Adaptation to ecosystems
Husbandry systems of carpet wool breeds and their importance in animal production
Genetic improvement and improved management
The carpet wool breeds of the Mediterranean have coarse, mixed wool fleeces with long straight outer coats and short undercoats. They are of a strong constitution and are naturally adapted to mountainous, hilly, aria or semi-desert regions. These breeds often face serious nutritional problems caused by high temperatures and drought but have developed physiological mechanisms for survival. Despite harsh conditions at times, these breeds provide the bulk of animal protein for the Mediterranean basin, while their milk represents a substantial proportion of world production.
The large category of carpet wool sheep includes a broad spectrum of breeds from the Mediterranean area, having in common a coarse, mixed wool fleece with a long, straight outer coat and short undercoat. In the majority of cases, this wool pattern corresponds to an animal of small or moderate size with a strong constitution, naturally adapted to the optimal use of marginal regions such as mountainous, semi-mountainous, aria or semi-desert areas, and with aptitudes for a mixed production of milk, meat and wool. This particular species of small ruminant, together with the indigenous goat, plays an important role in Mediterranean animal production and has been closely associated with agricultural development and human civilization over the past millennia.
Guided by Mason's classification (1967), the authors have included in this review the carpet wool dairy breeds, the Zackel breeds and the fat-tailed as well as the thintailed sheep. Although they are shown in Fig. 1 and Tables 1 and 2, data and descriptions have been summarized on production and reproduction traits of the main sheep breeds as they are found from the Iberian peninsula to the European, Asiatic and African borders of the Mediterranean basin. The Merino, semi-fine wool, demi-coarse wool and uniform wool breeds are not included, although they undoubtedly contribute a great deal to the ovine population and production of the Mediterranean region. Among the other types omitted are the prolific dairy breeds, originating from small islands and constituting valuable isolated genetic material.
1. Carpet wool sheep breeds of the Mediterranean basin (Europe) - Races à laine à tapis du bassin méditerranéen (Europe) - Razas ovinas de lana tapicera de la cuenca mediterránea (Europa)
1. The coarse wool sheep breeds of the Mediterranean - Races ovines à laine grossière de la Méditerranée - Razas ovinas mediterráneas de lana gruesa
2. Carpet wool sheep breeds in the Mediterranean basin (Asia, Near East, North Africa) - Races à laine à tapis du bassin méditerranéen (Asie, Proche-Orient, Afrique du Nord) - Razas ovinas de lana tapicera de la cuenca mediterránea (Asia, Cercano Oriente, Africa del Norte)
2. Carpet wool breeds as a proportion of total sheep population in the Mediterranean basin - Part des races à laine à lapis dans le cheptel ovin du bassin méditerranéen - Razas ovinas de lana tapicera como proporción de la población ovina total en los países de la cuenca mediterránea
Despite modernization of agricultural production in the Mediterranean basin over the past decades, carpet wool sheep breeds continue to play an important economic role in the agricultural activities of rural societies because of their natural adaptation to marginal rangeland under difficult grazing conditions (Zervas, Boyazoglu and Hatziminaoglou, 1983).
The growing season in the Mediterranean climate is from May to October. As a consequence, sheep production has traditionally been matched to the seasonal pattern of grass-growing with lambing in the spring, and the production of carpet wool sheep is also heavily dependent on grazing. In many cases, almost 90 percent of the nutrients supplied to sheep come from natural pasture. The synchronization of high feed requirements (ewes in lactation) with grass growth in the spring was the result of a long natural adaptation process of these breeds.
High temperatures, often accompanied by extreme drought, caused nutritional fluctuations with regard to quantity and quality of grass and fodder in this xerothermic (arid/subarid) region of the world. This climatic influence on the growth of grass, which is fibrous in nature and at times non-existent, and on the appetite behaviour of the grazing animals, lowers the level of absorption and utilization of dry matter and very often creates serious nutritional problems for these breeds: their productivity (prolificacy, milk production, lamb growth rate) fluctuates according to the nutritional level of the grazing land. Moreover, the region's variable rainfall produces frequent periods of grass shortages, an unsuitable situation for productive sheep breeds. The carpet wool breeds, however, have adapted perfectly to these conditions. By developing several physiological mechanisms that allow them to cope with periods of nutritional deprivation, they have survived and remained productive.
