3.1. Animal Traction
3.2. Manure Supply
3.3. Crop Residue Supply
3.4. Introduction of Improved Forages and Ley Farming
3.5. Socio-economics and Complementarity
Complementarity is defined by McINTIRE et al. (1992) as the supply of inputs from one sector to the other, such as using draught animal power and manure in crop production or crop residues as feeds. The benefits of crop-livestock interactions are many. Animal traction could improve the quality and timeliness of farming operations now done by hand, thus raising crop yields and incomes. The transfer of nutrients from pasture to cropland through manure contributes considerably to the maintenance of soil fertility. Livestock provide a least-cost, labour-efficient, route to intensification through their role in nutrient cycling. Keeping animals on the farm could also provide a use for other resources such as crop residues, which might be wasted in the absence of animals. Livestock sales would generate cash to buy inputs such as fertilisers. The positive environmental impacts accruing from crop-livestock interactions will now be emphasised.
Draught animal power has a long history of use in highland areas in sub-Saharan African countries such as Ethiopia, and is widely employed in The Gambia. Animal traction is also important in smallholder farming systems in Asia. Draught animals can assist farmers to improve soil tillage and introduce soil conservation practices such as terracing, ridging and the broadbed-furrow system; operations that are unlikely to be undertaken with hand-cultivation. Improved tillage requires extra power, for which resources of hand-labour are presently inadequate. Animals can provide this power. The lower compaction resulting from land preparation using animal traction, compared to tractor ploughing, also reduces the erosion hazard. In Asia, hillsides have been levelled into terraces for rice fields initially and then re-levelled using draught animal power annually, to ensure even spread of water and its redistribution to lower paddies. Without such a system, erosion of rice fields would make farming unsustainable within a few years. Heavy-textured problem soils such as Vertisols have a high water-storage capacity and become waterlogged during much of the growing season. To tap the inherent fertility of these soils, draught animal power can be harnassed to shape broadbeds and improve surface drainage. In India, KAMPEN et al.(1981) found that water run-off from a broadbed-furrow system was 44 percent lower than that from fallowed land during three months of the cropping season.
The vast majority of farmers in South-East Asia do not have resources to replace draught animal power with tractors. The environmental benefits and economic savings to Asian nations through the use of draught animal power has been highlighted by RAMASWAMY (1985), who estimated that it would take 30 million tractors to replace some 300 million draught animals used on small farms. The use of renewable animal power instead of non-renewable fossil fuels and tractors has, amongst other things, reduced carbon dioxide and carbon monoxide emissions into the atmosphere.
The advantages of ridge tillage have been summarised by SRIVASTAVA et al. (1993) and include:
· Savings in labour as only 50 percent of the field is disturbed in comparison to general ploughing.In studies with pearl millet, KLAIJ and HOOGMOED (1989) found that ridging using animal traction increased grain yields by 35 percent and straw yields by 57 percent over fertilised, hand-weeded plots. These increases resulted from superior stands and better early growth. Since weeds were less of a problem, weeding time was also reduced.
· Enhancement of soil fertility through concentration of topsoil and incorporation of ash and organic residues into the seedrow zone.
· Decreased run-off losses, increased infiltration and conservation of water in the root zone.
· Control of wind and water erosion.
· Enhancement of root penetration in shallow soils.
It would be difficult to measure the direct effects of animal traction on environment. Positive effects would operate through the improved soil conservation practices mentioned above, and could be estimated using the indicators described in Section 6.2.
Soil fertility depletion is a major constraint to agriculture, notably in farming systems in sub-humid and humid areas, and where the use of artificial fertilisers is low. Even when inorganic compounds are applied, cereal yields may not be maintained under continuous cultivation on nutrient-poor sandy soils with a low buffering capacity (BATIONO et al., 1989). The use of mineral fertilisers alone leads to a decrease in base saturation, pH and the occurrence of aluminium toxicity. Examples of the extent of yield decline as a result of soil fertility depletion are given by SRIVASTAVA et al. (1993). Organic materials applied in bulk can improve soil texture, promote better absorption of moisture, reduce run-off and prevent crusting of the soil surface (WATTS PADWICK, 1983). Even small quantities of organic materials can bring about marked improvements in the cation exchange status of soils. Manure is also valuable in reversing the deterioration in soil structure in sodic soils, characterised by high contents of exchangeable sodium and low permeability.
