The Role of Large Ruminants

by B.R. Joshi

INTRODUCTION

Large ruminants (cattle, buffalo) play a vital role in the whole agricultural system and so have a large influence on the rural economy of Nepal. It can be said that large ruminants are the backbone of Nepalese agriculture and maintain the livelihood of the 93% of the population which depends upon agriculture.

In general, agriculture in Nepal is subsistence-based and very complex, with a mutual interdependency of crops, livestock and forest resources. It is the large ruminants which are central to these complex farming systems. It has been estimated that agriculture contributes about 60% of the country's real GDP (Gross Domestic Product) and about one third of exports. It also provides employment for more than 90% of the national labour force. The 3.1 million hectares of land under cultivation in Nepal are mostly cultivated using animal power. Various crops are grown among which rice is the dominant crop (52%) followed by maize (19%), wheat (15%) and millet/barley (6%) (DFAMS, 1990).

Livestock products like milk, meat, and hides, contribute 15% of national GDP, and 28% of the agricultural GDP. Additionally, livestock are an integral part of the agricultural production system, providing almost all of the draught power and fertilizer (manure) for cultivation. Oxen and male buffaloes are also used for hauling and transport of agricultural inputs and products. In the hills and mountains, mules, yaks, sheep and goats make an important contribution as pack animals.

Milk and milk products are a major source of animal protein in the Nepalese diet, and livestock and livestock products are an important source of household cash income (about 20% of total income) especially in the hills and mountains (Nepal Rastra Bank, 1988).

A survey of cash income and labour use for livestock enterprises in Nepal (Table 1) reveals the situation in different regions of the country.

Table 1: Household income and labour utilisation pattern in Nepal.

Source: Nepal Rastra Bank (1988).

In another survey in a midhill village of western Nepal, livestock and livestock product sales contributed only 8.6% of total income per household. However, it was found to be contributing some 67% of the total income from agricultural activities (Shrestha et al, 1991).

CONTRIBUTION OF LARGE RUMINANTS TO THE FARMING SYSTEM

Large Ruminants as a Source of Farm Power

It has been estimated that there are 2.7 million draught animals in Nepal, 94% of which are cattle bullocks (Rajbhandari and Pradhan, 1991). Cattle, buffalo, yak, chauries, ponies, mules and donkeys are all used as draught animals, but the use of large ruminants in particular is most valuable as a means of farm power in Nepal.

58% of cultivated land and 62% of the rural population are found in the hill and mountain areas (LRMP, 1986). Only a limited area is accessible by road and the majority comprises small, fragmented and steeply terraced land where modern agricultural equipment could not be used efficiently. Traditionally, only male animals are used for draught purposes, and only male members of the family operate them usually in pairs. Almost all land preparation is performed by animals, and will be for a considerable period of time to come.

Oli (1984) estimated that the draught power used for cultivation in Nepal was equivalent to about 1.37 million kilowatts (KW) of energy and the contribution of those animals was worth about NCR 1300 million at 1984 prices. Oli (1985) also stated that draught animals are used for work in the hills for about 62 days/annum and in the Terai for about 130. The average number of working hours/day recorded for a pair of working oxen during the rice planting season in the hills was eight hours and the area ploughed 0.25 ha/day. The number of working hours/day for a pair of swamp buffalo for the same work in the Terai was about seven hours and the area ploughed 0.37 ha/day.

It is apparent that draught animals are not utilised efficiently throughout the year, except for agricultural practices (ploughing, threshing) in most parts of the country. Oli (1985) estimated that 78% of food energy consumed by the draught animal is used simply to maintain it during unproductive periods of the year. This is estimated to be equivalent to 24.8 megajoules of metabolisable energy (MJME) for maintenance, and an additional 15 MJME for work for hill Zebu oxen (Oil, 1985)

However, in the higher Himalayan regions, yaks, chauries and even Zebu animals are used, not only for agricultural practices, but also as pack animals.

The time required for a pair of hill Zebu oxen to plough land for different crops has been investigated, and the data are presented in Table 2.

Oli (1985) estimated that farmers are providing a ration with an average feeding value of 47 MJME, so that energy is not a limiting factor in draught animal performance. However, this study was undertaken in the Eastern hills of Nepal and may not represent the situation prevailing throughout the country.

Some work has been done to compare the efficiency of local hill Zebu oxen and Jersey crossbreds (Pearson, 1990). It was reported that the absence of the hump in the crossbred oxen had no effect upon the position of the yoke, or on pulling the plough, and that the Jersey crossbreds, particularly the longer legged ones, had a higher rate of work than local oxen.

Table 2: Average time taken by a pair of hill Zebu oxen to accomplish work on different crops.

Source: Oli (1985)

Large Ruminants as a Source of Manure

Traditionally, crop production in the Nepalese hills is based upon the recycling of organic residues from arable cultivation as animal feed and bedding to provide manure and compost. Estimates of compost application on farmers fields are quite variable with a range of 3–21 mt/ha (Heuch, 1986). This variation is possibly due to the variable factors governing the use of compost by farmers, e.g. cropping pattern, variety of crop grown, land type, distance of land from the compost source, size of land holding, number of animals, labour availability, and access to forest resources. Taking the nutrient content of compost to be 0.5%, 0.3%, and 3% of N, P and K respectively (ICAR, 1986), the amount of soil nutrients supplied by a compost application of 21 mt/ha would be 84, 63 and 63 kg of N, P and K per hectare.

The amount of plant nutrients removed from soil by the two major cropping patterns in the hills is presented in Table 3. From this table, it is clear that the amount of plant nutrients removed from the soil by crops in the maize - wheat system is just matched by the compost (maximum level) applied by farmers. However, the amount of compost is not sufficient for the maize - rice - wheat system.

Given the present situation of declining soil fertility, crop productivity cannot be sustained. To meet the national requirement for food, a doubling of the productivity of the staple crops by the year 2000 A.D. is needed. This will be feasible only if there is a rapid expansion in double and triple cropping using high yielding varieties (Sthapit et al, 1989), which in turn will only be made possible by an increased rate of inorganic fertilizer use.

In view of the existing situation of poor infrastructural development for fertilizer transport and distribution, and the limited purchasing power of hill farmers, these targets are not realisable in the near future, and hill farmers will have to continue to depend upon traditional sources of nutrients for the maintenance of soil fertility, and to intensify agricultural production.

The main contributors of FYM for composting are the large ruminants, the number of which is already at a critical level in the midhill region of Nepal. An increase in their numbers is not desirable.

Table 3: Quantity of plant nutrients removed from the soil by the maize - wheat and maize - rice - wheat cropping pattern.

Note: a - Average National yields in the hills for 1981/82.
b - Straw yields are calculated on the basis of a 0.4 harvest index.
c - Calculated from data of Kunwar and Dutta (1970).

Source: Balogun (1989)

The amount of dung produced per animal per day depends upon the amount of forage fed to the animals. In general, most of the animals in Nepal are underfed throughout the year except during the monsoon period. Oli (1987) estimated the total output of dung (fresh weight) from cattle and buffalo per day to be about 10 kg and 12 kg respectively giving a total of 3.6 tonnes and 4.3 tonnes of dung per animal per annum respectively. Assuming the animals are stall fed, the annual output of N, P and K for cattle would be 12.6kg N, 5.4kg P and 7.2kg K, and from buffalo 15kg N, 6.5kg P and 8.6kg K.

Table 4: National herd composition.

Source: LMP (1991)

Assuming the dung output of young animals to be half that of adults, the total dung output from cattle and buffaloes would be about 33 million mt which is equivalent to 116,500 mt N, 50,000 mt P and 66,500mt K. (Dung NPK = 0.3% – 0.4%, 0.1% – 0.2%, 0.1% – 0.3% - ICAR, 1986). At current prices the value of dung from cattle and buffaloes, in terms of fertilizer for the country, would be equivalent to US$ 58.75 million. In reality, it is reasonable to assume that half of this potential manure production would not be available to crops because most of the cattle population in the country is raised under a semi-stall fed system and the dung produced during the grazing periods is not fully utilised for crop production.

