Five myths in breeding

Retired after many years at the University of the Free State’s animal science department, Prof Gert Erasmus talks to Roelof Bezuidenhout about five myths in breeding and explains why there’s no such thing as the “ideal” animal.
Issue Date: 16 February 2007

- Advertisement -

Retired after many years at the University of the Free State’s animal science department, Prof Gert Erasmus talks to Roelof Bezuidenhout about five myths in breeding and explains why there’s no such thing as the “ideal” animal.

Myth one: an animal’s true genetic potential can only be realised by optimum feeding This may be true, but what use does it have in animal breeding? It’s better to talk about ”genetic ability” because under sub-optimum feeding it’s of equal, if not greater, importance. If people want to breed animals with a low sensitivity to environmental changes, they must select in an environment opposite to the change that is desired. If they want to increase weaning or other body weights and fleece weights, they must select under poor feeding conditions. But if people want to decrease fibre diameter they must select under good feeding conditions. This is known as Falconer’s paradox.

Myth two: two or more traits that are unfavourably genetically correlated cannot be improved simultaneously This would only be true if the correlation were unity (perfect). The truth is that an unfavourable genetic correlation makes simultaneous improvement more difficult, but not impossible. A good example is the correlation between body weight and fibre diameter. For years it has been generally accepted that fine wool sheep can only be small.

- Advertisement -

Today there are many examples of very large fine wool Merinos in SA. Breeders have succeeded in decreasing fibre diameter while increasing body weight. The most efficient method of achieving this is to select on combined best linear unbiased prediction (BLUP) breeding values of the traits. The current misconception that animal breeding cannot be reduced to a single figure is giving way to the realisation that this is also difficult, but indeed highly possible and even essential. Simultaneous selection for more than one trait has become an important and advanced multi-disciplinary science with strong genetic and economic undertones.


Myth three: breeders should stick to a specific bloodline within a breed Doing so would be extremely silly because genetic diversity supplies the building blocks that are essential for genetic improvement. Breeders could even use material from other related breeds to improve specific traits. The only proviso is that a clear definition of selection criteria must be strictly adhered to. However, there is much genetic variation within the Merino that this shouldn’t be necessary in this breed.

The recently completed MSc study of Pranisha Buduram, using DNA markers, shows the Merino to have the largest genetic variation of all the sheep breeds in SA. Genetic diversity is a gift and should be nurtured. Beware of an over emphasis of breed standards and breed purity. A breed is not a static entity, but a process, as Theodosius Dobzhansky so aptly phrased it. Prof Almero de Lange (previously at the UFS’s Department of Animal Science) correctly states that breed standards are normally a description of what has been achieved in a breed and not what to strive for.

Myth four: measurement and recording of production (and reproduction) performance is the sole saviour of the stud breeding industry The opposite view – that it poses a threat to the industry – is also a misconception. Measurements (or scores, if some traits cannot be measured) are imperative, but if they are not measured correctly, accurately and meaningfully analysed and processed, they can be vastly misleading. Animals can only be genetically compared within so-called contemporary groups where all the non-genetic factors such as age, sex and nutritional and managerial level are identical. It stands to reason that breeders must strive for few, large and uniform contemporary groups.

The fact that animals can only be compared within contemporary groups makes performance testing on its own rather useless. In an effort to circumvent this problem, the Merino industry introduced group breeding schemes, veld ram clubs, control tests, a centralised progeny test and the show for measured production. But BLUP has made all these largely redundant and they should eventually be phased out. It is, however, important to realise that these efforts were instrumental in giving breeders insight into how animals could be genetically improved in practice. Breeders that have gone through these schools should in future be the most effective in applying BLUP breeding values, simply because they understand the basic principles. Making performance testing compulsory gives a breed society and its members the false complacency that they are scientifically correct and advanced.

Myth five: breeders should strive for an “ideal” animal “Ideal” means there cannot be anything better. The “ideal” animal is readily described in meaningless descriptions as having a broad chest, wide mouth, straight back, strong nose and soft face. However, it is impossible to attach actual figures to these nondescript terms. The aim of selective breeding should be to shift the mean of the population in the desired direction for every important trait. Unfortunately the observation of Helen Newton Turner that “the image of the ideal animal has at best been shaken but not shattered” still holds true.

Contact Prof Gert Erasmus on (028) 735 2419. |FW