The supply of manure can be a major benefit of crop-livestock integration. In principle, manuring allows higher crop output from a smaller area, thereby preserving pasture from the encroachment of additional cropping and alleviating land competition between crops and livestock. In zones with a low population density, fallows are the principal means of maintaining soil fertility in peasant agriculture in sub-Saharan Africa and the Andean zone of South America. However, at higher population densities, where there is no fallow, manure is often the only means of maintaining organic matter levels and sustaining soil fertility in cropping systems in developing countries. The results of manuring trials conducted in tropical Africa are given by WATTS PADWICK (1983), SANDFORD (1989) and McINTIRE et al. (1992). In Asian countries such as Thailand, manure has sustained rice yields at 1.5-2 t/ha for 4-5 decades with the minimal use of artificial fertilisers (GUZMAN and PETHERAM, 1993). In Java, Indonesia, farmers collect feed refusals in pits beneath their animal barns. Refusals combine with faeces and urine falling through slatted floors to produce manure-compost. This is ranked by farmers as one of the most important outputs from livestock production. In the uplands of Java, 90 percent of the fertiliser used on smallholdings is manure-compost. It is essential to the sustainability of some of the most intensive cropping cycles in the world (TANNER et al., 1993). In permanent highland cultivation systems, the production of manure is a major reason for keeping livestock. In Israel, where dairy herds are housed for 12 months, manure is collected and used on arable ground.
The quantity and quality of animal excreta will depend on the number of animals, the species, the amount and type of feed consumed, and the interaction between the form in which nutrients are ingested and the digestive processes taking place in the animal (ROMNEY et al., 1994). Purchased concentrates, for example, that are economically viable for dairy cattle, can raise nitrogen levels in the excreta. By controlling animal movements, adjusting stocking rates in a grazing situation or storing and distributing manure from animals kept in confinement, a farmer can influence the proportion of nutrients to be returned to the soil in a given location. The management of the soil and of the returned nutrients will influence the losses incurred after application and the quantity potentially available to the crop. Sources of nutrient losses have been reviewed by ROMNEY et al. (1994).
In sub-Saharan Africa, in areas of low cropping intensities, manure is often obtained through paddocking contracts, with cattle being kept on farmers fields in rotation. In such cases, herders have little or no land of their own and, therefore, do not require large amounts of manure for themselves. Accordingly, there is transfer of nutrients from the grazing lands to the cropping areas, which can offset the loss of crop residues for incorporation into soil as a result of their utilisation by livestock. In West Africa, the exchange of manure for crop residues for livestock feed is widespread. In the Andean highlands of Ecuador, some farmers sow oats and vetch, and feed these crops to livestock confined on a plot of land for several weeks or months as a regular part of the crop rotation. Potatoes are usually planted after manure application.
In the humid zones of South-East Asia, where integrated crop-livestock-aquaculture systems prevail, pig manure is drained and the clear effluent applied as fertiliser to the rice paddies and vegetable plots (EUSEBIO, 1980). The solid component is used for biogas production. In Eastern Europe, ducks and fish are farmed together for five to six years, at which point a build-up of organic matter leads to a decline in fish production (SMITH, 1990). The ponds are then drained and crops such as rice, grown for two to three years, utilising the accumulated nutrients, before the duck-fish system is recommenced. In Colombia, in the humid tropics, PRESTON and MURGUEITIO (1993) have developed an intensive sustainable system incorporating sugar-cane, multiple-purpose trees and water plants as sources of biomass to feed livestock (pigs, ducks, sheep) and fish, as well as providing fuel for the farm. The pigs and sheep are confined and their excreta recycled through plastic-bag biogas digesters, ponds (for aquatic plants) and earthworms, providing organic fertiliser and humus for the crops. The leguminous trees also fix nitrogen. Soil erosion, a serious problem in some tropical grazing systems, is avoided. Many small farmers in Colombia are introducing or using either some or all of the elements of this integrated farming system.
Ultimately, the benefits of manuring will depend largely on nutrient transfers from non-cropped grazing land. Except in sparsely-cultivated areas, the livestock numbers required to support continuous cropping cannot be maintained by local pastures without external inputs in regions such as semi-arid West Africa.
It would be difficult to measure manure production and utilisation in the practical situation. However, the effects on soil properties such as cation exchange capacity and organic matter content means that these parameters can be used as indicators to monitor the positive effects of manuring (see Section 6.2).
In many developed countries and most developing countries, residues from crops grown for human food are a major source of nutrients for livestock in the dry season. At this time of the year, levels of crude protein and phosphorus in the residues of fertilised crops can be two or three times higher than those available from native pasture. Crop residues can compensate for the loss of natural grazing as a result of increased cultivation and burning, and help reduce the overstocking of rangeland which can cause land degradation. As a result, in some areas, higher livestock densities may be frequently associated with higher cropping intensities, negating the effects of land competition.
