IMPROVEMENT AND CONSERVATION OF BUFFALO GENETIC RESOURCES IN ASIA

S. Sivarajasingam 1/

1. INTRODUCTION

The contribution of buffalo to the Asian agrarian economy is considerable by way of milk, meat and draught power production and as a source of security that requires minimum inputs. The domesticated buffaloes in Asia, representing 98 percent of the world buffalo population, are broadly grouped as the river and swamp types. The former which constitute approximately 69 percent are found predominantly in the Indian subcontinent. In India they supply 59.3 percent of the total milk produced.

The multipurpose swamp buffalo (30 percent) predominates in other parts of Asia especially China, Indonesia, Philippines, Thailand and Vietnam providing draught power and meat in rice growing areas and milk in other regions. The Mediterranean type predominant in west Asia, represent about 1 percent. The buffalo distribution in Asia is given in Table 1 by country. It is sufficient to mention here that the wild buffalo such as Anoa of Celebes, Tamarao of Mindoro and the Ami or Indian wild buffalo do exist and could provide potential genetic resources for further investigation.

Table 1 BUFFALO POPULATION IN ASIA BY COUNTRY, 1984

Source: FAO

The trends in the growth of human and buffalo populations in Asia are given in Figure 1. There is a steady increase in the human population while that of the buffalo is not sustained at a similar pace. Buffaloes produce 45 percent and 31 percent of all milk and meat produced in 1984 by large ruminants in Asia, respectively (FAO statistics).

Figure 1. Trends in buffalo and human populations in Asia (Adapted from Mahadevan, 1983, Proc. Preconf. Syrap. of the 5th World Conf. on Anim. Prod., Tsukuba, Japan and FAO, Production Yearbook 1970-1983).

Buffaloes have been domesticated over several centuries. However, they have been subjected to genetic manipulation such as selection and crossbreeding only during recent decades. By virtue of the fact that it is Often conSidered a neglected species, much of its genetic variation may not have been lost except through natural selection in the domestic environment. However, two areas of concern need to be mentioned. Firstly, a complete documentation and evaluation of the variability and characterization of the various breeds and strains are lacking. Secondly, the animals, the swamp type in particular, are being displaced from their traditional ecosystems due to changing farming practices, agricultural intensification and inadequate allocation of multiplication facilities. These problems are worsened by their low reproductive rates.

This paper is mainly concerned with the potential of the buffalo germplasm which is the basis for improvement, effective methods of conservation and utilization and constraints of buffalo rearing in Asia. This will be discussed under the following headings:

2. Buffalo Performance Characteristics

2.1 Milk yield and length of lactation

2.2 Weight characteristics

2.3 Size characteristics

2.4 Carcass characteristics

2.5 Breeding efficiency

2.6 Draught characteristics

3. Genetic Improvement of Buffaloes

3.1 Selection for size

3.2 Breeding and selection for milk

3.3 Breeding and selection for beef

3.4 Breeding and selection for draught

3.5 Multipurpose breeding strategies

4. Crossbreeding River and Swamp Buffaloes

5. Conservation, Improvement and Utilization

5.1 Genetic conservation - why?

5.2 Genetic conservation - how?

5.3 Improvement within conservation

6. Conclusion

2. BUFFALO PERFORMANCE CHARACTERISTICS

The potentials for improving the river and swamp buffaloes depend on the existing genetic variability within and between breeds, standards of health, feeding and management and infrastructure for recording of production data. A considerable amount of evaluation or documentation studies or both have been reported on buffaloes under varying environments. A brief review of the various production traits is given below only to show overall characteristics of the buffalo breeds.

2.1 Milk Yield and Length of Lactation

Milk is an important source of animal protein (including essential amino acids), vitamins and minerals. Efforts to improve buffaloes and cattle have been undertaken in many developing countries where the rural poor are largely dependent on livestock. A summary of yields of buffalo breeds is presented in Table 2.

