Last update: November 19, 2010 09:20:49 AM E-mail Print

 

GENETIC ANALYSIS OF THE CRADOCK FINE WOOL MERINO STUD

 

W.J. Olivier1, J.J. Olivier2, S.W.P. Cloete3 and J.B. van Wyk4

 

1Grootfontein ADI, P/Bag X529, Middelburg (EC) 5900, South Africa

2ARC: LBD (Animal Production), P/Bag X5013, Stellenbosch 7599, South Africa

3Institute for Animal Production: Elsenburg, P/Bag X1, Elsenburg 7607, South Africa

4Department of Animal, Wildlife and Grassland Science, University of the Free State,

PO Box 339, Bloemfontein 9300, South Africa

 


INTRODUCTION

The increase in the demand for fine wool resulted in an increase in the production of fine wool. The number of flocks in South Africa where selection for decreased fibre diameter was practiced thus increased markedly. In some instances, it was even the only selection objective, regardless of the effect on the other production traits. It is, however, important to maintain a balance between the economically important traits. This is particularly true in South Africa, where meat production contributes more than 75% of the farm income of Merino farmers. The South African wool industry requested the Department of Agriculture to do research on the production of fine wool. This has led to the establishment of the Cradock fine wool Merino stud. The purpose of this study is to estimate genetic parameters for the production, reproduction and subjectively assessed wool and conformation traits in this flock. These results will also be used to quantify the effect of selection for reduced fibre diameter on the other production and subjectively assessed traits.

 

MATERIAL AND METHODS

Data. The Cradock Fine Wool Merino Stud was established in 1988 as described by Olivier et al. (1999). Ewes were bought from Merino farmers with the finest clips throughout South Africa and four fine wool rams were imported from Australia. Data collected on 5747 ram and ewe hoggets born within this stud from 1988 to 2004 were used for the analysis of the production and subjectively assessed wool and conformation traits. Lifetime reproduction was assessed with data of 1278 ewes born in the stud from 1988 to 2002.

 

Statistical analyses. The means and standard deviations for the production, reproduction and subjectively assessed traits were obtained with the PROC MEANS-procedure of SAS, and significance levels for the fixed effects were obtained with the PDIFF-option under the PROC GLM-procedure of SAS (Littell et al., 2002). The hogget production traits analyzed included 15-month age body weight (BW), clean fleece weight (CFW), mean fibre diameter (MFD) and staple length (STL). The subjective traits assessed on a linear scale from 1 to 50 were wool quality (WQ), variation over the fleece (VAR), staple formation (ST), conformation of the front quarters (FQ) and overall body conformation (CON) (Olivier et al., 1987). The reproduction traits analyzed included total number of lambs born (NLB) and weaned (NLW) and total weight of lamb weaned (TWW, calculated from weaning weight of lambs corrected for gender). Several fixed effects (year of birth, sex, rearing status, age of dam in years, number of lambing opportunities and year of birth of the ewe), as well as the hogget age as a linear regression were tested. Only effects and interactions which had a significant effect (P<0.01) on a specific trait were included in the final operational model.

 

The estimation of the genetic parameters and breeding values were done with ASREML (Gilmour et al., 2002). Log likelihood ratio tests were done to determine the most suitable model for the estimation of (co)variance components for each trait. The most suitable model for BW and CFW included both the direct additive genetic variance and the maternal additive genetic variance. For MFD, the direct additive genetic variance and the maternal permanent environmental variance were included in the model. Only the direct additive genetic variance was included in the models for STL, WQ, VAR, ST, FQ and CON. The direct heritabilities and genetic trends for the respective traits were obtained from univariate analyses, while the genetic and phenotypic correlations were obtained from bivariate analyses.

 

RESULTS AND DISCUSSION

The means (standard deviation; SD) for BW, CFW, MFD and STL of the hoggets were 61.10 (11.57) kg, 4.40 (1.02) kg, 19.18 (1.52) µm and 102.98 mm. The means (SD) for WQ, VAR, ST, FQ and CON of the hoggets were 30.72 (8.20), 36.76 (6.90), 29.76 (6.05), 27.35 (6.73) and 29.05 (7.07). The means for the lifetime NLB, NLW and TWW were 4.46 (1.62), 3.70 (1.52) and 92.98 (35.99) kg respectively.