The importance of crop residues for feed in the West Asian and North African countries is described by NORDBLOM (1988). In Asia and Latin America, McDOWELL and HILDEBRAND (1980) have estimated that crop residues and by-products contribute 30-90 percent of livestock feed on small mixed farms. In sub-Saharan Africa, the contribution of crop residues to feed intake in ruminants has been discussed by SANDFORD (1989). On an annual basis, the proportion of feed derived by cattle from crop residues may be as high as 43 percent, increasing to 80 percent in the peak months of the dry season. Utilisation rates for cattle are higher than those for small ruminants, reflecting differences in preference between species and the priority given to large ruminants, especially those used for traction purposes. Some 60-100 percent of the leaves are utilised compared to 40 percent of the stalks; an indication of the higher nutrient content of the leaves.
Crop residue availability can be estimated from grain yields using multipliers, and can be related to livestock numbers and visits (KOSSILA, 1988). However, it is unlikely to be useful as a measurable indicator of direct environmental impact at local level.
Livestock can provide entry points for practices that promote sustainability, such as the introduction of improved forage species as leys into rotations based on annual arable crops and also into perennial plantation crops. In addition to their feeding value for livestock, which is well-documented, forages (especially legumes) make an important contribution to soil fertility and erosion control. Grasses and legumes, including woody species, can be used in a variety of erosion control measures such as vegetative hedges, strip cropping and cover cropping.
Vegetative hedges of densely-planted bunch grasses or shrubs, established on the contour at regular intervals, decrease run-off velocity, increase infiltration of water into the soil and facilitate sedimentation and deposition of eroded material by reducing the carrying capacity of overland flow. In the technique known as alley-cropping, annual food crops are grown in alleys between two adjacent hedgerows of trees and shrubs, usually legumes (KANG et al., 1990). The hedgerows are cut back at planting and periodically pruned during cropping to prevent shading, and to reduce competition with the associated food crops. The International Livestock Centre for Africa (ILCA) extended the concept of alley-cropping to include livestock by using a portion of the hedgerow foliage for animal feed. In regions such as sub-Saharan Africa, it is essential that fodder production systems for ruminants should not impinge excessively on land or labour required for crop production, because of the overwhelming importance and dominance of crops relative to livestock. Integration of legume fodder within the farming system through alley-cropping offers a means of achieving this as well as linking livestock production with arable crop production. In sub-Saharan Africa, the concept of alley-farming has been developed most in the humid areas of Southern Nigeria.
The inclusion of leguminous trees for livestock feed can also have important effects on the maintenance of soil fertility due to their efficient nutrient cycling. Legumes such as Leucaena leucocephala are capable of symbiotically fixing appreciable quantities of nitrogen, as well as supplying other nutrients. The yields of nutrients from prunings of L. leucocephala and Gliricidia sepium in alley-cropping systems in Nigeria are shown in Table 2. The effects of addition of prunings from the hedgerows on maintenance of soil fertility under 6 years of alley-cropping in Nigeria are presented in Table 3. With continuous additions of L.leucocephala to the soil, higher organic matter and nutrient levels are maintained than in soils receiving no prunings. Increases in organic matter under alley-farming enhance soil biotic activities. Also, the addition of prunings increases soil moisture retention, reduces run-off and minimises erosion. On sloping land, the woody hedgerows form a solid barrier to reduce wind and water erosion.
The effects of Leucaena and Gliricidia mulch on maize yields and soil chemical parameters on four farms in three villages in Nigeria are reported by LARBI et al. (1993). Table 4 shows the effects of mulching (0-100 percent of prunings) on soil properties. As more of the total prunings are applied as mulch, so soil chemical characteristics increase. Mean maize yields, averaged over three years for the four farms, were 1.68 t/ha (zero mulch), 2.45 t/ha (50 percent mulch) and 2.95 t/ha (100 percent mulch). Thus, cultivation of leguminous trees for livestock feed can also benefit the environment.
Since prunings applied close to planting the crop generate the greatest responses in terms of cereal yields, in integrated crop-livestock systems preplanting prunings of hedgerows can be used as mulch and part or all of the later prunings can be removed for animal feed without depressing current crop yield. In the above study, LARBI et al. (1993) concluded that half of the prunings could be used for livestock without seriously affecting the crop. In Bali, Indonesia, NITIS (1985) has developed a three-strata system. This comprises a herbage layer (including Panicum maximum, Cenchrus ciliaris, Stylosanthes hamata, S. guianensis Centrosema pubescens), a shrub layer (Leucaena leucocephala, Gliricidia sepium) on the border of the herbaceous layer, and a tree layer (Hibiscus tillaceous, Lannea corromandilica, Ficus peacelli) interplanted with the shrubs. The herbage layer provides the main feed during the wet and early dry season, the shrubs are mainly used during the dry season, and foliage and young shoots of the tree species are cut towards the end of the dry season.