The genetic material of these breeds consists of a multitude of more or less productive populations which make optimal use of the scarce Mediterranean-type vegetation. They adapt well to the contrasting topography; severe climate; dry, rocky, shallow or salty soils; and the degraded mountainous and hilly rangelands resulting from overgrazing in some areas and undergrazing in others. Under these conditions, carpet wool sheep are often the only possible form of animal production. However, the nutritional demands (Flamant and Morand-Fehr, 1982) of pregnant ewes of these breeds reach a maximum when the winter shortage of grass is most severe.
As a consequence, these breeds' reproduction rates are very low on average, as are their prolificacy, fecundity, lamb survival and growth rates - factors which also vary with the severity of prevailing conditions. Nevertheless, the carpet wool breeds are adapted to make optimal use of the classic Mediterranean grazing land, which offers a variety of ecosystems, and to provide the bulk of animal protein for the Mediterranean basin's population (Flamant et al., 1976).
There is a great diversity in the importance of carpet or mixed wool sheep of the Mediterranean area. In the European and eastern part of the basin, the percentage of this type of sheep over the total ovine population rises from a mere 10 percent in the Iberian peninsula to 85 to 90 percent in the Balkans. In the Asian, Near Eastern and North African countries, mixed wool sheep prevail, with an overall share of 80 to 95 percent (see Fig. 2). This distribution pattern results from the fact that the fraction of carpet wool breeds is inversely proportionate to that of the Merino genotype.
The husbandry systems for virtually all breeds show common features and manifest traditional patterns, including constant efforts to secure year-round grazing wherever possible. They consist of basically poor pastures and supplementary feeding in a restricted number of cases. Hand-milking is the general rule and, if it exists at all, housing is only primitive.
Transhumance of the flocks from the lowlands to high-altitude pastures or midway areas has been a common practice in all Mediterranean regions. However, the recent intensive exploitation of the plains for agricultural crops has helped force the flocks away from their traditional winter habitats and toward the edge of plains, where the possibility of grazing is very limited. Cases in point include Corsica and central Greece.
It should be noted that nomadism still exists in many areas of the Near East and North Africa. Statistically speaking, the size of the flocks (50 to 300 head) is rarely higher than the family enterprise potential. The improved Awassi of Israel are not included in this more or less generalized type of management (semi-intensive to extensive). All the elements of an intensive management system (modern housing, feeding based on limited or zero grazing, mechanization) exist in Awassi production. There are a few other cases of sheep production under intensive feeding and housing conditions: for example, the Langhe of Italy.
The carpet wool sheep population includes many breeds which show considerable variation in their reproductive traits, not only among breeds but also among individuals within breeds.
One such important trait is the growth rate of lambs, which varies considerably among carpet wool beeeds but which is generally regarded to be slow. As a consequence, replacement lambs in most of these breeds are usually weaned at two to three months of age, corresponding to a liveweight of approximately 20 kg. Because of this slow maturation, lamb ewes are bred for the first time at an age of 16 to 18 months and quite often at two years of age.
Ewes have a long reproductive season and never a substantial anoestrous period (Folch and Cogne, 1983). Sexual activity undergoes a considerable decline in the spring and most ewes are mated in late summer to early autumn. Consequently, in the majority of cases the lambing season is between February and April. Although the conception rate (fertility) at first mating differs considerably among breeds, it is regarded to be satisfactory and at times high. Lambing percentage (prolificacy) generally ranges from 100 to 130 in the different breeds and depends heavily on seasonal and nutritional fluctuations.
The ratio of breeding rams used during the mating season is less than 5 percent (five rams per 100 ewes). Furthermore, some of the rams used are rather old. In practice, rams are not easily replaced because their value is relatively high and their teeth remain in good condition for a longer period than those of ewes. Old ewes are usually culled at approximately six years, largely because their teeth are by that time worn out by hard grazing conditions. As a result, the average replacement of ewes is 25 percent.
Carpet wool breeds living under xerothermic conditions (high temperatures, low - precipitation and extreme droughts) and kept under extensive management conditions are less prone to diseases that are often experienced by animals raised under more intensive conditions. Non-parasitic respiratory diseases are few and, in most areas, generally unknown. Parasitic diseases caused by roundworms and tapeworms (in lambs) are the major problem encountered in these breeds (Michel, 1976). It is therefore recommended that these breeds be dosed two to four times per year, according to need. with broad-spectrum anti-worm remedies.