Animal urine is another potential source of soil nutrient. Borowski and Liebhardt (1983) reported that urine is very rich in nitrogen and potassium. The value in cattle urine has been reported to be 0.9 – 1.2% N, a trace level of P and 0.5 – 1.0% K. Urine output per kg body weight is in the range 17–45 ml/kg body weight. The average bodyweight of Nepalese cattle and buffaloes can be taken to be about 200kg and 300kg respectively. Hence, average urine production per day per animal would be about six and nine litres in cattle and buffaloes respectively.

Urine as a source of fertilizer has been investigated at Lumle. The effect of a 1:1 water:urine mixture top dressing on the marketable yield of leafy vegetable crops was found to be a significant increase of 81.7% over an untreated control, and of 23.7% over a top dressing of urea (Joshi et al, 1990).

If the 50% of urine that is wasted could be exploited as fertilizer, an additional 10,000 mt nitrogen and 55,000mt potassium would be added to the amount of soil nutrients. At the present pricings in Nepal this would be equivalent to US$ 12m worth of chemical fertilizer.

However, as previously mentioned, the manure obtained from animals has not been exploited to its maximum potential for various reasons. First, animals are not stall-fed throughout the year due to the shortage of forage and watering difficulties. Second, animal housing is inappropriate, especially for the collection of urine, and third, inadequate bedding materials are provided in the animal sheds due to the limitation of forest resources.

Large Ruminants as a Source of Animal Protein

The contribution of large ruminants as a source of animal protein to the rural as well as to the urban population of the country is very significant. Consumption of milk and milk products in Nepalese food is traditional. Meat from buffaloes constitutes about 65% of the total meat production (94,000mt) of the country (DFAMS, 1990). Per capita consumption of animal protein in Nepal is given in Table 5.

Table 5: Per capita consumption levels, current and targets.

Source: DFAMS, (1990)

Table 5 clearly shows the need to increase livestock production to fulfil the basic need requirements of a population increasing at the rate of 2.1% per annum (See Table 1, Chapter 3). FAO (1988) estimated the animal protein supply for Nepal to be 7.8 gm/ha/day for the period 1984–86. This is considerably below the minimum animal protein requirement in the diet. In a survey in the Koshi Hills of Nepal, Marsh (1985) reported that in rural areas only 18% of the milk produced is consumed as milk or curd, with 19% being sold as milk or curd and 72% used for the production of ghee (churned butter). Ghee, as a non-perishable commodity, is the main livestock product sold for cash by the average Nepalese farmer, although only 60% of the ghee produced is actually sold (Marsh, 1985).

The by-product of ghee production is “butter milk” which is consumed by the family. Although cash income from the sale of ghee is considerably less than from the sale of fluid milk, the milk protein present in the butter milk remains in the family for home consumption and thus compensates for the lower income from the sale of ghee.

Large Ruminants as a Means of Cash Generation

Livestock enterprises constitute a major source of income for the average household in Nepal. A multipurpose household budget survey (Nepal Rastra Bank, 1988) showed that, of the total household income, livestock sources contributed 21.2% in the mountains, 19.7% in the hills and 9.7% in the Terai. In a household survey in a mid hill village of western Nepal, Shrestha et al (1991) reported that only 12.9% of annual cash income came from agricultural activities, of which livestock enterprises contributed 8.6%. The major livestock product for cash generation is ghee followed by live animals. However, in urban and peri-urban areas where milk collection facilities exist, the sale of fluid milk is a major source of income. The Dairy Development Corporation (DDC) is the major buyer and it controls about 65% of the market by volume. During 1986/87, 15 million litres were collected by the DDC, representing 2% of total annual production. Milk is collected from some 50,000 farmers with the average volume procured per farmer being 1.0–1.5 litre/day (LMP, 1991).

On a national scale, livestock enterprises earned a total of 1.6 million rupees during 1987/88 which accounted for 64% of total agricultural exports (Nepal Rastra Bank, 1988). The major commodities for export from large ruminants are ghee, live animals, and hides and skins. These represent about 50% of the total export of livestock commodities. The export of livestock and livestock commodities from large ruminants is presented in Table 6. Along with these tangible benefits for the household and/or country, there are considerable intangible benefits from livestock in the rural household.

Table 6: Export of livestock and livestock products from large ruminants during 1987/88 (estimated at 1988 prices).

Source: DCFTS (1988)

While a milking buffalo is cared for as a family member, the sacred cow is kept, not only for the production of draught animals, manure and milk, but also for religious purposes. Cattle dung and urine has religious significance, and is utilized in religious ceremonies.

In the rural household, livestock is the main agricultural enterprise utilised in times of urgent need. In a study on the ways in which Nepalese households in the hills cope with economic crisis, Nabarro et al (1989) found that the sale of animals plays a crucial role. Economic stress on households arises through illness or death in the family, harvest failure or environmental hazard such as fire, landslide or earthquake. In response to the unexpected, families sell one or two animals to generate cash, usually large ruminants (oxen and cows) in preference to smaller animals (goats, pigs).

Gatenby et al (1990) estimated the annual income from the sale of milk from one buffalo to be Rs 1650 (US$ 59) and from one cow to be Rs. 1000 (US$ 36). Colin and Falk (1979) reported that for small scale farmers, the sale of livestock and livestock products accounts for 30% of all on farm income.

It can be concluded that livestock rearing in Nepal, and especially in the hills and mountains, is multipurpose and crucial to the overall economy and well being of the farm household. Moreover, livestock rearing is a means of sustaining crop productivity and hence, ensuring the livelihood of farmers.

POPULATION AND BREED COMPOSITION

The large ruminant population in Nepal during 1989/90 was 6.2 million cattle, 3.0 million buffaloes and 11,000 yaks/chauries (DFAMS, 1990). The total livestock population places Nepal amongst the countries having the highest livestock population per unit of land area. The livestock density per unit of arable land in different ecozones is presented in Table 7.

Table 7: Livestock density per unit of arable land (head/km²).

Source: DFAMS (1990)

From Table 7 it is evident that the livestock population is concentrated in the hill and mountain regions of the country. This situation is further illustrated in Table 8. The number of livestock units per household in the hills and mountains is more than double that of the Terai, which reflects their economic significance in the region.

Table 8: Availability of livestock units per household and per hectare of cultivated land between the ecological regions.

Source: Chitrakar (1990)

The available statistics suggest that the livestock population in Nepal and its pressure on the ecology of the country is considerable, and perhaps the excessive drain on natural resources will lead to an ecological and environmental imbalance. This problem is much more acute in the hills and mountains. The current 9.62 million livestock units in the mountains and hills are already an excessive burden on the land. Rajbhandari and Shah (1981) have estimated that, given the existing level of productivity and land use pattern, the hills of Nepal could efficiently maintain only 2.78 million livestock units. The excess population is competing for limited resources leading to declining productivity of individual animals and deterioration of the environment.

However, despite this overstocking, Nepal is still a net importer of livestock and livestock products which suggests that the individual productivity per livestock unit in Nepal is very low. Buffalo meat, which accounts for an estimated 65% of total meat consumed, is mostly derived from imports of buffaloes from India. Similarly, a considerable number of small ruminants are imported from Tibet and India. The Department of Customs and Foreign Trade Statistics data showed that Nepal imported 130,000 buffaloes and 177,000 sheep and goats for meat consumption in 1987/88. Thus, the total deficit in trade for meat animals was about Rs. 108 million (US$ 2.4m). Furthermore, Nepal imports a substantial amount of milk powder and condensed milk to cater for the needs of tourists and the urban population.