Strip-cropping divides farmland into long narrow strips that cut across the path of erosive forces of running water. In this system densely-grown crops with a closed canopy alternate with strips of open-row crops with a tall and loose canopy structure. Strips can be established on the contour, around the borders of fields as single or double-row barriers or as buffers of permanent vegetation. In addition to erosion control, cultivating land in alternative strips regenerates soil fertility, improves soil structure and restores productivity. The biomass produced in the strips can be fed to livestock. Some commonly grown legumes sown in strip-cropping practices are given by SRIVASTAVA et al. (1993).
The use of leguminous cover crops is a long-established practice in plantations of coconuts, rubber and oil palm (THOMAS, 1978). Cover crops control weeds and soil erosion and improve soil fertility. Plants absorb the energy of raindrops, reduce splash erosion and the velocity of water run-off. The roots help bind the soil particles. The importance of leguminous cover crops in erosion control is illustrated by the observation of ALVARADO (1982) in the Colombian Amazon. Over a 32-month period, the accumulated soil losses were recorded as 54.5 t/ha for bare soil, 30.8 t/ha for maize grown on land prepared by conventional tillage, but only 2.8 t/ha under a cover of Pueraria phaseoloides. At Sungei Tekam in peninsular Malaysia, on a 10 per cent slope, the run-off coefficient (run-off expressed as a fraction of the rainfall received) averaged circa 0.21-0.23 on bare soil, whilst in the first year of establishment for Pueraria phaseoloides and Centrosema pubescens it decreased from 0.17 to 0.06-0.08 as the canopy developed (HONG, 1978). Cover crops of tropical grasses and legumes can have a positive effect on soil organic matter content, total nitrogen levels, water retention and transmission properties, and decrease bulk density in the surface layer of soils (WILSON et al., 1982). Livestock are easily integrated into plantation agriculture where they graze the understorey of improved forages (SHELTON et al., 1987).
Legumes are able to fix significant amounts of nitrogen. This element is made available to crops by decay of plant material and returned to the soil via the activity of the soil fauna or in dung and urine deposited by animals grazing the legumes. Compared to grass litter, legume litter has more nutrients and is released faster (RAO et al., 1992). Perennial forage legumes in the tropics usually fix between 20 and 180 kg N/ha annually, depending on the proportion of legume in the pasture which, in turn, is influenced by a range of environmental factors (HENZELL, 1968; THOMAS, 1973). However, shrubs such as Leucaena leucocephala are potentially capable of fixing up to 300 kg N/ha (DART, 1994). Nitrogen fixation of the order of 224-336 kg is possible with temperate clovers at high altitudes in East Africa (MORRISON, 1966), whilst the amounts fixed in Western Europe have been estimated at between 65 and 280 hg/ha/year (FRAME and NEWBOULD, 1986). Subsequent crops can benefit from the nitrogen fixed by legumes when grown in rotation. Maize sown after fodder-banks of Stylosanthes guianensis or S. hamata of one to three years duration has outyielded crops sown after bush-fallow or grown continuously (MOHAMED-SALEEM, 1986). Higher soil nitrogen levels and higher water infiltration rates were recorded, which reduced run-off and the erosion hazard. Infiltration rate in an area continuously cropped was only 15 mm/h compared to 49 mm/h after a three-year fodder-bank. Soil bulk-density was also lower after fodder-bank establishment (1.32 compared to 1.77 g cm-3). The amount of nitrogen fixed by the legumes was also significant. To obtain crop yields equivalent to the various legume-fallow treatments, 32-110 kg N/ha would have to be applied to soil cropped for three years. Similar observations have been made by TARAWALI (1991). Maize sown in well-managed fodder-banks produced maximum yields at 60 kg N/ha compared to 120 kg N/ha for bush-fallow or continuously cultivated soil. In the first year of cropping, 45 kg N/ha had to be applied to maize grown outside fodder-banks to produce the same grain yield as unfertilised maize following a good legume pasture. However, the improvement of fallow lands with selected forages is only possible in areas of low population densities. The role of Stylosanthes as a forage and fallow crop was the subject of a recent regional workshop in West Africa (de LEEUW et al., 1994).
