Lamb wool shedding is a good predictor of ewe wool shedding.

Interest in reducing labour costs due to shearing has led to development of breed types that shed their wool naturally. Selection at young ages can facilitate response. Reliability of predictions of adult from lamb wool shedding (WS) is thus key in the design of breeding programmes to increase shedding. Our objectives were to estimate heritabilities and genetic relationships between WS measured once in lambs and repeatedly in ewes and to assess the accuracy of lamb WS EBV to predict ewe WS EBV based on a multi-trait threshold or a repeatability model. Data were 4,971 lamb and 3,335 ewe WS records on a Romanov, White Dorper and Katahdin composite flock. For the multivariate model, WS heritability ranged from 0.47 ± 0.03 in lambs to 0.59 ± 0.04 at 1 year of age. For the repeatability model, WS in adult ewes was moderately heritable (0.50 ± 0.03) and repeatable (0.60 ± 0.02). Genetic correlations were 0.72 ± 0.04, 0.65 ± 0.05, 0.50 ± 0.09 and 0.51 ± 0.09 between lamb WS and 1st through 4th record, respectively. Given the moderately high heritability and high correlations between WS performance in lambs and ewes, selecting animals early in life would effectively increase WS in crossbred flocks.

Estimating heritability of wool shedding in a cross-bred ewe population.

Low wool prices and high production costs in sheep systems have resulted in the introduction of genotypes that shed wool into flocks to reduce shearing costs. Wool shedding occurs naturally in a few breeds and can be incorporated by cross-breeding. The opportunity to enhance shedding through selection depends on the extent of genetic variability present. Genetic and environmental parameters for wool shedding for ewes from a three-breed composite population were estimated using Bayesian inference. Data on 2025 cross-bred ewes, including 3345 wool shedding scores (WS) and 1647 breeding weight (BW) records, were analysed using bivariate and, for WS, univariate animal repeatability models. Breeding weight was included to account for possible selection bias. Breeding weight was moderately heritable and highly repeatable with means of 0.317 and 0.724, respectively. Under both models, WS was found to be moderately heritable and repeatable with means of 0.256 and 0.399, respectively. Based on a cumulative link model and contingency table analysis, age and reproductive activity influenced the extent of WS (p < 0.05). Given that WS is moderately heritable, selective gain in WS can be achieved.