Last update: August 12, 2011 02:53:25 PM E-mail Print

 

Breeding plans for South African Angora goats

 

M.A. Snyman, J.J. Olivier & D. Wentzel

Grootfontein Agricultural Development Institute, Middelburg, 5900

 

Introduction

Over the years, South African Angora goats have continuously been selected for increased mohair production. Considerable progress has been made in this regard, as is evident from the fact that the Angora goat is at present the most efficient fibre producing small stock breed in the world. However, the inability of Angora goats to survive and produce under sub-optimum conditions is also well known. Energy supplementation, in the form of starch, is an effective way to circumvent this problem. This practise has, however, become too expensive, especially during times when mohair prices are low. Furthermore, preliminary studies indicated a possible negative relationship between fibre production potential and the physio-endocrine mechanisms influencing hardiness in small stock (Herselman, 1990; Herselman et al., 1993).

 

The above mentioned facts, together with the relative price difference between fine and strong mohair, have convinced Angora goat stud breeders to review their existing breeding policy. The Department of Agriculture has subsequently been requested to develop a scientific breeding plan to meet their specific needs. In the reviewed breeding policy, the emphasis on increasing fleece weight, has shifted to increasing body weight, while either decreasing or keeping fibre diameter constant. In order to achieve this, selection should be based on a selection index into which these traits are incorporated. Before a selection index can be constructed, heritabilities and genetic as well as phenotypic correlations for the traits in question, need to be known. Estimates of heritabilities and genetic correlations for the economically important traits in South African Angora goats are, however, unknown.

 

Owing to the fact that young rams are sold at 14 to 16 months of age while carrying their third fleece, they have to be performance tested at 8 to 9 months of age at the second shearing. The question arises whether superior animals could be identified accurately when testing is done at this relatively young age.

The objectives of this study were, firstly, to estimate genetic parameters  for body weight, greasy fleece weight and mean fibre diameter measured at 8 to 9 months of age in South African Angora goats, and secondly, to construct selection indices to support the revised breeding policies of some of the Angora goat stud breeders.

 

Material and Methods

The data used for this study were collected through the Angora goat performance testing pilot scheme run by the Department of Agriculture at the Grootfontein Agricultural Development Institute. The pilot scheme was initiated in 1991. Data of kids born from 1990 until 1993 in 17 different studs were used in the analysis. Traits analyzed were body weight (BW), greasy fleece weight (FW) and mean fibre diameter (FD).

 

Performance testing of these traits were done at 8 to 9 months of age on the second shearing for both ram and ewe kids. Body weight was recorded 1 to 2 days after shearing. Mean fibre diameter was determined from midrib samples by means of a Fibre Diameter Analyzer. Greasy fleece weight was recorded just after shearing and corrected to 180 days' hair growth prior to data analysis. Pedigree information of the performance tested animals was obtained from the Angora Goat Stud Breeders' Society of South Africa.

 

Genetic parameters were estimated under animal models, using the DFREML programme of Meyer (1989, 1991a, 1991b). In an animal model, all relationships between all animals are taken into account when estimating genetic parameters, as opposed to a sire model, where full pedigree information can not be utilised.

 

For the calculation of repeatabilities, data were collected on the 1991- and 1992-born ewes of eight stud breeders at the second, third and fourth shearings.

 

Results

The mean, as well as the minimum and maximum values for each trait are summarized in Table 1.

 

Table 1. Description of the data set

  Body weight Fleece weight Fibre diameter
Mean 27.2 kg 2.4 kg 29.9 µ
Minimum 8.5 kg
0.7 kg 17.2 µ
Maximum 54.0 kg 4.9 kg 43.5 µ
CV 28.1 % 28.1 % 11.3 %


From Table 1, it is evident that large variation exists in the stud industry with regard to body weight, fleece weight and fibre diameter. This variation is mainly due to environmental and management differences between studs.

 

Heritability and correlations

Estimates of heritability and genetic and phenotypic correlations between body weight, fleece weight and fibre diameter are presented in Table 2.