Infectious agents causing foetal mortality, abortion, stillborn lambs and ram infertility have been recognized as another major cause of losses in these breeds. Chlamydiosis, brucellosis, vibriosis, salmonellosis, listeriosis, toxoplasmosis, tick-borne fever and border disease have been identified as the most prevalent diseases which cause reproductive losses in these breeds. Lamb deaths occurring during the neonatal period can result from Escherichia coli, Clostridium welchii, rotavirus and coccidiosis.
However, most of these diseases can be controlled and the likelihood of outbreaks can be greatly reduced by introducing appropriate vaccination programmes and correct hygiene in lambing areas for each flock of sheep.
The mixed wool breeds constitute the classical example of triple-purpose sheep (milk, meat, wool). The contribution of any one of these three products is not on its own sufficient in terms of gross revenue for the producer. Therefore, because of its specific quality, limited quantity and the on-off recession in the international wool trade, carpet wool virtually represents a by-product. The skin, on the other hand, contributes very little to the farmer's income: in many instances, for example, in Greece, it has no value for the producer although it is valuable for the buyer of the animal.
Milk. The production of sheep milk in the Mediterranean region reached 4.46 million tonnes in 1989, representing a substantial proportion of world production. The importance of mixed wool breeds for sheep milk production in the Mediterranean basin becomes apparent when it is taken into consideration that these breeds alone produce about three million tonnes of milk, making up over 67 percent of the region's production.
Milk of such origin is considered rich in butterfat and protein content, varying between 6 to 8 and 5 to 5.5 percent, respectively. However, its consumption as a raw product is very limited and its marketing as yoghurt has not been sufficiently developed.
To make the best use of it, the bulk of sheep milk production is transformed into various types of cheese, reflecting the specificity and peculiarity of each producer country or region. The contribution of sheep milk to gross farm/flock income varies widely and is directly influenced by various factors: the level of production; competition from other products such as lamb; regional methods of cheese preparation; and current market prices. In Greece the share of sheep milk in farm income is 50 to 65 percent (Laga, 1986); in France 35 to 50 percent (PATRE, 1986); in Italy 70 to 80 percent (Mason, 1967); and in Turkey 24 percent (Açil and Demirci, 1983).
Meat. The type of lamb produced by the Mediterranean mixed wool breeds differs from one region to another. In the European part of the Mediterranean, milk lamb dominates. This is a lamb weaned and slaughtered after six weeks to two months of suckling. Stemming from the need for early weaning and milking to secure as much marketable milk as possible, this practice is encouraged by the traditional consumer preference of young, small carcasses.
It also results from the inability of the breeds to produce heavier carcasses of desirable quality and without excess fat under traditional feeding conditions. In Asian and African countries of the Mediterranean basin, on the contrary, sheep meat comes from lambs slaughtered at a much heavier liveweight or from mature animals (mutton). The value of this product to the gross income of sheep farmers varies and relates to that of milk production: Turkey. 54 percent (Açil and Demirci, 1983); Greece, 30 to 45 percent (Laga, 1986); Italy, 40 to 60 percent (Mason, 1967).
Wool. The carpet wool type shows great variation in its characteristics: fleece weight 1 to 3 kg, staple length 7 to 30 cm, quality number 36's to 56's, yield 40 to 70 percent. However, its contribution to the gross income appears to be uniform, at least for the European region of the Mediterranean.
Wool does not exceed 5 percent of total farm income. The product's highest value is obtained through its use in local factories and handcraft workshops, for carpet manufacturing or for the production of other similar commodities.
Carpet wool breeds in mountain pasture - Peloponnese, Greece - Races à laine a tapis sur pâturage de montagne (Péloponnese, Grèce) - Razas ovinas de lana tapicera en pasturas de montaña, Peloponeso (Grecia)
Carpet wool breeds grazing in a forest area - Peloponnese, Greece - Races à laine à tapis pâturant sous couvert forestier (Péloponnèse, Grèce) - Razas ovinas de lana tapicera en pasturas en una zona de bosque, Peloponeso, Grecia
Grazing in shrubland, Crete - Brebis sur pâturage arbustif (Crète) - Ovejas en pastura en un área de matorral (Creta)
Carpet wool breeds on poor pasture, Crete - Races à laine à tapis sur parcours pauvre (Crète) - Razas ovinas de lana tapicera en pasturas áridas (Creta)
For approximately 30 years after the 1940s, genetic improvement efforts were mainly focused on the Awassi breed of Israel and the Sarda of Sardinia, Italy, and gave very positive results in increasing milk production. In the latter case, the overall husbandry system did not deviate much from the traditional pattern, except for the implementation of some intensification measures such as the use of mechanical milking and improved management of the natural pastures and hilly grazing lands. In Israel today, there are reported to be 25 000 Awassi ewes, while in Italy there are some 90 000 Sarda ewes.