THE NATIONAL HERD OF LARGE RUMINANTS

The national herd of large ruminants consists mainly of Bos indicus type cattle, and buffalo with a predominant swamp ancestry. Both species have suffered from generations of indiscriminate breeding and are now of a non-descript type, though well adapted to the rigours of the environment.

Cattle

Some attempts have been made to describe and define the native cattle and buffalo into distinct breeds. Epstein (1977) classified Nepalese cattle according to body size and other physical characteristics. He described the cattle breeds in the Terai as Zebu cattle of Terai, Ponwar, Kherigarh, Bachaur, and Hariana, and associated each group with various regions of the Terai. This may reflect the importation of these animals from the adjacent Indian plains.

Similarly Epstein (1977) classified hill cattle into different breeds according to region. Hence, there are Siri cattle from Bhutan which were bred with the black hill Zebu cattle, the crossbreed being known as Kachha Siri. This breed is predominant in the hill districts of Eastern Nepal.

Throughout the hill region in mid-Nepal the black hill Zebu is the predominant breed and it is suggested that they resemble the black or red Purnea hill Zebu of North-East Bihar. The body weight of the adult bull is 200–250kg and of adult cows is 120–160kg. Hill cattle are larger in the hills of the Mahabharat range than in the foothills of the Himalayas. In the far-west hill districts of Achham, another distinct Zebu bred, the “Achhami” has been described. This breed is smaller than the black hill Zebu cattle, the adult bull weighing 150kg and adult cow 100–120kg.

In the high Himalayan zone and rain shadow areas, a small number of humpless cattle are found. These animals are identified as Lulu (Western Region) and Kirko (Eastern Region). The animals are small with the adult bull weighing 200–250kg, and cows weighing 100–150kg. The total population of these animals was estimated to be only about 30,000 (Joshi, 1982).

The Yak (Bos grunniens) and the crossbred chauries are present in the high Himalayan region of the country. Joshi (1982) has described their distribution in the sixteen high-Himalayan Districts of Nepal, and their annual migration pattern. He estimated that the total population of yaks and chauries (excluding Lulu) to be about 25,000 head. However, more recently, DFAMS (1985) estimated the population to be 11,000 head indicating a sharp decline.

Buffalo

Buffaloes in Nepal are reared in areas from the tropical Terai to the subarctic Himalayan region where, during the summer buffaloes can be found grazing at altitudes of 3500–4500m asl.

Most native buffalo in Nepal are small and sturdy, with adults weighing 270–450 kg with a short horn. They are dark grey or black in colour with good hair coats. The predominant swamp ancestry is indicated by the chevron markings seen in many specimens (FAO, 1977). However, with the introduction of buffalo improvement programmes in the country, the riverine buffalo breeds, Murrah and Swati have also been introduced. In the districts bordering India, most of the buffaloes are now either Murrah or Murrah crossbreds.

Nepal also has a small population of wild buffalo, the Arni (Bubalus arnee) now mostly maintained in wildlife reserves.

Cockrill (1974) regarded indigenous buffaloes with long laterally spreading horns as derivatives of swamp buffalo. However, it is generally agreed that there are no distinct buffalo breeds in Nepal and that all Nepalese buffaloes are of Indian origin.

Epstein (1977) classified the buffaloes of Nepal into three regional types namely Terai, hilly midlands, and Himalayan Mountains on the basis of the size of their horns and the length and colour of their coats.

The swamp buffalo of the hills have also been classified into two types, the Lime and the Parkote (Shrestha and Sherchand, 1988). These authors estimated the populations of Parkote to be 0.5 million, Lime, 1.5 million, Murrah cross 0.4 million, and Terai Buffalo 0.6 million. Although buffaloes in Nepal are described as being of variable type, the distinguishing features are not very well described or documented.

PRODUCTION PERFORMANCE OF INDIGENOUS LARGE RUMINANTS

Though different breed characteristics have been described for the cattle and buffalo populations, the different production characteristics in terms of milk, meat and other outputs have not been reliably measured. Studies conducted so far regarding milk production characteristics have described the native animal as a nondescript local. Hence, the recorded performance of animals reflects the performance of native stock of a particular area, and variation in management, feeding, and the recording system itself has led to variable findings.

Shrestha (1989) estimated the average lactation yield of Nepalese cattle to be 332kg and for buffaloes to be 754kg. Pradhan et al (1989) combined data from various sources, and reported the lactation yield of local cattle and buffaloes in the midhills to be 300 and 1350 litres respectively.

In the Eastern hills of Nepal, Shrestha et al (1988) collected on-farm data for cattle and buffalo productivity, and reported the average lactation milk yield of native cattle and buffalo to be 445kg and 836kg respectively. In a similar on-farm study in the mid western hills of Nepal, Joshi et al (in press) found the lactation milk yield of local cattle and buffalo to be 344 litres and 810 litres respectively. With respect to lactation length and calving interval, Shrestha et al (1988) reported the lactation length of local cattle and buffaloes to be 292 days and 326 days, whereas Joshi et al (in press) have found it to be 242 and 284 days respectively. Pradhan et al (1989) reported lactation length to be 150 and 200 days for local cattle and buffaloes, which is considerably lower than the values reported by the other workers. The reason for these large variations are difficult to explain.

Calving interval of local cattle and buffaloes has been reported to be 19 months and 20–21 months respectively (LMP, 1991), 18–20 months and 16–20 months respectively (Oil and Morel, 1985), and 15 and 18 months respectively (Joshi et al, in press).

With such variable data, it is difficult to draw valid conclusions regarding the productivity of Nepalese cattle and buffaloes under various environmental and management regimes. It is even more difficult to characterise the productivity of the different types of cattle and buffalo reported within the country. However, it is evident that the average milk yield production per animal is very low for native cattle and buffaloes. This may be the result of poor nutrition, health, or genotype either individually or in combination. The performance of native animals under improved nutritional status has not yet been satisfactorily evaluated. In a study in the Trisuli Watershed Area, Field and Pandey (1969) reported that the feeding of commercially available concentrate to local buffaloes at the rate of 0.5 kg per litre of milk produced, increased the yield of animals by 18% but it was not economic.

Since the highest concentration of cattle and buffaloes is in the midhill region of the country, the greatest pressure from unproductive stock is also on the environment of this area. The situation has become critical with the cattle population, because of legislation banning the slaughter or export of live animals. The composition of the national herd of large ruminants is presented in Table 9.

Although an increase in the population for the hills and mountains is not anticipated, an increase in the total population will drain the limited resources available over the whole country. The rate of population increase for cattle and buffaloes is presented in Table 10.

Table 9: Composition of cattle and buffalo population and estimation of unproductive animals.

* L.U. = 300 kg body weight.
Source: Rajbhandari and Pradhan (1991)

EXISTING MANAGEMENT SYSTEMS

Management of large ruminants in the hills is governed by factors such as cropping intensity, availability and proximity of forest resources, animal species and productive stage, labour availability and animal population per household. Livestock rearing in general, and the rearing of large ruminants in particular, is dependent upon the overall farming system of the area. The farming systems at different altitudes are dependent upon temperature, irrigation and other interrelated factors, and will vary. For the large ruminants, three management systems predominate.

Transhumance system

This system is adopted in the high Himalayan areas where herds of cattle and buffalo migrate from one place to another throughout the year. This system utilises forage resources available from the alpine pastures during the monsoon season, and crop stubble of fallow land during the winter season. During the upward and downward migration, the undergrowth in the forest region is the major forage source.

The large ruminant animals involved in the system are the yak and cattle, but in some areas, buffalo are also included.

Table 10: Livestock population increase assumptions.