 

Table 2. Estimates of heritabilitya and geneticb and phenotypicc correlations between body weight, greasy fleece weight and mean fibre diameter


  BW FW FD
BW 0.34 0.70 0.62
FW 0.57 0.22 0.55
FD 0.54 0.57 0.30

a Heritabilities on diagonal

b Genetic correlations above diagonal

c Phenotypic correlations below diagonal

 

Heritability estimates of 0.34, 0.22 and 0.30 were obtained for body weight, fleece weight and fibre diameter respectively. These estimates fall within the ranges reported for Turkish (Yalcin, 1982), New Zealand (Nicoll et al., 1989) and Australian (Gifford et al., 1991) Angora goats for body weight (0.13 to 0.50), fleece weight (0.13 to 0.45) and fibre diameter (0.11 to 0.51).

 

Phenotypic correlations estimated between BW and FW (0.57) as well as between FW and FD (0.57) in this study, are similar to those estimated by Delport (1987) and Nicoll et al. (1989), but higher than other reported estimates (Yalcin, 1982; Gifford et al., 1991). The phenotypic correlation of 0.55 estimated between BW and FD, however, is higher than any other reported phenotypic correlation between BW and FD.

 

Estimates of genetic correlations obtained between BW and FW (0.70) and between BW and FD (0.62) in the present study, are considerably higher than corresponding estimates in the literature. For FW and FD, however, the genetic correlation of 0.55 estimated in this study is lower than that of 0.75 and 0.98 reported by Shelton & Bassett (1970) and Nicoll et al. (1989) respectively.

 

Although these differences can possibly be ascribed to differences in the respective data sets and different models of analysis used, it is more probable that they are the result of actual differences in phenotypic and genetic correlations between the different populations.

 

From these results it is evident that selection for any of the traits analyzed, will lead to a correlated increase in the other two traits. In respect of the revised breeding policy of increasing body weight, decreasing or keeping fibre diameter constant and maintaining fleece weight, these correlations are unfavourable. Selection for increased body weight, for instance, will lead to an unwanted correlated increase in fibre diameter as well as in fleece weight. It is therefore obvious that the only way in which body weight could be increased genetically, without increasing fleece weight and fibre diameter simultaneously, would be to incorporate these traits into a suitable selection index.

 

Repeatabilities

Repeatabilities calculated between the second and third, second and fourth, as well as third and fourth shearings are summarized in Table 3.


Table 3. Repeatability estimates for body weight, fleece weight and fibre diameter

Trait Between 2nd & 3rd shearings
Between 2nd & 4th shearings
Between 3rd & 4th shearings
Body weight 0.83 0.72 0.77
Fleece weight 0.56 0.46 0.60
Fibre diameter 0.82 0.73 0.81


Repeatability estimates (Table 3) obtained for body weight and fibre diameter between the different shearings were very high, which imply that selection of superior animals can be done successfully on the second shearing. For fleece weight, however, substantially lower repeatabilities were calculated, and it therefore appears that fleece weight measured at 8 to 9 months of age does not provide an accurate prediction of the animals' future hair production.

 

Selection indices

Genetic parameters estimated from the data that had been collected during the first year (1991) were used at that time for the construction of preliminary selection indices. The genetic parameters estimated in the present study, are based on much more data and are therefore more accurate than those used for the construction of the preliminary selection indices. However, the old and the new selection indices select virtually the same animals.

 

The following two selection indices were constructed using the more accurate genetic parameters estimated in the present study :

The first selection index (SI:BWFD) will increase body weight, decrease fibre diameter and keep fleece weight constant. In this index, equal importance was assigned to body weight and fibre diameter.

                                    SI:BWFD = (13 x BW) + (4 x FW) - (23 x FD)

 

The second index (SI:BW) will increase body weight and keep fleece weight as well as fibre diameter constant.

                                    SI:BW = (3 x BW) + (15 x FW) - (1 x FD)

 

These two selection indices are used by most of the breeders who participate in the performance testing scheme. Other selection indices can be constructed on request.

 

If selection based on these selection indices is carried out consistently for a ten year period, for example, the theoretical progress achieved with SI:BWFD would be a 6.4 kg increase in body weight, a 3 micron decrease in fibre diameter, with no change in fleece weight. In the case of SI:BW, a 9 kg increase in body weight, without any changes in fleece weight and fibre diameter is theoretically possible over a ten year selection period. These responses represented an annual progress of 1 to 2 percent, which is the theoretical average for most biological traits in farm animals.

 

These theoretical optimum gains are, however, never achieved in practice. The primary reason for this is the fact that a lot of other traits are also taken into consideration when selection is done. Furthermore, in the calculation of theoretical gains, it is assumed that selection of replacement ewes is also based on this index. This is, however, rarely possible and selection is usually done on rams only.

