Last update: November 25, 2010 11:02:09 AM E-mail Print

 

CORRELATIONS OF SUBJECTIVE FLEECE TRAITS WITH PRODUCTION

AND REPRODUCTION IN AFRINO SHEEP


M. A. Snyman*, W. J. Olivier, P .J. Griessel & J. A. N. Cloete1

Grootfontein ADI, Private Bag X529, Middelburg Cape, 5900

1 Carnarvon Experimental Station, Carnarvon, 8925

 

* Corresponding author

Tel : + 27 49 842 1113

Fax : +27 49 842 4352

e-mail: php echo mailMe(

 


1.  Introduction

The ultimate aim of sheep breeding research is to provide estimates of the parameters required to construct a genetic improvement plan leading to improved viability, productivity and profitability of the specific enterprise. Genetic parameters for objective production and reproduction traits have been estimated, and it is known how to incorporate these traits into the breeding plan. In practice, however, a lot of emphasis is placed on subjective conformation and fleece traits during the selection of breeding sires and dams. In many instances, animals are culled on the basis of these traits, but this occurs at the expense of the economically important traits such as reproduction. Very little is known regarding the heritability and genetic relationships of these subjective traits with the economically important traits such as body weight, fleece weight, fibre diameter and reproductive performance.

The Afrino is a white woolled mutton sheep breed kept in the more extensive sheep grazing areas of the RSA. The selection objectives of the breed is to produce good quality slaughter lambs at a relatively early age from the veld, and to produce a reasonable amount of good quality fine wool. As with every woolled sheep breed, there are a lot of subjective fleece traits, not necessarily of economic importance, that are taken into account during selection of replacement sires and dams. One such trait that is discriminated against in Afrino sheep, is creeping belly (extent to which wool creeps up the sides). Other subjective conformation and fleece traits on which emphasis are placed, are general conformation of the head, softness of face covering, extent of pigmentation in the face and on the ears, crimp definition, density of fleece, evenness of fleece, straightness of the top line, hocks and pastern joints.

For the purpose of this paper, the heritabilities of and genetic and phenotypic correlations amongst the subjective fleece traits, as well as pigmentation and softness of face, were estimated. Genetic and phenotypic correlations of these traits with body weight, objective fleece traits and reproduction were also estimated. This was done in an attempt to clarify the importance of traits such as creeping belly in the selection programme of Afrino sheep, as well as other white woolled breeds.

 

2.  Material and methods

2.1  Data

Data used for this study were collected on the Carnarvon Afrino flock, and include data records of 3291 animals, the progeny of 127 sires and 772 dams, born from 1986 to 1998. At 14 - 16 months of age, 7 subjective traits were assessed on a linear scale ranging from 1 to 50 in all young rams and ewes. These traits were softness of face covering, extent of pigmentation in the face and on the ears, softness of fleece, crimp definition, density of fleece, evenness of fleece (in terms of fibre diameter, length, and crimp definition) and creeping belly (extent to which belly wool creeps up the sides). The scale of assessment for these traits is summarized in Table 1.

Table 1. Linear scale for assessment of subjective traits in Afrino sheep

 

Scale of assessment a

 

1

 25

 50

 Softness of face (FACE)

 Hard

 Average

 Very soft

 Pigmentation (PIGM)

 None

 Ideal

 Excessive

 Softness of fleece (SOFT)

 Hard

 Average

 Soft

 Crimp definition (CRIM)

 Poor

 Average

 Ideal

Density of fleece (DENS)

 Poor

 Average

 Ideal

 Evenness of fleece (EVEN)

 Uneven

 Average

 No variation

 Creeping belly (CBEL)

 High up the sides

 Average

 None

a    1 - 10 = Poor

  11 - 20 = Below average

  21 - 30 = Average

  31 - 40 = Above average

  41 - 50 = Excellent

 

Data records on weaning weight, body weight at 9- and 12-months of age, as well as clean fleece weight, fibre diameter and staple length measured at 15 months of age were also available for the same animals. Reproduction data of 686 ewes born from 1986 till 1997 used for the study, include lifetime number of lambs weaned and lifetime total weight of lamb weaned.

 

2.2  Statistical analysis

Heritability estimates for the subjective traits were obtained by fitting single trait animal models, including only direct additive effects, under the ASREML programme (Gilmour et al., 1997). The fixed part of the models included fixed effects for year of birth, rearing status and sex of the animal, as well as age of the animal (linear) and age of dam (quadratic). 

Genetic and phenotypic correlations among all traits were obtained by fitting bi-variate animal models under the ASREML programme. Details regarding model specification for body weight, fleece traits and reproduction can be obtained from Snyman et al. (1995) and Snyman et al. (1997). Effects included for analyses of the subjective traits were the same as those included for estimation of heritability.

