Selection for clean fleece weight in Merino sheep. I. Direct response to selection

Abstract

Two closed selection flocks of Merino sheep consisting of 100 ewes mated to five rams each were established at the Agricultural Research Station, Trangie, N.S.W., in 1951. One flock was selected for high clean fleece weight (Fleece plus) and the other for low clean fleece weight (Fleece minus). Responses to selection in these flocks over approximately five generations were measured against a randomly selected control flock (100 ewes mated to 10–25 rams). All flocks were selected from the same base population (1700 ewes and 150 rams) and were maintained in the same environment.

The response patterns were asymmetrical. In the Fleece plus flock there was an initial increase of approximately one standard deviation in the first two generations but no consistent response was observed in the following three generations. This is shown by realized heritabilities of 0.53 and 0.11 for ewes in the first two, and subsequent generations respectively, the corresponding figures for rams being 0.65 and 0.02. In contrast to this pattern, response in the Fleece minus flock was approximately linear with overall realized heritabilities of 0.44 and 0.38 for ewes and rams respectively.

The average level of inbreeding increased between 4 and 5% during selection, and the pooled partial regression of inbreeding coefficient on clean fleece weight independent of the effects of selection was –0.033 kg per 1% inbreeding. This would not have been sufficient to cause a cessation of response after the first two generations.

Sampling variation and genotype x environment interaction are possible causes of this response pattern. An additional explanation is suggested which is based on the fact that clean fleece weight is a complex character in which two components, fibre density and cross-sectional area, have a negative genetic correlation between them which may become stronger under selection. It is shown that this could account for a large proportion of the reduction in realized heritability. Thus it is suggested that breeding plans should aim to control fibre diameter while selecting for high clean fleece weight if continued response is to be obtained.