Genetic relationships between visual and objective measures of carcass composition in crossbred lambs

The aim of this study was to estimate generic and phenotypic (co)variances between objective measures and carcass visual scores, as a test of the potential value of visual scores in selection programmes to improve carcass composition in crossbred lambs. In each of 1986, 1987 and 1988, 22 Suffolk rams were chosen with either high or low scores on an index designed to increase lean growth rate. These rams were joined with 18 to 20 crossbred ewes each and their lambs were grown on grass to one of three target live weights (35.5, 41.5 and 47.0 kg) for slaughter The carcasses of 1881 lambs were visually scored for overall conformation and fatness using the standard Meat and Livestock Commission methods. Additionally, a more detailed 15-point scale assessment of conformation and a direct visual score of subcutaneous fat on the carcass were taken on 1252 lambs during the latter 2 years of the study. Carcass composition was estimated by dissection of a shoulder joint into lean, fat and bone. The possibility of combining data collected on lambs slaughtered at each of the three target live weights, for the estimation of genetic parameters was investigated. Results indicated that heritability estimates for a trait using data collected within each of the slaughter groups were homogeneous. Genetic correlations between records collected for a trait within each of the slaughter groups were not significantly different from one. These results indicated that data collected at each of the target slaughter weights could justifiably be combined. Heritability estimates were generally higher for shoulder tissue proportions (0.3) than for visual scores (0.2). Genetic correlations between all conformation scores and tissue proportions were not significantly different from 0 and therefore of little or no value in predicting carcass composition. Genetic correlations between visual scores of fat and both tissue proportions and ratios were generally high (around 0.65). These results suggest that fat scores collected on crossbred animals could be valuable in purebred selection programmes where improving carcass composition of the crossbred generation is the underlying objective.