PAPERS PRESENTED AT THE EXPERT CONSULTATION (Contd.)

C. PRACTICAL ISSUES FOR THE CONSERVATION AND IMPROVEMENT OF PRIORITY BREEDS WITH GLOBAL REVIEWS OF SPECIES (Contd.)

A GLOBAL REVIEW OF THE GENETIC RESOURCES OF SHEEP AND GOAT BREEDS

R.W. Ponzoni¹

1 Introduction

There are 1190 and 557 million sheep and goats in the world, respectively. Table 1 shows the break-up by economic class and region according to FAO (1991). Of the total, 53 and 94 per cent of the sheep and goats are kept in developing countries, respectively, where they make an important contribution to the well being of human populations as a source of meat, milk, fibre and hides. Note the disproportionate number of goats in developing countries compared to that in the developed ones.

In the past, developing countries have shown a marked tendency to rely on imported ‘improved’ sheep and goat breeds for genetic improvement. This has sometimes resulted in a reduction in the number of stock of the indigenous breeds, and in extreme cases in a threat to their future existence. There are, of course, other factors that may also put some breeds at risk, such as natural disasters or pressure from other forms of land use. More recently, however, there seems to be a greater awareness of the need to identify, preserve and improve local breeds perceived as possessing attributes that could be valuable now or in the future. This change of attitude can be partly explained by the efforts made by organizations such as FAO in the conservation of global genetic resources, but it is also probably due to the frequent failure of introduced breeds to perform satisfactorily. In harsh environments, which can include special disease risks, the indigenous ‘unimproved’ breeds have often turned out to be better suited to the local conditions.

This paper consists of three main sections. The first section is a review of publications presenting descriptive information about sheep and goat breeds, with special emphasis on those in developing countries. The second section reviews reports that have specifically pointed to indigenous sheep and goat breeds that are worthy of attention. Finally, the third section lists those breeds that available information suggests should be the target of research and development programmes in the immediate future. The overall aim of the paper is to discuss a number of practical issues related to the conservation and improvement of indigenous sheep and goat breeds.

¹ Department of Agriculture, Box 1671, G.P.O., Adelaide, S.A., Australia 5001.

Table 1
Sheep and goat numbers (1000 head)^A

^A Source: FAO 1991

2 Review of publications on sheep and goat breeds

Currently, FAO statistics and most country census do not provide information on breeds. Sometimes information about numbers of a particular breed can be derived indirectly from local statistics when the breed is confined to a restricted area. However, more commonly, information on particular sheep and goat breeds cannot be obtained unless a special effort is made with that purpose.

One of the problems faced when trying to make decisions about priority sheep and goat breeds is the lack of a data base containing an inventory of breeds, with the necessary characterization in terms of productivity, outstanding features, population size, etc. Answering questions about the need for maintenance of various breeds without such an inventory is extremely difficult. Table 2 lists key publications describing sheep and goat breeds. The list is by no means exhaustive and it concentrates mainly on developing nations. In addition to the work cited in the table there have been numerous articles dealing with specific breeds. These will be referred to in the next section, only in case they discuss a breed that is considered worthy of immediate attention for some reason.

In table 2 the studies have been grouped according to their scope, that is, whether they attempt to cover the whole world, a particular country or region, or a specific type of sheep or goat. In some instances the studies are merely of a descriptive nature, with little or no comment regarding the relative worthiness of the breeds or their risk status. Nevertheless, as a whole the information provided is a very good basis for the creation of an inventory of sheep and goat breeds worthy of attention on a global basis. Much of the information called for by the descriptor lists for sheep and goats (FAO 1986c) can be found in these studies. Computerization of this information would make the search of breeds according to any relevant criterion (e.g. risk status, heat tolerance, etc.) a relatively simple task. At present, any attempt at choosing breeds on the basis of specific criteria is a very difficult and time consuming task, except perhaps for a small number of very knowledgeable people.

Table 2
Key publications describing sheep and goat breeds

Scope	Title of studyA	Species		References
		Sheep	Goats
World	Chapters 2 & 3. World survey of sheep breeds			Ryder and Stephenson 1968
	Sheep of the world			Ponting 1980
	Wild goats and their domestication			Mason 1981a
	Breeds			Mason 1981b
	Sheep and man			Ryder 1983
	World dictionary of livestock breeds			Mason 1988
Country or region	Declining breeds of Mediterranean sheep			FAO 1978
	The sheep breeds of Afghanistan, Iran and			Yalçin 1979
	Turkey			FAO 1981
	Animal genetic resources in Latin America			Bhat et al. 1981
	Animal genetic resources in India
	Evaluation of animal genetic resources in Asia			SABRAO 1980, 1981
	and Oceania			FAO 1982
	Sheep and goat breeds of India			FAO 1985a
	Livestock breeds of China			FAO 1985b
	Sheep and goats in Pakistan			Gatenby 1986
	Appendix II. Breeds of sheep in the tropics			FAO 1986a
	Sheep and goats in Turkey			FAO 1986b, 1987a, 1989a
	Small ruminants in the Near East, vols. I, II and			Devendra and McLeroy 1987
	III			FAO 1989b
	Goat and sheep production in the tropics
	Animal genetic resources of the USSR
Type of animal	Trypanotolerant livestock, vols. I and II			FAO 1980a, b
	Prolific tropical sheep			FAO 1980c
	Hair sheep of Western Africa and the Americas			Fitzhugh and Bradford 1983
	The Awassi sheep			FAO 1985c
A For space reasons some titles have been shortened. The complete titles are given in the REFERENCES section.

The reports cited in this section are of three different types. Firstly, there are those that draw attention to particular indigenous breeds, either because they are at risk, or because they possess specific valuable attributes, or for both reasons. Secondly, there are reports on live animal preservation programs currently underway in various parts of the world. It is reasonable to assume that the breeds included in such programs would be at risk and (or) have specific valuable attributes. Thirdly, a number of personal contacts were made during the preparation of this review. The information obtained in this way is quoted as a personal communication.

Table 3 lists sheep and goat breeds that have been identified as worthy of attention. The information is grouped into four regions (Africa, Latin America, Near East and Far East). These correspond with currently accepted FAO regions (FAO 1991) for developing countries. Regions classified by FAO as developed were not included in the review. Two interesting features emerge from table 3. One of them is that the number of sheep breeds listed is about three times greater than the number of goat breeds, even though the number of sheep in developing countries is only 1.2 times greater than the number of goats. This feature is probably a reflection of the difference in the amount of interest in the two species shown by researchers. The other feature revealed by table 3 is that several of the breeds listed are related to each other in some way, and could therefore be considered together as a single population from the point of view of implementation of preservation, research and development programs. A brief discussion of the relationship among some of the breeds follows.

