LIVE ANIMAL EVALUATION OF TWO ULTRASONIC PROBES AS ESTIMATORS OF SUBCUTANEOUS BACKFAT AND CARCASS COMPOSITION IN PIGS

Ultrasonic probe data recorded on a total of 186 Lacombe barrows and gilts were obtained by two operators using the Krautkramer USM2 and Scanoprobe 731A. Backfat measurements were taken, prior to marketing, at approximately 90 kg at locations 5 cm off either side of the midline on the mid-back and on the loin, corresponding to Canadian Swine ROP policy. These measurements were then repeated on the carcass. Correlations between average probe backfat (PF) and average probe site carcass fat (CF) were similar for both operators using the USM2 (r = 0.78, 0.84) and the 731A (r = 0.78, 0.82). The two units accounted for similar proportions of the variances of percent carcass lean (R2 = 0.35, 0.35) and of percent carcass fat (R2 = 0.45, 0.47). Consideration of the effects of sex and weight increased the R2 value by 2–6% for percent carcass lean and 6–7% for percent carcass fat. The USM2 (PF = 19.2, 20.4 mm) produced results in closer agreement with CF (20.0 mm) than did the 731A (PF = 18.9, 19.4 mm), for each operator. However, greater bias appeared to be associated with the USM2 than with the 731A, for measuring PF as an estimator of CF. Overall, it was judged that both machines were equivalent in their ability to measure subcutaneous backfat and to predict carcass composition from the live animal. Considering the additional cost of the USM2, purchase of this machine could not be justified on the basis of performance. Key words: Swine, ultrasonic probes, backfat, live animal evaluation, carcass evaluation