The Ontario Pork Carcass Appraisal Project was initiated in November, 1990 as a result of a recommendation from a broad-based industry steering committee and an industry-wide campaign to secure the necessary funding. Dr. Gordon Bowman provided much of the initial vision for the project and it was through his persuasion and cajoling that many of the rest of us came to share his vision. We would also like to recognize the role of Dr. Brian Kennedy who formulated many of the project's design parameters but who did not live to see the project completed. Funding and contributions-in-kind totaling almost $1.5 million were received from all segments of the Ontario pork industry. Contributors are acknowledged earlier in this report.

The Ontario Pork Carcass Appraisal Project was undertaken with the following objectives:

• to assess carcass composition and meat quality traits in the major swine breeds in Ontario
• to determine the phenotypic, genetic and environmental parameters for carcass and meat quality traits
• to examine the use of live animal measures to evaluate carcass and meat quality
• to evaluate the feasibility of a genetic improvement program for carcass and meat quality and to develop a program if feasible

Over the course of the 3½ year project, it became evident that the project provided a number of other opportunities for collaborative research. As such, the scope of the project was expanded significantly to include the following :

• an analysis of androstenone and skatole levels undertaken by Dr. Jim Squires at the University of Guelph
• an evaluation of the PSS genotype of each pig and its' relationship with carcass and meat quality characteristics. An initial 1162 PSS gene assays were undertaken by Dr. Peter O'Brien of the University of Guelph and 1700 assays were completed by VITA-TECH Canada Inc. who also extracted DNA for the project's DNA bank.
• an assessment of the chemical composition of pork loins and bellies undertaken by Dr. Andre Fortin of the Centre for Food and Animal Research, Ottawa
• a trained taste panel analysis of the sensory characteristics of pork loins undertaken by Dr. Les Jeremiah of the Lacombe Agriculture and Agri-Food Canada Research Centre and Dr. Elizabeth Gullet of the University of Guelph

Experimental Design

From November, 1990 to March, 1994, a total of 3117 carcasses were evaluated. Fifty-seven different Ontario seedstock breeders provided the pigs for evaluation coming from a total of 118 different herds. A list of breeders is provided in the acknowledgments section of this report. The majority of pigs were of purebred origin, however, a small number of crossbred pigs were evaluated in the final stages of the project. In total there were 1388 Yorkshire, 794 Landrace, 570 Duroc, 285 Hampshire, 65 Hampshire x Duroc and 15 Lacombe carcasses in the project. Results are presented here for the four principal breeds.

All pigs were raised at the New Dundee Central Test Station. The test station consisted of four rooms, each with a capacity of 128 pigs. A different room was filled each month on an all-in/all-out basis. Pigs began test at an average weight of 30 kg. A test group consisted of four pigs (two boars, a gilt and barrow) from the same litter. Pigs were raised in pens of two - i.e. two littermate boars in one pen and the littermate gilt and barrow in another. Upon arrival at the test station, groups were randomly assigned to pens. While on test, all pigs were fed a standard 18.4% crude protein ration with a lysine level of 1.16%.

While on test, pen feed consumption was monitored. All pigs were weighed and probed at two different weights (approximately 90 and 100 kg) with an A-mode Medimatic probe. Duplicate (right and left side) backfat measurements were taken at two sites as follows:

• the point directly above the outside curve of the last rib of the pig, 5 cm lateral to the spine of the pig
• 15 cm posterior to the point directly above the outside curve of the last rib, 5 cm lateral to the spine of the pig.

Age and backfat thickness were adjusted to standard weights of 90 and 100 kg. Average feed consumption was recorded when pigs were weighed off test and feed conversion was calculated on the basis of weight gain of the true sibs in the pen while on test.

