Maghembe (eds). Agroforestry Research and Development in Malawi. MoA. Lilongwe, Malawi. p 93-100.

Munthali J. T. and Dzowela B. H. 1987. Inventory of livestock feeds in Malawi. In: J A Kategile, A N Said, and B H Dzowela (eds). Animal feed resources for small-scale livestock producers. IDRC, Ottawa, Canada. p 61-69.

Munthali J. T., Musa F. A. and Chiwayula C. L. K. 1993. Smallholder dairy development in Malawi. In: J A Kategile and S Mubi (eds). Future of Livestock Industries in East and Southern Africa. ILCA, Addis Ababa, Ethiopia. p 143-148.

National Statistical Office, 2000. GoM, Lilongwe, Malawi

NRI, 1998. Soil Fertility in Malawi: a review of policies, productivity and perceptions. NRI, Chatham, UK

Rattray J. M. 1960. The grass cover of Africa. FAO Agricultural Studies No 49. FAO, Rome

Roberts N. 1988 Dambos in development: management of a fragile ecological resource. J Biogeography 15: 141 – 148

Saka A, Bunderson W. T. and Maghembe J A. (eds) 1991. Agroforestry research and development in Malawi., MoA, Lilongwe, Malawi.

Savory R and Breen J. A., 1979. The production of Leucaena leucocephala in Malawi: Utilisation. Working Paper No 22. FAO/UNDP Project MLW/75/020. Lilongwe, Malawi

Thomas D. 1976. Effects of close grazing or cutting on the productivity of tropical legumes in pure stands in Malawi 


The author of this paper can be contacted aa follows:

Dr. L. Reynolds, Manor Farmhouse, Huish Champflower, Taunton TA4 2EY, UK

Tel/fax: (+44) 01984 624915

[The paper was drafted in October 2000 and was edited by S. G. Reynolds]