Last update: September 2, 2011 02:40:21 PM E-mail Print


Performance testing on recording schemes for mohair production


S A Fleece Testing Centre (ADSRI)

Private Bag X529

Middelburg C P




The Angora goat is one of the most specialised of all livestock breeds. It is kept almost entirely for the production of mohair, a much sought-after luxury fibre. Shelton (1984) rightly states that it may be considered as one of the success stories in animal breeding because of its present relative efficiency of production. The increase in the demand for mohair, and the resultant higher prices realized, has led to considerable interest in the breed, which presently constitutes less than 5% of the world population of goats. In some countries Angora flocks are being developed by grading up from non-Angora foundations (Shelton, 1984). This practice will inevitably lead to the production of inferior mohair and a lower average price. The high price of mohair has, at least in South Africa, also led to Angoras spreading to areas where they are naturally not well adapted. Manipulation of the environment through the provision of artificial shelter, the use of hormones and the increased provision of supplementary feed have, even in the traditional Angora areas, become a profitable method of stimulating production and are increasingly being practised. Angora goats are in some cases even being kept under systems of zero grazing. Speculations on the long-term viability of these high input production systems are not the purpose of this paper. The realities of the production system, however, have a direct bearing on breeding objectives and hence on what provision must be made for in a performance-recording scheme.

The objectives of performance recording, the traits to be considered and the in1plementing of a recording scheme are henceforth discussed, bearing possible changes in production system in mind but assuming that the object will always be mohair production.



A comparison between breeding objectives in Angora goats and milk-producing goat breeds highlights the vast diversity of the goat as a specialist producer of goods in the service of mankind. In milk-producing breeds, milk production is of primary economic importance whereas hair covering can be regarded as a "fitness" trait important in thermal regulation of the body and thus survival. In Angora goats the hair covering is of primary economic importance with milk production being a "fitness" trait related to reproduction. Darwinian fitness or adaptive value is by definition "the relative capacity of carriers of a given genotype to transmit their genes to the gene pool of the following generations" (Dobzhansky, 1959).

Traits related to fitness only are therefore important in a breed even if they have no saleable value. These traits will naturally stabilize at an optimal level in any given environment, but, in developed livestock industries, changes in environment are as much part of the production system as changes in genotype, as illustrated in the case of Angoras. Recording production and fitness related traits are therefore essential for selection and effective management.

Genetic improvement in the past has relied almost entirely on visual selection. Shelton (1984) reports some genetic improvement in fleece weights in Texas (USA) but emphasises that the rate of improvement could be considerably enhanced by a more scientific approach to breed improvement programmes. The same can most probably be said for other mohair-producing countries, including South Africa. Selection cannot be confined to a single trait (amount of mohair or fleece weight), but will naturally have to include quality (price)-related and fitness-related traits. When multiple traits are considered, their inclusion in a selection program will depend on their relative economic importance, genetic variation, heritability and genetic and phenotypic correlations with other traits. Additionally, for a performance recording scheme, it must be decided which traits require monitoring and which parameters will serve best as an aid in management.

Information regarding the parameters needed to formulate breeding objectives is, with the possible exception of the work done by Aritürk & Yalcin (1979), sketchy and some important information is completely lacking. Shelton t 1984) furthermore emphasises the fact that genetic parameter estimates theoretically apply only to the flock or population from which they were calculated. The Angora populations of the world can be seen as resembling Wright's t 1940) "island model" with little or no migration, different environments and different gene frequencies. It could therefore be dangerous to base any recommendation on results obtained elsewhere.



Fleece traits

Table 1 presents the mean, standard deviation and coefficient of variation of 19 fleece traits measured on 12-month-old goats (four months hair growth) in a run-of-the-mill South African Angora flock.


Very little is known about the relative importance of these traits in determining price, processing performance or the value of the end product. However fibre diameter and, to a lesser extent, lock length (or fibre length) has been identified as having a marked influence on price (Van der Westhuysen, 1982) on topmaking and spinning performance (Strydom & Gee, 1985) and on yarn and fabric properties (Hunter, Smuts & Gee, 1985).

No genetic parameters are presently available for South African Angora goats. Some phenotypic parameters have however been calculated (Delport & Erasmus - unpublished). Repeatability estimates for some traits, together with the comparative estimates of Yalcin et al, (1979), are given in Table 2.


There is very little difference in the estimates for comparable traits, the only exception being yield, which could be due to the very high yield of the South African flock. Of interest is the moderate to high repeatability for fleece weight and fibre diameter.

Phenotypic correlations estimated for 12 objectively measured traits is supplied in Table 3 and some of these will be discussed later.


