Masters Theses

Date of Award

3-1987

Degree Type

Thesis

Degree Name

Master of Science

Major

Food Science and Technology

Major Professor

H. Dwight Loveday

Committee Members

Jim Riemann, Frank Masincupp, E. R. Lidvall

Abstract

Two hundred twenty feeder pigs (80 U.S. Nq. 1, 80 U.S. No. 2 and 60 U.S. No. 3 feeder pias) were purchased through a Tennessee graded feeder pig sale. Feeder pigs were then delivered to the University of Tennessee Plateau Experiment Station feeder pig facility where the pigs were subjectively evaluated by trained evaluators on a scale of 1 to 5 for: frame height, frame length, overall frame, propensity to fatten, body width, belly depth and bone. Objective feeder pig traits included: initial weight, length from poll to tail, body girth posterior to front leg as well as ultrasonic measurements of fat depth at the first rib, last rib and last lumbar vertebrae and an estimate of tenth rib longissimus muscle depth. A photographic slide of each pig was taken and the following life scale measurements were obtained from the developed slide: hock length, body depth, height from floor to belly, height of midpoint between shoulder and tail and length from point of shoulder to tail. A scatter plot of poll to tail length versus initial weight was generated to allot similar pigs to three slaughter endpoints of 91, 104 or 118 kq (SG I, SG II and SG III, respectively). Prediction equations were developed to estimate average daily gain and lean gain per day on feed by the subjective and objective measurements. The variable sex accounted for a large degree of variation for both average daily gain and lean gain per day on feed. In addition the mean separations tables show that barrows gained faster than gilts (P <.05). Body types, either assessed subjectively or objectively, were poor predictors of average daily gain and lean gain per day on feed. A statistical approach of canonical variate analysis was then used to attempt to transform the subjective and objective variables into a smaller composite variables to predict slaughter grade and percent lean cuts. Pooled slaughter groups were only slightly successful in predicting slaughter grade and percent lean cuts. There was and improvement in the prediction of slaughter grade and percent lean cuts as hogs reached heavier weights and expressed their genetic potential. Only a small percentage of the variation in slaughter grade and percent lean cuts was explained. The predictions were superior to the predictions of carcass merit by feeder pig grades. Finally, an attempt was made to characterize fat deposition patterns of slaughter hogs. None of the subjective measurements affected the fat deposition nattern (P >.05). Objective measures bone dimension (p <.04) and body depth {P <.09) both, influenced fat deposition patterns. The relationship of bone and fat deposition was not expected and is difficult to explain.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS