Last update: December 2, 2010 08:41:58 AM E-mail Print


Adaptations in the South African Merino industry

to successfully compete with mutton breeds



1 ARC: Animal Improvement Institute, Middelburg, 5900, South Africa.

2 Elsenburg Agricultural Development Institute, P/Bag X1 Elsenburg, 7607, South Africa.



The diminishing ratio between clean wool and mutton prices forced the South African Merino breeder to put more emphasis on mutton production in relation to wool production. This was achieved, firstly, by selecting for plain bodied sheep, which have higher growth rates and are more fertile than developed types, especially under adverse environments. Higher reproduction and growth rates were also achieved through indirect selection for higher body weights or direct selection for fertility. Evidence is presented that substantial genetic progress can be made in reproduction without serious disadvantages to other production traits. In order to select a hardier animal, more and more rams are being reared and selected under natural production environments. Currently there are 16 veld ram clubs and 27% of rams sold during 1997 at public auctions, were from these clubs. In order to increase the efficiency of production, scientific breeding policies are adopted by ever increasing number of stud breeders. The data of animals from 18 studs, with genetic ties between them, were recently used for the first national multi-trait BLUP analysis. The mechanisms for further genetic evaluation of Merinos, including reproduction, are well in place and will ensure that Merino sheep can successfully compete with dual purpose and mutton breeds.



The Merino is the major sheep breed in South Africa, but its numbers have declined substantially over the last few decades relative to other breeds. During the period 1960 to 1964 the Merino represented 78% of all sheep in South Africa. The comparative percentages for non-wool and other wool sheep were 12% and 10% respectively (Abstract of Agricultural Statistics, 1996). For the period 1990 to 1994 the Merino represented 59%, non-wool sheep 25% and other wool sheep 16% of the total sheep population in South Africa. It could be argued that external factors such as a changing farming environment (more irrigation) or climatic changes are related to this change. The main reason, however, is the diminishing price ratio between wool and mutton. Changes in the ratio between clean wool and mutton prices over the past 40 years are illustrated in Figure 1.


The ratio declined from a maximum of nearly 5:1 during 1956 to a minimum of 1.5:1 during 1971. After that the ratio fluctuated around 2:1 with a maximum of 3.3:1 during 1988 and a minimum of 1.4:1 during 1992. The effect of a varying wool: mutton price ratio on the relative importance of mutton production versus wool production is illustrated in Table 1. The relative importance of higher nett reproduction rates (Iambs weaned per ewe mated) in times of a narrow wool: mutton price ratio is obvious. With a 70% weaning percentage, the clean wool: mutton price ratio must be near to 2.5:1 or higher, before wool contribute 50% or more of total income. In Table 2 the gross margin of a Merino farm in the Karoo region of South Africa with a carrying capacity of 1600 SSU is compared to a mutton production unit, with the same carrying capacity, at different reproduction rates and wool: mutton price ratios. An increase of 20% in reproduction resulted in a 5-7% increase in gross margin within the Merino breed. At a price ratio of 1.5:1, a mutton enterprise had a gross margin of between 9 and 25% higher than that of the Merino enterprise. With a price ratio of 2:1, mutton sheep must have a 20% higher reproduction rate than Merinos to break even. This result was obtained from average production data and will differ from farmer to farmer. From these results, however, it is clear that the Merino enterprise should have achieved a reproduction rate of higher than 90% to successfully compete with a mutton production enterprise over the last 10-20 years.



According to De Klerk et al (1983) the average number of Iambs weaned per ewe mated in Merino flocks in South Africa during 1982 was 70%. Merino farmers thus had to increase their reproduction rate to financially survive the low wool: mutton price ratios of the last number of years. To complicate wool farming further, the main wool producing areas were also affected by severe droughts, putting further stress on a ewe flock, and resulting in lower reproduction rates. In an economic environment that favours mutton production, this aspect had a marked negative effect on Merino farming enterprises.



The saying, "adapt or die" has become very relevant in the Merino industry (and specifically the stud industry) over the last 10 years. With a 70% weaning rate the number of surplus animals available to be sold is only 10%. It was therefore of the utmost importance to increase the reproduction rate of Merinos. Supplementary feeding can largely enhance reproduction. Due to high costs of concentrates the only viable way was to change the Merino genetically. The adaptations in the Merino industry, in order to compete with other dual purpose and mutton breeds, could be categorised into three areas. The first was to increase the reproduction rate of the animal, the second, by selecting under natural production environments and thirdly by applying more scientific breeding policies.


