Three groups (n = 147) of New Zealand mixed breed lambs averaging 170 d of age and 31.7 kg in weight were killed after a diet of pasture to determine whether the total depth of soft tissues over the 12th rib 11 cm from the dorsal midline (GR) could be measured in live lambs with sufficient accuracy to warrant its use as a selection tool for breeding flock replacements. Relationships among live and carcass measurements and carcass chemical composition also were determined. An ultrasonic measurement of GR in the live lambs was a more accurate predictor of carcass GR (r = .87) and percentage carcass fat (r = .80) than was a measurement of GR made with a needle (r = .80 and .67, respectively). Both measurements were sufficiently accurate to permit culling of over-fat lambs from breeding flock replacement prospects. The best single indicator of percentage carcass fat (r = .87) was a shoulder fat measurement, followed closely by carcass GR (r = .85). Both were superior to USDA yield grade for estimating carcass chemical composition in these young, lightweight lambs. These two measurements also were most highly related to percentage carcass protein (r = -.78 and r = -.77, respectively). These results indicate possibilities for improving the method of evaluating the composition of U. S. lamb carcasses.

Two sets of correlated responses were studied in five Romney selection lines and their controls. The lines were from one experiment selecting for yearling body weight or yearling fleece weight since 1967, from a second experiment selecting for weaning weight or yearling fleece weight since 1973, and from a third experiment selecting for yearling body weight since 1973. Overall, the five selection lines were evaluated whilst being run together with controls at a single site. The responses were faecal egg count (FEC) after weaning, using lamb crops born in 1990-93, and food intake using 1993-born 7-month-old ram lambs. Four or five new sires were used per line per year, with 1471 and 232 animals being recorded for FEC and food intake respectively. The three body weight lines were each 20-22% heavier than their respective controls in March (6 months of age) and May. Average FEC over this time period was significantly greater by 63% in one of these three lines whereas there was no difference from the control in the other two body weight lines. The two fleece weight lines were 10-11% heavier in March-May, with 34-35% heavier yearling fleece weights, than their respective controls, whereas average FEC was significantly greater (by 35 and 46%) in both selection lines. Significant correlated increases in estimated food intake were recorded in all three body weight lines and in one of the two fleece weight lines, and these increases were proportional to the liveweight response. The implications of the correlated responses and their variability are discussed.

An experiment was carried out to study direct and indirect responses to selection in Romney sheep. There were two selection lines, one selected for greasy fleece weight (GFW) and the other for liveweight (LW), maintained alongside a control line (CO). Data from lambs born in 1967-90 were analysed to quantify the correlated responses to selection. By 1990, there had been an average of nine generations of selection with an average generation interval of 2.7 years. The correlated responses analysed included yearling and ewe greasy and clean fleece weights, washing yield, yearling and ewe fleece quality traits (staple length, mean fibre diameter, fibre diameter variation, loose wool bulk, Commission Internationale de l'Eclairange (CIE) Y value (brightness) and CIE Y-Z value (yellowness)), birth weight, weaning weight, yearling liveweight, ewe pre-mating weight, and ewe reproductive rate (fertility, litter size, weaning percentage, and lamb survival). Correlated responses were estimated as the deviation of selection lines from the CO line, and genetic correlations were obtained by restricted maximum likelihood techniques. Direct responses measured in the GFW and LW yearlings were 1.7 and 1.2% per year, respectively. Correlated annual responses were 0.7% in yearling liveweight in the GFW line and 0.2% in yearling fleece weight in the LW line. Corresponding annual changes in ewe performance were 1.5% in greasy fleece weight and 0.5% in pre-mating weight in the GFW line, and 0.0 and 1.2% respectively in the LW line. Mean fibre diameter increased in yearlings and ewes of the GFW line, but did not change in the LW line. There were small but significant increases in staple length in both lines. Loose wool bulk among GFW yearlings was reduced by 0.12% per year but did not change significantly in ewes. Yearlings and ewes of the LW line both showed an increase in bulk (0.15 and 0.43%, respectively). Relative to the CO flock, CIE Y-Z among yearlings increased significantly by 1.35% per annum and 0.25% per annum in the GFW and LW flocks respectively. Among ewes it increased in the GFW line whereas the opposite occurred in the LW line. There was no significant change in percentage ewes lambing as a result of GFW or LW selection. Litter size increased and lamb survival to weaning decreased significantly, but there was a net increase per year in lambs weaned per ewe lambing (0.27% per year for the GFW line and 0.48% per year for the LW line). A95024

No abstract

An experiment was carried out to study the responses to selection for yearling fleece weight and liveweight in Romney sheep. There were two selection lines, one selected for fleece weight (FW) and the other for liveweight (LW), and a control (CO) line was maintained. Data from lambs born 1967-90 were analysed to quantify the direct responses to selection. By 1990 there had been an average of 9 generations of selection with an average generation interval of 2.7 years. Realised heritabilities and selection responses were estimated from the deviation of the selected lines from a control line and by restricted maximum likelihood (REML) techniques. Realised heritabilities were 0.36 ± 0.04 for both fleece weight and liveweight respectively when estimated using the regression of response on selection differential. The corresponding REML estimates of heritabilities were 0.39 ± 0.03 for fleece weight using data from FW + CO and 0.42 ± 0.03 for liveweight using LW + CO. The direct responses to selection, derived from deviation from CO line, were 1.20 kg (39%) and 11.9 kg (28%) for fleece weight and liveweight respectively. The corresponding annual rates of response were 0.047 kg/year (1.5%/year) and 0.53 1 Deceased A93069