COMPARISON OF METHODS FOR ESTIMATING CARCASS FAT-CONTENT OF YOUNG CHAROLAIS BULLS IN PERFORMANCE TESTING STATION

With the purpose to integrate carcass composition in the selection cri teria applied in performance testing stations, accuracy of four in viv o techniques for predicting carcass fatness was evaluated on a sample of young Charolais bulls (n = 136) that were relatively homogeneous (s tandard deviation of carcass fat content: 2.2%) and representative of young bulls performance tested in France. Carcass composition was esti mated from dissection of the 6th rib and the weight of internal fat. I n vivo measurements were: (a) handling scores, (b) fat thickness measu red on scans performed at the 10th and 13th thoracic and 3rd lumbar ve rtebrae, (c) speed of ultrasound through the back and the hind limb, ( d) the diameter of subcutaneous adipose cells in a biopsy removed from the top of the thigh. Except for the latter technique, each value was the average of two measurements taken at proximate locations. The tec hniques were compared on their ability to explain variation in carcass fat in addition to that using live weight in a model with contemporar y group effect fitted. Adipose cell diameter and the speed of ultrasou nd through the back ranked first and explained 42% of the variance in fat content. The speed of ultrasound through the hind limb explained s lightly less: 31%, while handling scores and scanning explained only 1 2% to 17%. Taking into account these differences in accuracy and diffe rences in repeatability and practicability of these four techniques, m easurement of the speed of ultrasound through the back appears to be t he method that is most potential for inclusion in performance tests in beef bulls. (C) 1997 Elsevier Science B.V.