The river buffalo, extensively used in the Indian subcontinent for milk production, has a production average between 1181 kg to 1934 kg with lactation lengths ranging between 283 and 313 days. The most widespread breed is of the Murrah type which has also been exported to several southeast and east Asian countries for crossbreeding with swamp buffalo. The latter breed, rarely used for milk production, produces less than 800 kg per lactation of about 250 to 330 days. Work on persistency and milk let-down in milch buffaloes is limited. It is a usual practise to allow calf suckling for about 30 to 40 seconds before milking to initiate milk let-down. Whether this is really necessary needs further investigation. Table 3 shows overall averages for buffalo milk constituents. Fat percent ranged between 7 to 10 with a mean of 7.5 amongst dairy breeds. Values for swamp buffalo were within a similar range but only a few samples were available. These values for buffalo are more than double that of cattle for which the mean is 3.7 percent. Protein percent is also higher than for cattle where mean is 3.5. The calorie value of buffalo milk is 31.5 percent higher than that of Bos taurus cows such as Friesian and Guernsey (FAO, 1959).

Table 2 LACTATION MILK YIELD AND LENGTH

Table 3 MILK CONSTITUENTS OF BUFFALOES

2.2 Weight Characteristics

Body weights at various ages of adult buffaloes are given in Table 4. Birth weight and subsequent weights are higher than indigenous cattle in these areas of Asia. Mature buffaloes over three years of age weigh between 450 and 800 kg. A number of comparisons have been made on weight gains between buffaloes and cattle, from which no definite conclusions could be made bearing in mind the low quality of inputs we need to consider reflecting the actual conditions of the smallholder farms. It is interesting to note here the trials of Shute (1966) in Trinidad where the daily gain of buffaloes was 0.21 kg compared to zero values for Jamaica Red cattle and Brahman cattle under poor pastures. The gains increased to 0.62 kg for buffaloes and 0.50 and 0.30 kg for the other cattle breeds respectively on moderate pastures.

Table 4 BODY WEIGHTS AT VARIOUS AGES (KG)

2.3 Size Characteristics

The milk buffaloes of India are strikingly larger than their swamp counterparts in east and southeast Asia where they are used for draught purposes as shown by the three body measurements in Table 5.

Table 5 BODY MEASURMENTS

2.4 Carcass Characteristics

In Table 6 some characteristics of buffalo carcass are given. Most of the work has been concentrated on the swamp type. Dressing percentage in most reports is below 50 percent whereas in cattle it is usually a little above 50 percent. However, the Murrah buffalo proved to be superior in many carcass traits (Ognjaovic, 1974) including a dressing percentage of 54.7 percent. There are considerable differences between the swamp buffalo reports due to the wide differences between reports, in the methods of characterization of the various cuts, variation in feeding, age and sex of the animals and regional genetic differences within the swamp buffalo.

Table 6 CARCASS CHARACTERISTICS

M = male, F = female, C = castrates

Ages range between 2-5 years.

A number of studies have compared carcass characteristics between buffaloes and cattle (Kissir et al.', 1969; Ognjanovic, 1974 and Charles and Johnson, 1972 and Charles, l982) and showed only marginal differences in quantity and quality indicating that the buffalo has a potential role to play in the beef industry of Asia.

2.5 Breeding Efficiency

A major concern in buffalo production is its low reproductive efficiency. The oestrous cycle in buffalo is similar to that of cattle although their external manifestation is not as strongly expressed as in cattle. Peak luteal levels of plasma progesterone are lower (1 to 2.5 ng/ml) and occur later in the cycle (Kamonpatana et al., 1979 and Jainudeen et al., 1982).

Overall reproductive performance compiled by a review of references is given in Table 7. Age at puberty amongst swamp buffaloes was reported to be at 2.8 years and first mating usually after the third year. However, these values largely depend on management factors. Among Murrah buffaloes, first signs of heat were also observed at the age of 2.8 years (Bhattacharya, 1954). Late first calving age and long calving intervals are common. Longer calving intervals are frequent especially among the swamp types used for draught purposes because they are rarely exposed to bulls and pregnancy when having a suckling calf running along is considered a nuisance in the field. Oestrous cycle varies between 20 and 28 days with some variation in the oestrous duration. Postpartum oestrus and bull fertility are also largely affected by seasonal variations.

2.6 Draught Characteristics

The draught power of the swamp buffalo has been reported in various countries in southeast Asia and China Table 8). They are often used in rice cultivation for ploughing. On the average they plough between 0.025 to 0.032 hectares of padi land per hour. Number of working days vary between two months and five months depending on the type of agricultural activity. Buffalo have been reported to be far superior to contemporary indigenous cattle in Taiwan (Ma, 1980). Buffalo can plough an area almost three times that covered by Yellow cattle per day. They also outlive their cattle counterparts by having a working life of 10 to 15 years compared to 6 to 12 years for cattle.