Source: LMP (1991)

Sedentary system

Ruminant livestock make daily grazing forays from the village, and return every evening. The main grazing areas during the summer are the scrubland and community grazing land around the village. The sedentary population consists of working oxen, dry buffaloes and a small number of cattle.

This system prevails at the lower altitudes of the midhills (900–1000m asl) and utilizes all the available forage in and around the village. The grazing area is usually degraded, and gully formation and soil erosion are evident. Though, the animals spend half their time grazing, most of the forage resource is based upon crop by-products and tree fodder during winter, and grasses and weeds from crop land during the summer, which are offered during evening and morning hours.

Stall-fed system

This type of animal rearing is found mainly in the low to mid-hills (400–900m asl) and peri-urban areas with milking buffalo, and exotic or crossbred cattle. It is governed both by the availability of community grazing land in the village, and by the steepness of the terrain, which may mean that other classes of livestock are also maintained under stall-feeding conditions. The system prevails in areas of intensive cultivation (three crop sites), where the availability of crop by-products is adequate to maintain the animals in winter. In addition to crop by-products, tree fodder, and grasses and weeds collected from the farm land also constitute an important forage source.

CONSTRAINTS ON THE EXISTING SYSTEMS

Excessive Population Depending on Limited Natural Resources

In the Nepalese hills, a complex mixed farming system is practised with the complete integration of crops, livestock and forest resources. Crop by-products and grasses from cultivated land, and fodder from the forest constitute the main forage source for animals. In return, animals contribute manure and farm power. In principle, this traditional system is ecologically sound and it has functioned well in the past. However, chiefly on account of ever-increasing human and animal populations, a tragic situation is now unfolding, resulting in the destruction of large tracts of land by erosion, land slides and the drying up of water resources. The sequence of events is illustrated below.

It has been estimated that the livestock stocking rate is three and thirteen times more than the carrying capacity in the steppe and open grazing lands of the mid-hills respectively (Rajbhandari and Shah, 1981). The mid-hills and mountains are capable of maintaining only 2.78 million livestock units efficiently, indicating that there is an excess of 2.2 million cattle (53%) and 1.0 million buffalo (50%). It has also been estimated that the total feed available as a percentage of requirement is 53.6%, which under the prevailing conditions will decline to 4.3% by the year 2000.

In a study of the migratory animal rearing systems and their relationship to forest resources in a high altitude village, Metz (1990) found that each household harvested 29mt (on a fresh weight basis) of fodder from the forest each year. The annual per capita use was found to be 4,692kg on a fresh weight basis. He also estimated that the fodder collected and hand fed to animals constituted only 30% of the total feed requirement of the animals, the rest of which was obtained from grazing in the forest and on marginal areas.

Metz (1990) concluded that the intense grazing and browsing of livestock is the most destructive part of this system on the forest, because it destroys the young seedlings and saplings of fodder tree species which are often the dominant trees of the forest. When grazing is combined with repeated lopping of the fodder trees, a definite sequence of forest degradation occurs. The dense, diverse oak and broad-leafed forests are continually deprived of their fodder species as these are progressively cut back to the trunk. When the leaf area becomes small, the tree dies and is cut for fuelwood. Continued grazing prevents the regeneration of favoured fodder species, and the non-palatable species begin to dominate the forest flora.

High Labour Requirement with Minimal Net Return

In a study in a mid-hill village, where animals were raised under sedentary and stall-fed systems, Shrestha and Evans (1984) estimated that the total labour requirement to raise an average of 7.37 animals per household was 80.3, 208.0 and 106.6 days per annum for men, women and children respectively. The time spent on various livestock related activities is presented in Table 11. This indicates that livestock rearing takes up 7.5 hours per day when labour invested by men, women and children is considered together. In the same study, the authors estimated the average annual income from livestock enterprises to be Rs. 1862 and the total expense to be Rs. 1373 at 1984 price values. Manure production was an added advantage.

Table 11: Estimated average annual labour input to livestock.

Source: Shrestha and Evans (1984)

Therefore, although livestock raising is an important and integral part of the farming system in Nepal, and requires considerable labour, the net return from the enterprise is minimal. It can be considered that livestock raising in the hills sustains crop productivity and the livelihood of the farmers, but at the expense of the environment, and unless other alternatives are identified it will continue to do so in the future.

Severe Forage Deficit

The availability of feed and fodder for livestock, especially during winter and early summer, is the major constraint on livestock productivity in all the existing systems. It has been estimated (DFAMS, 1990) that, of the total livestock population in the hills and mountains, 80.7% is made up of large ruminants. The feed requirement and supply situation for the livestock population is presented in Table 12. These data suggest that there is a serious shortage of nutrients for animals. Hopkins (1983) estimated the energy balance for livestock feeding (maintenance only) on different land holdings for the hill districts of Koshi Zone. The data are presented in Table 13.

Table 12: Estimated TDN (total digestible nutrients) requirement and supply

Source: LMP (1991)

From Table 13, it is clear that farmers with small land holdings are in an acute energy deficit situation, and must rely heavily on communal land and forest to make up the deficit. In the hills of Nepal, 62.3% of all households have less than 0.4 hectares of land (Chitrakar, 1990). It is understandable that most of the farming families depend on natural forest to supply energy to their animals.

Table 13: Energy balance on land holdings of different sizes.

Information on the number of animals by species, age and production category is not accurate, and the nutritional requirements of different species, as well as the influence of seasonal and agro-climatic factors, are not well documented. The extent and nutritive value of different types of vegetation, the utilisation of available animal feeds and crop by-products, and the influence of season on feed quality is also poorly recorded. It is obvious that there is an acute shortage of animal feed and animals lose body condition during winter. This deficit is more serious for the large ruminants, and as the number of unproductive stock (especially cattle) is large, the limited feed is shared between the productive and unproductive animals. Therefore, most of the large ruminant population in Nepal is never fed to their basic requirement, leading to a gradual decline in their productivity over generations.

In order to improve livestock productivity, a serious commitment will need to be given to the increase in total feed supply, to improving feed quality, and to overcoming seasonal feed deficiencies.

Diseases and their Effects

A high incidence of disease and parasitism is also an important factor reducing the productivity of the existing management systems, and is further aggravated by poor nutrition. Many major diseases and parasites of livestock are endemic to Nepal, and there are probably many more yet to be diagnosed.

In the hills of Nepal, the animal diseases considered to be of major importance in cattle and buffalo are rinderpest, haemorrhagic septicaemia, foot and mouth, and helminth parasitic diseases. While the incidence of the prevailing diseases and the economic losses caused by them have not yet been fully evaluated, estimated economic losses caused by some of the major diseases of large ruminants in Nepal are presented in Table 14. These estimated losses were calculated in terms of reduced productivity and increased mortality. It is evident that diseases are a major constraint on improving productivity. Though the incidence and frequency of various diseases in different animal production systems have not been studied in depth, it is generally considered that the incidence of infectious diseases is similar under all the existing management systems. However, for parasitic diseases, prevalence was found to vary with altitude and management system (Joshi, 1988).

It is also reported that some diseases viz. Enzootic Bovine Haematuria are limited to grazing animals. A mean rate of prevalence of 0.79% has been reported in the cattle population of the Eastern Nepalese hills (Mahato, 1986). Diseases related to plant poisoning are more frequent in grazing animals, whereas metabolic disorders are more common in stall fed animals.

In general, it can be said that information regarding the prevalence of various animal diseases in different systems, and their effect on production is lacking. What is certain is that they play a major role in reducing productivity in all animal production systems.

Table 14: Estimated loss due to disease (in million Rupees).