 

Results

3.1  Mean and coefficient of variation

The means and coefficients of variation for all the traits are given in Table 2.

Table 2. Mean and coefficient of variation for traits analysed

 

MEAN

CV (%)

Softness of face covering (FACE)

35.48

15.05

Pigmentation (PIGM)

23.55

32.58

Softness of fleece (SOFT)

33.06

22.26

Crimp definition (CRIM)

27.07

33.73

Density of fleece (DENS)

34.81

16.09

Evenness of fleece (EVEN)

34.17

17.90

Creeping belly (CBEL)

38.65

28.89

Weaning weight (kg) (WW)

27.75

12.03

9-month body weight (kg) (W9)

43.29

9.35

12-month body weight (kg) (W12)

55.33

8.47

15-month clean fleece weight (kg) (CFW)

2.03

17.79

Fibre diameter (Fm) (FD)

20.73

6.44

Staple length (mm) (SL)

79.58

15.27

Total weight of lamb weaned (kg) (TWW)

118.97

30.31

Number of lambs weaned (NLW)

3.88

33.13

From Table 2 it is obvious that animals in the Carnarvon Afrino flock were above average for all the subjective traits assessed, with the exception of CRIM, which was average. An ideal score was given for PIGM.

Averages for creeping belly by year, sex and rearing status are summarised in Table 3, while the average yearly creeping belly scores for the rams and ewes are depicted in Figure 1.

Table 3. LS-means of creeping belly by year, sex and rearing status

YEAR

LSMEANS

SE

1986

35.31

1.48

1987

37.05

1.49

1988

31.39

1.45

1989

37.17

1.45

1990

40.78

1.45

1991

46.71

1.46

1992

44.84

1.49

1993

42.58

1.49

1994

45.81

1.46

1995

42.28

1.46

1996

36.69

1.44

1997

35.93

1.43

1998

37.71

1.46

SEX

LSMEANS

SE

Ram

37.32

1.52

Ewe

41.80

2.08

REARING STATUS a

LSMEANS

SE

11

39.46

0.38

21

38.24

0.85

22

38.43

0.24

31

41.46

6.87

32

41.51

2.94

33

38.25

1.77

a : 11=: Born as single, weaned as single; 21 = Born as twin, weaned as single; etc.

 


Fig. 1. Relationship between year and creeping belly score for rams and ewes

From Table 3 it is evident that creeping belly scores differed between years. In the drier years, the animals tended to have finer fleeces and more creeping bellies. Furthermore, rams generally had more creeping bellies than their ewe contemporaries. The influence of year on the appearance of creeping belly is also illustrated in Figure 1. The difference in creeping belly score between rams and ewes was less in some years than in others. On two occasions, rams even had less creeping bellies than the ewes. Rearing status did not have any significant influence on creeping belly.

 

3.2.   Heritability of and correlations amongst subjective traits

Heritability estimates of and genetic and phenotypic correlations amongst the subjective traits are summarised in Table 4.

Table 4. Heritability estimates (SE) of and genetic and phenotypic correlations (SE among various subjective traits a

 

 

 

FACE

 

PIGM

 

SOFT

 

CRIM

 

DENS

 

EVEN

 

CBEL

 

FACE

 

0.23 (0.04)

 

0.10 (0.09)

 

0.20 (0.09)

 

0.05 (0.10)

 

-0.05 (0.11)

 

0.18 (0.12)

 

0.03 (0.09)

 

PIGM

 

0.05 (0.02)

 

0.50 (0.04)

 

-0.09 (0.07)

 

-0.05 (0.07)

 

-0.03 (0.09)

 

0.07 (0.10)

 

0.09 (0.07)

 

SOFT

 

0.11 (0.02)

 

0.00 (0.02)

 

0.51 (0.04)

 

0.80 (0.03)

 

-0.64 (0.06)

 

0.81 (0.05)

 

-0.52 (0.05)

 

CRIM

 

0.04 (0.02)

 

0.03 (0.02)

 

0.62 (0.01)

 

0.47 (0.04)

 

-0.61 (0.07)

 

0.62 (0.07)

 

-0.67 (0.05)

 

DENS

 

-0.05 (0.02)

 

-0.02 (0.02)

 

-0.40 (0.02)

 

-0.36 (0.02)

 

0.26 (0.04)

 

0.02 (0.12)

 

0.51 (0.07)

 

EVEN

 

0.04 (0.02)

 

0.01 (0.02)

 

0.32 (0.02)

 

0.40 (0.02)

 

0.05 (0.02)

 

0.28 (0.04)

 

-0.27 (0.10)

 

CBEL

 

0.00 (0.02)

 

0.00 (0.02)

 

-0.35 (0.02)

 

-0.45 (0.02)

 

0.31 (0.02)

 

-0.15 (0.03)

 

0.37 (0.04)

 a Heritability on diagonal, genetic correlations above diagonal and phenotypic correlations below diagonal

 

From Table 4 it is evident that the heritability estimates for the various subjective traits ranged from moderate (FACE, DENS and VAR) to high (PIGM, SOFT, CRIM and CBEL). Genetic response to selection based on individual phenotype of the latter four traits, could therefore be achieved.