Following the Roquefort pattern, trials to establish genetic improvement schemes were introduced during the last few decades to the Pyrenean, the Manech and the Basque-Béarn breeds, with 70 000 ewes officially recorded. Proven rams (after progeny testing) contribute substantially to the upgrading of the population. Similar improvement efforts in different stages of development have been undertaken for the Corsican breed (comprising approximately 5 000 controlled milk ewes), the Spanish breeds (Churra, Lacho and Manchega, a total of 90 000 controlled ewes) and the Greek breed, Karagouniko (approximately 20 000 animals).
Needless to say, efforts to implement modern genetic improvement methods apply to only a small percentage of the total population of breeds in the Mediterranean region. It is well recognized that, to tackle problems generally encountered in implementing genetic improvement schemes in sheep and goat production, elaborate technical and substantial financial support is required. Under such circumstances, alternative efforts are channelled toward the improvement of environmental conditions (housing, feeding, sanitation, management), with increased productivity as the main target.
Another important element of intensification is the processing of sheep milk by its transformation into quality cheeses. Such cheeses secure high revenues: for example, Roquefort cheese from Corsican milk; ossau-iraty cheese from Pyrenean breeds, pecorino romano from the Sarda and various Gruyère-type cheeses from Greek mountainous breeds.
These sheep breeds have adapted well to marginal and poor grazing regions, and production generally takes place under difficult environmental conditions. The sheep graze alone or together with other ruminants (goats or even cows), making the best possible use of the region's typical Mediterranean vegetation. In 90 percent of cases, these grazing lands are described as poor and fibrous (chaparral), arid or semi-desert, and thus, are not usually used for any kind of intensive agricultural activity. Furthermore, it should be emphasized that sheep production in the Mediterranean basin, and particularly the traditional exploitation of the dual- or triple-purpose breeds, is closely linked to human history and cultural developments. It is well-known that the various civilizations of the Mediterraean area developed in close association with the olive tree, the vine, the palm-tree and also with sheep and goats (Boyazoglu, 1985a).
The production system consequently should not only be judged by qualitative criteria but its socio-economic role must also be taken into consideration (Flam ant and Morand-Fehr, 1982). Over the past few years, with an increasing tendency in almost all of the Mediterranean countries to abandon poor areas, the particular role played by small ruminants in retaining the rural populations acquires greater significance.
Whatever improvement plans are proposed, the particular nature of the region should be taken into account, including the polymorphism of the environment, the existence of different flora and types of sheep and systems of management as well as the varying technical and scientific methods that will be used to support improvements.
Table 1 shows the fluctuation of milk production of Mediterranean carpet wool breeds between 50 and 150 kg. Exceptions to this rule are the improved Awassi of Israel and the Italian Sarda. with an average milk production of close to 300 kg for the former and 200 kg for the latter. Where milk plays the most important role in the production system, or is at least equally important as lamb meat, attempts were made to upgrade with East Friesian breeds. Because of acclimatization problems, those attempts were not successful (Flamant et al., 1982). Suitable methods for the genetic improvement of these populations (Boyazoglu, 1985b; Zervas, Boyazoglu and Hatziminaoglou, 1983) are:
· selection in nucleus breeding flocks;
· progeny testing in flocks, using natural mating or artificial insemination; and
· progeny testing in stations.
Recently, important selection and improvement schemes have been implemented successfully in France (Pyrenean breeds), in Israel (Awassi) and Sardinia, Italy (Sarda). The schemes are based on effective milk control networks in these areas. Progress in selection and improvement has been made in the Spanish breeds, Churra and Lacho, and recently in the Greek breed, Karagouniko. For other breeds, because of the limited number of controlled animals or the small size of the flock, as is the case for Comisana and Langhe, it is difficult to establish relevant selection programmes. In the great majority of cases, however, where sheep production takes place entirely under extensive conditions, genetic improvement pursued by elaborate means such as progeny testing cannot be justified as a primary objective. Alternative solutions entailing fewer constraints in their implementation should be sought: nucleus breeding units, for instance, the main objective of which is the diffusion of selected rams, descended from above-average parents with desirable phenotypic characteristics (Boyazoglu, 1985b; Zervas, Boyazoglu and Hatziminaoglou, 1983).