 

Considering the fact that fleece weight measured at 8 to 9 months of age is not a very accurate predictor of future fleece weights, and the fact that most breeders only want to maintain fleece weight, it was decided to investigate the possibility of selection indices from which fleece weight is excluded. These indices are as follows :

 

                                    SI:BWFD2 = (13 x BW) - (24 x FD)

                                    SI:BW2 = (13 x BW) - (15 x FD)

 

In the case of the first selection index, exactly the same animals are selected in exactly the same order when either SI:BWFD or SI:BWFD2 is used. The selection indices SI:BW and SI:BW2 also select the same animals, but not in exactly the same order. It seems therefore that fleece weight does not need to be included in the selection index in the case of these two specific indices. However, fleece weight must still be measured, and indices for fleece weight for the individual animals can still be calculated, to allow the culling of animals with extreme fleece weights if the breeder wishes to do so. Fleece weight data will furthermore also be needed for BLUP-analysis and to monitor the genetic trend.

 

Conclusion

The data set used for this analysis incorporated data from leading studs, in terms of ram sales, in South Africa. It would therefore be reasonable to assume that the genetic parameters estimated, apply to the entire Angora goat stud industry, and possibly to a lesser extent to the commercial mohair industry as well.

 

From the results it appears that the economic important traits in South African Angora goats are low to moderately heritable. Selection can therefore be done on the individuals own performance. However, increased genetic gains in respect of these traits can be achieved if selection is based on BLUP of breeding values. A study done by Belonsky & Kennedy (1988) has shown that, for a trait with a heritability of 0.30, selection based on BLUP of breeding value can increase gain by 20%, when compared to selection based on the individuals own performance only.

 

The relatively high correlations estimated between body weight, fleece weight and fibre diameter in this study, is a cause for concern. The high genetic correlation estimated between BW and FD is of special interest, as increasing body weight is the most important aspect of the revised breeding policy of many stud breeders. Selection for an increase in body weight will lead to an unwanted correlated increase in fibre diameter as well as in fleece weight. The only way in which genetic progress in the desired direction would be achieved, will be through selection based on selection indices into which these traits are incorporated.

 

 

References

BELONSKY, G.M. & KENNEDY, B.W., 1988. Selection on individual phenotype and best linear unbiased predictor of breeding value in a closed swine herd. J. Anim. Sci. 66(5) : 1124 - 1131

DELPORT, G.J., 1987. Phenotypic parameters for production characteristics of Angora goats. Angora Goat and Mohair Journal 29(1) : 10-17

GIFFORD, D.R., PONZONI, R.W., LAMPE, R.J. & BURR, J., 1991. Phenotypic and genetic parameters of fleece traits and live weight in South Australian Angora goats. Small Rum. Res. 4(3) : 293-302

HERSELMAN, M.J., 1990. Die energiebehoeftes van Angorabokke. MSc(Agic), Univ. Stellenbosch

HERSELMAN, M.J., OLIVIER, J.J. & WENTZEL, D., 1993. Varying fibre production potentials under veld conditions. Karoo Agric. 5(1) : 8-10

MEYER, K., 1989. Restricted maximum likelihood to estimate variance components for animal models with several random effects using a derivative‑free algorithm. Genet. Sel. Evol. 21 : 317‑340

MEYER, K., 1991a. Estimating variances and covariances for multivariate animal models by restricted maximum likelihood. Genet. Sel. Evol. 23 : 67‑83

MEYER, K., 1991b. DFREML : Programs to estimate variance components by restricted maximum likelihood using a derivative‑free algorithm. User notes, Ver. 2.0

NICOLL, G.B., BIGHAM, M.L. & ALDERTON, M.J., 1989. Estimates of environmental effects and genetic parameters for live weights and fleece traits of Angora goats. Proc. New Zealand Soc. Anim. Prod. 49 : 183-189

SHELTON, M. & BASSETT, J.W., 1970. Estimate of certain genetic parameters relating to Angora goats. Texas Agricultural Station Research Report (PR-2750) : 38-41

YALCIN, B.C., 1982. Angora goat breeding. Proc. 3rd Int. Conf. Goat Prod. Disease, Tucson, Arizona

 

Published

Angorabok- en sybokhaarblad, 38(1) : 23-31