Jacques P. Chesnais
Canadian Centre for Swine Improvement Inc.


The swine industry is of major importance to the Canadian agricultural sector. About 17.6 M hogs were produced in 1995, accounting for close to $2.2 billion in farm cash receipts, more than 15% of the total agricultural product. Of these $2.2 billion, $190 M correspond to the export of live hogs (mainly to the US).Exports of pork meat have increased considerably over the last few years. In 1995, they amounted to about $1 billion (mainly to the US, Japan, Mexico). Canadian pork is known as a superior product on these markets.

Swine production in Canada is characterized by a high level of investment, and is generally carried out in a total confinement system. To produce in these intensive conditions, animals must be able to sustain themselves on concrete or slatted floors. They must grow quickly, convert feed efficiently and perform well in a high density situation. Producers are paid on the basis of carcass leanness and weight, therefore there is a strong incentive to produce lean pigs.

Selection is based on a balanced approach, i.e. one that aims to improve over-all animal productivity. Selected traits include those contributing to efficiency of production (e.g. growth rate, feed efficiency, physical soundness), to quality of product (leanness, carcass and meat quality), and to efficiency of reproduction (litter size, farrowing interval, longevity). Although feed efficiency is monitored on groups of animals in test stations, progress for this trait is mostly achieved through indirect selection for lean, fast-growing pigs. Experience proves that this is more efficient than direct selection for feed efficiency (Kennedy et al, 1991). Until recently, selection for carcass and meat quality was based primarily on leanness and freedom from the halothane gene. However, the number of traits being considered is expanding rapidly. Selection on reproduction traits is also evolving, with the introduction of EBVs for a number of these traits (official EBVs for litter size, pilot EBVs for farrowing interval, research on longevity traits).


The suppliers of swine genetic material in Canada can be divided roughly into three groups:

Breeding herds on the Canadian Swine Improvement Program (CSIP) form one of the largest super-nucleus schemes in the world (about 12,000 breeding sows). The national program is based on a series of tools which allow participants to identify animals, record performance data and evaluate breeding stock jointly. The Program is based on the concept of independent verifiable data and publicly available results (all EBVs can be accessed directly on the data base APigs on the Web@). Participants can compare their own genetics to that of other breeders on the program, and acquire (mostly through AI) the genetic material they need to enhance their own selection program. In this fashion, breeders and breeding companies in the program benefit from each other=s selection efforts.

The Canadian program consists of a set of integrated components: the animal identification program, the home test program, the test station program, the sow productivity program, the genetic evaluation program, and the health monitoring program.

The swine improvement program is delivered to breeders in Canada through a two-tier structure. The Canadian Centre for Swine Improvement (CCSI) is responsible for developing national standards and computing national genetic evaluations (national EBVs). The provincial or regional improvement centres are responsible for the direct delivery of the program to breeders. They have a staff of accredited technicians who capture data (weight, probe and sow productivity records) and produce reports for the breeders. Interim EBVs for each animal are computed the day of the test using a micro-computer program. Parent EBVs from the national system are used to account for genetic differences across herds, so that interim EBVs can be compared from herd to herd.

National standards are established for a number of swine improvement activities: accreditation of technicians, data collection, recording and reporting, record formats for inclusion in the national data base, genetic evaluation methods, health monitoring, nutrition guidelines for test station animals, development of software, etc. This is an evolving process as new components are continually added to the program, e.g. meat and carcass quality standards, health monitoring plan, etc. These responsibilities, which were formerly carried out by Agriculture Canada, are now handled by CCSI.

The national identification system is an essential part of CSIP. All animals on the program and their parents must have positive identification based on the standards established by the Canadian Swine Breeders Association (tatoos or ear notches). In most cases, the animals= pedigrees can also be verified and traced back through the national registration system administered by the Canadian Livestock Records Corporation.

The home test program is a key component of the genetic improvement system. More than 100,000 pigs are currently in the program. Pigs are tested for backfat thickness and age at 100 kg. Backfat measurements are taken in 4 locations (right and left, mid-back and loin) using an ultrasonic machine. In many herds, blood samples are also taken to identify any halothane gene carriers with the halothane gene probe. The frequency of the halothane gene in the Canadian swine breeding population was relatively low to begin with (1 to 2% of reactors) and is being progressively reduced with the use of the gene probe. The elimination of the halothane gene will result in lower levels of stress-induced mortality and higher meat quality (less PSE).