Changing the reproduction rate.

The negative relationship between skin fold score and reproduction in Merino sheep, especially in an adverse environment, is well documented (Atkins, 1980). It was therefore no surprise to notice a trend within the stud industry towards a plainer animal. This change towards a plainer South African Merino was documented by Poggenpoel & Van der Merwe (1987) who found a reduction of 29% in skin fold score in four studs over an average of 6.8 years of selection. This fact is further demonstrated in the genetic trend for skin fold score in 18 studs participating in the national Merino breed analysis (Figure 2).

In order to improve the mutton component of the enterprise, more emphasis was put on body weight than on fleece weight during selection. As reviewed by Brien (1986), body weight is generally positively correlated to reproduction.

Snyman et al (1997b) found genetic correlations of 0.60 to 0.80 between two-tooth body weight and weight of Iambs weaned after three parities in three Merino experimental flocks. It can be concluded that the mutton production side of Merino farming was emphasised through direct selection for body weight and indirect selection (through higher body weights) for a higher reproduction rate. Herselman et al (1997) illustrated in three selection flocks that the amount of clean wool per unit of body weight is negatively correlated with the total weight of Iambs weaned over three parities. The type of Merino favoured by most commercial breeders is a big plain-bodied animal (higher reproduction and growth) with finer wool.

Little scientific evidence is available to prove this trend and its result on reproduction rates. It is, however, accepted that a weaning percentage (Iambs weaned per ewe mated) of 85-90% is easily achieved under natural environments. The question is raised whether this is due to improvement in management and the environment, or rather to the genetic improvement of reproduction rate. The answer is most probably that both these factors played a role. The effect of better nutrition on reproduction is well documented. The potential of genetic improvement of reproduction is illustrated by the following experiment:

A divergent selection experiment for ewe multiple rearing ability was started during 1985 at the Tygerhoek experimental station near Riviersonderend in the Southern Cape of South Africa. Ewes from the same base population were randomly allocated to one of two lines in 1985. Ewe and ram replacements in these lines were subsequently selected on ranking values for maternal multiple rearing ability (Cloete & Durand, 1994; Cloete et al., 1997) in a High ( ) and a Low (-) line. In the line, replacements were preferably descended from dams that reared more than one Iamb per lambing opportunity (i.e. reared twins at least once). Progeny of dams that reared less than one Iamb per mating (i.e. failed to Iamb or lost all progeny born at least once) were preferred in the -line. Since their establishment in 1985, the two lines were maintained as a single flock.


These results accord well with those published by Atkins (1980), showing that purposeful selection of Australian Merinos for reproduction led to similar advantages. It also confirms earlier work (Cloete & Durand, 1994) on another South African Merino flock.

It is important to consider correlated responses in any selection programme aimed at enhancing the economic viability of sheep farmers. In this experiment, averaged EBV's for two-tooth live weight in the line increased by 0.67% p.a. (r=0.55) relative to mean - line performance (Figure 5). Fold score was reduced by 2.45% p.a. (r=0.90) in the line, compared to mean line performance. Clean fleece weight and fibre diameter was not affected to the same extent. The trends for live weight, clean fleece weight and fibre diameter are consistent with genetic correlations discussed by Snyman et al (1997b). Atkins (1980) similarly found that selection on fold development had an adverse effect on ewe reproduction.

From these results it is clear that there is considerable potential for the Merino to increase its reproduction rate without serious negative effects on other qualitative and quantitative production traits. It is also clear that one cannot expect a high wool production and reproduction rate in an environment with limited resources. The relative importance of reproduction, in comparison to wool weight as well as higher price premiums of finer wool, is the main motivation why the South African breeder (and flock farmer) prefers a large animal (reared under natural production environments) with finer wool.


Selection under natural environments

The second area of change was the raising of stud rams under natural production environments. This trend began by the establishment of the first veld ram club during 1984. In a typical veld ram club, young rams enter at an age of 4-6 months with a starting weight between 30-50 kg. The rams are then tested for a minimum of 180 days on good veld (mostly without any supplementation). Traits measured at the end of the test phase included body weight, average daily gain, clean fleece weight, mean fibre diameter, clean fleece yield and staple length. The growth of veld ram clubs over the last decade was exceptional and presently there are 16 of these clubs, co-ordinated by the Merino Breeders Society. From Figure 6 it is also clear that the sale of rams through veld ram clubs are increasing, while that of official rams sales are declining. Currently, approximately 27% of rams sold on public auctions were from veld ram clubs.

