Avaliação de carcaças de suínos da raça Large White utilizando medidas convencionais

Abstract

With the objective of establishing the measures that best define thre quality of the carcass of swine, data of meat yielt (RCARNE), fat yielt (RGORD), dressing percentage (RC), carcass lenght by the Brazilian Method of Classification of Carcass (CCMB), carcass lenght by the American Method of Classification of Carcass (CCMA), average backfat thicness (ETM), backfat thickness at 4 cm from the dorsal (P1), backfat thickness at 6,5 cm from the dorsal line (P2), backfat thickness at 8 cm of the dorsal line (P3), area of loin eye (AOL), lean:fat ratio (RCG) and ham yield (RP), obtained in 664 swine of the Large White breed were estabilished. Initially, seeking to elucidate what the importance and possible relation structures existing among the variables, the multivariate cluster analysis was used. Afterwards, the multiple regression linear models was used. The analysis was performed by Proc Reg of the SAS package. To evaluate the importance of the variables studied on carcass quality in first phase, the partial determination coefficient of each variable (Type II) and in a second phase, the adjusted equation selection procedure called "backward" were adapted. It was verified that, except for RC, all the measures of carcass classification evauated presented significant correlation with RCARNE and RGORD. The measures ETM, P1, P2, P3 and RCG were correlated inversely with RCARNE and directly correlated with RGORD. The opposite was observed for the measures CCMB, CCMA, AOL and RP. It was verified that to evaluate carcass quality in terms of meat yield, only the measures ETM, AOL, RCG and RP are enough and in terms of RGORD, besides these previous measures, the measurement of P1 is necessary. The measures of area of loin eye and backfat thickness as proposed by MBCC proved to be the ones which best evaluated the amount of meat and fat in the carcass.