Characterisation of the divergence of the Elsenburg Merino resource flock.

Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2012.The Elsenburg Merino flock has been divergently selected for the ability of ewes to rear multiple offspring since 1986. Updated genetic trends for reproduction are reported for the Elsenburg Merino resource flock. The objective was to determine whether genetic trends estimated previously for the Elsenburg Merino Resource flock changed significantly with the introduction of genetic material from the industry to the high (H) line. All analyses included the full pedigree file, consisting of 6547 individuals. Heritability estimates were 0.08 ± 0.02 for number of lambs weaned and 0.11 ± 0.02 for corrected weight of lamb weaned. The ewe permanent environment variance was estimated at 0.09 ± 0.02 and 0.11 ± 0.02 for number of lambs weaned and for corrected weight of lamb weaned, respectively. Genetic trends for the H and low (L) lines were divergent (P < 0.05) for all reproduction traits during the period prior to the observed breakpoints. Progress for number of lambs weaned in the H line stabilised after 1999 while a decline in response for weight of lamb weaned in the H line occurred after 2003. The change points may result from reduced selection intensity during the formation of reciprocal crossbred lines, or the introduction of unrelated industry sires in the H line. The pedigree was analysed and inbreeding trends computed for the H and L lines with the aim to test the significance of inbreeding within the lines. The software packages used for the statistical analyses were ENDOG v4.8 and POPREP web analysis software. The average inbreeding coefficients (F) were 1.47% and 0.73% for the divergently selected H and L lines. The rate of inbreeding (ΔF) per generation was 0.5% for the H line and 0.6% in the L line. The overall rates of inbreeding per generation were different in the H and L lines but within acceptable levels. The L line, however, showed an unwanted recent increase in inbreeding that will need to be considered in future. Since 2003, part of the Elsenburg Merino breeding flock was subjected to structured reciprocal within-breed crossing. Lamb survival traits and ewe reproductive performance of purebred (H and L) and reciprocal crosses (HxL and LxH) were evaluated using least squares analyses. Levels of heterosis were also assessed. The mean survival of the two crossbred lines was notably superior to the midparent value in absolute terms, although the contrast did not reach statistical significance (P = 0.098). Further research is required to establish whether this within breed heterosis for lamb survival can be exploited to decrease lamb losses. Reproduction, number of lambs born (NLB) and number of lambs weaned (NLW) in the H line was higher than in the L line (P < 0.05) while the two crossbred lines were intermediate and different from both the H line and the L line (P < 0.05) from the analyses of annual reproduction and overall “lifetime” reproduction across three lambing opportunities. Individual heterosis for annual reproduction was estimated at 2.2% for NLB, 13.8% for NLW and 8.5% for corrected weight of lamb weaned (TWW), with the estimate for NLW reaching significance (P < 0.05). Corresponding estimates for total production over three lambing opportunities were 8.7% for TNLB, 19.1% for TNLW and 13.8% for TTWW, with the estimate for NLW reaching significance (P < 0.05). Ten RAPD markers were used to study molecular divergence between the H and L lines. Phenotypic data on the lifetime reproduction of ewes born in 1999 and 2000 indicated that reproduction in the H line ewes was markedly higher than that of L line contemporaries (P < 0.01). The RAPD assay, conducted on 15 ewes from each line, used eight primers and produced 87% polymorphic loci. The mean coefficient of genetic differentiation between lines (Gst) was estimated to be 0.25. In conclusion, the H and L lines were shown to be divergent for genetic trends and levels of inbreeding. The derived estimates of heterosis may also be used to infer divergence between the lines and significant molecular divergence proven using RAPD assays