Published online by Cambridge University Press: 02 September 2010
Forty-one Texel and Texel-Oxford sires were selected over 5 years (1978-82) on high and low adjusted 100-day weight from 174 ram lambs artificially reared from birth and performance tested. The correlated responses to selection were measured in 1103 crossbred progeny, out of an experimental group of crossbred ewes, the progeny being naturally reared in field conditions. Hierarchical analysis of variance was performed, fitting the usual fixed effects and interactions. Co-heritabilities (rAh2h2; rA is the genetic correlation between traits with heritabilities h12 and h22) were estimated from offspring-parent regression and from the realized responses to selection. The selection differentials achieved (1·70 standard deviations in 100-day weight between high- and low-weight sires) were lower than expected, due to mating difficulties, mortality and other requirements for the stock. The co-heritabilities for growth traits were generally positive but small (mean 0·08; s.e. 0·08). Though not statistically significant, lambs slaughtered at a fixed weight from the high-weight rams were slaughtered earlier with lighter and leaner carcasses, as expected. Similarly, lambs slaughtered according to fat cover from high-weight rams were slaughtered later with heavier slaughter and carcass weights. Generally, the co-heritabilities for carcass traits were not significantly different from zero for either slaughter group. On the basis of these results, selection on ram 100-day weight with artificial rearing may not offer any advantage over natural rearing in the improvement of lamb growth and carcass traits.