Ultrassonography is an important option to evaluate carcass composition and quality, as it is a non-invasive technique which quantifies different tissues in live animals. The objective of this study was to estimate the correlation between measurements taken in vivo, using ultrasound and skinfold thickness, in sheep carcasses from Santa Inês breed. Eighty-one males, aging 8 to 18 months, and weighting on average 31 kg live weight were used. The measures taken in vivo by ultrassonography were longitudinal diagonal length, longitudinal rib eye area, transversal diagonal length, transversal rib eye area. Hot carcass weight, carcass yield, carcass fat thickness, carcass length, carcass rib eye area and commercial retail cut weight (leg, loin, shoulder, ribs and neck) were determined. The analyses of variance, main components, polynomial regression and correlation were carried out using SAS® statistical program. There was no difference between castrated and intact animals for carcass quality and components, indicating that castration does not provide better carcass quality. Animals with higher leg, shoulder and rib weights had lower skin weight, wither height and weight of abdominal organs, indicating a more desirable body type for selection. Body weight, body length and rib eye area by ultrasound (in vivo) can predict the rib eye area, leg weight, carcass length, hot and half carcass weight. However, loin, neck and rib weights, carcass yield and weight of abdominal organs cannot be predicted by these in vivo measurements. The measurements taken with skinfold showed no significant correlations with carcass measures, which indicates that they are not efficient for prediction of carcass components and therefore should not be used.
carcass yield; castrated; correlation; regression; rib eye area; skinfold