Latin American sheep breeds such as Bahama Native, Barbados Blackbelly, Morada Nova, Pelibuey, Red African, Santa Inés and Virgin Island White are all hair sheep derived from those taken to America from West Africa (Fitzhugh and Bradford 1983) during the seventeenth century (Devendra and McLeroy 1987). Despite the differentiation that has taken place over time (e.g. Bahama Native, Barbados Blackbelly and Virgin Island White are more prolific than the other breeds) there are still some remarkable phenotypic similarities among them, and also with West African sheep. For example, Santa Inés are larger than the rest of American hair sheep, and said to be derived from crosses between Morada Nova and Bergamasca. Nevertheless, Fitzhugh and Bradford (1983) point out the great similarity between Santa Inés and the Sahelian type of West African sheep.

The four Latin American goat breeds listed in table 3 are from North Eastern Brazil. Many observers consider that these represent different colour types, rather than true ‘breed types’ (Shelton and Figuereido 1981). Apart from sharing some common ancestry, with the exception of the Marota breed, the other three (Canindé, Moxotó and Repartida) are similar in form, function and apparently in performance (Shelton and Figuereido 1981).

In the Near East, Libyan Barbary sheep are considered the prototype for the Barbary sheep type of Northern Africa, to which the Barki (Egypt) also belongs (Devendra and McLeroy 1987). Barbary sheep are coarse-wooled and fat-tailed. They are a multiple purpose type of sheep, producing fibre, meat and some milk.

The Chios breed of Greece, renowned for its milk production and prolificacy gave origin to the Sakiz of Turkey (Mason 1988).

The Damascus, Nubian and Zaraibi goats all belong to the ‘Nubian Type’ (Mason 1988). These are dairy goats, with a Roman nose and long lop ears. Devendra and McLeroy (1987) state that the Damascus goat found in Syria, Lebanon and Cyprus gave rise to the Zaraibi of Egypt and to the Nubian goat of Sudan.

Considering relationships among breeds is important when making decisions about which ones merit support for preservation, research and development programs. With limited resources it appears that it would be unwise to support two or more breeds that are closely related and have similar characteristics.

Table 3
Sheep and goat breeds identified as worthy of attention

4 Suggested breeds for immediate implementation of support programs

Given the relatively large number of sheep and goat breeds that have been identified as worthy of attention in preservation, research and development programs (table 3) it is clear that attending to all of them immediately will not be possible. Under such circumstances an overall strategy involving short term and long term activities appears to be required. In the short term, breeds about which sufficient information is available to make a positive judgement about their value should be the target of programs commencing as soon as possible. At the same time, but with a longer term view, activities directed at the establishment of a computerized data bank on all indigenous breeds should be undertaken so that other threatened and (or) potentially useful populations are identified. This latter activity would provide a sound basis for choosing breeds for future programs.

The terms of reference for the present review established that four or five breeds had to be recommended to receive immediate support for a preservation, research and development program. The suggested list of sheep and goat breeds is presented in table 4, together with a brief justification for the choice. In a few instances breed support programs have already been initiated.

The breeds were chosen more often because of some special and valuable characteristic they possess, than because of their current risk status. However, it may be worthwhile making a distinction between developed and developing countries in this respect. In the latter countries even livestock breeds that from a numeric point of view might appear quite ‘safe’, are often at much greater risk than their counterparts in developed countries. Threats to indigenous breeds in developing countries include natural disasters, expansion of other forms of land use, human population growth, and indiscriminate crossbreeding with ‘improved’ breeds. Crossbreeding is sometimes undertaken because of fashion, social pressure and prestige, rather than because of ‘economic-genetic’ reasons. Furthermore, in developing countries, even what may be considered the most important livestock breeds have very seldom been studied with the depth that important breeds in developed countries have. Therefore, the judgement of the risk status of breeds in developing countries requires more than simply increasing the minimum number of the various risk categories.

A number of criteria were used in the selection of the breeds listed in table 4. One of them was, of course, that the breed possessed one or more highly desirable attributes from the point of view of adaptation and (or) productivity. Another very important consideration made was in relation to the potential impact of the work in case it resulted in a favourable outcome. Preferred breeds were those in which preservation and improvement could have the potential for influencing relatively large populations, not only in the country in which the project was developed, but also in other countries with the same or with very similar breed types. Finally, an attempt was made to achieve some balance across world regions. The sheep and goat breeds listed in table 4 satisfy these criteria.

The literature cited in tables 2 and 3 contains ample information about all the breeds listed in table 4, which includes a brief justification for their choice. Therefore, only a few clarifying comments about some of the breeds will be made here.

Table 4
Sheep and goat breeds for immediate implementation of support programs

4.1 Sheep

Djallonké sheep (and Fouta Djallon goats) were suggested because of their ability to survive and reproduce in trypanosomiasis affected areas. Furthermore, it appears the area of West Africa to which these breeds are adapted would have a substantial amount of unused carrying capacity (FAO 1980a, Ch. 5). The name Djallonké sheep was preferred to West African Dwarf because as Fitzhugh and Bradford (1983) point out, these sheep are relatively small but they are not achondroplastic. Similarly, the name Fouta Djallon goats was preferred because these goats may (Mason 1984) or may not (Hall 1991) be achondroplastic. Note that the breed names Djallonké and Fouta Djallon are collective names that refer to a variety of ‘types’ distributed throughout the region. Studies on the genetic similarity among these ‘types’ would be of importance in establishing whether the whole population may be treated as a single breed or whether it should be subdivided.

Fitzhugh and Bradford (1983) have argued that with the exception of the Blackhead Persian (fat-rumped) all other American hair sheep could be considered as belonging to the same ‘genetic type’. In table 4 the Pelibuey was chosen as representative of that type. It was threatened by planned and by indiscriminate crossbreeding for some time, but it is now relatively safe (Baffi-personal communication). Because the island of Cuba is free from a number of diseases, livestock developed there should be acceptable from an animal health point of view to many other countries within and outside the region. The Cuban Pelibuey is not a uniform population. There is enough variation in coat colour, type and performance, which should enable the development of strains selected for different objectives (e.g. reproduction for a maternal line, growth rate for a terminal sire line).

