Last update: November 18, 2010 03:33:49 PM E-mail Print




M. A. Snyman, M.J. Herselman & C. Stannard

Grootfontein Agricultural Development Institute, Private Bag X529, Middelburg (EC), 5900

eMail:  Gretha Snyman



The difference in profitability of sheep breeds remains one of the most controversial discussion issues among farmers and is one of the aspects which receives little comment from researchers and scientists. The reason for this is that the full implication of such a comparison is seldom taken into account. Nonetheless, farmers are continuously changing from one breed to another. The decision to change from one breed to another is mostly due to short-term financial reasons and current market trends favoring either wool or meat. For example, when changing from wool sheep to mutton sheep, the first mutton income follows quickly on the last wool income, with favourable short-term cash flow implications. This is, however, a once-off situation, as the following mutton income will also be on a routine yearly basis.


The issue of less selective grazing behaviour of Dorper sheep is also frequently mentioned. This is based on the relative improved animal performance (reproduction and growth), which is frequently observed by farmers when changing from one breed to another. Research results in various areas in the Karoo have shown that there are very little differences in grazing behaviour and diet selection of different sheep breeds (Botha et al., 1983; Du Toit et al., 1994; Du Toit, 2000).


In a study conducted on Eastern Mixed Karoo veld (Grootfontein Agricultural Development Institute) with Merino and Dorper sheep over several seasons, it was found that there was a 97 %, 92%, 75% and 97% overlap in diets selected by the two breeds during the winter, spring, summer and autumn seasons respectively (Du Toit, 1998).  On the whole over all the seasons, Merino sheep concentrated their grazing more on grass, while Dorper sheep concentrated their grazing more on the woody plant component of the veld, i.e. karoo bushes. From these it was clear that Dorper sheep were the more generalist grazers of the two breeds, grazing more of the woody karoo bushes, while Merino sheep were the more selective grazers, grazing more soft-leaved grasses and karoo bush leaves and thin karoo bush twigs (Du Toit, 1998). This could be a possible explanation for the relative improved animal performance observed when changing from Merino to Dorper sheep. In such cases there is a relative large volume of plant material available for Dorper sheep, which was not utilized by the wool sheep. Initially, Dorper sheep will perform very satisfactory, but this improved performance will not be sustained over a long period, as were indeed observed by various farmers in practice. Consequently, the initial improved performance of the Dorper sheep was most likely achieved at the cost of the natural vegetation.


Another aspect of breed comparisons, which contribute to wrong conclusions, is the inaccurate large stock unit (LSU) values assigned to various small stock breeds by the Act on the Conservation of Agricultural Resources (Act 43 of 1983)  (Department of Agricultural Economics and Marketing, 1985). If these LSU-values are used for the purpose of breed comparison, income per hectare of mutton sheep tends to be over-estimated relative to that of wool sheep. Furthermore, if ewe numbers are based on current legislation when changing from wool to mutton sheep, the natural resource will suffer due to an over-estimation of Dorper ewe numbers (Herselman, 2000).


Research done on the Carnarvon experimental station and the Tarka conservation area (near Hofmeyr), indicated that there were relatively small differences in income between Afrino, Dorper and Merino sheep over the long term (Snyman et al., 2000). There were, however, relatively large differences between the two experimental localities. The experimental layout of the above-mentioned study was such that all available ewes in the flock were mated every six months. This places some limitations on the extrapolation of these results to other farms. Despite the aforementioned report indicating towards relative small differences in income from different sheep breeds, various producers still raised their concern with regard to the influence of topography and veld type on breed choice in a specific ecological zone.


A study was therefore conducted to investigate the relative performance of different sheep breeds in two different topographic/veld types, on the same farming enterprise in the Middelburg district of the Eastern Cape Province. One part of the farming enterprise, namely, Grootvlei, is characterized by typical “Vlakte”-veld, which is normally regarded as a better veld type than the mountainous veld which was the predominant veld type of Twistkraal, the other part of the farming enterprise. At the time, Grootvlei was utilized for Dorper farming and Twistkraal for Merino farming. Apart from Merino (wool breed) and Dorper (meat breed) sheep, which were already run on Twistkraal and Grootvlei respectively, Afrino (dual purpose) sheep were also included for the purpose of this study.