Three trends have developed in the Mediterranean basin: production of milk Lambs in the European subregion: production of heavier lambs in the Asian subregion; and the tendency to produce mutton from older animals in the African subregion. This status quo is closely linked to local traditional consumer preferences and will not change drastically in the foreseeable future. The limited quantitative and qualitative meat capacities of carpet wool breeds do not allow for the production of prime 25 to 30 kg liveweight lambs after an early weaning and fattening period. The consumer accustomed to the milk lamb is reluctant to accept such a carcass because of the excess deposition of fat and reduced tenderness (Zervas, Boyazoglu and Hatziminaoglou, 1981).
It is clear that maximizing the meat potential of triplepurpose carpet wool breeds cannot be accomplished on the basis of prototypes used for increasing milk production. It should also be remembered that, in countries where sheep meat selection programmes give fruitful results, the heavy lambs come from specialized meat breeds or large-format breeds of mixed aptitudes and with a substantial percentage of merino blood. Better meat production from carpet wool breeds could be obtained by making optimal use of the existing variability in carcass quality; gradually increasing the average liveweight of slaughter lambs and implementing suitable fattening regimes; and the use of industrial cross-breeding. Encouraging results have been obtained by crossing meat rams and indigenous Mediterranean ewes, although there are many issues to be settled before the large-scale implementation of such schemes becomes possible (PHILOETIOS, 1981).
The wool originating from carpet wool breeds is essentially a by-product, contributing very little to the gross income of the enterprise. Nevertheless, wool production plays an important role in the economy of regions where carpet production represents a traditional occupation. In fact, in addition to the dynamic carpet industry found in Turkey, wool craft industries flourish in various regions such as Sardinia, Morocco, Greece, Portugal, Yugoslavia and Egypt. One of the main factors that has contributed to the development of this activity has been the increase in tourism and a subsequent growth in demand for woollen products (Boyazoglu, Zervas and Hatziminaoglou, 1985). In these countries, the major controversial issue regarding carpet wool, low-priced as an animal product but valuable as an industrial raw material, is the fact that the carpet and other traditional industries of woollen commodities often import raw material from overseas countries of the Southern Hemisphere. The renewed preference for natural wool fibre in fashion and the rising prices of synthetic fibre will eventually strengthen the importance of production of mixed wool breeds without, however, their becoming of major economic significance. Given certain circumstances, selection within pure breeds to increase wool production could be a possible solution. More radical measures such as the introduction of merino blood are not recommended because they would effect an unwelcome alteration of the characteristics of these breeds.
In the broader Mediterranean area, and under conditions that may definitely be described as difficult, carpet wool breeds that are very well adapted to the environment play a particular role from an economic, sociological and cultural point of view. Consequently, production of these breeds appears to be an important factor in maintaining the stability of what is a very sensitive ecosystem.
Within a context where new scientific accomplishments are adopted with difficulty, costly genetic improvement schemes have their place only where appropriate techno-economic support is available and where goals are very clearly defined. The gradual implementation of simple improvement methods, such as the selection of nucleus flocks, could constitute an alternative means of genetic improvement.
The improvement of environmental conditions seems to bring about the most spectacular results in increasing productivity. These efforts should concentrate on problems relating to the improvement of grazing lands and hygienic conditions of the ovine populations. Certainly, the intensification of management techniques, wherever possible and economically sound, could also help a great deal in obtaining favourable results. Such techniques may include the mechanization of milking; improved housing; the increased use of supplementary feeding; and the maximum exploitation of sheep milk through its conversion into high-quality cheeses. The recently introduced techniques of hormonal manipulation can also contribute positively to increased productivity of the multiple-purpose carpet wool sheep.
Açil, F.A. & Demirci, R. 1983 The productivity and profitability of sheep breeding in Turkey Proc. Int. Symp Prod. of Sheep and Goats in the Mediterranean area, Ankara, 1983 p 350-368. EAAP.
Associazione italiana allevatori. 1985. L'ovinicultura italiana nel contesto mondiale ed europeo: possibilità di sviluppo. Editcoltivatore.
Associazione nazionale della pastorizia. 1972. Razze: consistenza, distribuzione. Indagine nazionale su alcuni aspetti degli allevamenti e delle produzioni ovine. Rome, Ministero dell'Agricoltura e delle Foreste (MAF).
Associazione nazionale della pastorizia. 1984. L'allevamento ovino. Rome, MAF.
Associazione nazionale della pastorizia. 1984. Le principali razze ovine e caprine iscritte ai libri genealogici nazionali. Rome, MAF.
