Effectiveness of morphometric measurements for improvement of the prediction accuracy of beef carcass composition

Investigations were conducted into the effectiveness of morphometric measur ements (MM) of beef carcass for the prediction of weights and percentages o f muscle and fat using 237 carcasses of two pure and five cross-bred breed types. In addition to the MM, side weight (CW) and cross-sectional (SM) mea surements were used for prediction by multiple regression analysis, singly or in combinations of these measurements. The cross-sectional measurements were made on three sections of a rib joint named R6, R9 and R12 which were cut at the anterior position of the 6th, 9th and 12th thoracic vertebrae. F or the prediction of weights and percentages of muscle and fat, the MM grou p was the best predictor. The CW was effective only for the improvement of tissue weight prediction. The CW + MM + SM(R12) and the MM + SM(R12) combin ations were best for the tissue weight and the percentage prediction, respe ctively. A synergistic effect was observed between MM and SM for the improv ement of tissue percentage prediction accuracy. Regrouping the data to thre e breed types by morphological differences and analyzing it by a principal component analysis improved the prediction accuracy. Carcass weight range w hich caused differences in carcass fat content did not affect the predictio n accuracy of muscle and fat percentages in the carcass when MM were used. (C) 1999 Elsevier Science Ltd. All rights reserved.