Kempster A J, Chadwick J P, Cue R I, Grantley-Smith M
Meat and Livestock Commission, PO Box 44, Queensway House, Bletchley, Milton Keynes, MK2 2EF, Great Britain.
Meat Sci. 1986;16(2):113-26. doi: 10.1016/0309-1740(86)90070-7.
Three trials, involving a total of 290 lambs, were carried out to examine the precision of probed fat and muscle measurements for estimating carcass composition in classification and grading schemes. The measurement positions and probes were not always the same in different trials but common measurements provided the basis for comparison between trials. Residual standard deviations (sd) for the prediction of carcass lean percentage averaged over trials are referred to in this Summary. Residual standard deviation for prediction from carcass weight was 3·5. The visual fat assessment currently used in the national Sheep Carcase Classification Scheme, operated by MLC, contributed significantly to the prediction (residual standard deviation = 3·0) but was less precise than a visual assessment of carcass subcutaneous fat content to the nearest percentage unit (SF(e)) (residual standard deviation = 2·6). Fat thickness measurements taken over the M. longissimus at the 12th rib with the Danish optical probe, a simple steel rule or the pig version of the Hennessy Grading Probe (HGP), showed similar precision to the classification fat class. An M. longissimus thickness measurement taken by the HGP did not add significantly to the precision. Probe fat measurements added significantly to the precision achieved with visual fat assessments (residual standard deviation with classification fat class = 2·8; residual standard deviation with SF(e) = 2·5). There would be advantages, therefore, in using both a visual fatness assessment and fat measurements in classification.
开展了三项试验，共涉及290只羔羊，以检验在分类和分级方案中，用于估计胴体组成的探测脂肪和肌肉测量的精度。不同试验中的测量位置和探头并不总是相同，但常规测量为试验间的比较提供了基础。本摘要中提及了试验中预测胴体瘦肉率的平均残差标准差（sd）。根据胴体重量进行预测的残差标准差为3.5。由肉类与家畜委员会（MLC）运营的国家绵羊胴体分类方案目前使用的视觉脂肪评估对预测有显著贡献（残差标准差 = 3.0），但不如对胴体皮下脂肪含量进行精确到最接近百分比单位（SF(e)）的视觉评估精确（残差标准差 = 2.6）。使用丹麦光学探头、简单钢尺或猪用版亨尼西分级探头（HGP）在第12肋处对背最长肌进行的脂肪厚度测量，显示出与分类脂肪等级相似的精度。使用HGP进行的背最长肌厚度测量对精度的提高并不显著。探头脂肪测量显著提高了视觉脂肪评估所达到的精度（分类脂肪等级的残差标准差 = 2.8；SF(e)的残差标准差 = 2.5）。因此，在分类中同时使用视觉脂肪评估和脂肪测量会有优势。