The study was conducted at Grootvlei and Twistkraal in the Middelburg district in the Eastern Cape province under natural conditions. The veld type is classified as False Upper Karoo (Veld type 36; Acocks, 1988). From April 2001 to July 2003, Afrino, Dorper and Merino ewe flocks, consisting of ±100 ewes each, were kept at both Grootvlei and Twistkraal. Hundred Merino ewes, as well as 100 Dorper ewes were bought from farmers in the Middelburg and adjacent areas. The rest of the Merino and Dorper ewes were obtained from the existing flocks at Twistkraal and Grootvlei. Afrino ewes were bought from different breeders. All ewes were tagged with an identity number after they were bought. Afrino rams were bought on the Afrino ram sales, while Merino and Dorper rams available at Twistkraal and Grootvlei respectively, were used in the Merino and Dorper flocks.


The ewes of all three breeds were managed as one flock at each location for the duration of the study, except during mating. Lambs were also run together and managed as one flock. Animals were kept under natural veld conditions and received no feed or mineral supplementation at any stage. The normal inoculation, drenching and tick control programs were followed.


Ewes originating from the different farmers were divided equally between Grootvlei and Twistkraal in such a way that the two flocks had a similar age structure. A system of two breeding seasons per year was followed. At the start of the experiment, 50 ewes of each breed were mated at each locality to 4% rams for a seven-week period from April to June 2001. All rams were tested for fertility before mating started. The second breeding season stretched from 15 October to 25 November 2001. Ewes mated at each locality included all ewes that did not lamb during September 2001, as well as an additional 50 Afrino ewes bought from breeders and 50 Merino and 50 Dorper ewes from the existing flocks at Twistkraal and Grootvlei. This mating system was followed till the March 2003-lambing season.


Data recorded

Data collected on the ewe flock for four lambing seasons (from September 2001 to March 2003) included records on number of ewes mated, number of ewes that lambed, number of lambs born and number of lambs weaned. The following data were also recorded for each lambing season: dam ID, lamb ID, date of birth, sex and birth status of the lamb, as well as weaning weight at 120 days of age. Furthermore, monthly body weight of lambs was recorded from five to ten months of age. At six months of age, all ewe lambs were classed and those not suitable for replacement purposes were declared surplus. All lambs were weighed monthly and the surplus ewe lambs, together with all the ram lambs, were slaughtered at the local abattoir as soon as they have reached slaughter weight (± 40 kg for Dorper lambs; ± 42 kg for Afrino and Merino lambs). Upon reaching slaughter weight, lambs were fasted over night and fasted body weight was recorded before slaughtering the next morning. Carcass weight, dressing percentage, V1- and V3-fat depth measurements and carcass grading were recorded.


Body weight of ewes was recorded before each mating season. Afrino and Merino ewes were shorn at the end of January 2002 and 2003. Fleece weight was recorded after shearing and a midrib wool sample was taken from each ewe for determination of fibre diameter, clean yield percentage, staple length, crimp frequency, duerden, coefficient of variation and standard deviation of fibre diameter.


Statistical analysis

Body weight and wool production data of ewes, as well as growth and slaughter traits of lambs, were analysed with Least‑squares procedures (Littell et al., 1991).


Results and discussion

Reproductive performance

The reproductive performance for the three breeds at Grootvlei and Twistkraal over the experimental period is summarised in Table 1. Reproductive performance recorded in this study falls within the ranges reported in literature for the specific breeds (Cloete et al., 2000; Snyman et al., 1993; Snyman et al., 2000; Olivier et al., 2002; Snyman & Olivier, 2002; Snyman, 2004).