The primary function of the swamp buffalo as a beast of burden has been greatly reduced due to mechanization and the introduction of double cropping in Malaysia, China, Taiwan and Thailand. However, they are salvaged in areas where fragmentation of farms has led to small units as in Thailand and the buffalo remains a significant component. More recently, swamp buffalo are being used to haul bunches of fruit in oil palm plantations in Malaysia. They are superior to mini-tractors in that they could reach all points of collection even on difficult terrain.

3. GENETIC IMPROVEMENT OF BUFFALOES

More than 70 percent of buffaloes are nondescript and are in the hands of villagers. Genetic improvement could only be realized if there is a simultaneous alleviation of feeding and management standards including proper recording and Al or a superior bull distribution network on the ground. Objectives, in line with the envisaged production system, should also be well defined. The buffalo indigenous to China and southeast Asia, covering a wide range of ecosystems, is generally grouped together as the swamp type although distinct types are recognized. The swamp buffalo has a multipurpose role and is usually confined to traditional farming systems. The relative priority of each role varies from region to region. It is important to identify those types of buffaloes that are more efficient in draught, beef production or adaptability so that breeding goals can be more precisely defined.

Recently in 1984 ACIAR organized a workshop on the evaluation of large ruminants for the tropics in Rockhampton (ACIAR, 1984). It was obvious, also in the case of the present review, that the information available was scanty and 'disconnected' in the sense that no breed could be meaningfully characterized and its potential compared with other breeds of buffaloes. It is useful to mention here that some of the workshop's

Table 7 REPRODUCTIVE CHARACTERISTICS

recommendations included identification of genetically different populations of the swamp buffalo, an increase in their draught efficiency, and numbers especially in Thailand and China and milk capacity in Philippines. A similar approach could also be extended to the buffalo of the Indian subcontinent with emphasis on milk and draught efficiency.

Table 8 DRAUGHT POWER PERFORMANCE

M = males, F = females, C = castrate

3.1 Selection for Size

Size is an important characteristic to be considered only in relation to efficiency of milk, meat or draught production and adaptation. The buffalo is at the larger end of the size scale among all domestic indigenous ruminant livestock and its surface area to body weight is therefore smaller compared to other species in the same environment, especially cattle. It has also been reported that buffaloes have a larger gastrointestinal volume than cattle (Moran and Wood, 1982) in relation to total body size. This has a bearing on feed intake which is in turn positively associated with rate of passage of feed and negatively related to digestibility. The complex relationship between size and other factors such as intake, digestibility, heat load and its dissipation within the river and swamp types needs to be further studied before breeding goals can be formulated. A curvilinear relationship was observed between body size and milk yield among Holsteins where sires that were just above average for size proofs produced daughters that yielded more milk than smaller or larger contemporaries (Sivarajasingam et al., 1984). A similar trend may be expected in the tropical environment.

3.2 Breeding and Selection for Milk

A considerable amount of work has been done in India including progeny testing for milk (Nagarcenkar, 1979; Gill, 1985 and Tiwana and Dhillon, 1985). The milk yield in 305-days and total yield increased from 1062 kg and 1120 kg in 1971 to 2346 and 2450 respectively after 12 years of selection through a progeny testing programme. Heritability for milk yield among river buffalo shows a medium to high value (Table 9). However, reproductive traits in Table 10 show lower estimates but higher than in most cattle breeds. These figures indicate much genetic progress could be achieved through selection of superior bulls for milk production. The current genetic limit for milk yield which is as high as 4000-4200 kg (5 animals) per lactation of 305 days is encouraging (Gill, 1985). It will be of interest to study the efficiency at these high levels compared to cattle of similar production.