Bradford (1989) points out that there is a potentially very large forage resource under tropical tree crop agricultural systems, which could be exploited with suitable genotypes. The Javanese Thin-tailed sheep are one such genotype. They are particularly interesting because the single gene for prolificacy segregating in the population should enable the development of strains differing in reproductive performance, and thus suited to production systems of different intensities.

The D'man breed from Morocco is a relatively minor breed in the country. Nevertheless, it has attracted attention because of its exceptionally large litter size, early puberty and short lambing interval (Bradford et al. 1989, Lahlou-Kassi et al. 1989). It appears that in the D'man the high litter size is transmitted in an additive manner (not via a single gene of large effect). Since the breed evolved in a very hot environment, apart from being valuable in itself, it could be useful for crossing with other breeds in similar environments.

The Awassi is the predominant breed in the Near East, where strains differing in milk production, size and wool quality can be found (FAO 1987a). The Awassi thrives under arid conditions and can tolerate extremely high ambient temperatures (Devendra and McLeroy 1987). Increasing our understanding of this breed so that its remarkable characteristics can be maintained and if possible improved further appears to be justified.

4.2 Goats

Comments on Fouta Djallon goats were made under the Sheep heading.

The Damascus or Shami goat originates in Syria, but it has spread to a number of countries of the Near East. It is believed to have given rise to the Zaraibi goat of Egypt and to be an ancestor of the Kilis breed of Turkey (Devendra and McLeroy 1987). The Damascus breed has high prolificacy and milk yields, combined with adaptation to hot environments. It has been identified as having great potential in the Near East (FAO 1987a).

The Jamnapari is the largest of the Indian goat breeds (Mason 1981b). It has contributed to the formation of the Anglo-Nubian breed (Bhat 1987). It is a dual purpose animal, useful for meat and milk, but it is mainly as a dairy animal that it is sought (Devendra and McLeroy 1987). Jamnapari goats have been taken to South East Asia, East Africa, the West Indies and South America to improve milk production (Mason 1981b, Devendra and McLeroy 1987). Bhat (1984) comments that there has been an alarming reduction of the number of animals of this breed. This decline, coupled with the valuable attributes of the Jamnapari have prompted suggestions that the breed is worthy of immediate attention.

The greatest concentration of goats in Brazil is in the North Eastern part of the country, in a drought prone hot semi-arid tropical environment (Figueredo et al. 1982). Of the ‘local’ breeds the Moxoto' is the most common one. It is a small, very good looking goat, highly adapted to the environment. It is kept as a meat animal and it produces a very valuable skin. As other goats in the region, it has often been indiscriminately crossed with imported breeds, such as the Anglo-Nubian and the Bhuj. Fears about declining numbers of the breed led to a small herd of Moxoto' goats being established in Brasilia, where a cryopreservation program started in 1989 (da Silva Mariante 1990).

The Boer goat of South Africa is derived principally from the goat of the Hottentots (Mason 1981b). The Improved Boer is a fast growing, well muscled and strong boned goat. It has good reproductive rate and milk production, but its greatest potential would seem as a terminal sire breed. In some production systems it could be used over small prolific dams to improve the growth rate and carcass attributes of their progeny. Boer goats have recently been introduced to Australia.

5 Final remarks

A note of caution in relation to the activities that support programs for sheep and goat breeds could entail may be appropriate. Very often preservation and breed development are mentioned as part of such programs, but research is seldom discussed. In simple terms breed development could be defined as the improvement of the breed's usefulness by genetic means, probably accompanied by its numerical expansion. With indigenous sheep and goat breeds, in most instances, the necessary information to implement a scientifically based program of genetic improvement is not currently available. Therefore, a period of intense research activities should often precede the implementation of genetic improvement programs. Intense selection for some simplified economic goal could result in the loss of valuable attributes in the population the program is trying to improve. For example, a common feature of many sheep and goat breeds adapted to tropical environments is their relatively small size, but very good reproductive rate. It is tempting to suggest that selection for greater growth rate should be recommended. However, intense selection in that direction would be risky, unless we knew the likely correlated responses in other important traits, such as reproductive rate or resistance to disease, and we had a clear understanding of the role played by body size in the adaptation of these breeds to harsh environments.

Support programs for indigenous sheep and goat breeds should have as one of the most important aims the establishment of resource flocks or herds (R F&H). These populations should have adequate size and design to enable the estimation genetic parameters, to provide experimental animals for physiological studies, and to undertake genetic studies at the gene and molecular levels. The range of characteristics recorded in the R F&H should be much greater than in normal genetic improvement programs so that correlated responses of interest can be predicted and genetic variation in potentially useful traits is uncovered (e.g. resistance to internal parasites, ability to hyperhydrate, etc.). The R F&H would ensure preservation of the breed, and could be a source of breeding stock. However, dissemination of the genes of the indigenous breed in question should be the aim, rather than intense selection in any direction. The latter should be delayed until such time as we have sufficient knowledge indicating that the pursuit of a particular goal will not result in the loss of the attributes that led us to preserve the breed in the first place.

At present, making suggestions about which sheep and goat breeds should be the target of support programs in the near future is a very difficult task. Injustices could be done simply because of ignorance about other breeds, or due to biased perceptions. It is hoped that in the future the situation will be such that making objective choices becomes easier, and that other valuable indigenous sheep and goat breeds can also be supported.

6 Acknowledgments

Dr G.E. Bradford, Mr R.C. Cardellino, Mr C.H.S. Dolling and Dr Helen N. Turner made many valuable suggestions during the preparation of this paper. They are, however, innocent of any omissions or errors of appreciation the paper could contain.

7 References

Acharya, R.M. (1990). Endangered livestock breeds in South Asia. FAO Animal Production and Health Paper No. 80, pp. 197–206.

Bhat, P.N. (1984). Breeding plans for improvement of indigenous breeds and species. FAO Animal Production and Health Paper No. 44/1, pp. 67–79.

Bhat, P.N. (1987). Principles for animal improvement in the tropics. Sheep and goats: Asian experiences. FAO Animal Production and Health Paper No. 66, pp. 89–124.

Bhat, P.M., Bhat, P.P., Khan, B.U., Goswami, O.B. and Singh, B. (1981). Animal Genetic Resources in India. Indian Vet. Res. Inst., Izatnagar, Public. No. 192.

Bradford, G.E. (1989). Animal agriculture research and development: Challenges and opportunities. Can. J. Anim. Sci. 69:847–856.

Bradford, G.E., Burfening, P.J. and Cartwright, T.C. (1989). Evaluation of production and reproduction of sheep, goat and alpaca genotypes in the Small Ruminant Collaborative Research Support Program. J. Anim. Sci. 67:3058–3067.

