Context. Mortality of new-born lambs is well known to have important implications for both animal production and welfare. Improving survival rates by genetic selection is very desirable, but the low heritability of survival traits challenges the prospect of useful genetic gain.
Aim. This study aimed to derive genetic and environmental parameters for lamb survival in the Elsenburg Merino resource flock. It also investigated correlations to possible indicator traits birthweight and birth coat score and reported genetic progress in breeding values for lamb survival following 33 years of divergent selection.
Methods. The flock was divergently selected for number of lambs weaned and was separated by the High (positive) and Low (negative) selection groups. The pedigree file identified 8138 lambs as the progeny of 273 sires and 2062 dams. The study considered total survival from birth to weaning (Tsv) that was also partitioned into perinatal survival to 3 days of age, and the remaining period. Variance components were derived by linear mixed models by using the ASREML® program. Genetic trends were derived by predicting mean breeding values for selection groups within each year and evaluated by fitting linear and broken-stick regression models.
Results. Predicted Tsv of H-line lambs (0.81 ± 0.01) was higher (P < 0.01) than that of L-line lambs (0.68 ± 0.01). Heritability was significant but low for survival traits (0.03-0.07), moderate for birthweight (0.16) and high for birth coat score (0.54). Genetic trends of the H-line trended divergently (P < 0.01) to the L-line for survival traits, but a changepoint (P < 0.01) in trend suggested that the H-line reached a selection plateau following 19-22 years of selection. Preceding this period, the rate of genetic change equalled ~1% of the mean for Tsv.
Conclusions. Despite the low heritability of survival, the genetic trends reported in this study contradicted the premise that genetic selection is not a worthwhile method to reduce incidences of lamb mortality.
Implications. It is recommended that lamb survival phenotypes should be recorded and incorporated into indices individually where possible, but composite traits are a viable alternative.