Table 9 HERITABILITY (h²) AND REPEATABILITY (r) OF LACTATION
YIELD AND LENGTH

Table 10 HERITABILITY OF SOME REPRODUCTIVE TRAITS

3.3 Breeding and Selection for Beef

An improvement programme for beef production has been limited to the last few years mainly in China and Thailand. Main characteristics were weights at weaning and later ages. Heritability for body weights are given in Table 11 and are generally medium to high as in the case of beef cattle.

information regarding river buffalo is limited but genetic variation is expected to be high. Response based on performance testing for growth and family selection for carcass characteristics will prove effective. As for dairy buffalo, implementation of improvement programmes with farmers will involve high costs and management difficulties. A possible solution is to establish test stations around the country to evaluate selected bulls for growth, carcass (using relatives), reproductive and draught characters. Top bulls are selected and used to improve the national herd. Embryo transfer technology could be a useful tool here to multiply bulls for natural mating in the absence of proper facilities for AI.

Table 11 HERITABILITY OF BODY WEIGHTS

3.4 Breeding and Selection for Draught

Breeding for genetic improvement involves retention of superior males for semen collection, and females for regular calf production. However, most of the buffaloes identified for draught are deprived of their normal reproductive activity. Males if not sold for beef, are often castrated. For the females, having to nurse a calf is considered a nuisance by the owner during ploughing. Even if the animals are fertile and given the opportunity to mate, the rate of success is expected to be low due to the low levels of feed quality and stress due to work in the field.

In many countries, the farm sizes are declining due to fragmentation resulting in an accompanied increasing scarcity of feed resources. Under such circumstances the decline in body size as a result of castration of larger animals for draught purposes may prove to be a compromise or even an advantage (Mahadevan, 1985). He cites the work of Vercoe et al. (1985) who favour more efficient utilization of small animals for draught purposes under situations of limited feed resources and selecting for superior heat tolerance individuals within the population. However, the choice of identification of existing strains that are heat tolerant, low in maintenance requirement and resistant to parasites and other diseases amongst the swamp buffalo population and using them in crossbreeding programmes may be more applicable in developing countries (Mahadevan, 1985).

3.5 Multipurpose Breeding Strategies

Although selection for site, milk, beef and draught has been discussed separately, they are not mutually exclusive and they are also not the only traits of concern. Buffaloes in smallholder and institutional farms are known to have slow reproductive rates and high mortality rates. These are largely due to environmental and partly genetic factors. As was mentioned earlier, more than 70 percent of the buffaloes in all countries of Asia can be considered nondescript. Well defined breeds and breeding programmes are only confined to institutional farms. This needs to be extended to improve the national herds hand-in-hand with development of a recording system and related infrastructure. However, costs of operating such a system based on conventional progeny testing will be enormous. An alternative system like the Irish progeny testing and selection programme (Cunningham, 1979) will be more effective. This system will also allow selection for total economic merit to include traits like reproductive efficiency and feed efficiency which are vital in buffalo production.

Breeding strategies tend to vary from country to country. In the Indian subcontinent, breeding for milk first and draught second will continue. In southeast Asian countries, except Malaysia, draught power is of prime importance followed by beef in most areas or milk in the Philippines. In Malaysia and to some extent in the Philippines and Sri Lanka, swamp buffaloes are significant contributors to the beef industry. However, declining buffalo numbers due to the advent of farm mechanization, their slow reproductive rates and few numbers compared to the indigenous Kedah-Kelantan cattle in Malaysia, the future of the swamp buffalo in this country is uncertain. A solution to this problem is to upgrade them using Murrah or Nili-Ravi or both breeds into a dual purpose dairy beef buffalo. This will further enhance the existing village ghee industry and other milk products for which there exists a substantial market. Attempts to conserve the buffalo as a beef animal especially in the Philippines, Thailand, Indonesia, Taiwan and China need to be studied. Besides the fact that their population is decreasing due to earlier mentioned mechanization, low reproductive rates and high mortality rates, the buffalo are also decreasing in size due to reasons discussed earlier in this paper.

Another area that has been extensively discussed, but little work has been done, is efficiency of feed utilization by buffaloes. Buffaloes are considered more efficient utilizers of coarse feeds than cattle but this has not been well documented.

4. CROSSBREEDING RIVER AND SWAMP BUFFALOES

Crossbreeding has been practised in almost all countries where swamp buffalo predominates i.e. China, Burma, Thailand, Philippines, Malaysia and Sri Lanka with the desire of improving milk yield capacity and size for work in the field. China started crossbreeding work as early as 1960 and produced some 45 000 crossbreds by 1977 (Wang, 1979) through AI. The crosses have been further upgraded with Nili-Ravi resulting in grades with 50, 25, and 25 percent of Nili-Ravi, Murrah and swamp buffalo levels of inheritance. They have been evaluated (Liu et al., 1985) and summarized in Table 12.