Regulated leys, characterised by fencing and grazing management, are widespread in farming practice in temperate and sub-tropical zones, but relatively rare in the tropics. In the temperate zone, sheep and beef cattle particularly graze the leys in rotations and the cereal stubbles, enhancing fertility through the deposition of dung and urine. A review of intensive arable-sheep systems has been made by NEWTON (1982). In South and South-East England, sheep are associated with cropping farms that produce over 50 percent of the wheat, barley and potatoes in the nation. Pasture species include perennial ryegrass, timothy, cocksfoot and white clover, and catch-crops such as rape, fodder radish and stubble turnips are often grown. Excess pasture production is conserved as hay and silage for winter feed. In Southern Australia, rotations of wheat and self-regenerating annual legumes such as subterranean clover, utilised by sheep, were developed in response to declining cereal yields and soil erosion. In Northern New South Wales, 2-3 years of cereals alternate with a 4-year ley, which sustains 5-15 dry sheep/ha, builds up organic matter and fixes nitrogen. In the former Soviet Union, rotations can be elaborate e.g. an 8-year system composed of 4 years alfalfa, 2 years annual grasses and 2 years maize. In the West Asia - North Africa region, Medicago species have been shown in trials to have a positive effect on soil factors when established in cereal-fallow rotations. However, attempts to introduce the successful South Australian technology onto farms in the region have foundered due to a combination of biological and socio-economic constraints (RIVEROS et al., 1993).
In sub-Saharan Africa, leys rotate with tobacco in Zimbabwe and with a variety of crops including wheat and pyrethrum in the highlands of Kenya. However, population pressures in countries such as Ethiopia have led to continuous arable cropping with serious implications for the maintenance of soil fertility. In South America, rotations including maize, soyabeans, dryland rice and the grass Brachiaria decumbens are common in the central savanna region of Brazil. In the savannas of Colombia, the development of new acid-soil tolerant, blast-resistant dryland rice varieties has opened a range of new production systems options including pasture-rice rotations and the simultaneous establishment of rice and grass-legume pastures. In the Andean zone and at lower altitudes in the countries of the Southern Cone of South America, alfalfa is the most important forage crop in leys at elevations up to 3200 m, followed by oats. Alfalfa is often sown in rotations with crops such as maize, potatoes, barley, peas and garden vegetables, sometimes under irrigation. In Colombia, between 2000-3000 m altitude, Kikuyu grass is widespread, whilst other species include perennial ryegrass, white and red clover. Over 4000 m elevation in the Andean region, one or two years cropping with Andean tubers (bitter potatoes, oca), Andean grains (quinoa), potatoes, barley, oats or broad beans is rotated with four to >10 years fallow. The grazing area is cropped and allowed to revert to native grassland utilised by sheep and camelids. This, in effect, is an unregulated ley system without the fenced sub-divisions of regular ley farming. Such a system is as vital to the restoration of soil fertility in peasant farming systems in the high Andes as it is in sub-Saharan Africa.
In arable soils, the longer the period in pasture the higher the water-stability of the soil aggregates and the more quickly can a seedbed with a better physical state be prepared. The longer an arable soil is intensively cropped, the longer it will take to restore its structure. In the temperate zone, experiments comparing the effects of length of ley on test crops of wheat showed increases in grain yield of 35 and 61 percent, respectively, after one or three years of ley, over continuous cropping (LEWIS et al., 1960). Similar results were obtained with kale. The ley improved soil physical conditions and the nutrient status with a greater supply of nitrogen. There was less carry-over of crop diseases and less weeds when a ley was included. The beneficial effects of leys on soil nutrient status can be replaced by inorganic fertilisers, but at a financial cost.
Crop rotations which include a ley are an excellent insurance against sodicity problems in sodic soils (high contents of exchangeable sodium and low permeability). Pastures build up the soil structure and improve its stability. The return of manure, when the leys are utilised, is also valuable in maintaining the permeability of the surface soil and reversing the deterioration in structure that occurs in sodic soils.
Recent evidence from Colombia suggests that improved grasses may also help to counteract the greenhouse effect by removing as much as 2 billion tons of carbon dioxide annually from the atmosphere (ANON, 1994). The perennial grasses (Andropogon gayatus and Brachiaria humidicola convert as much as 53 tons of CO2 per hectare annually to organic matter. When sown with legumes, the grasses are able to fix even more CO2.
In improved pastures, a decline in the proportion of sown species (especially the legume content) could be used as an indicator of grassland degradation (see Section 6.1), along with changes in soil properties (see Section 6.2).