Body weight of ewes

Body weights of ewes recorded before each mating season are summarised in Table 2 for the three breeds at the two localities. Body weight of ewes before mating was higher at Grootvlei than at Twistkraal, while Dorper ewes were heavier than Afrino ewes, which in turn were heavier than Merino ewes. Afrino ewes in this study had lower body weights than those recorded for Afrino sheep at Carnarvon (60.0 to 64.7 kg) and Tarka (66.4 kg) (Snyman et al., 1993; Snyman et al., 2000; Snyman, 2004). Body weight of Dorper ewes at Grootvlei and Twistkraal also falls within the lower ranges of 50.0 kg to 74.0 kg cited in the literature (Cloete et al., 2000; Snyman et al., 2000; Snyman & Olivier, 2002). Merino ewes in this study also weighed less than reported body weights in literature (Olivier et al., 2002; Snyman et al., 2000).


Table 1. Reproductive performance of Afrino, Dorper and Merino ewes at Grootvlei and Twistkraal over the experimental period










Number of ewes in the flock







Number of matings







Number of lambs born







Number of lambs weaned







Lambs born per 100 ewes (%)

151.6 a

129.9 b

110.1 a b

168.0 a b

139.7 b

106.5 a

Lambs weaned per 100 ewes (%)

132.0 a

100.0 a

93.7 a

153.3 a b

114.7 b

96.4 a

Lambs weaned per 100 lambs born (%)




91.2 a

82.1 a b

90.5 b

Average number of matings per ewe







a b Values with the same superscript differ (P<0.05) between breeds at the same locality


Table 2. Body weight of ewes before mating




Average / Breed (kg)

Average / Locality

51.7±1.3 a

49.9±1.3 a





52.3±1.4 b




56.6±1.4 b




43.4±1.4 b

a   =  Values with the same superscripts differ significantly (P<0.05) between localities

b    =  Values with the same superscripts differ significantly (P<0.05) between breeds


Wool production

Wool production is summarised in Table 3. Merino ewes produced on average 1.3 kg more clean wool, which was 0.6 μm stronger than that produced by Afrino ewes. From Table 3 it is clear that both Afrino and Merino ewes at Grootvlei produced less wool with a lower fibre diameter than the ewes at Twistkraal. Wool production as well as fibre diameter of Afrino and Merino ewes at Grootvlei and Twistkraal were lower than those recorded in other studies under similar conditions (Snyman et al., 1993; Snyman et al., 2000; Olivier et al., 2002; Snyman, 2004).


Table 3. Wool production of Afrino and Merino ewes at Grootvlei and Twistkraal








Greasy fleece weight (kg)

1.74±0.06 a b

3.43±0.07 a b

1.92±0.06 a b

3.67±0.07 a b

Clean fleece weight (kg)

1.15±0.05 a b

2.47±0.05 a

1.26±0.05 a b

2.68±0.05 a b

Clean yield (%)

65.6±0.6 a

72.9±0.7 a b

65.6±0.6 a

74.0±0.7 a

Fibre diameter (μm)

20.1±0.2 a b

20.7±0.2 a b

20.6±0.2 a b

21.2±0.2 a b

CF (%)



97.6±0.3 a

96.7±0.3 a

SD (μm)





CV (%)





Staple length (mm)

64.0±1.6 a b

83.1±1.8 a b

70.2±1.5 a b

89.3±1.7 a b

Crimp frequency

14.8±0.3 a

12.8±0.3 a b

14.6±0.3 a

12.2±0.3 a b


101.5±1.4 a b

97.2±1.6 a b

108.2±1.3 a b

101.2±1.5 a b

a Values with the same superscript differ (P<0.05) between breeds at the same locality

b Values with the same superscript differ (P<0.05) between localities for the same breed


Growth performance of lambs

Growth performance of the lambs is presented in Table 4. Dorper lambs were the heaviest at weaning, followed by Afrino and then the Merino lambs at each locality. The same tendency was observed for 6- to 9-month body weights at Twistkraal. At Grootvlei, there were no differences in 6- to 10-month body weight between Dorper and Afrino lambs, both being heavier than the Merino lambs (P<0.05). A significant locality effect was observed for all body weights recorded, where lambs at Grootvlei were heavier than those at Twistkraal. Growth performance of lambs accords well with those cited in literature for the three breeds (Snyman et al., 1995; Cloete et al., 2000; Snyman et al., 2000; Olivier et al., 2002; Snyman & Olivier, 2002).