Casey, N.H. and Van Niekerk, W.A. (1988). The Boer goat. I. Performance testing, reproduction and milk production. Small Rumin. Res. 1:291–302.

da Silva Mariante, A. (1990). Endangered livestock breeds in South America. FAO Animal Production and Health Paper No. 80, pp. 213–230.

Devendra, C. and McLeroy, G.B. (1987). Goat and sheep production in the tropics. (Longman Scientific & Technical, U.K.).

FAO (1978). Declining breeds of Mediterranean sheep. FAO Animal Production and Health Paper No. 8. FAO, Rome.

FAO (1980a). Trypanotolerant livestock in West and Central Africa. Vol. 1 - general study. FAO Animal Production and Health Paper No. 20/1. FAO, Rome.

FAO (1980b). Trypanotolerant livestock in West and Central Africa. Vol. 2 - country studies. FAO Animal Production and Health Paper No. 20/2. FAO, Rome.

FAO (1980c). Prolific tropical sheep. FAO Animal Production and Health Paper No. 17. FAO, Rome.

FAO (1980d). Animal genetic resources conservation and management. Report of the FAO/UNEP Consultation. FAO, Rome.

FAO (1981). Recursos genéticos animales en America Latina. FAO Animal Production and Health Paper No. 22. FAO, Rome.

FAO (1982). Sheep and goat breeds of India. FAO Animal Production and Health Paper No. 30. FAO, Rome.

FAO (1985a). Livestock breeds of China. FAO Animal Production and Health Paper No. 46. FAO, Rome.

FAO (1985b). Sheep and goats in Pakistan. Animal Production and Health Paper No. 56. FAO, Rome.

FAO (1985c). The Awassi sheep - with special reference to the improved dairy type. FAO Animal Production and Health Paper No. 57. FAO, Rome.

FAO (1986a). Sheep and goats in Turkey. FAO Animal Production and Health Paper No. 60. FAO, Rome.

FAO (1986b). Small ruminants in the Near East, vol. I. FAO Animal Production and Health Paper No. 54. FAO, Rome.

FAO (1986c). Animal genetic resources data banks. 2. Descriptor lists for cattle, buffalo, pigs, sheep and goats. FAO Animal Production and Health Paper No. 59/2. FAO, Rome.

FAO (1987a). Small ruminants in the Near East, vol. II. FAO Animal Production and Health Paper No. 55. FAO, Rome.

FAO (1987b). Animal genetic resources. Strategies for improved use and conservation. FAO Animal Production and Health Paper No. 66. FAO, Rome.

FAO (1989a). Small ruminants in the Near East, vol. III. FAO Animal Production and Health Paper No. 74. FAO, Rome.

FAO (1989b). Animal genetic resources of the USSR. FAO Animal Production and Health Paper No. 65. FAO, Rome.

FAO (1991). Production yearbook. Statistics series No. 99.

Figuereido, E.A.P., Simplicio, A.A., Bellaver, C. and Pant, K.P. (1982). Evaluation of goat breeds in the tropical North - East Brazil. I. A study of birth-related traits of native and exotic goat breeds. Pesq. Agropec. Bras. 17:643–650.

Fitzhugh, H.A. and Bradford, G.E. (1983). Hair sheep of Western Africa and the Americas. (Winrock International, U.S.A.).

Gatenby, Ruth M. (1986). Appendix II. Breeds of sheep in the tropics. In: Sheep Production in the Tropics and Sub-Tropics, pp. 269–292. (Longman, U.K.).

Hall, S.J.G. (1991). Body dimensions of Nigerian cattle, sheep and goats. Anim. Prod. 53:61–69.

ILCA (1991a). Genetic resistance to endoparasites and other diseases in small ruminants - ILCA research. SRNET Newsletter No. 21, pp. 9–11. (ILCA, International Livestock Centre for Africa).

ILCA (1991b). ILCA (International Livestock Centre for Africa) 1990: Annual Report and Programme Highlights. ILCA, Addis Ababa, Ethiopia.

Iniguez, L., Sanchez, M. and Ginting, S. (1991). Productivity of Sumatran sheep in a system integrated with rubber plantation. Small Rumin. Res. 5:303–317.

Lahlou-Kassi, A., Berger, Y.M., Bradford, G.E., Boukhliq, R., Tibary, A., Derqaoui, L. and Boujenane, I. (1989). Performance of D'Man and Sardi sheep on accelerated lambing. I. Fertility, litter size, postpartum anoestrus and puberty. Small Rumin. Res. 2:225–239.

Mason, I.L. (1981a). Wild goats and their domestication. In: Goat production, pp. 35– 55. (Edited by C. Gall, Academic Press, U.K.).

Mason, I.L. (1981b). Breeds. In: Goat production, pp. 57–110. (Edited by C. Gall, Academic Press, U.K.).

Mason, I.L. (1984). Goat. In: Evolution of domesticated animals, pp. 85–99. (Longman, U.K.).

Mason, I.L. (1988). A world dictionary of livestock breeds, types and varieties. (C.A.B. International, U.K.).

Ngere, L.O. (1987). Principles for indigenous animal improvement in the tropics -African experiences with sheep and goats. FAO Animal Production and Health Paper No. 66, pp. 125–135.

Patterson, H.C. (1991). Performance of four breeds of hair sheep and their crosses at Blenheim Station, Tobago, West Indies. In: Proc. Hair Sheep Research Symposium, pp. 33–40. (Edited by S. Wildeus, Univ. of the Virgin Islands, St. Croix, U.S. Virgin Islands).

Ponting, K. (1980). Sheep of the world. (Blandford Press, U.K.).

Ryder, M.L. (1983). Sheep and Man. (Duckworth, U.K.).

Ryder, M.L. and Stephenson, S.K. (1968). World survey of sheep breeds. In: Wool growth, pp. 25–205. (Academic Press, U.K.).

SABRAO (1980). Procs. of SABRAO Workshop on Anim. Genet. Resources in Asia and Oceania (Tsukuba, Japan, Sept. 3–7, 1979). Published by Tropical Agric. Research Centre, Tsukuba, Nekken Shiryo No. 47.

SABRAO (1981). Evaluation of animal genetic resources in Asia and Oceania. Procs. 2nd SABRAO Workshop on Anim. Genet. Resources, Kuala Lumpur.

Setshwaelo, L.L. (1990). Live animal conservation projects in Africa. FAO Animal Production and Health Paper No. 80, pp. 135–141.

Shelton, M. and Figueiredo, E.A.P. (1981). Types of sheep and goats in Northeastern Brazil. Int. Goat and Sheep Res. 1:258–268.