Source: Hassel et al (1979)

Poor Productivity Potential of Indigenous Large Ruminants

Milk production of cattle and buffalo in Nepal is very low. Lactation yields of indigenous cattle have been previously presented. The long calving interval and late sexual maturity of these animals make them even more uneconomic. Low productivity, limited marketing opportunity and a dependency on dwindling forage resources has led to an interdependent relationship between crops and livestock. Any development programmes must consider all the interrelated factors, so that extreme commercialisation of one enterprise does not affect the existence of another.

Milk production of indigenous animals under good nutritional conditions has not been extensively studied. The few studies conducted so far have indicated that although milk production of animals on a high nutritional plane increased, it was not viable economically. Field and Pandey (1969) recorded a total increase in lactation yield of 423 litres at a cost of 1350kg of concentrate feed, and concluded that, except for intensively selected milk-producing buffalo, concentrate feeding at the level given was not economic.

In a recent study on the milk production potential of local cows under intensive management, Dhaubhadel et al (in press) reported that animals maintained on ad lib green grass and commercial concentrate at the rate of 2kg per kg milk produced, improved in body condition considerably, but not in milk production. Average daily milk production was only 1.4 litres, with the highest daily yield recorded being 2.6 litres. Assuming that the milk yield from two quarters is generally fed to the calf, the actual yield obtained by the farmer was only 0.7 litres/day. Hence, at present pricing levels a net loss of Rs. 10/animal/day was recorded for concentrate feeding only.

These preliminary findings suggest that the milk production potential of native cattle and buffalo, even under good nutrition is poor. To improve the productivity of the individual animal, genetic improvement of existing stock is a necessity. The slaughter of cattle is prohibited by law in Nepal, leading to a relatively large number of unproductive cattle in the population. This prohibition is not applicable to buffaloes although the slaughter of female animals is theoretically not allowed. However, this ban is not enforced and, in fact, a large proportion of slaughtered buffaloes are old, unproductive cows. The buffalo meat production industry has not been developed in the country and a large number of male calves are slaughtered at birth. It was recorded in villages in the Western hills that the mortality rate in male buffalo calves was 55.5%, whereas it was only 8.7% in female calves. This indicates that a large number of male calves were intentionally killed at birth to save the milk needed for their rearing. Feed and fodder scarcity, and disease are considered by farmers to be the major reasons for not raising male buffalo calves which are not generally used for draught power in the small terraces of the mid-hills. In the Terai, where buffalo bulls are used for ploughing and, in the high Himalayan villages where forest resources are more easily accessible, male buffalo calves are reared and used for draught power and meat respectively.

The deliberate slaughtering of male buffalo calves in the mid-hills has led to the gradual decline in genetic potential within the native population. Calves from high yielding mothers are slaughtered, whereas the calves from poor yielders, which do not let down their milk without suckling, are reared and are eventually used for breeding.

Poor Marketing Structure

Poor marketing structure and the difficulties of transport and communication have hindered the development and growth of specialised industries concerned with livestock production. Farmers are forced to maintain unproductive stock with multiple objectives. Due to a lack of transport facilities, most of the milk produced in the country is not marketed, and only 2% of total annual milk production was collected by the Dairy Development Corporation during 1986/87 (LMP, 1991). The most valuable marketable commodity is churned butter or ghee, which is non-perishable and can be transported long distances.

The inaccessibility of large areas of the country, means that intensive commercial production is centered around peri-urban areas, and in the rural areas, a large number of multipurpose animals are reared with little direct economic benefit accuring to the farmer.

STRATEGY FOR SUSTAINABLE DEVELOPMENT OF LARGE RUMINANTS

Sustainable development of large ruminants in the hills of Nepal will depend to a large extent upon improving the productivity of the individual livestock unit, and on the reduction in the number of unproductive animals. In this way, the severe pressure on natural forest resources will be minimised, and they could be utilised in a sustained manner for productive animals.

As previously discussed, the genetic potential of native cattle and buffalo for milk production, even under good nutritional status, is very limited. Therefore, genetic improvement of the native stock is an important requirement if increases in productivity are sought. However, Nepalese hill agriculture will continue to depend upon animal power and manure for a considerable time to come. Therefore, any genetic improvement programme should not ignore these essential characteristics, so that crossbred progeny of native stock remain equally acceptable to local requirements.

Improvement in health and in particular, the development of disease prevention strategies (infectious and parasitic) should be adopted rationally, based upon the epidemiological pattern of the disease for a given environment or region.

The impact of an organised marketing system on livestock improvement programmes is well known. However, state regulations should ensure the economic interests of farmers are not exploited by businessmen.

Thus, the theme of sustainable development of large ruminants in Nepal has to be based upon the following guiding principles:

• Improvement in animal productivity without offsetting the animal's traditional role.

• Reduction in dependency on natural forest resources for livestock rearing, and quantitative and qualitative improvement of forage production from farm and marginal land.

• Development of management, feeding and disease control strategies.

Productivity Improvement per Animal Unit

Selection within the indigenous population and the introduction of suitable exotic breeds for upgrading the existing large ruminant population are the two possible breeding options available for improving productivity. Since the genetic potential of indigenous stock even under adequate nutritional management was found to be very limited (Field and Pandey, 1969; Dhaubhadel et al, in press), selection within the indigenous population, though recommended by Rajbhandari and Pradhan (1991) has practical limitations, such as the non-existence of performance records, the small herd size of individual farmers, and the ultimate problem of animal disposal (especially for cattle). In the Nepalese context, the most appropriate method of improving productivity in large ruminants will be the introduction of suitable exotic blood into the native population, so that the adaptive traits of native stock could be blended with the productive traits of the exotic stock.

Some efforts have already been made with regard to the breed development of native cattle and buffalo, although a definitive policy regarding the maintenance of a suitable exotic breed at desired blood levels is lacking at present. The Jersey and the Murrah breeds have been most widely used so far in cattle and buffalo breed development programmes, and although it was probably an historic or geopolitical reason that these breeds were selected at the beginning, studies conducted later have supported the suitability of these breeds. The result of a study conducted at Khumaltar Livestock Development Farm (KLDF, 1979) showed the superiority of the Jersey cross over other crossbreeds (Table 15).

Table 15: Average milk production (305 days) of 50% cross and local Nepali cattle.

Source: KLDF (1979)

Government policy is still to recommend the Jersey as the exotic breed for small farm milk enterprises (Shrestha, 1989). It has been used for the upgrading of native tropical breeds worldwide and, indeed the small bodysize with lower maintenance requirement, early sexual maturity, and high milk butterfat content make it the logical choice for cross-breeding in areas of forage deficit (Clinch, 1991).

In most of the hills and midhills of the country the performance of Jersey crossbred cattle under farmers' conditions has been very encouraging (Table 16).

Table 16: Comparative performance of local and Jersey crossbred cattle under farmers' management.

Table 16 indicates that Jersey cross cattle are well suited to the hill environment of Nepal. The belief of some farmers that the lack of a hump could adversely effect the crossbred's draught potential is unfounded, because draught performance has been found to be superior to that of native oxen, and the absence of a hump does not affect the position of the yoke or the way in which the oxen pull when ploughing (Pearson, 1991). Indeed, farmers actually showed a slight preference for Jersey crossbred stock in the surveyed area of Koshi Hills (Gatenby et al, 1990).

A three-tiered breeding system has been suggested for Nepal by Yazman (1990), in which the Jersey crossbred would be developed as a milk/draught power breed for the mid-hills, whereas a crossbreed based on tropical dairy breeds such as the Sahiwal or Red Sindhi may be more appropriate for the Terai and lower hills. Finally, a breed based on temperate dairy breeds such as the Holstein/Friesian may find a niche in peri-urban areas.

