Last update: August 17, 2011 01:39:19 PM E-mail Print




1University of Fort Hare

2Grootfontein College of Agriculture



The composition of the South African wool clip has changed dramatically over the past forty years. Fine wools dominated the clip during the fifties but have been reduced to insignificant levels during the eighties (figure 1). Not only did the percentage of fine wool decline but there was an alarming increase in the 23 micron and stronger wools recently (figure 2).

This came about primarily because it paid to produce stronger wool (Jordaan,1969, Minnaar,1969). There has always been a strong perception that increasing fibre diameter was a sure way of increasing production, and this is borne out by recent research. Earlier studies such as De Klerk & de Lange (1970) did not support this notion. They investigated the phenotypic relationship between fibre diameter and wool production on 892 unselected two tooth Merino ewes from two experimental flocks and five farms in the southern Orange Free state. It was found that production increased by only about one-tenth of a kilogram for everyone micron increase in fibre diameter. This is a relatively weak association and furthermore individual ewes' production scattered widely around the trend line. Furthermore, the genetic correlation between clean fleece mass and fibre diameter is low, usually less than 0,2 (Erasmus 1989, and Rogan 1984).



More direct evidence for the existence of a much stronger association between production and fibre diameter than referred to above comes from calculating the regression of BLUP breeding values for clean fleece mass on BLUP breeding values for fibre diameter. This was done for all individuals born in the Grootfontein stud from 1966 to 1985 (N = 5347) after correcting for genetic trend. The regression coefficient was found to be 0,49 + 0,043 with a correlation coefficient of 0,15. These results suggest that there is a much stronger underlying trend than formerly expected, although individuals still scatter widely around the trend line. The trend being that clean fleece mass changes by almost half a kilogram for everyone micron increase in fibre diameter.

Confidence in these results are strengthened from two other sources. Poggenpoel (1984) found a correlation between flock means of 0,64 with regard to clean fleece mass and fibre diameter. This is a fairly high correlation. Also, in the experiment reported on by Bezuidenhout (1991) and discussed in paragraph 3 below, it was found that the Grootfontein ewes produced 7,5 kg clean wool with a diameter of 25,8 micron compared to 4,4 kg and 21,1 micron by the Finewool flock under the same conditions. This represents a reduction of 0,66 kg per micron. The corresponding figure with regard to the progeny of these two flocks was 0,64 kg per micron.

During the sixties the late Mr JC de Klerk became concerned about the trend towards strong wool in the South African clip and wrote to many buyers to ask their opinion on this. The following are a few pertinent extracts from some of their replies (de Klerk & de Lange, 1970).


- Port Elizabeth Wool and Mohair Buyers Association Ltd:

"…. we buyers are of the opinion that the breeding policy undertaken by growers has been to a very great extent responsible for the present trend. We are well aware that breeders and growers of Merino wool have been campaigning towards the production of bigger bodied sheep, producing heavier weights of fleece wool and we agree that such a policy has served the grower well over the last decade or so, but - and this is a big "but" - we consider that this campaign had gone too far and the production of the type of wool for which this country has been famous for over so many generations has been substantially reduced to a level which should not on any account be permitted to continue."

- Martin & Co (Bradford):

“Your particular problem is very dear to my own heart…. However we well appreciate the difficult situation for growers of merino wool during the past ten years or more….. Consequently the tendency has been for farmers to overcome this problem by increasing the weight of fleece per animal by growing stronger wools…. We believe that the fine cloth trade in the world will continue to expand and that synthetics are not a threat to that particular end product if you take the other end of the scale and look at 60/64's etc. you find that these qualities are produced in quantity in many other countries…. So, therefore, any grower in the Cape of such qualities is going to have to stand up to far more competition than if he were concentrating on the finer end of the merino structure….South Africa can find themselves with a relatively high cost of production but small return due to the price they receive being a world price and not a special one for that particular market as we believe is the case in the finer end."

- Top Manufacturers Firm, USA:

"It is understandable why the breeders and growers want to produce bigger bodied sheep which will also produce heavier weights of fleece wool but from our viewpoint we feel that the trend has gone too far and will take many years to collect."

- Japan Wool Spinners Association, Tokyo:

“To meet the demand for wool products in higher quality and fashion for lighter weight wear, finer quality wools are bought by all users. 64’s and 66's and finer wools are consequently in keen demand, the latter is especially in South Africa. Against this trend in requirements, to our great regret, your wool clip is getting stronger. Broader quality wools are also required, but they are more available in quantity and variety in other wool sources than your countr. "

The message was therefore consistent: South African wool growers, stay with your speciality product, namely fine wool. Don't join the mass-production sector of the spectrum of merino type wools. Unfortunately their arguments were clad in terms of market share and not price. Although a small price difference in favour of fine wool did exist at the time this alone was not a strong enough incentive to prevent the movement towards strong wool. Warnings by Verbeek (1969), Marx (1981; and various others, went largely unheeded.