A trait of real interest is lustre, as it is well known that the light-reflecting property of mohair is one of the main attributes of this fibre and distinguishes mohair from other keratin fibres. Van Rensburg & Maasdorp (1985) showed that the lustre value of mohair is dependent on the mean fibre diameter and the chemical treatment applied. The highly significant (P ~ 0,01) correlation of r = 0,689 between fibre diameter and lustre in Table 3 (untreated mid-side samples) is in accordance with these findings. As can be seen in Table 1, the phenotypic variation in lustre is very small, giving little scope for selection. Furthermore, the repeatability (Table 2) was found to be low (t = 0,219). Since repeatability estimates can also be used to determine the upper limits of heritability, this is also bound to be low and very little progress in this trait can be expected with direct selection.

It seems therefore that, in addition to fleece weight, mean fibre diameter is the most important fleece trait to be considered in selection. The genetic correlation between lustre and mean fibre diameter needs to be estimated to determine a possible correlated response. It should however be borne in mind that the average fibre diameter of a flock is also, to a large extent, a function of its age structure and, therefore, increasing the proportion of kids and young goats could cause a meaningful decrease in average diameter.

A disconcerting aspect is the positive correlation between fibre diameter and fleece weight. A limited number of genetic correlations have been calculated on reasonably small data sets and the results are highly variable and to some extent contradictory (Shelton, 1984). Correlated responses to selection are therefore difficult to foresee, but the generally high phenotypic correlations obtained do suggest that it will be difficult to obtain a high positive selection differential for fleece weight with a negative selection differential for fibre diameter when simultaneously selecting for the two traits. The positive correlations between body weight, fleece weight and fibre diameter further complicate matters.

Since the yield of South African mohair is generally very high, as is the correlation between greasy and clean fleece weight (r = 0,990), selection could safely be based on greasy fleece weight. However, since fleece samples must be scoured for fibre diameter determination, clean fleece weights can be used with little extra effort.

Apart from fleece weight and mean fibre diameter, a range of subjectively assessed traits are recorded through the breed society and will be listed later.


Body traits and reproductive rate

Although body size and growth rate have little or no direct economic value in Angora goat production, they are seen, for example by Shelton (1984), to be a major concern in selection programmes. The reason is that body size and development contribute directly to improved reproductive rate, survival in times of stress and probably fleece weight. The objective is, however, most likely not a maximum value but merely an "adequate" or intermediate optimum value. More research into this aspect is still needed but it seems as if such an intermediate optimum for young does would be in the region of 30 kg (Shelton, 1965). Although reproductive performance presently improves beyond this point, the extra income is unlikely to offset the additional cost when achieved by additional feed and a decrease in efficiency of feed conversion is likely to counter any genetic improvement in body weight beyond this or some other optimum owing to an increase in maintenance requirements. Shelton (1984) considers most Angora does in the USA to be below the optimum body weight. Although the figure in South Africa is not as low, supplementary feed levels are probably higher and under normal grazing conditions in a normal season with no supplementation, the proportion of young goats achieving this optimum will be very low. It does seem therefore that selection for size and growth rate is presently warranted especially if the present favourable input: output cost ratio should narrow.

As in any livestock production system, reproduction rate is of prime importance in Angora goats and this is probably the single aspect requiring the most immediate attention. Although selection for increased body weight would possibly be the easiest way of increasing reproduction rate genetically in the short term (that is if the Angora is presently at a sub-optimum body weight), performance-recording schemes offer the opportunity for more sophisticated selection procedures with better long-term prospects. Accepting the principle that the bearing of the kid (or kids) by a doe is as much a function of production as is the production of the mohair fleece leads to a concept worth considering in selection, viz. total doe productivity, which can be defined as the total production (weight) of these two products, weighted according to their relative economic value. Unlike most other breeds, however, total weight of kid(s) weaned is not directly linked to income (as fleece weight is) and this requires that a minimum value be placed on the weaning weight(s) of the kids produced by a doe. It can be argued that the fleece of the kid produced is to some extent also a function of the reproductive and mothering efficiency of the doe. This is probably valid and, as the income derived from mohair is most important, one is tempted to express all measures of efficiency in terms of mohair produced. Since it can be accepted that the mohair production of the kid has a significant sire effect, the total weight of kid(s) produced is in all probability a better or less confounding measure of reproductive efficiency. The higher mohair production, which is inevitable if the kids are fostered to a higher weaning weight, is however an important advantage which must be included in the relative economic values of the doe's own fleece weight and the weaning weight of her kid(s). Research into this aspect is currently underway.