Similarly, buffalo breed upgrading programmes have been based upon crossbreeding between the riverine Murrah breed and the native swamp buffaloes. Milk production of the F₁ progeny has been recorded in various regions of the country. Performance monitoring results obtained so far indicate that Murrah × swamp crossbred buffaloes are more productive than native swamp buffaloes even under traditional management, but the results are not as dramatic as with crossbred cattle. In the Eastern Hills, the crossbred buffaloes were found to produce 210kg more milk in a lactation than the native swamp buffaloes (Shrestha et al, 1988), whereas the increment recorded in the Western Hills was 397kg per lactation (Joshi et al, in press). The comparative performances of native swamp buffaloes and Murrah crossbreds are presented in Table 17.

This table clearly indicates that milk production can be increased by the crossbreeding of native swamp buffaloes with the riverine Murrah breed. However, this upgrading programme might not be as successful at higher altitudes where age at first calving and calving interval were found to be greatly increased compared to lower altitudes (Shah, 1982).

Crossbreeding Programmes

It is clear that the potential for milk production can be greatly increased by crossbreeding. Rajbhandari and Pradhan (1991) estimated that milking animals constitute only 11.6% of all cattle, and 27.5% of all buffalo. By upgrading the lactating female populations of cattle and buffalo by 1 %, milk production from the two species will increase by 0.53% and 0.16% respectively. They stated that to have a significant impact, an upgrading programme should cover at least 10% of total lactating females, i.e. approximately 67,000 cattle and 68,000 buffaloes, which would produce 11,700 and 8,520mt of additional milk for the country, on the assumption that the average milk yield for upgraded cattle and buffalo was 500kg and 900kg respectively.

Since extensive performance evaluation of different breeds and crossbreds has not yet been performed in the hill environment, it is dangerous to extrapolate results from various breeding programmes elsewhere. However, the results obtained from the Jersey and Murrah breeds are reasonably satisfactory, and have provided enough increased potential which could be exploited even further if adequate management (feeding and health care) systems could be adopted. This is even more important where the crossbreds have been found to perform adequately in the other important agricultural roles that farmers have for their stock.

Table 17: Comparative performance of hill swamp buffaloes and Murrah x swamp buffaloes under farmers' traditional management.

Constraints upon Breeding Programmes

To carry out breeding programmes on a wider scale, considerable infrastructural and organisational development is required, so that the programmes achieve the desired objectives. Because of the problem of transport and communication, the national A.I. service (mostly in cattle) is confined to 31 mainly low altitude districts, out of a total of 75 districts in Nepal. Since its inception in 1960, only 110,700 animals have been inseminated artificially, and the national target for the year 1989/90 was only 17,000 animals. The A.I. programme only covers a limited population of cattle and buffalo, and does not have a serious impact on national production (Shrestha, 1989). Problems related with A.I. at present are irregular and unreliable supplies of liquid nitrogen and frozen semen, and the lack of a national policy on the use of a particular breed. Furthermore, the poor conception rates recorded in some areas has reduced the confidence of farmers in the programme.

Improvement not only of the technical aspects of A.I., but also of the management of the programme are required. A seasonal A.I. programme is possible in the hills where cattle and buffalo show a definite breeding pattern (Table 18).

From Table 18, it can be seen that around 80% of cows, and 80% of buffaloes come into heat during the period August to November. This could be exploited for seasonal A.I. programmes. The experience in Lumle's Extension Area shows that with an effective extension programme, farmers can be convinced of new technologies and ideas, and participate enthusiastically. This was evidenced by the farmers' willingness to allow crossbred calves to suckle all milk produced by the dam, because they realised that the small amount of milk produced by the local cow was insufficient to rear the calves properly.

Table 18: Breeding pattern of local cattle and buffaloes in the hills of western Nepal.

Source: LAC (1985)

Therefore, although A.I. is a desirable method of upgrading local stock, it is not considered a viable option for most of the country. Breeding programmes will therefore have to depend upon natural mating with exotic sires. Natural mating systems require a regular supply of good sires, but bulls provided by the Government into the villages are regarded as “no man's” property, and are not well cared for. High rates of mortality are encountered and because of the topography and terrain of the hills, a bull may mate with only 20–30 cows in a season for cattle and 70 cows for buffalo (LAC, 1983). This indicates that a very large number of bulls need to be stationed in villages in order to cover a significant proportion of the animal population.

The recent concept of a group or committee approach adopted by the government extension services and NGO's, where farmers with a common interest for improvement, form an interest group, does solve this problem to some extent. The experience of the Livestock Extension Programme of Lumle Agricultural Centre shows that when communities are united and have a common goal, the launching of livestock development activities is not a daunting problem. The only support required is the regular follow-up of the programme, so that minor problems can be resolved in good time. Support at the inception of the committee, in terms of the provision of a breeding bull and the training of one enthusiastic, mature person in emergency animal health care, is sufficient to operate the programme smoothly.

The committee (group) itself forms rules and regulations regarding the management of their property, and levy charges on various livestock-related activities so that funds can be raised for bull rearing. In many areas, community land has been allocated and planted for forage production to serve as winter feed for the breeding bulls. Village Livestock Development Committees can act as a launching pad for other Livestock Development Programmes in the villages, and if organised properly are very successful in achieving results. The success rate of village committees in the Lumle area is nearly 90% (LAC, 1991).

Regular follow-up, the exchange of genetic material, and the monitoring of the performance of the programmes is essential, so that improved opportunities are identified and can be exploited.

Improvement in Feed Supply

Improvement in genotype in favour of productive traits will increase the potential of an animal to produce. Whilst this may lead to an increase under the prevailing management conditions, the full potential of the improvement can only be realized by a concurrent improvement in nutrition and management (Clinch, 1991). To meet the requirements of productive animals, it will not only be necessary to increase the quantity of the forage available but also to improve the quality of the forage. To make the system sustainable, dependency upon natural forests will have to be minimised and community pastures managed properly. The deficiency of TDN for the animal population in the mid-hill region has already been discussed and an acute shortage is evident. Interventions for forage improvement and by-product utilisation are discussed in Chapter 7.

Improvement in Management, Feeding and Disease Control Strategies

Among the existing management systems of large ruminants in Nepal, transhumance and sedentary systems of animal management are a cause of concern from the environmental point of view. Stallfed systems if based entirely on forest resources, as in high hill villages, will not be sustainable in the long run. Overgrazing of animals has often been blamed for the soil erosion problems the country is facing at present. The annual soil erosion rate for forest cover land, scrubland, grazing land and terraced land is 8 mt/ha, 15 mt/ha, 30 mt/ha and 10 mt/hectare respectively (Fleming, 1978). Similarly, the annual soil erosion rate is estimated to be 9.4 mt/ha for pasture land and 34.7 mt/ha for overgrazed grassland (Mulder, 1978 cited by Shrestha, 1984). Although these values cannot be generalised for the situation throughout the country, they do indicate that overgrazing, which is a major consequence of the sedentary systems predominant in the mid-hill regions of the country, is a significant cause of soil erosion. As grazing areas are community-owned, an attitude of “no man's property, hence no one's responsibility” prevails.

The maintenance of cropland as fallow during the winter is one of the reasons that farmers continue to adopt a sedentary or transhumance system, as they consider it an opportunity to utilise the crop stubble and at the same time manure their fields in-situ. This system of animal management is now regarded as one of the major constraints upon environmental protection programmes. In the high Himalayan areas, farmers have traditionally practised transhumance systems for generations, rearing large herds of cattle and buffaloes, which are a problem, when initiating tree or grass planting programmes on fallow land. It is the migratory herds of large ruminants which are accelerating deforestation, together with the human quest for fuelwood and fodder (Karki, 1984).

Rajbhandari and Pradhan (1991) have calculated that at present, the high Himal and high mountain regions of the country are in a positive animal feed balance situation, perhaps the result of utilisation of alpine pastures and forest undercover in the adjoining Himalayan forests. However, it seems that the system cannot be maintained in the long run, because other forest products required by an increasing population will substantially add to the destruction caused by the high animal populations.