Table 4. Body weight of Afrino, Dorper and Merino lambs at Grootvlei and Twistkraal
















Weaning weight (kg)

28.1±1.2 a

31.0±1.3 a

22.1±1.2 a

26.9±1.2 a

30.7±1.1 a

21.3±1.2 a

Average daily gain: birth to weaning (g/day)

251±12.6 a

280±13.6 a

193±13.0 a

240±12.6 a

284±12.1 a

183±13.1 a

6 mo weight (kg)

32.9±1.2 a

34.7±2.0 b

26.1±1.6 a b

28.8±1.5 a

33.7±1.9 a

23.9±1.6 a

7 mo weight (kg)

37.6±1.3 a

39.6±2.1 b

31.2±1.7 a b

34.3±1.6 a

37.5±1.9 a

29.8±1.6 a

8 mo weight (kg)

40.6±1.3 a

42.2±2.1 b

33.1±1.7 a b

34.6±1.6 a

39.3±1.9 a

30.0±1.6 a

9 mo weight (kg)

43.4±1.3 a

43.7±2.0 b

34.8±1.6 a b

37.2±1.5 a

41.3±1.9 a

33.1±1.6 a

10 mo weight (kg)

45.4±1.3 a

44.8±2.1 b

36.1±1.7 a b

39.8±1.6 a

42.5±2.0 b

34.5±1.7 a b

a b  Values with the same superscript differ (P<0.05) between breeds at the same locality

* Number of lambs weaned


Slaughter traits of lambs

Slaughter traits for the lambs are summarised in Table 5. From Table 5 it is evident that Dorper lambs reached slaughter age earlier than Afrino and Merino lambs. Dressing percentage was also higher in Dorper lambs. The lower V1- and V3-measurements of Dorper lambs could most probably be ascribed to their earlier slaughter age. Afrino and Dorper lambs in this study were slaughtered at a higher average age than the ranges recorded in literature (Badenhorst et al., 1992; Cloete et al., 2000; Snyman et al., 1993; Snyman et al., 2000; Snyman & Olivier, 2002). However, differences in slaughter age could most probably be ascribed to differences in slaughter weight, as is clearly illustrated in the different values cited in the review of Cloete et al. (2000). Other carcass traits fall within the ranges reported in the literature.


Table 5.  Slaughter traits of Afrino, Dorper and Merino lambs at Grootvlei and Twistkraal
















Slaughter weight (kg)

42.9±0.7 a

40.9±0.8 a


42.1±0.7 a

39.2±0.7 a b

42.3±0.7 b

Slaughter age (days)

268±9.2 a

248±11.1 a

337±11.2 a

295±9.6 a

236±9.9 a

348±10.3 a

Carcass weight (kg)




19.3±0.2 a

19.2±0.2 a

18.5±0.2 a

Dressing %




46.5±0.8 a

46.8±0.9 b

44.5±0.9 a b

V1 (mm)




4.4±0.4 a

4.0±0.4 b

3.8±0.4 a b

V3 (mm)

2.7±0.3 a


1.8±0.3 a

3.0±0.3 a

2.4±0.3 a


a b  Values with the same superscript differ (P<0.05) between breeds at the same locality

* Number of lambs slaughtered


The cumulative percentage of lambs slaughtered for each breed is depicted in Figures 1 and 2 for the September- and March-born lambs respectively. From these figures it is also evident that Dorper lambs were slaughtered earlier than Afrino lambs, with Merino lambs taking the longest time to reach slaughter weight. A definite seasonal effect is also evident from Figures 1 and 2, where the autumn-born lambs (Figure 2) reached slaughter weight earlier than spring-born lambs (Figure 1).