The National Angora Goat Performance and Progeny Testing Scheme was launched as a pilot scheme in South Africa in 1983 (Delport & Erasmus, 1984). It was decided to exploit the unique breed structure of the South African Angora which can be described as a two-tier hierarchical structure with only one family group (Delport, 1982). Delport (op cit) divided the multiplier studs into "elite" daughter and daughter studs. This classification, as well as the source of rams of each category, is given in Table 4.


The parent stud and five "elite" daughter studs were selected as co-workers in an effort to develop a practical and efficient performance-testing scheme. Other breeders are however not excluded.

The scheme is run as an on-farm performance-recording scheme. A central testing station is currently not envisaged, but instead existing pedigree information across studs will be utilized as discussed later. Central testing facilities will, of course, have to be considered should efficiency of feed conversion be identified as a valid selection criterion.


INPUTS - the scheme caters for three input lists:

For reliable estimates of weaning weight it is suggested that the first shearing of the kids should coincide as closely as possible to weaning and that kids be weighed after a 12 hour starvation period.

These traits are recorded as part of the scheme for the purpose of establishing their possible contributions to income per head and aspects of functional efficiency and also to monitor any possible correlated responses to selection on measured performance.


OUTPUTS - The scheme provides two outputs for use in mass selection:


The following annually updated reports are also supplied:


Information is only starting to accumulate but it is already clear that the arrangement is not only suited for efficient dissemination of any genetic improvement that might accrue through performance recording, but also provides the opportunity for estimating genetic, environmental and phenotypic parameters. Pedigree records have been kept by all the studs for some time. Augmenting these with records of performance affords the opportunity of implementing Henderson's best linear unbiased prediction (BLUP) of breeding values. As all the studs in question are genetically tied through the parent stud, an across-stud evaluation can be done. A multiple trait reduced animal model can be applied, as the analysis will initially be limited to the six most important studs. The variance and covariance estimates needed for BLUP and obtained from this mixed model will further contribute greatly to understanding the genetics of the Angora goat.



Lack of reliable information can be seen as the major problem in formulating breeding aims and implementing a scientific breeding programme for Angora goats. An attempt has been made to define some of the information needed and some interim guidelines have been given. It is hoped that this will in some way contribute to research on Angora goat breeding.



ARITüRK, E. & YALCIN, B.C., 1979. Genetic and environmental aspects of Angora goat production. Final technical report A22-AH-3. Univ. Ankara.

DELPORT, G.J., 1982. The breed structure and a genetic analysis of the Angora goat in South Africa. M.Sc. Agric. treatise. Univ. Stell. (English summary).

DELPORT, G.J. & ERASMUS, G.J., 1984. Breeding improvement of Angora goats in South Africa. Proc. 2nd Wld Congr. Sheep and Beef Cattle Breeding. Pretoria. 393-398.

DELPORT, G.J., 1986. Fenotipiese parameters vir produksiekenmerke van Angorabokke. Karoo Agric 3 (8), 39-42.

DOBZHANSKY, T., 1959. Genetics and the origin of species. Columbia, New York.

HUNTER, L., SMUTS, S. & GEE, E., 1985. The effect of mohair fibre properties on yarn and fabric properties. Proc. 7th Int. Wool and Textile Res. Conf. Tokyo. 105-115.

SHELTON, M., 1965. The relation of size to breeding performance of Angora does. Texas Agric. Exp. Station Res. Rep. (PR-2339) 18 - 19.

SHELTON, M., 1984. Breeding and improvement of Angora goats. Proc. 2nd Wld Conf. Sheep and Beef Cattle Breeding. Pretoria. 353-361.

STRYDOM, A. & GEE, E., 1985. The effect of fibre properties on the topmaking and spinning performance of Cape mohair. Proc. 7th Int. Wool and Textile Res. Conf. Tokyo. 75-84.

VAN DER WESTHUYSEN, J.M., 1982. Mohair as a textile fibre. Proc. 3rd Int. Goat Conf. Tucson. 264-267.

VAN RENSBURG, N.J.J. & MAASDORP, A.P.B., 1985. A study of the lustre of mohair fibres. Proc. 7th Int. Wool and Textile Res. Conf. Tokyo. 243-252.

WRIGHT, S., 1940. Breeding structure of populations in relation to speciation. Amer. Nat. 74,232 - 248. Y ALCIN, B.C., 1982. Angora goat breeding. Proc. 3rd Int. Goat Conf. Tucson. 269-278.

Y ALCIN, B.C., ARITüRK, E., IMERYUZ, F., SINCER, N. & MUFTU OGLU, S., 1979. Genetic and environmental aspects of Angora goat production. 2. Phenotypic and genetic parameters for the important production traits. Istanbul Univ. Vet. Fak. Derg. 5. 19-34.



Karoo Agric 3 (9)