The most critical area in terms of animal feed deficit is the mid-hills, where a shortage of 40.5% dry matter, 55% total digestible nutrients and 68% crude protein has been reported (Rajbhandari and Pradhan, 1991). These authors stated that this acute shortage of animal forage drives animals from this zone up to the high mountain areas. In the mid-hill region, community grazing lands are usually overgrazed to the extent that virtually no forage can be obtained from them. This has not only led to soil erosion, but also to a loss of about 50 % of total manure production. Faeces deposited by animals whilst grazing are washed away by rain or dry out, and so are lost.

Forest resources in most of the mid-hill areas are in short supply and so the only available option for developing forage resources will be the proper utilisation of these community grazing lands. Hopkins (1983) stated that by protecting naturally-occurring grassland, and thus preventing overgrazing, productivity of pasture could be increased from one to six mt/ha/annum.

Considering all these factors involved in livestock management, it has become apparent that the sedentary transhumance system of raising large livestock in the hills is no longer sustainable, since it leads to serious environmental degradation. The only acceptable system for the future will be the stall feeding system for which forage resources are self produced, conserved and utilised properly.

The effect on stall-feeding management of animal productivity can be illustrated in the following case study:

Milk production of Jersey-cross cattle under farmers' traditional management in the high hill region has been reported to be 1100 kg/lactation or about 3.5 kg/day (Joshi et al, in press). Jersey-cross cattle were supplied to farmers and their productivity under stall-fed management was recorded. The findings to date are summarized in Table 19.

This study is still continuing, but the results suggest that the milk yield of Jersey-cross cattle under stall-fed conditions are substantially higher than under a sedentary or transhumance system, without any extra inputs. The productivity of around six litres per day can be obtained under farmers' conditions and with some improvement in feeding there is scope for further improvement. For the Nepalese hills this can be regarded as low input: moderate output sustainable cattle development. The benefits of the stall-feeding system are outlined below.

Table 19: Milk production of Jersey-cross cattle under farmers' stall-fed management.

(July-December 1991 records only)

* Figures in parentheses indicate days for which milk yield was recorded.
** Concentrate “Kundo” is prepared from maize flour, rice bran, kitchen waste and salt.

• Only productive animals are reared and high productivity per animal unit can be maintained, leading to a higher level of animal production and economic return.

• Less effect on the environment, because the natural resources are allowed to regenerate.

• Large amounts of manure production as the potential 50% loss during grazing is avoided. With the incorporation of bedding material the amount of compost can be substantially increased. Urine can also be utilised as a valuable source of fertilizer, where collection facilities exist.

• Better care and management for animals can be provided, with reduced incidence of parasitic diseases.

• Labour requirement of women (grass collection) can be minimised so reducing their work load.

• Organisation of the marketing system could be better developed.

This change from traditional practices is only possible when there is an overall understanding and co-operation on the part of the whole community regarding the advantages of the proposed system. This can only be achieved through a community or group approach, and requires considerable infrastructural, organisational and financial support from outside agencies. The advantages of the change must be quantifiable and visible to the community, so that the newly adopted change remains in place. Some of the disadvantages of the stall-feeding system are described below.

• Forage resources available at present are insufficient for stall feeding. Programmes to increase annual forage production must be provided concurrently.

• Marketing systems for animal products are lacking at present and should be developed so that farmers become market-oriented.

• Animal housing requires improvement which implies a financial commitment from the farmers.

• Collective ownership and the utilization of resources developed need to be formalised.

• Oestrus detection may become a problem in tethered animals, leading to longer calving intervals.

• Manure transport to the field may increase labour requirement. However, utilisation of oxen for pack use might solve this problem.

Diseases associated with intensive management may be more prevalent, but in the stall fed system, disease control programmes, especially vaccination against infectious diseases and strategic drenching against parasitic diseases would be possible. Those combined with the improved nutrition from forage improvement programmes, allied to the better genetic potential of the animals, could greatly increase the productivity of the animals in the country and increase the income of the individual farming family.

In conclusion it can be said that, considering the advantages and disadvantages, perhaps the future of livestock rearing in Nepal, depends upon a change in the management system from transhumance/sedentary to a stall-fed system, which would be more sustainable in the fragile environment of the hills.

REFERENCES

Balogun, P. (1989). A discussion of crop cut estimates of rice, maize and millet yields in the ECA. LAC Technical Paper 6. Lumle Agricultural Centre, Kaski, Nepal.

Borowski, A.M. and Liebhardt, W.C. (1983). Biological aspects of composting. In: Proceedings of Workshop on Resource Efficient Farming Methods for Tanzania, held at Morogoro, May 16– 20, 1983, pp 47–57.

Chitrakar, P.L. (1990). Planning, Agriculture and Farmers. Strategy for Nepal. Mrs Ganesh Devi Chitrakar, Kathmandu, Nepal pp 21–36.

Clinch, N.J.L. (1991). Review of breeding strategies for large ruminants in the Extension Command Area of Lumle Regional Agricultural Research Centre. LAC Review Paper 02/91. Lumle Agricultural Centre, Kaski, Nepal.

Cockrill W.R. (1974). The Husbandry and Health of the Domestic Buffalo. FAO, Rome, Italy.

Colin, S. and Falk, A. (1979). A survey on the socioeconomy of the Koshi Hills Area - guidelines for planning an integrated rural development programme; KHARDEP Report 3. Koshi Hills Area Rural Development Programme, Dhankuta, Nepal.

DCFTS (1988). Department of Customs and Foreign Trade Statistics, 1987/88. His Majesty's Government of Nepal, Kathmandu, Nepal.

DFAMS (1988). Department of Food and Agricultural Marketing Services. His Majesty's Government of Nepal, Kathmandu, Nepal.

DFAMS (1990). Department of Food and Agricultural Marketing Services. His Majesty's Government of Nepal, Kathmandu, Nepal.

Dhaubhadel, T.S., Joshi, B.R., Clinch, N.J.L. and Kadariya, R.K. (in press). Milk production of local cows under intensive farm management. Lumle Agricultural Centre, Kaski, Nepal.

Epstein, H. (1977). Domestic Animals of Nepal. Holmes and Meier Publishers, Inc, New York/London.

FAO (1977). The Water Buffalo. FAO Animal Production and Health Series No. 4, pp 171–173.

FAO (1988). Animal Production Yearbook, 1988. FAO, Rome, Italy.

Field, D.I. and Pandey, K.P. (1969). Livestock on-farm trials. In: Surveys and Demonstration for the Management and Development of Trisuli Watershed. Pasture, Fodder and Livestock Development. Project Report No. 15, UNDP/FAO/HMG, Nepal, pp 44–46.

Fleming, W.M. (1978). Phewa Tal Watershed Management Project. Livestock and environmental issues. (cited by Shrestha, 1984).

Gatenby, R.M., Pearson, R.A. and Limbu, T.B. (1990). A survey of local and Jersey crossbred draught oxen in the hills of East Nepal. PAC Technical Paper 98. Pakhribas Agricultural Centre, Dhankuta, Nepal.

Gatenby, R.M., Shrestha, N.P. and Joshi Y.R. (1990). The benefits of animal-keeping in the Koshi Hills. PAC Technical Paper 137. Pakhribas Agricultural Centre, Dhankuta, Nepal.

Hassel et al (1979). Impact of animal health services on the National Economy. Bull. of Vet. Sci. and Anim. Husb., Nepal 12, 60–64. (cited by Shrestha, 1984).

Henk, P.D. (1988). Productivity of local, crossbred and Murrah buffaloes - a field research in the hills of Nepal. Department of Tropical Animal Production, National Agricultural University, Wageningen, The Netherlands.