Shkolnick, A., Maltz, E. and Gordin, S. (1980). Desert conditions and goat milk production. J. Dairy Sci. 63:1749–1754.

Singh, L.B. and Dolling, C.H.S. (1970). Improvement of sheep and wool in India by selection within indigenous breeds. Wool and woollens of India, March 1970, pp. 15– 18.

Yalçin, B.C. (1979). The sheep breeds of Afghanistan, Iran and Turkey. Report of FAO/UNEP Project FP/1108-76-02 (833). FAO, Rome.

Yaochun, C. (1990). The endangered livestock breeds in East Asia. FAO Animal Production and Health Paper No. 80, pp. 207–212.

A GLOBAL REVIEW OF THE GENETIC RESOURCES OF PIGS

L. Ollivier and M. Molénat¹

1 Introduction

The pig is a major source of protein for human populations. According to FAO 1990 statistics it accounts for about 40% of the world production of meat. Regional variations however are important, as pig meat only represents 6% of the total meat production in Africa, whereas in China this percentage reaches 85%; Europe is a major area of production with about 51% percent (excluding the former USSR).

The pig is mainly fed on grain, which developed countries overproduce, while developing countries are badly in need of such food resources. Though it is impossible to forecast what will be the future balance in the allocation of resources between man and pigs, the evolution of pig will have in any case to be directed towards the most efficient use of whatever resources are available (Epstein and Bichard, 1984). Management of pig genetic resources should be envisaged in such a perspective. Various aspects of domestication, the present use of pig breeds and their conservation have already been dealt with in several reports, such as those by Epstein and Bichard (1984), Molénat and Legault (1986), Jonsson (1991), King (1991) and Steane (1991). This paper reviews the global situation of the pig genetic resources to-day, the objectives assigned to the management of such resources and the identification of priority breeds which deserve preservation/development action. Emphasis will be put on indigenous breeds in 3 world regions, where the need for international support can be considered the most urgent, namely Asia, Latin America and tropical Africa.

¹ INRA-Station de Génétique Quantitative et Appliquée, 78352 Jouy-en-Josas Cedex, France.

2 Management of genetic resources in pigs: general considerations

Potential resources include a wide spectrum of pig populations, which may be classified, according to a typology suggested by Lauvergne (1982), into four categories by increasing degree of evolution : (1) wild (or feral) populations, (2) traditional (or indigenous) populations, (3) standard breeds, and (4) selected lines (and so-called “new breeds”). Though a somewhat different classification has been retained by Mason (1988) in his dictionary, one can easily find in this compilation of over 500 pig breeds listed, representatives of all four categories of populations with a marked predominance of categories (2) and (3).

Management of genetic resources is usually understood in a longer term perspective than is the case for current breeding schemes oriented towards meeting rather near-term requirements. The goal is, in a long-term perspective, to maintain genetic variability, be it quantitative or qualitative (Ollivier and Lauvergne, 1988). In pigs, like in other livestock species, there is at present no indication that genetic variability is at risk for the traits usually considered by farmers and given present conditions of breeding (Fredeen, 1984). The highly polygenic nature of such traits, combined with systems of matings which largely favoring outbreeding, is a preventative against any significant decay of genetic variability. In addition, between-breed variation is generally thought to be the result of different frequencies of genes shared in common by several populations rather than of presence/absence of individual genes. In contrast, the risk of loss of variability may be greater for qualitative variation, especially when it has a monogenic basis. The difficulty here is that our genetic knowledge of pig populations is presently too limited to allow tracing, with the required accuracy, the population-gene combinations really at risk.

Another, and perhaps more important, concept underlying genetic resources is genetic flexibility. Even assuming that genetic variability makes possible, in theory, changes in any population in any direction for any trait, flexibility in pig evolution is somewhat constrained by the multiplicity of objectives to be considered. Each one of these objectives needs an appropriate allocation of the limited opportunities for selection and is also constrained by the differing amounts of genetic variability depending on the trait considered. Between breed diversity plays an essential role by allowing changes to be made rather quickly to cope with new situations. A recent example is the repopulation of Haiti after the outbreak of African swine fever in 1983. The use of a combination of a Criollo pig from a neighboring island (Guadeloupe), a local European breed (Gascon from France) and the highly prolific Chinese breeds (Taihu) quickly provided genotypes well adapted to the variable and generally extremely harsh rural environment (Delatte et al., 1991). Because of their low heritability, traits related to fitness, such as fertility, longevity, general disease resistance, are particularly valuable criteria for deciding on priorities for conservation/development action.

Adaptation in pigs should probably be considered rather differently from that of ruminants as pigs, except in a few situations, are much less dependent upon the vegetation of a particular region than are cattle or sheep. In many cases, it will be preferable to introduce an improved pig breed rather than to establish an improvement programme in a native breed. In his discussion of dietary adaptation, King (1991) stresses the lack of significant interaction in many genotype-environment experiments, though he expresses caution on the need for more experiments with indigenous and improved breeds. In such experiments, as stressed by Molénat and Legault (1986), the improved breed × harsh environment combination is often missing, and this “empty cell” is usually difficult to fill. With regard to climatic adaptation, the success of feral populations shows the good adaptability of improved genotypes, and there is again surprisingly little scientific evidence on particular adaptation to hot or cold environment.

Our limited knowledge of several aspects of the pig's biology calls for further scientific investigations and some extreme genotypes may be of great utility for that purpose. Such is the case, for instance, with the European Wild Pig and the Chinese Meishan in the European pig gene mapping project (Haley and Archibald, 1991). A final point in this regard is the preservation of a world heritage, though in this respect the pig has a much less fashionable image in most of our societies than other farm animals.

A proper assessment of the available resources is of fundamental importance. Breeds have to be catalogued and evaluated. Information on pig breeds can be found in scientific literature and, more recently, following FAO, European and Asian (for example SABRAO) initiatives, regional and global data banks are being established. The catalogue by Mason (1988) provides an overview of the distribution of pig breeds across the world. As shown in table 1, over 300 breeds are currently exploited, the largest numbers of breeds being found in Asia and Europe. More detailed information can be found in several data banks, such as the EAAP/FAO Global Animal Genetic Data Bank (Simon, 1990), which now includes information on 68 European and 10 Chinese pig populations, and the Nordic Data Bank for Farmed Animal Genetic Resources. Various private or state organizations in U.S.A., Canada and India can also be consulted (D. Simon, personal communication). Information on regional data banks in developing countries is unfortunately limited. Breed comparison evaluations are being carried out in several countries, often using exotic and local breeds, but there is little connection between different countries. In that respect, the world survey of Sutherland et al. (1985) merits being extended to more breeds and countries.