One to five days prior to slaughter, all pigs were evaluated by real-time ultrasound technology using the Aloka 500V Echo Camera. A total of six different sites were evaluated using this technology:

• Shoulder: measurement of fat depth only
• 3rd/4th last rib (referred to here as 10^th rib): measures of fat depth, muscle depth and muscle area
• Last rib: measures of fat depth, muscle depth and muscle area
• Loin site (15 cm posterior to the last rib site): measures of fat depth, muscle depth and muscle area
• 10th to 15th rib (longitudinal): measures of fat depth and muscle depth
• Ham (biceps femoris): measures of fat depth, muscle depth, b. femoris area and ham "v" fat

Pigs were slaughtered at an average weight of 106.7 kg. A portion of the boars were selected for the monthly test station sale, therefore not all boars were evaluated for carcass traits. On the day of slaughter, all pigs were weighed and tattooed and delivered to Quality Meat Packers in Toronto. Pigs were rested for a minimum of 3 to 4 hours prior to slaughter. Carcasses were graded at the 3rd/4th last rib, 7 cm off the midline of the split carcass with a Destron PG-100 fibre-optic probe and estimated lean yield was determined from the following formula:

Estimated lean yield = 55.07 - (0.377 x fat depth) + (0.052 x muscle depth)

Occasionally, carcasses would be sorted to the held rail for further inspection or removal of trimmable demerits. Demerits removed from the left side of the carcass were recovered and identified so that they could be compensated for during later dissection of the side. Two hot carcass weights were recorded for each carcass. The first included the head, leaf lard and kidneys and was used to determine dressing percent and carcass index. The second hot weight (and a corresponding cold weight - 20 hours post-slaughter) were taken after the removal of these components and recorded to provide a comparison from which moisture loss could be determined. Carcass index was determined according to the 1991 Canadian Hog Carcass Grading Settlement System using Estimated Lean Yield and Hot Carcass Weight (including head, kidneys and leaf lard).

After leaving the kill floor, leaf lard was removed from all carcasses and weighed. Salivary glands were collected from all boars and identified - samples were later analyzed for androstenone levels as part of a co-operative study. Prior to chilling (20 hours at 2C), carcass length and three midline backfat measurements were recorded for each carcass. Following chilling, the left side of each carcass was delivered to the University of Guelph Meat Lab for further analysis. Sides were separated according to the guidelines of the 1978 Agriculture Canada Livestock and Poultry Production Division Pork Carcass Cutability Specification Guide. Separation of sides was done with an electric or manual hand saw and all components were weighed on a Weigh-Tronix WI-125 scale and recorded to the nearest 0.02 kg. A detailed description of the dissection procedure as well as the meat quality and carcass composition measures is presented in Appendix 1 - Figures 1 to 5.

Collaborative Studies

• One carcass from each test group was randomly selected for further analysis which included sensory evaluation of the longissimus dorsi, chemical analysis of loin and belly samples and additional belly measurements to assess lean to fat ratios.

• Sensory evaluations were carried out in two different labs. The first 225 samples were evaluated at the University of Guelph, while the remaining 611 samples were assessed at the Agriculture and Agri-Food Canada Lacombe Research Station. Differences in methodology between the two sites were large enough to warrant separate analysis of the two data sets. Methods are described in Appendix 1 - Figure 6.

• During dissection of the loin, three 1 inch samples were removed as follows: one from each end and one from the middle. After grinding, representative samples were prepared for chemical analysis. Selected bellies were subjected to additional measurements to determine lean and fat deposition. Following measurement of these variables, homogenous belly samples were prepared for analysis of chemical composition. Preparation methods are outlined in Appendix 1 - Figure 7.

• During dissection of the loins, a small sample (less than 5 grams) of subcutaneous fat was recovered from each carcass and stored in scintillation mini vials at -85 C. In addition, a comparable tissue sample was also removed from the adductor muscle. The samples were later used to determine the pigs PSS (Porcine Stress Syndrome) genotype through DNA testing of the amino acid sequence of the gene responsible for this condition. DNA was extracted from approximately 2200 fat and tissue samples and is maintained in a DNA-bank at the University of Guelph.

• Salivary glands were routinely collected from all boars at the abattoir and later analyzed for androstenone levels. In addition, a 100 to 150 gram sample of subcutaneous fat was recovered from each boar and then stored at -30C for later analysis of skatole and androstenone levels, as part of a co-operative study described earlier.

Future Directions

The Ontario Pork Carcass Appraisal Project may be the most comprehensive pork carcass evaluation project undertaken to date, involving in excess of 250 measurements on each of 3117 carcasses of identifiable origin. The analyses reported in these proceedings represent only a fraction of the potential projects that could be addressed with this database. The data is held in trust at the University of Guelph for the benefit of the Ontario swine industry and will continue to provide the necessary direction for future research and development opportunities in genetic evaluation and carcass grading systems for many years to come.