Estimation of carcass and growth composition in the growing pig.

Abstract

In the 1st of 2 experiments to evaluate different methods of estimating body composition in the growing pig, 80 castrated Dutch Large White X Dutch Landrace [male] piglets, from 16 litters and aged 10-11 wk, were assigned to 14 treatment groups involving various combinations of housing system (in metabolic cages, in groups in a calorimeter, or individually in a pen), feeding (ad lib. or 85 or 74% of ad lib.) and weight at slaughter (25, 65 or 105 kg). In the 2nd experiment, 16 [male] and 16 [female] Dutch Landraces from 4 litters born under specific-pathogen-free (SPF) conditions were used. Half the piglets were reared under SPF conditions, and the remainder were suckled by a host sow and reared under normal farm conditions until 6-8 wk of age, when they were placed in groups in calorimeters and were fed an irradiated diet at approx. 80% of the mean intake capacity estimated in experiment 1. Eight pigs were slaughtered at the onset of the experiment, and the remaining 24 at 105 kg body weight. Nitrogen balance was measured frequently during growth, and backfat thickness was measured ultrasonically before slaughter. After slaughter, the carcasses were dissected anatomically and into commercial cuts, and the tissues were analysed chemically. Detailed results are tabulated. In the live animal, ultrasonic measurement of backfat thickness gave the best estimate of body composition (Rsuperscript 2 = 0.6), the residual standard deviation for muscle percentage being 1.9. With increasing percentage of muscle in the carcass, the proportion of protein in the carcass and in the total body increased, and the proportion of chemical fat decreased. The variation in protein percentage was less than that in fat percentage. Regression analysis indicated that the muscle percentage in the carcass can be estimated accurately from a linear combination of percentage of ham + loin and carcass weight, the residual standard deviation being 0.9 and the residual coefficient of variation 1.5-2.0%. Estimates of muscle percentage based on backfat thickness, protein percentage in the carcass and fat percentage were less accurate than the above regression, the coefficients of variation being approx. twice as large. Values of measurements of nitrogen deposition by nitrogen balance and by comparative slaughter differed, the latter technique yielding the more accurate results. (Abstract retrieved from CAB Abstracts by CABI’s permission)