The crossbreds in the above report had good conformation, a massive body structure with well develped hindquarters, with an average daily gain of 0.8 kg on grass. Average fat content of crossbred milk was 7.5 percent. The temperament of triple crosses as superior but the same could not be said for the halfbred, Murrah x swamp crosses. Reports on crossbred performance, although on a smaller scale, have also been reported in the Philippines (Eusebio, 1975), Taiwan (Liu, 1975), Sri Lanka (Jalatge, 1980) and Nepal (Keshary and Shrestha, 1980). These reports had lower milk yields ranging from 492 to 956 kg than the work in China shown in the Table 12. However, the results do indicate genetic potentials of crossbreeding to improve the swamp buffalo for milk and meat. In this context, it is also relevant to note that an FAO/UNDP project in collaboration with the Philippine Council of Agricultural Resources, Research and Development (PCARRD) has an ongoing evaluation of the progeny resulting from mating of different breeds/strains of river buffaloes with the Philippines carabao for use as draught, milk and meat animals (Mahadevan, 1985). Preliminary results showed an increase of 32 percent in birth weight, an increase by 100 percent in weight at 18 months (average 300 kg) and a 3 to 4-fold increase in milk yield (1200 1 in 300 days) than the native carabao (Ranjhan, 1985).

Table 12 CHARACTERISTICS OF CROSSBRED BUFFALOES
(adapted from Liu, 1985)

Crossbreeding between swamp and river buffaloes raises an area of concern. The differences in the chromosome numbers between swamp (2n = 48) and river (2n = 50) (Fischer and Ulbrich, 1968) buffaloes may have an effect on the fertility of the offspring considering synaptic possibilities in the F₁s resulting in some genetically, unbalanced meiotic products that degenerated. This was revealed by Bongso et al. (1983) where a large proportion of degenerating spermatocytes and abnormal spermatids were found in testicular biopsies of F₁ hybrids. The F₁ produced by river and swamp matings had a chromosome number 2n = 49 (Fischer and Ulbrich, 1968 and Bongso and Jainudeen, 1979). Reports by Bongso et al. (1984) showed further segregation resulting in three F₂ populations (2n = 48, 49 and 50), two populations (2n = 49 and 50) when backcrossed to river buffalo and two populations (2n = 48 and 49) when backcrossed to the swamp buffalo. These interesting findings were however limited to small sample sizes. It is now necessary to relate these genotypes (different chromosome numbers within different levels of exotic inheritance) to fertility and extent of heterosis for production traits in smallholder farms and large commercial ones.

5. CONSERVATION, IMPROVEMENT AND UTILIZATION

The buffaloes of Asia have evolved within their ecosystem over several centuries and have thus acquired adaptive characteristics and still remain useful in food production. The review of literature above though not exhaustive, and previous meetings on Animal Genetic Resources Conservation (FAO, 1981, FAO, 1983 and SABRAO, 1981) have revealed that although a wealth of information has accumulated, large gaps exist in the total characterization of buffaloes and other species. For instance, efficiency of the buffalo for milk, meat, draught or multipurpose on high and poor quality roughages and byproducts is limited or absent. Such information is vital in agricultural planning strategies and allocation of animals and breeding programmes to various farmers and farming systems. The buffalo of Asia are generally lumped together as the swamp type although wide variation is recognized. Blood markers are useful in the characterization of breed structure and the relationship between the varieties of the buffalo population. Blood markers, especially those related to membrane antigen, may be of value for understanding, control and eradication of diseases (Braend, 1981).