PIGM and FACE had no genetic correlation with any of the subjectively assessed fleece traits. High genetic correlations were estimated amongst the subjective fleece traits, with the exception of DENS and EVEN. Generally, animals with softer fleeces had better crimp definition, their fleeces were more even, less dense and had larger creeping bellies.

The phenotypic correlations amongst the subjective fleece traits were similar, but somewhat lower in magnitude, than the corresponding genetic correlations.

 

3.3.  Correlations of subjective traits with body weight, fleece traits and reproduction

Genetic correlations of reproduction, body weight and fleece traits with the various subjectively assessed fleece traits are summarised in Table 5.

Table 5. Genetic correlations (SE) of reproduction, body weight and fleece traits with various subjective traits

 

 

 

TWW

 

NLW

 

WW

 

W9

 

W12

 

CFW

 

MFD

 

SL

 

FACE

 

0.49 (0.24)

 

0.42 (0.21)

 

-0.08 (0.13)

 

0.02 (0.10)

 

-0.06 (0.10)

 

0.40 (0.08)

 

-0.22 (0.08)

 

0.28 (0.10)

 

PIGM

 

-0.17 (0.25)

 

-0.15 (0.21)

 

0.13 (0.10)

 

0.12 (0.07)

 

0.07 (0.07)

 

-0.02 (0.06)

 

0.08 (0.06)

 

0.04 (0.08)

 

SOFT

 

0.01 (0.34)

 

-0.13 (0.29)

 

-0.22 (0.11)

 

-0.07 (0.08)

 

-0.09 (0.08)

 

0.02 (0.07)

 

-0.80 (0.03)

 

0.00 (0.09)

 

CRIM

 

0.45 (0.36)

 

0.38 (0.29)

 

-0.08 (0.11)

 

0.06 (0.08)

 

0.09 (0.08)

 

-0.02 (0.07)

 

-0.42 (0.05)

 

-0.09 (0.09)

 

DENS

 

-0.08 (0.53)

 

-0.02 (0.47)

 

0.11 (0.13)

 

0.14 (0.10)

 

0.11 (0.10)

 

0.22 (0.08)

 

0.48 (0.06)

 

-0.38 (0.09)

 

EVEN

 

-0.50 (0.33)

 

-0.32 (0.27)

 

-0.08 (0.14)

 

-0.04 (0.10)

 

-0.03 (0.10)

 

-0.02 (0.09)

 

-0.57 (0.07)

 

-0.19 (0.10)

 

CBEL

 

-0.33 (0.23)

 

-0.25 (0.20)

 

-0.26 (0.10)

 

-0.31 (0.08)

 

-0.35 (0.07)

 

0.53 (0.05)

 

0.41 (0.05)

 

0.27 (0.08)

 

Due to the limited number of reproductive records, standard errors for the genetic correlations were relatively high. These results should therefore be regarded merely as an indication whether positive or negative relationships exist between traits. High genetic correlations were estimated between the reproductive traits and FACE. No significant genetic correlations were estimated between reproduction and PIGM. Genetic correlations variable in sign and magnitude were estimated between the fleece traits and reproduction. EVEN and CBEL were negatively correlated with TWW and NLW, while CRIM was positively correlated with reproduction.

Of the subjective fleece traits, CBEL had the highest genetic correlation with body weight, ranging from -0.26 " 0.10 for WW to -0.35 " 0.07 for W12. The other fleece traits (SOFT, CRIM, DENS and EVEN), as well as FACE and PIGM, were not significantly correlated with body weight.

FACE had favourable relationships with CFW, MFD and SL. Genetically, fleeces with lower fibre diameter were softer, had better crimp definition, were less dense and more even, and had more creeping bellies. Staple length was negatively correlated with DENS (-0.38 " 0.09) and EVEN (-0.19 " 0.10), but positively with CBEL (0.27 " 0.08).