The situation changed dramatically in the eighties. The price gap between fine and strong wools widened spectacularly (figure 3) and suddenly there was a huge demand for finewool rams.




A general widespread reduction in fibre diameter over a broad front in the stud industry will take time. Firstly because the industry is no longer dominated by a small group of breeders from the same geographic region as was the case during the transition towards strong wool (Erasmus, 1977). The same dedicated effort towards achieving a specific goal might therefore be more difficult to realize. Secondly, although fibre diameter has a high heritability and should therefore respond well to within flock selection, the genetic variation is generally only about half of that observed for traits like body mass or fleece weight. In other words the opportunity for selection is much lessened.

Actual standard deviations of breeding values were calculated from data from the studies reported by Olivier (1989). The data consisted of animal model BLUP estimates of breeding values of 2695 eighteen month old rams and ewes, corrected for fixed effects and genetic trend. The standard deviation of breeding values were found to be 2,0 kg for body mass, 0,29 kg for clean fleece mass, 0,78 micron for fibre diameter and 0,5 units for fold score. The genetic variation for fibre diameter obtained in this manner is slightly higher than the expected value of 0,62 assuming a heritability of 0,45 and multiplying that with the phenotypic standard deviation found by Poggenpoel (1984).

The variation in breeding value therefore suggests that in practice one could probably at the very best hope for a genetic selection differential of approximately two standard deviations or 1,5 micron. With ewes it is not feasible to measure fibre diameter objectively so that the expected response per generation is likely to be no more than about three-quarters of a micron.



Van der Merwe and Poggenpoel (1990) reports on the differences in breeding value of 40 different merino flocks by comparing the progeny of home-bred rams with those of rams from the Typerhoek control flock. In this particular group of flocks the maximum difference between means is about three and a half micron. Unfortunately, according to Poggenpoel (1990) the correlation between phenotype and breeding value in the case of these flocks was only 0,27. In this particular case identifying the flocks with the potential to breed fine would therefore pose a major problem if they had not been tested against the control flock. Results discussed in the following paragraph do however suggest that where fineness was the primary selection objective, such flocks tend to breed relatively fine even on a very high plane of nutrition.



The history of this flock is described by Anonymous (1990). Briefly, four fine wool Australian rams were bought and mated to 444 fine wooled ewes purchased from 30 different South African breeders from all over the country. These breeders were identified in the following manner. With the help of the NWGA the finest clips in the different harbours were identified and the owners approached to make ewes available. Ewes were subjectively classed, fibre diameter was objectively measured on mid-rib samples, and the finest ewes were bought. The numbers bought varied from 10 to almost 20 per stud, except in one case where 54 ewes were bought from the same person. After fifteen months on irrigated pasture at the Halesowen Research Farm near Cradock the mean fibre diameter of the ewes was 21,19 micron. The four Australian rams measured 19,1, 19,1, 19,4 and 20,7 microns respectively under Grootfontein management conditions.

This Finewool flock (F) was used in different trials conducted on the veld at Grootfontein and on irrigated pastures at Cradock (lucerne, small grains, ryegrass). They were compared with two other flocks: a flock representative of the Grootfontein stud (G) and a flock consisting of selected extra-strong ewes (S) from the same stud. Table 1 contains some interesting data for both the ewes and their progeny from different crossings.


Table 1.  Mean fibre diameter in micron of the Fine (F), Grootfontein (G) and strong (S) flocks on veld and irrigated pastures (1990 data).

F ewes 21,19 ±1,64  
E ewes 26,19 ±1,37  
S ewes 29,07 ±1,62  
FxF progeny 20,17 ±1,18 *  
GxG progeny 23,24 ±1,27 * 21,1 **
FxS progeny 21,92 ±1,07 *  
FxG progeny   19,1 **


*     10 months old

**   12 months old


Two very important observations can be made about these results. First, it is very encouraging to note that the upper limit to fibre diameter in the fine wool flock seems to be about 21 micron and this on a very high plane of nutrition. This tendency to resist becoming very strong is further demonstrated in figure 5, which clearly shows that even change with age is not so pronounced in the fine wool flock than in the Grootfontein and strong wool flocks. Fibre diameter increased from 6 months of age to 22 months old by 6,3 micron in the G flock, by 3,7 micron in the FxS flock and by only 1,8 micron in the F flock (Bezuidenhout,1991).