Table 1. An overview of pig breeds, types and varieties across the world (from Mason, 1988)

In practice, genetic resources are managed using classical breeding plans, implying evaluation of either individual breeding values (for pure breeding) or of population mean values (for crossbreeding schemes), as reviewed by Glodek (1991). For conservation, live populations and cryoconservation can be used, as in other farm species, with the limitation that recovery of frozen embryos is not yet feasible in pigs (Steane, 1991). A.I. and the use of frozen semen have an important role to play (e.g. Walters and Hooper, 1990). Breeding can be managed at the farm level, provided minimal management conditions are met, whereas conservation techniques, to be reliable in the long term, should be envisaged within state or international institutions.

3 Regional survey

3.1 Asia

About half of the world pig population is located in Southeast Asia, with an inventory of around 400 million heads. China represents the most of it with about 350 million, followed by Vietnam with 13 million. The enormous Chinese reservoir of genetic diversity for pigs, among other farm animals, was first drawn to the world's attention by Epstein (1969) and later documented by Legault (1978) and Cheng (1984). In addition to results collected in China, research findings on a few local Chinese breeds imported by several western countries are also available.

China has a long history of domestication, going back more than a hundred centuries (Zhao, 1990). It is therefore likely that domestication in China started not later than and independently of the domestication center of Near-East Asia. A long tradition of pig husbandry under various climatic and geographic conditions is probably a major cause of the presently observed genetic diversity. From reports based on a comprehensive investigation launched in 1981 by the Chinese government (Zhang et al., 1987), the number of native breeds officially registered is about 50, to which 25 improved breeds (including imported foreign breeds) should be added. The local breeds belong to 6 different types : North China, Central China, Shanghai area, South China, Southwest China and Plateau.

Phenotypic variability among breeds is extremely large. In an attempt to characterize Chinese breeds by their phenotypic traits, Li and Enfield (1989) were able to classify 75 breeds into 6 clusters, mainly based on litter size and body weights. As shown in table 2, the range between extreme type averages is about 8 piglets for litter size (from 6.4 to 14.5) and 209 kg for adult body weight (from 33 to 242 kg).

Table 2 Characterization of Chinese breeds of pigs
(from a cluster analysis by Li and Enfield, 1989)

Among Chinese breeds, the Taihu group certainly deserves priority attention, because of the high prolificacy (going along with corresponding maternal abilities such as teat number) and the early sexual maturity of those breeds. Their merit under Chinese production conditions has been confirmed under intensive husbandry conditions in several other countries: see Legault and Caritez (1983) for one of the earliest evaluation of the Meishan and Jiaxing breeds outside China. These are really “improved” breeds as far as reproduction goes, compared to western breeds: the difference in litter size amounts to about one phenotypic standard deviation (sd), whereas for female sexual precocity the difference exceeds 4 sd, according to a recent French study (Després et al., 1992). Following the French importation of 1979, importations have followed in Hungary, Holland, Japan, U.K. and U.S.A.. Recent results can be found in the Toulouse symposium (Molénat and Legault, 1990).

However, owing to their extremely low potential for lean growth, the benefit from including those breeds into crossbreeding schemes, in order to exploit their complementarity with lean types, is highly dependent on the premium paid for leaner pork by the market. Present market conditions in China, as well as in many developing countries, still make crosses with Taihu breeds quite attractive, whereas in most developed countries, the loss in market value incurred by each individual pig is not compensated by the reduction in cost of piglet production.

According to a recent report by King (1992), the Taihu group includes seven genetically distinct breeds, some of them divided into strains (e.g. Meishan). The group has an estimated 600,000 breeding animals, distributed as shown in table 3, with a tendency to an increase in popularity of the Erhualian breed. None of those breeds is really in danger of extinction. Breeding is well organized and structured, and a clear hierarchy of herds exists. The danger for the future is an increased consumer demand for lean meat, which would put pressure on those breeds for crossing with exotic breeds. Taihu breeds will then be exposed to the risk of genetic dilution of their favorable reproductive characteristics, a risk which may be increased by a massive use of A.I. in the Taihu area.

Table 3 Distribution of the Taihu breeding animals (from King, 1992)

Opportunities for meeting other requirements are also reported in Chinese breeds, such as resistance to extreme cold weather in the Min breed of North China, which appears to combine this favorable trait with high prolificacy. Special food-seeking abilities are mentioned for the Tibetan pig, which combines dwarfism with adaptation to cold, alertness and highly developed digestive organs able to utilize shrubs, stems, roots and seeds of wild plants.

Vietnam, the second largest pig producing country in Asia, relies on both intensive production systems and household production, the latter however representing 80% of the total output. Vietnam also shares with South-East Asia some unique features in the role played by the pig: fat as a source of essential fatty acids, manure as fertilizer for rice cultivation and frequent association with fish production. Vietnam has an abundance of local breeds, several extremely early reproducing, among which the Mong Cai is considered as a genetic type of interest for the future and currently under investigation in several state farms (Molénat and Tran The Thong, 1991).

3.2 Latin America and Caribbean zone

Indigenous (or so called native) pigs of this region actually derive from early importations. According to Epstein and Bichard (1984), pigs were introduced from China in the 15th century, and the colonists in the 16th century brought along both Celtic and Iberian types. A rather detailed survey of indigenous pigs in Latin American has been given by de Alba (1972). To-day indigenous breeds play a role depending on the production system. Beside intensive production based exclusively on improved breeds such as Large White, Landrace or Duroc, Latin America has an important sector of smallholder production, with moderate productivity and low inputs, which relies on local breeds either pure or used in crosses with improved breeds. Backyard pigs raised for self-consumption are also a constant feature of meat production in that region of the world and Criollo types usually serve that purpose: see Le Mentec (1970) and Canope and Raynaud (1981) for evaluations of the Criollo pig of Guadeloupe.

Half of the pigs of that region live in Brazil. This is also the only tropical country in this area which devotes significant efforts to study and to improve its local breeds. In 1986, a project for evaluating national breeds of pigs was started in South Brazil, under the auspices of the Agricultural Research Corporation (EMBRAPA) and of the National Center for Genetic Resources (CENARGEN). A survey covering seven breeds in danger of extinction is currently underway and preliminary results are shown in table 4. State support is provided for four breeds, Moura, Caruncho, Pirapitinga and Piau, and conservation nuclei for the latter breed have been established in the 3 southern provinces of Brazil. Crossbreeding experiments with improved breeds are carried on in the Santa Catarina and Parana provinces. The objective is to obtain more precise information on the specific characteristics of those rather numerous local breeds, their nutritional requirements, their ability to survive in harsh environmental conditions and their possible resistances to diseases.