No significant phenotypic correlations were discernable between reproduction or body weight and any of the subjectively assessed traits. Of the objective fleece traits, only MFD had some significant phenotypic correlations with the subjective fleece traits. These were similar in sign, but smaller in magnitude than the corresponding genetic correlations, where fleeces with lower fibre diameter were softer (-0.80), had better crimp definition (-0.42), were less dense (0.48) and more even (-0.57), and had more creeping bellies (0.41).

 

4.  Discussion

From the selection objectives for Afrino sheep, it follows that selection should be aimed at increasing total weight of lamb produced (TWW), increasing the animal's own or direct growth performance (WW, W9), decreasing or maintaining fibre diameter (MFD) and at least maintaining fleece weight (CFW). 

Softness of face cover and the extent of pigmentation on the face and ears are two of the subjective traits on which a lot of emphasis is placed during selection. The results of this study indicated a favourable relationship between  softer faces and reproduction and fleece production. Selection for softer faces would therefore not adversely affect the economically important traits. The same applies for the extent of pigmentation on the face and ears, as no noteworthy genetic correlations between PIGM and any of the other traits analysed, were estimated.

Since 1988, selection for wool production in the Carnarvon Afrino flock focussed on improving crimp definition and evenness of fleece, while decreasing fibre diameter. No selection was done for amount of wool produced. The results of this study indicate that fleeces with lower fibre diameter had better crimp definition, were less dense and more even, and had larger creeping bellies. The genetic correlations estimated amongst these traits imply a strong genetic relationship amongst them.

This is clearly illustrated through the genetic trends in fibre diameter, creeping belly and crimp definition, depicted in Figure 2. From these it is obvious that selection for decreased fibre diameter and improved crimp definition in the Carnarvon Afrino flock lead to the appearance of more creeping belly.

 


Fig. 2. Genetic trends in fibre diameter (MFD), crimp definition (CRIM) and creeping belly (CBEL) in the Carnarvon Afrino flock

Decreasing fibre diameter and improving crimp definition are two of the most important aspects in the breeding plan of Afrino breeders. At the same time, breeders discriminate strongly against creeping bellies, even to the extent of culling animals superior in terms of the economically important traits. This practice should be reviewed, as indiscriminate selection against creeping bellies will result in increased fibre diameter and decreased crimp definition.

Other unfavourable correlations, in light of the current practice of discriminating against creeping bellies, are the genetic correlations estimated between CBEL and reproduction and body weight at all ages. Animals with higher body weights and ewes with higher reproductive rates also had larger creeping bellies. The economic implications, in terms of loss of genetic gain in body weight and reproductive rate when selection is done against creeping bellies, are obvious. However, the implication on wool income if less selection emphasis is placed on creeping belly, should be investigated further. Most probably this would be far less detrimental than anticipated, owing to the fact that fleeces with creeping bellies are usually finer, and would therefore fetch higher prices. It is proposed that only animals with extreme creeping bellies be culled.

Other subjective fleece traits analysed would not be influenced to any major extent when selection is done for increased body weight or reproduction. The possible exception is evenness of fleece, which is negatively correlated with the reproduction traits.  A similar conclusion, albeit based on phenotypic correlations, was drawn by Olivier et al. (1997) for Merino sheep.

 

5.  Conclusion

There is a fine balance between the various production functions which should be kept in mind when constructing and implementing a breeding plan for any sheep breed. This is especially important in dual-purpose sheep breeds where it is expected from the ewes to produce  good quality slaughter lambs, as well as a suitable amount of good quality fine wool. It is therefore important that selection priorities be based on economic value of the traits, and that breeders should learn to place more emphasis on traits of economic importance, and less on fancy traits. With the exception of one or two, the subjectively assessed traits would not be influenced detrimentally when selection is based on the economically important production traits.

As far as creeping belly is concerned, more data, especially reproductive data, is needed to quantify the economic implications if less selection pressure is placed against creeping belly. In the meantime, it is proposed that only animals with extreme creeping bellies be culled.

 

6.  References

GILMOUR, A.R., THOMPSON, R, CULLIS, B.R. & WELHAM, S.J., 1997. ASREML, Biometric Bulletin, 3, NSW, Agriculture, p 92  

OLIVIER, J.J., CLOETE, S.W.P. & SNYMAN, M.A., 1997. Relationship between type and production traits in South African Merino sheep.  Proceedings 35th SASAS congress, Nelspruit, 1-3 July

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. Livest. Prod. Sci., 44(3) : 229-236

SNYMAN, M.A., OLIVIER, J.J., ERASMUS, G.J. & VAN WYK, J.B., 1997. Genetic parameter estimates for total weight of lamb weaned in Afrino and Merino sheep.  Livest. Prod. Sci., 48(2) : 111-116

 

Published

Grootfontein Agric 3(1) : 7-11