 

The heritability estimates (h²) for BW, CFW, MFD and STL were 0.50, 0.54, 0.63 and 0.46 respectively (Table 1). Safari et al. (2005) reported generally lower h2 estimates for BW and CFW, but similar values than that obtained in this study for MFD and STL. Groenewald et al. (1999) and Olivier et al. (1997) reported lower h² estimates for BW, CFW and MFD in the National Merino Progeny Testing program of South Africa and two Merino flocks respectively.  These authors reported higher and similar h² estimates for STL. Snyman et al. (1996) reported similar h² estimates for BW and MFD, but a lower estimate for CFW in a Merino flock run under extensive conditions. Similar h² estimates for BW, CFW and MFD were reported for Afrino sheep (Snyman et al., 1995). Estimates of h² of WQ, VAR, ST, FQ and CON reported in the literature by Groenewald et al. (1999) and Olivier et al. (1997) were lower than the values obtained in this study. The h² estimates for NLB, NLW and TWW were 0.17, 0.10 and 0.09 respectively. Snyman et al. (1998) reported similar h² estimates for TWW over three parities for two Merino flocks, while the estimate for the third Merino flock was higher. Safari et al. (2005) reported similar h2 for NLW and TWW, but a lower h2 for NLB. The maternal heritability (h2m) for BW and CFW were 0.07 ± 0.02 and 0.08 ± 0.02 respectively and the maternal permanent environmental effect for MFD was 0.02 ± 0.01. The h2m of these traits were similar than the respective values derived by Safari et al. (2005) from the available literature.

 

It is evident from Table 2 and Figure 1 that the genetic trend for MFD was negative. Furthermore, it is evident from Figure 1 that there was only a slight decrease in MFD from 1988 until 1996. This was due to the fact that the animals were small with unsatisfactory conformation; therefore, more emphasis was placed on BW than on MFD for the first few years. Since 1997, more selection emphasis could be placed on MFD, as the BW and conformation of the animals were at a satisfactory level.

 

Table 1. The direct heritabilities (on the diagonal), genetic (above the diagonal) and phenotypic (below the diagonal) correlations of production, reproduction and subjectively assessed traits

Trait

BW

CFW

MFD

STL

WQ

VAR

ST

FQ

CON

NLB

NLW

TWW

BW

0.50

0.06

0.24

0.20

0.07

-0.11

0.00

0.67

0.81

0.16

0.06

0.53

CFW

0.25

0.54

0.17

0.51

0.40

-0.14

0.57

-0.11

0.16

-0.21

-0.38

-0.13

MFD

0.11

0.17

0.63

-0.02

-0.47

-0.60

0.63

0.17

0.22

0.33

0.40

0.43

STL

0.13

0.36

0.02

0.46

0.31

-0.43

0.02

0.24

0.44

-0.08

-0.11

-0.04

WQ

0.04

0.22

-0.33

0.20

0.50

0.43

-0.46

-0.04

0.07

 

 

 

VAR

-0.02

-0.07

-0.29

-0.15

0.34

0.34

-0.51

-0.24

-0.36

 

 

 

ST

0.11

0.37

0.38

0.04

-0.26

-0.20

0.40

0.02

0.11

 

 

 

FQ

0.46

0.05

0.12

0.15

0.01

-0.07

0.10

0.51

0.89

0.03

0.03

0.34

CON

0.60

0.23

0.15

0.26

0.08

-0.10

0.19

0.72

0.55

0.05

-0.06

0.35

NLB

0.10

-0.04

0.12

-0.04

 

 

 

0.07

0.12

0.17

0.93

0.91

NLW

0.04

-0.08

0.12

-0.04

 

 

 

0.09

0.09

0.81

0.10

0.90

TWW

0.11

-0.06

0.14

-0.04

 

 

 