Boletim pecuário. 1981 Produção de ovinos e caprinos em Portugal Ano, XLVII. Lisbon.
Boyazoglu, J.G. 1985a. Breed differences in productivity: the milk breeds. World Anim. Sci., vol. 72.
Boyazoglu, J.G. 1985b Milk recording, breeding and selection schemes for milk breeds. World Anim. Sci, vol. 72.
Boyazoglu, J.G., Zervas, N. & Hatziminaoglou, J. 1985. Sheep production and breeding in the Mediterranean basin. Strategies for the future. Int. Arid Land Res. and Dev. Conf., Tucson, AZ, USA.
Epstein, H. 1977. The Awassi sheep in Israel. World Rev. Anim. Prod., 13: 19-26.
FAO. 1977. Mediterranean cattle and sheep in cross-breeding. Report of the first FAO Expert Consultation on Breed Evaluation and Cross-breeding. Animal Production and Health Paper No. 6. p. 30-35. Rome, FAO.
FAO. 1985. FAO Production Yearbook 1984, volt 38. Rome, FAO.
Flamant, J.C., Boyazoglu, J.G., Casu, S., Espejo, M., Valls Ortiz, M. & Zervas, N. 1976. Inventaire et exploitation du patrimoine héréditaire des populations ovines méditerranéennes. Options Méditerranéennes, (35): 57-65. (Institut agronomique méditerranéen de Zaragoza.)
Flamant, J.C., Boutonnet, J.P., Dyrmundsson, D., Jankowski, S., Morand-Fehr, P., Robinson, J.J., Treacher, T. & Valls Ortiz, M. 1982. Sheep and goats. Livestock Production Science, 9(1/2): 163-196. Amstersdam, Elsevier Scientific Publishing Company.
Flamant, J.C. & Morand-Fehr, P. 1982. Milk production in sheep and goats. In Sheep and goat production. p. 275-295. Amsterdam, Elsevier Scientific Publishing Company.
Folch, J. & Cogne, Y.1983. Caractéristiques reproductives des races ovines méditerranéennes. Proc. Int. Symp. Prod. of Sheep and Goats in the Mediterranean Area, Ankara, 1983. p. 66-90. EAAP.
ITOVIC (Institut technique de l'élevage ovin et caprin). 1985. Les races laitières des Pyrénées. Races ovines, p. 236-242.
ITOVIC. 1985. La race Corse. Races ovines, p. 148-150.
Laga, V.1986. Systèmes d'élevages ovins laitiers transhumants de Grèce centrale. Résultats des élevages et performances de la rece Vlachico (thesis).
Mason, I.L. 1967. Sheep breeds of the Mediterranean. Farnham Royal, UK, CAB FAO.
Michel, J.F. 1976. The epidemiology and control of some nematode infections in grazing animals. Advances in Parasitol., 14: 355-397.
Ministerio de Agricultura, Pesca y Alimentación. 1985. Catálogo de razas autoctónas españolas. 1-Especies ovina y caprina. Madrid, Dirección general de la producción agraria.
PATRE. 1986. La revue de l'élevage ovin. No. 339.
PHILOETIOS. 1981. Résultats de l'expérience de croisement industriel entre races bouchères nord-européennes (Charmoise, Ile-de-France, Texel) et les races locales méditerranéennes pour la production d'agneaux de boucherie, réalisée en coopération entre cinq équipes de recherche. Options Méditerranéennes, 111:15-37. (Institut agronomique méditerranéen de Zaragoza.)
Rosenfeld, P.P. 1982. Israel's dairy industry. Israel Dairy Board.
Rubino, R., Nardone, A., Carena, A. & Matassino, D. 1983. I sistemi produttivi ovino e caprino nel mezzogiorno d'Italia. Prod. Anim. 2: 125-1 67.
Yalcin, B.C. 1979. The sheep breeds of Afghanistan, Iran and Turkey. FAO/UNEP, Project FP/1108-76-02 (833).
Zervas, N., Boyazoglu, J.G. & Hatziminaoglou, J. 1981. Croisement des brebis Serres et Karagouniko avec des béliers frisons (données préliminaires). Options Méditerranéennes, 111: 53-60.
Zervas, N., Boyazoglu, J.G. & Hatziminaoglou, J. 1983. The potential of Mediterranean sheep breeds for milk and meat production: strategies of improvement. Proc. Int. Symp. Prod. of Sheep and Goats in the Mediterranean area, Ankara, 1984. p. 1-23. EAAP.