Heuch, J.H.R. (1986). The quantity of compost applied to farmers field and its relevence to forest management in the mid hills of Nepal. LAC Technical Paper 86/11. Lumle Agricultural Centre, Kaski, Nepal.

Hopkins, N.C.G. (1983). The problem of fodder and its alleviation in the Eastern Hills. Bull. Vet. Sci. and Anim. Husb., Nepal 12, 17–25.

ICAR (1986). Handbook of Agriculture. ICAR, New Delhi, India, p 105.

Jansen, W. (1990). The use of Jersey bulls for increasing milk production in the Koshi Hills. In: Livestock in the Hills of Nepal, p 128. Pakhribas Agricultural Centre, Dhankuta, Nepal.

Joshi, B.R. (1988). Prevalence of fascioliasis (liverfluke) in cattle and buffaloes in the mid-western hills of Nepal. Journal of the Institute of Agriculture and Animal Science 91, 111–114.

Joshi, B.R., Karki, N.P.S., Kadariya, R.K. and Gurung, D.B. (in press). Milk production of local and 50% Murrah crossbred buffalo under farmers' traditional management in the western hills of Nepal. Lumle Agricultural Centre, Kaski, Nepal.

Joshi, B.R., Karki, N.P.S., Kadariya, R.K. and Gurung, D.B. (in press). Milk production of local and 50% Jersey crossbred cattle under farmers' traditional management in the western Hills of Nepal. Lumle Agricultural Centre, Kaski, Nepal.

Joshi, D.D. (1982). Yak and Chauri Husbandry in Nepal. HMG Press, Kathmandu, Nepal. Joshi, K.D., Sthapit, B.R., Pound, B. and Gurung, J. (1990). Research thrust: a multidisciplinary research approach to generate sustainable technologies. LAC Technical Paper 26. Lumle Agricultural Centre, Kaski, Nepal.

Karki, N.P.S. (1984). The experience of Lumle Agricultural Centre in the role of livestock in hill agricultural practices and its improvement. Paper presented at the Livestock Workshop, Lumle Agricultural Centre, 17–19 January, 1984.

KLDF (1979). Khumaltar Livestock Development Farm, Annual Report of Progress Department of Livestock Development, Kathmandu, Nepal.

Kunwar, J.S. and Dutta, N.P. (1970). Soil Science in Agriculture Yearbook. Indian Council of Agricultural Research, New Delhi, India, pp 201–283. (cited by Balogun, 1989)

LAC (1983). Lumle Agricultural Centre, Annual Report 1982/83. P.O. Box 1, Pokhara, Gandaki, Nepal.

LAC (1985). Lumle Agricultural Centre, Annual Report 1984/85. P.O. Box 1, Pokhara, Gandaki, Nepal.

LAC (1991). Lumle Agricultural Centre, Annual Report 1990/91. P.O. Box 1, Pokhara, Gandaki, Nepal.

LMP (1991). Livestock Master Plan for Nepal. His Majesty's Government of Nepal/Asian Development Bank/ANZDEC.

LRMP (1986). Land Resource Mapping Project. Land Utilisation Report, July, 1985.

Mahato, S.N. (1986). Preliminary observations on Enzootic Bovine Haematuria in the hills of Eastern Nepal. Bull. Vet. Sci. and Anim. Husb., Nepal 14, 36–41.

Marsh, D.K.V. (1985). Livestock production in the Koshi Hills. KHARDEP Evaluation Section, Koshi Hills Area Rural Development Programme Report, August, 1985, p 18.

Metz, J.J. (1990). Conservation practices at an upper elevation village of West Nepal. Mountain Research and Development 10, 7–15.

Mulder, R.P. (1978). Erosion plot measurement in Phewa Tal catchment. (cited by Shrestha, 1984).

Nabarro, D., Cassels, C. and Paut, M. (1989). Coping strategies of households in the hills of Nepal: Can Development initiatives help? IDS Bulletin 20, 68–74.

Nepal Rastra Bank (1988). National Statistics, 1987/88.

Oli, K.P. (1984). Utilisation of draught animal power in the hill agricultural system of Nepal. Paper presented at the Livestock Workshop, Lumle Agricultural Centre, 17–19 January, 1984.

Oli K.P. (1985) Draught animals in the hill agricultural system. In: Livestock in the Hills of Nepal. (Eds. A.M. Morel and K.P. Oli). Pakhribas Agricultural Centre, Dhankuta, Nepal.

Oli, K.P. (1987). On-farm research methodologies for livestock development at Pakhribas Agricultural Centre. PAC Working Paper 03/87. Pakhribas Agricultural Centre, Dhankuta, Nepal.

Oli, K.P. and Morel, A.M. (1985). Livestock production in the Eastern Hills of Nepal. In: Livestock in the Hills of Nepal. (Eds: A.M. Morel and K.P. Oli). Pakhribas Agricultural Centre, Dhankuta, Nepal.

Pearson, R.A. (1990). A comparision of the work performance of Jersey crossbred and local oxen in the Koshi Hills of East Nepal. PAC Technical Paper 130. Pakhribas Agricultural Centre, Dhankuta, Nepal.

Pearson, R.A. (1991). A comparision of Jersey crossbred and local oxen as draught animals in the Eastern Hills of Nepal. Australian Journal of Animal Science 4, 31–40.

Pradhan, P.M., Lohani, M.N. and Nembang, L.B. (1989). Study of dairy production in Nepal. Promotion of Livestock Breeding Project, Hariharbhawan, Kathmandu, Nepal.

Rajbhandari, H.B. and Pradhan, S.L. (1991). Livestock development and pasture management. Background papers to the National Conservation Strategy for Nepal. The World Conservation Union (IUCN).

Rajbhandari, H.B. and Shah, S.G. (1981). Trends and Projections of Livestock production in the Hills of Nepal. Seminar on Nepal's Experience in Hill Agricultural Development, 30 March - 3 April, 1981.

Shah, S.G. (1982). Buffalo production in Nepal. Bull. Vet. Sci. and Anim. Husb., Nepal 10/11, 41–48.

Shrestha, T.P. (1984). Livestock and Environmental Issue. Paper presented at the Livestock Workshop, held at Lumle Agricultural Centre, 17–19 January, 1984.

Shrestha, S.L. (1989). Research needs in cattle and buffalo breeding. In: Proceedings of the Workshop on Research Needs in Livestock Production and Animal Health in Nepal, held in Kathmandu, Nepal, January 1–7, 1989. (Eds. J.A. Yazman and K.P. Oli) pp 9–21.

Shrestha, R.L.J. and Evans, D.B. (1984). The private profitability of livestock in a Nepalese hill farming community. Agricultural Administration 16, 145–158.

Shrestha, P.K., Gurung, H.B. and Amatya, L.K. (1991). Household baseline study of Kewrebhanjyang off-station research site. LAC Technical Paper 11/91. Lumle Agricultural Centre, Kaski, Nepal.

Shrestha, N.P., Oli, K.P. and Gatenby, R.M. (1988). Milk production of local and Murrah crossbred buffaloes and local and Jersey crossbred cows in the hills of Eastern Nepal. PAC Technical Paper 97, Pakhribas Agricultural Centre, Dhankuta, Nepal.

Shrestha, N.P. and Sherchand, L. (1988). Role of livestock in Nepalese farming systems. Paper presented at VI World Conference of Animal Production, Helsinki, Finland from 27 June–1 July, 1988.

Sthapit, B.R., Balogun, P. and Pound, B. (1989). Development of biofertilizer technology in the hills of Nepal: the experience of Lumle Agricultural Centre. LAC Technical Paper 22. Lumle Agricultural Centre, Kaski, Nepal.

Yazman, J.A. (1990). Agricultural Research and Production Project, End of Tour Report. ARPP Project, Kathmandu, Nepal. USAID Project No. 367–0149.