5.1 Genetic Conservation - Why?

Conservation of live specimens of buffaloes or other livestock consumes sizable manpower, valuable space and costs besides demanding proper planning skills. However, the buffalo needs to be conserved for the following reasons in brief:

1. They possess adaptive characteristics to thrive in the stressful environment which could be lost through dilution and intensive selection for production traits.
2. They also possess the ability of converting poor quality feed resources into meat, milk and working capacity in the field.
3. The genetic variability should be maintained which is the basis for genetic improvement for the future. Fortunately, buffaloes, unlike cattle, have not undergone massive selection and crossbreeding until only very recently. There is still genetic variability to be salvaged.
4. The opportunity could be lost to exploit heterosis through cross breeding.
5. The future expectations of our buffaloes are unknown or unknowable. Selection goals in cattle breeding in the past have changed. From single trait selection we have turned to selection on total merit, when we may have lost some of the negatively correlated valuable genes. Selection for conformation traits are beginning to show their importance in relation to lifetime stayability and production. With raising production costs, we are now looking at yield per unit dry matter or energy input. In this respect the buffalo has an important role. The future spectrum of diseases and feed availability is unknown. Therefore we need to maintain the present variability or even increase it.
6. Finally, we ask ourselves, do we have the right to destroy, or even neglect our indigenous germplasm collection which rightfully belongs to our children and grandchildren who may find greater uses for them.

5.2 Genetic Conservation - How?

Without going into details, the following points need to be highlighted.

1. Live animals. An actively breeding population should be maintained, perhaps, each line or variety in a different farm to reduce costs. Two major advantages of live animal conservation are a) they are always available for immediate utilization in the event of any setbacks in the upgraded population; b) they are constantly exposed to new strains of diseases and their resistance evaluated. Such live animals would also contribute to education and to community awareness of the indigenous fauna. Cost of maintenance is often argued to be high. This may be an exaggeration overlooking their cheap maintenance costs, better longevity, lower veterinary costs besides revenue from milk and meat. A major problem that needs to be defined is the type of selection to be practised without altering the genetic variability. It is suggested here that both random selection as well as overall merit (with equal weighting for each trait) should be practised with minimum intensity.
2. Cryogenic storage. This is convenient and cheap and further work needs to be done. Another advantage is that the genotype will not be subjected to genetic drift. A disadvantage is that the animals, especially in the case of females, have a time lag when live adults are urgently required.
3. DNA genetic material storage. This is a useful tool but certainly not an immediate task.

Conservation by the last two methods has been discussed at length and mode of action outlined (FAO, 1983).

5.3 Improvement within Conservation

It is evident that the buffalo plays an integral part of our farming system and its numbers have been maintained or increased. Various river breeds have been documented and evaluated. Strains of swamp buffalo have been observed but attempts to characterize them genetically have yet to be made. Conservation should begin with a proper sampling technique to represent the existing variability and in sufficient numbers. In these populations for conservation, selection should be minimal and to maintain population size constant, either random culling or culling based on total merit should be practised. However, for the national herds genetic improvement could be achieved through selection and crossbreeding. For both, germplasm collection and national herd data banks are essential to monitor their genetic progress.

6. CONCLUSIONS

The total buffalo population now stands at 122 million in Asia. The swamp buffalo population is decreasing in some countries, especially Malaysia, Thailand, Philippines and Taiwan. Growth and carcass characteristics of buffalo and cattle are comparable. Draught capacity and the working lifespan of buffalo is superior to that of cattle. There is a management tendency in smallholders to deprive larger swamp buffaloes from producing offspring leading to a high likelihood that the buffalo body size may be affected. The river buffalo is larger than the swamp. Interrelationship between size, growth, milk yield and draught power needs to be further studied also with respect to feed efficiency and adaptation. The former produce about 1800 kg milk per lactation of 305 days. First calving age, intercalving periods and postpartum oestrus are longer than in cattle and perhaps this is a physiological phenomenon associated with a longer (310 days) buffalo gestation period. Feed efficiency with complete characterization of the buffalo breeds is timely with data banks having interregional links for exchange of information and material. The various strains of swamp buffalo need to be identified. Their special capabilities and adaptation to the particular niche has to be defined. Conservation of valuable breeds and strains of buffaloes is reemphasized before genetic variation is"diluted. Finally, crossbreeding and its advantages are being pursued in various countries and show preliminary prospects of genetic improvement of swamp buffaloes in spite of initial setbacks of differences in chromosome numbers between the river and swamp buffaloes and the genetic imbalance in germ cells resulting from meiosis.

ACKNOWLEDGEMENT

The author wishes to thank the Director-General of MARDI for permission to attend the meeting and present this paper.

REFERENCES

^1/Livestock Research Division, MARDI, P.O. Box 12301, General Post Office, 50774 Kuala Lumpur, Malaysia.