The second striking feature observed was that the oft-repeated warning against extreme selection pressure for fineness because of its supposed effect of increasing variability in the offspring, is not substantiated by the data. The coefficient of variation of fibre diameter in the crossing of extreme opposites, i.e. F x S, was 4,9% compared to 5,4% in the purebred Grootfontein progeny.


Since between flock selection provides a more potent mechanism than ordinary within flock selection to change the composition of the South African clip it is worth looking at the rams being licensed for AI. Van der Merwe & Poggenpoel (1990) describes the programme fully and presents results from the first four groups of rams tested. With the exception of one year the variation in breeding value for fibre diameter do not seem to be high, probably because it was not a primary criterium for the selection of rams to be tested. The authors also discuss some policy changes that are having a negative influence on the usefulness of the progeny test program. If correctly applied, however, this programme could become a powerful agent for change in the desired direction.


De Lange (1981) suggested that an upper limit to production may well have been reached in many studs, and recommended a shift in emphasis to the concept of ewe productivity. It is interesting to note (Olivier, 1989) that more than 20 years of selection in the Grootfontein stud have increased clean fleece production by about one-third of a kilogram, body mass by about 3 kg, and fibre diameter by about 0,2 micron. Accounting for the effect of a slight increase in fibre diameter and body size the response in wool production is markedly less than what was expected in terms of the heritability and the existing variation.

Herselman (1990) adds a new dimension to the argument by pointing out that the energy required by an animal for fibre production is four times the energy required for growth. He therefore suggests that reducing fibre production will increase adaptedness to nutritional stress.

It should be kept in mind however, that any form of selection for a high production of a special product puts pressure on the animal. For example dairy breeds are prone to milk fever but not beef breeds. In theory there exists a limit to raising production when it becomes impractical to cope with the associated problem managerially.


All the arguments discussed in this paper points to a single conclusion: a concerted effort should be made to reduce the disproportionate amount of strong and overstrong wool currently produced in South African Merino clip. To summarise the main arguments:

Within flock selection would take longer to reinstate the former composition of the clip than what it took to effect the change-over to the present state. The process can however be speeded up by the widespread application of between flock selection and artificial insemination.



ANONYMOUS 1990. Vorderingsverslag Subprojek K5312/07/4 van die Departement van Landbou-ontwikkeling.

BEZUIDENHOUT AG, GREYLING AC & OLIVIER JJ. 1991. Evaluation of fine wool Merino sheep on cultivated pastures under irrigation.

DE KLERK JC & AO DE LANGE. 1970. The South African Merino wool clip - present and future. Farming in SA. August 1970.

DE LANGE AO. 1981. The Merino ram breeding industry in the eighties. Merino Breeder's Journal, Oct. p. 13-21

ERASMUS GJ. 1977. Die teelstruktuur van die SA Merino. M.Sc.(Agric) verhandeling UOVS.

---------------------- 1989. A mixed model analysis of a selection experiment with Merino sheep in an arid environment. PhD. thesis UOFS.

HERSELMAN MJ. 1990. Energiebehoeftes van Angorabok M.Sc.(Agric) verhandeling US.

JORDAAN TB. 1969. Merino Breeders Journal.

MARX FE. 1981. Die gehalte van die SA Merinoskeersel oor 30 jaar. Karoo Agric 2(1) p. 13-14.

MINNAAR JS. 1969. Merino Breeders Journal.

OLIVIER JJ. 1989. Genetic and environmental trends in the Grootfontein Merino Stud. PhD. thesis UOFS.

POGGENPOEL DG. 1984. Breeding value differences between commercial Merino flocks. Proc. 2nd world Congress. Sheep & Beef Cattle Breeding, Pretoria. p.711-717.

--------------- 1990. The correlation between stud phenotypic means and estimated breeding values. Proc. 2nd World Merino Congress. Pretoria. p. 4.1.

ROGAN IM. 1984. Selection for wool production. Proc. 2nd World Congress. Sheep & Beef Cattle Breeding, Pretoria. p. 367-380.

VAN DER MERWE CA & POGGENPOEL DG. 1990. Hoëteg kleinvee en veselproduksie. Hand. SAVDP Kongress. p. V1-V30.

VERBEEK W. 1969. Fynwolproduksie baie belangrik. Die Wolboer, Sept.



Proceedings 30th SASAP congres