Veld ram clubs have played a major role in changing the perceptions of commercial and stud farmers namely, that a good ram has to weigh 120 plus kg and shear over 20 kg of wool. Presently many sires in leading studs were bought at a veld ram sale. These clubs also played a very important role in the understanding and interpretation of performance-tested results by commercial and stud breeders. Rams sold on veld ram clubs usually appear robust (virile) after some time on the buyer's farm and do not need to be pampered. The growth rate of the rams in natural environments has recently also been receiving more attention during selection of stud as well as flock rams.

The change to veld-reared animals also leads to the establishment of so called veld-shows. Rams and ewes compete in different weight and age classes. A prerequisite is that the animals must not be prepared in any way (trimming of fleeces, etc.). Interest in this type of show is increasing due to the fact that stud and flock farmers compete on equal grounds. This type of show is very important in establishing a closer collaboration between flock and stud farmers and to narrow the gap in differences in breeding objectives between stud and commercial farmers.

Demonstration projects with ewes from different farmers on a central property (measuring the financial output obtained from wool and mutton) also played an important role in changing the perspectives of many farmers. It was generally believed that the most economical sheep should have certain subjective characteristics. In numerous ewe projects it was illustrated that an economical ewe was the one that successfully rear her Iambs regardless of the amount or the fineness of the wool. The economic differences between individual sheep belonging to the same participant (within flock variation), was greater than the differences between the means of ewes belonging to different participants (between flock variation). This shows that there is ample scope for genetic improvement in production traits on a within flock, as well as on a between flock basis.


Application of more scientific breeding policies

Scientific breeding plans were initially adopted by only a handful of breeders. However, many stud breeders apply the different genetic improvement programs run by the Department of Agriculture and currently by the Animal Improvement Institute (All) of the ARC. The South African Fleece Testing Centre was established during 1966 and the National Woolled Sheep Performance and Progeny Testing Scheme during 1970. Currently 53% of all stud breeders participate in this Scheme. The benefits of applying a more scientific breeding plan in a traditional Merino stud were clearly demonstrated by Olivier et al (1995).

During the last 20 years the Merino Breeders Society implemented plans to accurately identify the between-flock genetic variation in production traits. These included the National Show for measured production (Grove et al, 1990), the evaluation of studs against an unselected control flock (Poggenpoel & Van der Merwe, 1987), veld ram clubs and progeny testing schemes. All these projects had their shortcomings which limit the impact on the stud industry. With BLUP most of these shortcomings are negotiated. In order to make use of this technology, the centralisation of data on a structured manner become very important.

It is for this reason that the Animal Improvement Institute (All) of the ARC, in conjunction with SA Studbook and the breed society, established and maintained a centralised integrated computer data basis. Parentage and performance data are captured by the breeder, either on computer or paper, and sent to the regional offices of the All. The data are then verified and sent to the INTERGIS (Integrated Genetic and Information System) where the data are stored. Within flock indices are calculated and send back to the farmer. Twice a year a National multiple-trait BLUP analysis is done on body weight, clean fleece weight, fibre diameter and pleat score. Depending on the data, weaning weight (direct and maternal), as well as total weight of Iamb weaned per ewe, will also be included. The availability of interim within flock EBV's is currently under investigation.

Eighteen Merino studs participate in the registration and performance-testing scheme through the INTERGIS. All these studs are linked genetically. A multiple-trait BLUP analysis was done, using these data, to obtained EBV's for approximately 38000 animals. The genetic trends in the measured traits are illustrated in Figure 7. From these trends it can be concluded that genetic progress is very slow. One of the reasons for this slow response on a national basis, is the fact that not all the breeders have the same breeding objectives.

The real value of a sire lies in the value of its progeny. The Merino Breeder's Society, in collaboration with the Department of Agriculture and AI stations started a progeny-testing scheme during 1985. Apart from most of the production traits measured and evaluated in this scheme, linear scores on 11 type traits were also used to calculate EPD's. A ram must have at least 15 progeny to qualify for a Golden AI Ram.

Furthermore, to qualify for golden merit status, a ram should not have more than 15% of its progeny culled for type or other non measurable traits like kemp, hairy fleeces, etc. To link the different tests, each new test should make use of link sires. To date more than 270 rams were evaluated. Stud as well as flock farmers in ever increasing numbers use these progeny tested rams.



