Last update: August 16, 2011 03:17:15 PM E-mail Print


SA Fleece Testing Centre internationally evaluated


G.J. Delport, A.J. Watermeyer and A. Botha

SA Fleece Testing Centre, Middelburg CP, 5900


The accuracy of wool analyses done by the SA Fleece Testing Centre has been of major interest over the past 2 years. The reason, namely that high prices were paid for finer wool, is a known fact. It is, however, still of basic importance that breeders maintain full confidence in the skill and accuracy by which analyses are carried out at the SA Fleece Testing Centre. For this reason laboratory procedures as laid down by the IWTO (International Wool and Textile Organization) are stringently followed. International monitoring tests are also performed on a regular basis. It will therefore be attempted to provide an indication of the value of these tests as well as of the standards maintained by the SA Fleece Testing Centre.



For the purpose of this paper it is necessary to make a clear distinction between the procedures followed for core samples of all types of wool (including lox, etc.) in commerce and midrib sampling procedures. The SA Fleece Testing Centre has participated regularly in the IWTO "Round Trials" for commercial test houses thanks to co-operation with the Wool Testing Buro. These tests require that the Centre measure the fibre diameter of one sample on a monthly basis. The result is then compared to an IWTO-reading. As published, the SA Fleece Testing Centre conformed to this commercial standard. For the sake of thorough investigation, it was decided to participate independently in the "Round Trials" for performance test laboratories by the Australian Wool plan (similar to the South African performance testing scheme).

This international test for performance test laboratories is done twice per year. Participants include 22 laboratories for fibre diameter testing and 18 for clean yield determination. The equipment used varies from sonic fineness testers and airflow meters (the standard method) to the latest technology, e.g. the FDA and OFDA equipment (still under test). It is clear that this test represents a thorough international comparison of the methods of the SA Fleece Testing Centre with all other available methods. It also includes tests by commercial test houses such as the AWTA laboratories in Australia and the New Zealand Wool Testing Authority.

This test requires that each laboratory receive 30 wool samples representing three replicates of 10 homogeneous samples prepared by the AWTA. These samples are then tested as part of the normal routine laboratory activity. The results are sent to the Wool plan executive responsible for proper statistical analysis.



The measurement are analysed statistically to obtain a measure of the repeatability of measurements by each laboratory. The results of all the laboratories for the sixth "Round Trial" is given in Table 1.


The 95 % reliability of a single test by one laboratory is indicated as 0,9% and 2,1 % clean yield. The reliability of the results by the SA Fleece Testing Centre (lab. no. 4 Table 1) which is indicated as 0,4 μ and 2,0 % clean yield, compares favourably with the overall confidence limits. The result of a single test by different laboratories (0,0 vs. 2,6 and-0,4 vs. 1,3 for clean yield percentage and fibre diameter, respectively) compares equally favourably with that of the other laboratories. It also serves as an indication of the high standard maintained by the SA Fleece Testing Centre.



The repeatability of tests at the SA Fleece Testing Centre (0,4 μ and 2,0 % clean yield) is of particular importance to South African Breeders since it determines the accuracy of selection in their flocks. The results above are, however, reassuring in that accurate comparisons between individual animals can be made by using official indices of the performance-testing scheme. Differences between measurements on different samples from a single fleece are much larger than differences ascribed to laboratory techniques only. Fortunately the breeder can control the accuracy of sampling on each sheep. Consequently the adverse effect of sampling variation can be limited in order to take sound decisions about the selection of breeding material.

The result also indicates that ram buyers frequently  overestimate the economic value of small differences in fibre diameter. It is, for example, easy to recall cases where large price differences occurred between rams similar in all respects except a less than half a micron difference in fibre diameter. Such decisions are illogical in terms of both genetic improvement and the precision of measurement.



The aim of the SA Fleece Testing Centre is to maintain and develop a complete stock improvement service. Development research is aimed at the effectiveness of both breeding and measurement aspects. A complete and accurate information system can only benefit the South African woolled sheep breeder.



Karoo Agric, Vol. 4, No 4, 1992 (16-17)