3.5.1. Animal traction
3.5.2. Manure supply
3.5.3. Crop Residue supply
3.5.4. Improved forages
3.5.5. Risk management and asset building
3.5.6. Agricultural intensification and crop-livestock interactions
The development of animal traction in sub-Saharan Africa has been slow and patchy and has not achieved the rates that planners and policy-makers predicted (McINTIRE et al., 1992). It has gained little acceptance in the humid and sub-humid zones partly as a result of disease (trypanosomiasis), and because bush-fallow agriculture, which predominates in these zones, is not well adapted to the technique (JONSSON et al., 1994). Elsewhere, under semi-arid conditions, acceptance has been greater but has relied upon several pre-conditions for its success. These include:
· Farming intensity and market access.PINGALI et al. (1987) argued that mechanisation is a response to growing population density and to changes in output and factor prices. A growing population density shortens fallow periods and raises annual labour input per unit area. Shorter fallow periods encourage weed infestation and increased labour demands from shorter fallows induce mechanisation to save labour. Increases in output, prices, wages or both will stimulate agricultural production and encourage the use of labour-saving techniques. They conclude that certain levels of population density and market access are almost always required for mechanisation to be successful.
· Feed competition.
· The development of contract work or hiring.
It is certain that, where animal traction has been adopted on a relatively large scale (e.g. semi-arid West Africa), it has been associated with the production of cash crops such as groundnuts and cotton. Access to markets for these products has been essential to pay for the technology, particularly where animals are not owned by cultivators. The introduction of the technology has led to area increases mostly at the expense of fallow and grazing land (PINGALI et al., 1987; JONSSON et al., 1994); by as much as 100 percent in some cases for individual farmers. The picture regarding yield increases is less clear as animal traction does not necessarily lead to more intensive production. However, improved timeliness may have a positive yield effect under some circumstances. BARTON (1987) reported a significant yield effect for deep ploughing before the rains under semi-arid conditions in Western Sudan. This was due to increased water infiltration. PINGALI et al. (1987) reported that it is often impossible to separate the yield effects of the introduction of animal traction from other influences such as increased use of fertilisers and the introduction of modern varieties. In general, it is assumed that the yield effects of the introduction of draught animals will not be sufficient to justify the investment in the technology without area increases and labour saving (McINTIRE et al., 1992).
Labour savings per unit of land are one of the major benefits of the adoption of draught animal technology and have been reported by PINGALI et al. (1987) in their review of the literature on the subject. Labour savings will, in theory, allow the release of labour from field operations for other tasks either on- or off-farm. However, despite labour savings per unit of land, labour use per farm often rises in response to the additional area cultivated and increased weeding operations. Therefore, in practice, mechanisation rarely releases labour for other farm tasks such as livestock production or enabled farmers to develop their off-farm income.
Although animal traction is generally regarded as an example of the complementarity that can be achieved through the combination of crop and livestock production there are also potentially some negative environmental and economic consequences of its adoption. These include declining soil fertility and pasture availability (see Section 6.4.4).
Although the use of draught animals can be seen to provide positive environmental and economic benefits in terms of savings in fossil fuels (i.e. less pollution and less demand on imported fossil fuels), generally speaking animal traction does not replace the use of tractors. Rather, it tends to replace human labour. It is, however, possible to calculate the potential benefits to the environment in reduced carbon dioxide emissions or the benefits to the economy in savings in foreign exchange if the work undertaken by tractors were performed by draught animals. It is perhaps most appropriate to make these calculations where the use of tractors and draught animals exists side by side, e.g. the Punjab, India and Pakistan (i.e. where the alternative power sources are available).
Crop-livestock farmers in sub-Saharan Africa and Asia rarely keep sufficient numbers of animals, or manage these animals in a way to provide enough manure and, therefore, nutrients to maintain fertility without the application of other techniques including fallowing, the growing of legumes, the incorporation of crop residues and the use of chemical fertilisers. Animals are kept for a number of social and economic reasons and the major objective of the farmer is often crop production to guarantee subsistence and income. The bulky nature of manure often results in less than efficient use, although, there are examples of improved use in the face of declining fallow periods and fertility (RAVNBORG, 1990). Whether the quantities are sufficient to maintain fertility is open to question.
In the humid zone of Africa, small ruminants predominate but they are rarely integrated with crop production, forage is not grown and the manure is not returned to cultivated plots (OKALI and SUMBERG, 1986). In Asia, ruminants are tethered and stall-fed depending upon the season and the availability of grazing land (DOYLE et al., 1986). Restricting the movement of animals obviously produces more usable manure.