0.14

0.15

0.77

0.96

0.09

SE’s for the heritabilities ranged from 0.03-0.08 and from 0.01-0.04 for phenotypic correlations. The SE’s for the genetic correlations between production and subjective traits ranged from 0.03-0.08, from 0.06-0.09 among the reproduction traits, and from 0.13-0.19 for the reproduction and hogget traits

 

Figure 1. Genetic trends for MFD compared to the other production and subjectively assessed traits

 

The genetic correlations obtained between MFD and LW, CFW, ST FQ, CON, NLB, NLW and TWW were unfavourable in terms of the selection objectives (Table 1). The rG reported by Swan et al. (1997) and Purvis & Swan (1997) for fine wool Merino sheep and Safari et al. (2005) between MFD and CFW and STL were unfavourable. Unfavourable rG between MFD and BW were also cited by Purvis & Swan (1997) and Safari et al. (2005). Cloete et al. (2002) cited unfavourable rG between MFD and NLW and TWW. Despite these unfavourable correlations, genetic change in BW, FQ, CON, CFW, STL and WQ were in the desired direction. It is evident from Table 1 that an improvement in FQ and CON scores would result in higher BW and TWW. Furthermore, the genetic correlations between WQ and MFD indicated that good quality wool tends to be finer with thinner staples and less variation over the fleece.

 

Table 2. The regression coefficients (b) and R² values applicable to the genetic trends

 

BW

CFW

MFD

STL

WQ

VAR

ST

FQ

CON

b

0.13

0.08

-0.09

0.11

0.13

0.06

0.00

0.12

0.14

R2

0.94

0.80

0.79

0.85

0.96

0.59

0.00

0.94

0.94

SE values for the regression coefficients (b) were 0.01

 

CONCLUSION

The breeding objective of this stud was to increase BW and STL, to maintain CFW and to decrease MFD.  This objective was largely achieved. This was done despite unfavourable genetic correlations between MFD and the other traits. It can therefore be concluded that selection for decreased fibre diameter will not have a negative effect on the other economically important traits, if the other traits are also included in the selection objectives, or monitored to detect possible unwanted correlated changes early.

 

References

Cloete, S.W.P., Greeff, J.C. and Lewer, R.P. (2002) Aust. J. Agric. Res.  53: 281-286

Gilmour, A.R., Gogel, B.J., Cullis, B.R., Welham, S.J. and Thompson, R. (2002) ASREML User’s Guide Release 1.0. VSN International Ltd, Hemel, Hempstead, HP11es, UK

Groenewald, P.G.J., Olivier, J.J. and Olivier, W.J. (1999) S. Afr. J. Anim. Sci. 29: 174-178.

Littell, R.C., Freud, R.J. and Struop, W.W. (2002) SAS-system for linear models, 4th Ed. SAS Institute. Inc. Cary, N.C., USA.

Olivier, J.J., Bezuidenhout, A.G., Greyling, A.C. and Cloete, S.W.P. (1999) Proc. Assoc. Advmt. Anim. Breed. Genet. 13: 62-65.

Olivier, J.J., Cloete, S.W.P. and Snyman, M.A. (1997) Proc. S. Afr. Soc. Anim. Sci. 35th Congress, Nelspruit, 1-3 July 1997. 

Olivier, J.J., Delport, G.J., Erasmus, G.J. and Eksteen, T.J. (1987) Karoo Agric. 3: 1-4.

Purvis, I.W. and Swan, A.A. (1997) Proc. Assoc. Advmt Anim. Breed. Genet. 12: 512-515

Safari, E., Fogarty, N.M. and Gilmour, A.R. (2004) Livest. Prod. Sci. 92:271-289

Snyman, M.A., Cloete, S.W.P. and Olivier, J.J. (1998) Livest. Prod. Sci. 55: 157-162.

Snyman, M.A., Erasmus, G.J., Van Wyk, J.B. and Olivier, J.J. (1995) Livest. Prod. Sci. 44: 229-236.

Snyman, M.A., Olivier, J.J. and Olivier, W.J. (1996) S. Afr. J. Anim. Sci. 21: 11-14.

Swan, A.A., Lax, J. and Purvis, I.W. (1995) Proc. Assoc. Advmt Anim. Breed. Genet. 11: 516-520