The changing of the appearance of animals, or rearing them under natural veld conditions, will not ensure the long-term survival of the Merino breed in South Africa. The commercial farmer sells products, the quantity and quality of which (as well as price) determines his financial survival. The genetic improvement of the efficiency of production is therefore of utmost importance, especially in the light of changing environments and fluctuations in product prices.

In the past, many fruitless discussions have gone into the pros and cons of fine wool versus strong wool. The amount of wool produced per unit of fibre diameter is of greater importance in the decision-making process. The results of veld ram clubs and ewe projects have shown that there is ample scope for improvement in this regard. Measurement and correctly implemented performance results can only assess this. The genetic trends in production traits of the national flock (Figure 7) demonstrated that some progress had been made in this regard. The viability of producing fine wool economically under the wide-ranging environments of South Africa is currently under investigation.

On stud level, reproduction has been largely ignored as a selection criteria for the last half century. Work in the Western Cape on the Merino suggests ample scope for the genetic improvement of reproduction in a purposeful selection programme. The challenge for the stud breeder and animal scientist is thus to construct breeding plans in which reproduction will receive proper attention. Snyman et al (1997a) proposed total weight of Iamb weaned per parity as a useful selection criteria for ewe selection. This proposition is well complimented by the results provided in Figures 3 and 4. The low heritability of reproduction traits and the fact that these traits can only be accurately measured at three or more years of age, does complicate selection. Even if there is little chance for accurate selection on reproduction rate, it is important to monitor it in order to identify possible disadvantageous correlated responses to selection for production traits. The National Performance Testing Scheme has been adapted to facilitate the recording of reproduction rate and to assist the breeder in his selection (or monitoring) of this important trait.

The mechanisms for the genetic evaluation of economically important traits on a national basis are well in place in the South African Merino industry. The correct application of this technology over a broad base within the Merino stud industry will not only greatly improve the efficiency of genetic improvement of individual studs, but also the industry as a whole.



Atkins, K.D. 1980. Selection for skin folds and fertility. Proc. Aust. Soc. Anim. Prod. 13:174-176.

Brien, F.D. 1986. A review of the genetic and physiological relationships between growth and reproduction in mammals. Animal Breeding Abstracts 54:975-997.

Cloete, S.W.P & Durand, A. 1994. A preliminary note on the reproduction of Merino sheep subjected to divergent selection on maternal values for Iambs weaned per ewe joined. S. Afr. J. Anim. Sci. 24:27-29.

Cloete, S.W.P., Scholtz, A.J. & Ten Hoope, J.M. 1998 A note on separation from one or more lamb(s) in Merino lines divergently selected for ewe multiple rearing ability. App. Anim. Behav. Sci., in press

De Klerk, C.H. & Duvel, G.H.,Terblance, E. Le F. 1983. Investigation into wool sheep farming in the RSA. SA Institute of Agricultural Extension, Pretoria

Grove, A.C., Poggenpoel, D.G., Van der Merwe, C.A., & Delport, G.J., 1990. Merino championships for measured production. Proc. 2nd World Merino Conference, Pretoria. Poster paper no 18.

Herselman. M. J., Olivier, J.J. & Snyman, M.A. 1997. Studies on small ruminant breeds with inherent differences in fibre production potential and ewe productivity. 1. Relationship between ewe productivity and wool production potential. S. Afr: J. Anim. Sci. (In press)

Olivier, J.J., Erasmus, G.J, Van Wyk, J.B. & Konstantinov, K. V. 1995. Response to selection on the BLUP of breeding values in the Grootfontein Merino stud. S. Afr: J. Anim. Sci.25:13-15.

Poggenpoel D.G. & Van der Merwe, C.A. 1987. Selection response with index selection in three commercial Merino flocks. S. Afr. J. Anim. Sci. 17:70-73.

Snyman, M.A., Olivier, J.J., Erasmus, G.J. & Van Wyk, J.B. 1997a. Genetic parameter estimates for total weight of Iamb weaned in Afrino and Merino sheep. Livest. Prod. Sci. 48:111-116

Snyman, M.A., Cloete, S.W.P. & Olivier, J.J. 1997b. Genetic and phenotypic correlations of weight of Iamb weaned with body weight, clean fleece weight and mean fibre diameter in three South African Merino flocks. Livest. Prod. Sci., submitted.