The level of intensity of crop residue use is closely correlated with the availability of alternative sources of fodder, in particular, natural grazing. For example, in Thailand, average land holdings are 4.28 hectares and 83 percent of this land is used for cereal production. The remainder is utilised for vegetable and fruit production with only 0.03 hectares available for grazing (KHAJARERN and KHAJARERN, 1984). This land is effectively waste, being field boundaries and bunds. Livestock contribute around 17 percent of farm income. The situation is similar in many other South-East Asian countries (DOYLE et al., 1986). Under these conditions, farmers have little alternative to the use of agricultural residues to maintain their livestock. This accounts for the efforts made in recent years to upgrade these residues by chemical treatment and supplementation (DIXON, 1985). As long as cattle and buffalo remain the source of power for farming operations in these countries there will be limited opportunities to intensify livestock production. It is not economically or socially feasible to replace crop production for human consumption with the development of fodder crops for livestock, where human population density remains high. However, there are exceptions eg. the highlands of Kenya.
There are examples of crops which have potential as both providers of human food and animal fodder. Groundnut hay is an important feed resource for livestock in West Africa. However, labour demands do not always allow the best use of these resources. For example, in the highlands of Ethiopia, oats is a crop which potentially can produce high-quality hay. However, it is not always harvested at its peak feeding value because of conflict with other cropping activities (BROKKEN and WILLIAMS, 1991). This results in lower - quality although lower - cost fodder. Groundnut hay is not always removed from the field in West Africa, where it can deteriorate or be grazed by roaming livestock, as labour and transport are in short supply and expensive.
As cultivation expands into natural grassland and increasingly the best grazing is enclosed for private use (QUAN et al., 1994), the importance of crop residues as livestock feed is likely to increase. However, McINTIRE et al. (1992) report an increasing tendency toward the privatisation of crop residues as crop farmers diversify into animal production and the availability of natural grazing diminishes. Thus, it is the specialist herders who are likely to become increasingly marginalised and forced to depend on rangeland unless they too begin to diversify into crop production. There is evidence that some pastoral groups are beginning to cultivate crops in response to the decline in availability of grazing (TOULMIN, 1983).
BROKKEN and WILLIAMS (1991) report that intercropping cereals with legumes is common practice in Southern Nigeria and that this technique reduces the opportunity cost of labour and land in producing forage. Intercropping forage legumes with cereals has been tested in Sudan (BUNDERSON et al., 1986), where it was concluded that research should concentrate on legumes with potential for both grain and forage production. Undersowing or intercropping have much greater chances of success than monocrop fodder production for several reasons. Land and labour are already allocated to the cereal crop, intercrops can suppress weeds and additional protection from livestock during the growing season is not required. In addition, intercrops of cereals and legumes are not alien concepts to many farmers in developing countries. Thus, these techniques have significant economic advantages over monocrop fodder and legume production.
One of the major constraints to the introduction of cultivated fodder, whether as a sole or undersown crop in Africa, is that animals often have free access to crop residues after harvest. Thus, if individual farmers were to sow fodders there would be no assurance that their own livestock would benefit (RAVNBORG, 1990). This situation may change as agriculture intensifies and farmers retain greater control over their crop residues. However, privatisation will probably initially manifest itself in the removal of residues from fields rather that the protection of the fields themselves through fencing.
In Africa fodder production is almost exclusively concentrated in the highland zones eg. Kenya, where specialised milk production is sufficiently profitable for forage crops to replace food and cash crops. Even under these conditions fodder production may be confined to the growing of perennial grasses on field bunds and roadsides. In the humid zones, alley-farming techniques developed in Nigeria for the provision of fodder and improvements to soil fertility have not gained wide acceptance (OKALI and SUMBERG, 1986), and there may be conflicts between the use of leaves for forage or mulching. Under semi-arid conditions sown forages face an acute labour conflict with arable crops and, therefore, must fit into specialised niches and markets. Some success has been reported with the introduction of fenced fodder banks for dry season grazing in an agropastoral system in Nigeria (BROKKEN and WILLIAMS, 1991).
Fodder production only becomes economically attractive where there exists an urban market for animal products, particularly milk. There are few examples of fodder production for meat or draught power in Africa (McINTIRE et al., 1992). PRASAD et al. (1987) report that the availability of a dairy marketing infrastructure in Andhra Pradesh, India, has led to farmers adopting a paddy-Napier grass cropping system producing rice and milk for local markets.
In Syria, under certain market conditions, farmers decide to graze mature barley with sheep rather than harvest the grain crop (NORDBLOM, 1983). If the relative prices for mutton are higher than those anticipated for barley, the crop will be grazed. Although not an example of forage production the possibility exists under this system to use barley as a fodder resource. Immature barley is often grazed by sheep during the winter in Syria and then left to mature as a grain crop.