Table 4. Brazilian pig breeds in danger of extinction (from Mariante, 1990)

Mexico also has several local breeds of interest, generally of small size and well adapted to various climates. The black hairless pig of Yucatan, in addition to being a natural miniature pig of about the same size as other artificially created strains (Panepinto et al., 1978), is reportedly well adapted to hot climates and to bulky diets. The Cuino miniature pig (10–12 kg adult weight), though nearly extinct according to Mason (1988), would deserve being conserved, as it is reported to be able to survive long periods of starvation under household conditions.

3.3 Tropical Africa

The pig population of tropical Africa is about 7.3 million, half of which is located in the coastal region of West Africa, from Senegal to Cameroon. The West African indigenous pig belongs to the Iberian type and includes such varieties as the Ashanti dwarf in Ghana, the Bakosi in Cameroon and the Nigerian native (Mason, 1988). This pig may have migrated from Northern Africa and Egypt (Pathiraja and Oyedipe, 1990), or, according to Epstein (1971) it would derive from early Portuguese imports in lower latitudes.

This type of pig is well adapted to the extensive conditions of the traditional village management, as it gets its food essentially from scavenging. It constitutes a valuable source of meat for the small-scale farmers of that region. In more intensively cultivated areas, indigenous pigs are kept under semi-intensive systems and fed agricultural by-products.

West African pigs grow slowly, produce small litters, partly because of an excessively early sexual precocity with a first farrowing at 8 month of age, but sows reproduce with regularity and may farrow 2.3 litters a year (Pathiraja and Oyedipe, 1990). There are indications that indigenous pigs may exhibit better heat and parasite tolerance than exotic breeds, and also better disease tolerance and trypanotolerance. However, these are field observations which would require the support of experimental evidence.

The best described genetic situation is that of the Nigerian native by Pathiraja and Oyedipe (1990). Between 1975 and 1985, the estimated pig population of Nigeria has increased from 867,000 to 1,050,000 heads, while a steady decline of the percentage of indigenous pigs has been observed, from 89 percent of the total in 1975 to 58 percent in 1985. As a consequence, in several major pig producing areas, indigenous pigs are virtually extinct. The same authors point to a possible contribution of inbreeding to the low productivity observed under traditional management systems, owing to the small size of individual herds.

In addition to Nigeria, where scientific investigations on native and crosses with exotic breeds have been carried on, work done in Zimbabwe (Pig Industry Board) and in Bénin (D'Orgeval, personal communication) should also be mentioned. As pig is a species of relatively minor importance in Africa, the lack of accurate statistics and basic production data, recognized as a major constraint for an adequate management of farm animal genetic resources in this part of the world (Setshwaelo, 1990), applies even more to African indigenous breeds of pigs.

3.4 Other regions

Local breeds of pigs in other parts of the world will be briefly mentioned here for the sake of completeness. In Europe, where each country has its own set of local breeds, the conservation of rare breeds has received considerable attention in recent years as well as reasonable support from various state and private institutions. Similar efforts are made in Canada and U.S.A., through Rare Breed and Minor Breed Conservancy associations. In India, a National Bureau of Animal Genetic Resources has been established and is engaged in conservation and management of native populations, and studies on local breeds of pigs are reported from several research teams. Australia and New Zealand have no local breeds as they rely on imported British breeds, some of which have become feral since their introduction on this continent. Particular mention should be made of the village pig in the civilization of Papua-New-Guinea.

4 Conclusion on priorities

Breeds eligible for priority support of development-conservation action are to be found in the three regions considered above: Asia, Latin America and tropical Africa. China should probably come first as it is to-day the most important and the most diverse gene pool in the world. In Zhao's (1990) words, the pig breeds of China are “invaluable treasures which belong not only to the Chinese people but also to the people all over the world”, and they should play an increasing role in the future improvement of the species. With Chinese experts fully realizing the value of their national breeds, and adequate breeding structures for implementing development projects of sufficiently long term, a most favorable situation exists. The Taihu breeds may be regarded as a valuable genetic material for addressing an important research problem, which is to determine whether lean growth can be improved while maintaining a high level of prolificacy. Those populations also offer an opportunity to check the increase in annual genetic gain accruing from the higher selection intensity/generation interval ratio allowed by their fecundity and sexual precocity (Bidanel, 1988). In that respect, the 10-year project for improving the Erhualian breed proposed by King (1992) should receive due attention. Such trials would gain from being replicated in at least another breed of the Taihu group (Meishan or Jiaxing) and an additional candidate could be the Min breed, which combines the reproductive characteristics of Taihu with more specific qualities of adaptation to cold environments.

The situation in Latin America is very different, as risks of extinction of valuable genetic material exist on a rather short-term. Brazil was among the first countries in Latin America to establish a national programme for conservation and evaluation of native livestock populations (Primo, 1987). This effort deserves recognition by the international community, which should ensure that it can be pursued, in spite of present economic difficulties. The native pigs of Brazil represent a suitable material for studies of survival qualities and general adaptation, of which smallholders can take advantage within production systems with very low inputs, typical of many tropical countries. The Brazilian Piau breed should be a valuable target of investigation, as it is generally considered as one of the best local breed for survival, growth and reproduction. More information is however needed on those breeds. A programme of evaluation of genetic distances can be recommended in order to more objectively justify priorities for conservation programmes.

Tropical Africa may be the region where the danger of extinction of local pig populations is highest, in the aggravating context of a limited number of such populations in the whole continent (see table 1). Measures of conservation of the West African pig are therefore urgently needed. Gene banks can here be recommended, as semen stores can be maintained at a much lower cost than live populations and with better prospects of continuity in most cases. Investigations on disease resistance, in particular to confirm the reported tolerance of local pigs to African swine fever, would be a further justification of the efforts needed to maintain this type of pig.

In practice, strong local support will obviously be needed for any long-term development project and national as well as regional initiatives should receive priority attention. The quality and the authority of the chosen local contacts will also be crucial to the success of any internationally coordinated project. The proposals presented in this report are to be considered as general guidelines, and they will certainly need further elaboration. As noted by de Alba (1972), an enormous gap exists in the annual output of slaughtered pigs relative to the inventory between tropical countries like Brazil, where a large part of the resources used by the pigs serves for herd maintenance, and developed countries like U.S.A. where proper housing and feeding allows a high proportion of the feed to be used to produce lean tissue. One cannot expect such a gap to disappear in a predictable future. A proper management of our genetic resources could however certainly contribute to narrowing it.