Figure 1. Cumulative percentage September-born lambs slaughtered


Figure 2. Cumulative percentage March-born lambs slaughtered



It was evident from the results of this study that differences in productive and reproductive efficiency of Afrino, Merino and Dorper sheep occurred between the two localities studied. When changing from one breed to another, or one commodity to another, the reproductive rate, mature body weight and fibre production of the two breeds under the farming conditions in question should be considered, as these are the primary factors determining profitability of a specific enterprise.



The authors wish to convey their sincere appreciation to the farm owner for permission to carry out the work, as well as for his financial contribution. The farm managers and farm aids of Grootvlei and Twistkraal and the responsible technicians and farm aids of GADI are also thanked for their assistance in the technical execution of the project.



Acocks, J.P.H., 1988. Veld types of South Africa. Memoirs of the Botanical survey of South Africa, 57, 1-146. Pretoria: Government Printers.

Badenhorst, M.A., Olivier, J.J., Cloete, J.A.N. & King, B.R., 1992. Produksie‑ en reproduksienorme vir Dorperskape in twee verskillende omgewings. Karoo Agric, 4(4), 9-11.

Botha, P., Blom, C.D., Sykes, E. & Barnhoorn, A.S.J., 1983. Comparison between the diets of small and large stock on mixed Karoo veld. Proceedings Grassland Society of South Africa, 18, 101-105.

Cloete, S.W.P., Snyman, M.A. & Herselman, M.J., 2000. Productive performance of Dorper sheep. Small Ruminant Research, 36, 119-135.

Department Of Agricultural Economics And Marketing, 1985. Act on the Conservation of Agricultural Resources (Act 43 of 1983). Government Gazette Nr R 2687, 6 December, 1985.

Du Toit, P.C.V., 1998. Diets selected by Merino and Dorper sheep in Karoo veld. Grootfontein Agric, 2(1), 15-22.

Du Toit, P.C.V., 2000. Wat vreet verskillende skaaprasse in die Dorre Karoo? Grootfontein Agric, 2(2), 12-13.

Du Toit, P.C.V., Blom, C.D. & Immelman, W.F., 1994. Diet selection by different small stock species in the arid karoo. Karoo Agric, 6(1), 25-27.

Littell, R.C., Freud, R.J., Spector, P.C., 1991. SAS-system for linear models, 3rd  Ed., Cary, NC.

Olivier, W.J., Olivier, J.J., Snyman, M.A., Pretorius, A.P. & Van Heerden, M., 2002. Production and reproduction norms of fine woolled Merino sheep on natural pastures in the Karoo. Grootfontein Agric, 5, 29-31.

Snyman, M.A., 2004.   Production and reproduction performance of the Carnarvon Afrino flock. Afrino Journal, 2004, 40-41.

Snyman, M.A.,  Erasmus, G.J., Van Wyk, J.B. & Olivier, J.J., 1995. Direct and maternal (co)variance components and heritability estimates for body weight at different ages and fleece traits in Afrino sheep. Livestock Production Science, 44(3), 229-236.

Snyman, M.A., Herselman, M.J., Cloete, J.A.N. & King, B.R., 2000. Bruto marge van Afrino, Dorper en Merinoskape. Grootfontein Agric, 2(2), 9-11.

Snyman, M.A. & Olivier, W.J., 2002. Productive performance of hair and wool type Dorper sheep under extensive conditions. Small Ruminant Research, 45(1), 17-23.

Snyman, M.A., Olivier, J.J., Cloete, J.A.N. & King, B.R., 1993. Produksienorme vir Afrinoskape in twee omgewings. Karoo Agric, 5(1), 25-28.



Grootfontein Agric 7 (1), 23-28