In India, WALKER and RYAN (1990) reported that the use of leguminous trees in alley-cropping systems was unprofitable because of a shortage of land, and the cereal monocrop produced a better return. A sole crop of leguminous trees could be profitable, but few farmers were willing to devote their entire resources to a single product as diversified systems spread risk and reduce uncertainty. The use of leguminous fodder trees has gained some acceptance in South-East Asia, but rather than being integrated with crop production they tend to be grown on waste land or on field boundaries and roadside verges. They provide important additional supplements for ruminants fed predominantly on crop residues, particularly in the dry season (BAMUALIM, 1985). However, rarely have these fodder trees replaced arable crops, and they remain secondary to the production of food for subsistence and sale.
Where manure and draught power are essential to the farming system various strategies are undertaken by farmers. BYERLEE and HUSAIN (1993) report that crop-livestock farmers in the North-West Frontier Province of Pakistan intercrop green fodder in food crops. They also plant maize at three to five times the optimal density for grain production, and then thin the crop throughout the growing season to provide green fodder for animals. Extension workers discourage this practice because it depresses grain yields. Yet, overall production is enhanced if the livestock production is taken into account.
Under many mixed farm circumstances livestock keeping allows for the accumulation of capital for investment or risk avoidance. Livestock are capital goods which are highly mobile and can be liquidated rapidly if economic incentives are unattractive, there is a need to purchase inputs for crop production or in times of crisis for the farm family (JARVIS, 1993). This is often particularly true for poor or small farmers in developing countries. SASTRY et al. (1993) report from Southern India that small farmers keep a higher proportion of livestock (cattle, sheep and goats) than larger farmers (measured by land area). Their income from livestock is also proportionally greater, they keep cows rather then bullocks, whose main function is draught on larger holdings, and they sell a higher proportion of smallstock which are mostly consumed on the larger holdings. Landless households realise most of their income from livestock sales and selling their labour. Income from crop production is much more uncertain than that from livestock as a result of crop failure and livestock provide stability to household income. This suggests that even if livestock income is secondary to crop income in terms of scale, it remains an important buffer in times of uncertainty.
Livestock play an important role in the subsistence mixed farming systems of north-east district, Botswana (MUKHERJEE, 1994). Livestock are essential for draught, manure supply, and household subsistence during that period when crops are sown and there is little grain left for human consumption. The consumption of livestock products (goat meat and biltong) increases during these periods. During times of drought livestock (goats) provide the only means of survival and are sold or exchanged for staple foods. Cattle are only sold under extreme circumstances. Similar conditions exist in Senegal, where livestock are critical to survival mechanisms following drought, other natural calamities or household distress. The sale of livestock (goats and poultry) provides a means of overcoming these difficulties (SCHOONMAKER-FREUDENBERGER, 1994). In Burundi goats and sheep are kept as a marketable reserve and as producers of manure under conditions of high human population density (BRANCKAERT, 1993).
Although offtake rates of livestock may appear to be low from traditional crop-livestock systems this may be a result of government intervention in producer prices (JARVIS, 1993) or reflect times of relative plenty. In times of difficulty the capital accumulated in livestock tends to be liquidated.
Thus, livestock contribute positively to the overall sustainability of crop-livestock farming systems, livestock often providing the resources for inputs to crop production and potentially providing the means for investment in land improvements (TIFFEN, 1992).
As population densities increase so agricultural production intensifies and the degree of integration and interaction between crops and livestock increases (McINTIRE, et al., 1992; RUTHENBURG, 1983). Crop land expands at the expense of fallow and pasture and farmers seek alternatives (manure) to fallow to maintain fertility. Labour use increases per unit of land (TIFFEN, 1992; PINGALI et al., 1987) as construction of fences, water control structures and erosion control measures takes place. Use of crop residues intensifies with cutting, storage and stall feeding and farmers replace hand cultivation with draught animal power and ultimately engine power.
As crop-livestock interactions become closer the unit cost of livestock production declines (manure becomes more valuable, and increased production of crop residue reduces feed costs) which allows higher stocking rates per unit of land which may increase competition for labour between livestock and crops. This increased competition for labour may be temporary as eventually intensification leads to private ownership of land and investment in land improvements. This in turn leads to increased use of perennial crops (trees) to produce products for sale and the planting of multipurpose trees (fuel and fodder) leading to further integration between crops and livestock (TIFFEN and MORTIMORE, 1992). These changes should lead to increases in household income unless a growing human population causes sub-division of holdings in the absence of alternative employment opportunities outside agriculture.
The final change in this process of development is a return to specialised crop and animal production (McINTIRE et al., 1992) under the pressure of human population growth and the development of profitable production options e.g. milk production, tea and coffee. Other factors which play a role in the development of specialisation are reduced transport costs, the availability of engine power for cultivation and declines in fertiliser prices and their replacement of organic fertilisers.