5 Acknowledgements

Information provided by C. Legault (Jouy-en-Josas), D. Simon (Hannover) and D. Steane (Rome) during the preparation of this report is gratefully acknowledged.

6 References

Bidanel J.P., 1988. Bases zootechniques et génétiques de l'utilisation en élevage intensif des races prolifiques chinoises. Cas du porc Meishan. Thèse de doctorat, INA-PG, 194 pp.

Canope I., Raynaud Y., 1981. Etude comparative des performances de reproduction, d'engraissement et de carcasse des porcs créoles et Large White en Guadeloupe. Journées Rech. Porcine en France, 13: 307–316.

Cheng P.L., 1984. Livestock Breeds of China. FAO Animal Production and Health Paper n^o46, Rome, 217 pp.

De Alba J., 1972. Productivité des races porcines indigènes et exotiques en Amérique Latine. Revue mondiale de Zootechnie, 4: 25–28.

Delatte J.J., Le Guyadec P., Le Duot P., Duclos J.M., 1991. Croissance et reproduction des porcs rustiques d'origine française selon le milieu d'élevage en Haiti. Journées Rech. Porcine en France, 23: 381–388.

Despres P, Martinat-Botte F., Lagant H., Terqui M., Legault C., 1992. Comparaison des performances de reproduction des truies appartenant à trois génotypes : Large White, Large White “hyperprolifique”, Meishan. Journées Rech. Porcine en France, 24 : 345–350.

Epstein H., 1969. Domestic Animal of China. CAB, England.

Epstein H., 1971. The Origin of the Domestic Animals of Africa. Vol. 2. Africana Publishing Corporation, New York.

Epstein H., Bichard M., 1984. Pig. In: Evolution of Domesticated Animals (I.L. Mason ed.), Longman, 145–162.

Fredeen H.T., 1984. Selection limits: have they been reached with pigs ? Can. J. Anim. Sci., 64: 223–234.

Glodek P., 1991. Evaluation and utilization of pig breeds. In: Genetic Resources of Pig, Sheep and Goats (K. Maijala ed.), Elsevier, 65–93.

Haley C.S., Archibald A.L., 1991. The Pig Gene Mapping Project (PiGMaP). Commission of the European Communities, DGXII, Biotechnology Division, Brussels, 22pp.

Jonsson P., 1991. Civilization and domestication of pigs. In: Genetic Resources of Pig, Sheep and Goats (K. Maijala ed.), Elsevier, 11–50.

King J.W.B., 1991. Pig breeds of the world: their distribution and adaptation. In: Genetic Resources of Pig, Sheep and Goats (K. Maijala ed.), Elsevier, 51–63.

King J.W.B., 1992. Genetic improvement and conservation of Taihu pig breeds in Jiangsu province (draft report).

Lauvergne J.J., 1982. Genética en poblaciones animales después de la domestication : consecuencias para la conservacion de las razas. 2nd World Congress Genet. Appl. Livest. Prod., 6: 77–87.

Le Mentec J.C., 1970. Etude monographique du porc local en Guadeloupe. Bull. tech. Inf. n^o251: 435–446.

Legault C., 1978. Particularités zootechniques des porcs élevés en République populaire de Chine. Bull. tech. Inf. n^o327: 115–125.

Legault C., Caritez J.C., 1983. L'expérimentation sur le porc chinois en France. I. Performances de reproduction en race pure et en croisement. Génét. Sél. Evol., 15: 225–240.

Li M.D., Enfield F.D., 1989. A characterization of Chinese breeds of swine using cluster analysis. J. Anim. Breed. Genet., 106: 379–240.

Mariante A. da S., 1990. Programmes for live animal preservation for Latin America. FAO Animal Production and Health paper n^o80, 119–126.

Mason I.L., 1988. A World Dictionary of Livestock Breeds Types and Varieties. CAB International, 348pp.

Molénat M., Legault C., 1986. Le porc dans les pays en voie de développement. Quelques pistes d'amélioration. Bull. tech. Inf. n^o406: 39–53.

Molénat M., Legault C., 1990. Symposium sur le Porc Chinois. INRA, Jouy-en-Josas, 262pp.

Molénat M., Tran The Thong, 1991. Génétique et Elevage du Porc au Viet-nam. Institut d'Elevage et de Médecine Vétérinaire des Pays Tropicaux, Maisons-Alfort, France, 115pp.

Ollivier L., Lauvergne J.J., 1988. Development and utilization of animal genetic resources. Proc. 6th World Conference Anim. Prod. Helsinki, 85–101.

Pathiraja N., Oyedipe E.O., 1990. Indigenous pigs of Nigeria. Bulletin d'Information sur les Ressources Génétiques Animales, n^o7, 67–68, FAO. Rome.

Panepinto L.M., Phillips R.W., Wheeler L.R., Hill D.H., 1978. The Yucatan miniature pig as a laboratory animal. Laboratory Anim. Sci., 28: 308–313.

Primo A.T., 1987. Conservation of animal genetic resources: Brazil national programme. FAO Animal Production and Health Paper n^o66, 165–173.

Setshwaelo L.L., 1990. Conservation of animal genetic resources in Africa. 4th World Congress Genet. Appl. Livest. Prod., 14: 449–454.

Simon D., 1990. Data banks and the conservation policy. 4th World Congress Genet. Appl. Livest., 14: 423–426.

Steane D.E., 1991. Conservation of genetic resources in pigs and sheep. In: Genetic Resources of Pig, Sheep and Goats (K. Maijala ed.), Elsevier, 145–155.

Sutherland R.A., Webb A.J., King J.W.B., 1985. A survey of world pig breeds and comparisons. Anim. Breed. Abstr., 53: 1–21.

Walters J.R., Hooper P.N., 1990. British experience of A.I. and its use in the conservation of rare pig breeds. 4th World Congress Genet. Appl. Livest. Prod., 14: 480–483.

Zhang Z.G., Li P.T., Chen Y.H., 1987. The Annals of Chinese Swine Breeds. Shanghai Scientific Publishing House (cited by Li and Enfield, 1989).

Zhao Z.L., 1990. The characteristics of Chinese pig breeds. In: Symposium sur le Porc Chinois (M. Molénat, C. Legault eds), INRA Jouy-en-Josas, 56–65.