The original purpose of the test station program was to compare boars from different herds in the same environment. Through such comparisons, the test stations increased the number of genetic "connections" in the population, improving the accuracy of national genetic evaluations. Boars were tested between 30 and 100 kg for backfat thickness, average daily gain on test, and group feed efficiency. Over the last two years, the role of the test station has changed. More and more, they are being used for research, for the evaluation of meat and carcass quality (through sib-testing), or for the testing of commercial animals. Segregated early weaning (SEW) is being introduced to reduce health risks and increase the accuracy of comparisons among animals from herds of different health status.

The sow productivity and management program was designed primarily as a tool to help producers with the management of their herds. However, it can also be used for the selection of highly prolific sires or dams, since genetic evaluations identify the top animals for litter size in the Canadian population. The program provides producers with statistics for the herd and for each sow on traits such as age at first farrowing, farrowing interval, number born per litter, mortality rate from birth to 3 days, mortality rate from 3 days to weaning, number pigs weaned per sow per year, etc.

The Canadian genetic evaluation program is one of the most advanced in the world. Designed in 1985 in cooperation with the late Dr. Brian Kennedy, of the University of Guelph, it was the first national swine evaluation system ever to use the Best Linear Unbiased Prediction method, now recognized as the method of choice for genetic evaluation of livestock. The evaluation program integrates information on each animal and all its relatives; it generates estimated breeding values (EBVs) that can be compared across herds; and it allows for the identification of the top breeding stock in the country.

The genetic evaluation program calculates EBVs for age and backfat thickness at 100 kg, litter size, a sire line index and a dam line index.

The sire line index is designed for the selection of terminal sire breeds or lines and combines age and backfat EBVs. The dam line index is designed for the selection of commercial sow breeds or lines. It combines EBVs for litter size, backfat thickness and age. Index weights are based on economic weights and on the standard deviations of EBVs for all traits.

Every three months, and starting in 1997, every month, national EBVs are computed for every pig, sire and dam using all data collected in Canada until the time of the evaluation. A sire and dam report is published after each evaluation. On-farm EBVs are computed on the day of probing for each tested pig, to permit their rapid selection. The on-farm EBVs are linked to the national evaluation system through the national EBVs of their parents.

The Canadian health monitoring program is based on strict government controls within Canada and at border points. As a result, the Canadian swine population is free from major swine diseases. In addition to federal health controls, there are several provincial programs that monitor health in individual swine herds or that analyze data from slaughter plant inspections (e.g. APHIN).


The Canadian genetic improvement system has had a large impact on swine productivity (see Figures 1 to 4). In the last five years, genetic progress was responsible for 2/3 of all progress for growth rate and 4/5 of all progress for backfat thickness. This clearly demonstrates the importance of genetics as a tool for making swine production more efficient.

For example, from 1980 to 1995, age at 100 kg in the Canadian Yorkshire breed has decreased by about 20 days. Genetic progress was responsible for 14 of these 20 days. Other factors (health, nutrition, management) were responsible for the other 6 days.

Similarly, backfat thickness in the Canadian Yorkshire breed has decreased by 4.4 mm from 1980 to 1995. Genetic progress accounts for 3.9 of these 4.4 mm It goes without saying that genetic has been the major factor in increasing lean yield in Canada. Results for both age and backfat were similar in other Canadian breeds.

Litter size does not respond to selection as readily as backfat or even growth rate. In addition, EBVs for litter size have not been available until fairly recently. Nevertheless, there has been a small genetic increase for litter size in the Canadian Yorkshire population (and in other breeds) since 1987. This trend is particularly encouraging since it demonstrates that selection for production traits has not negatively affected litter size.

The Canadian genetic improvement system has paid off and is continuing to do so.

Based on Dr. B. Kennedy's research, and using the current rates of genetic change, the benefits arising from swine selection in Canada are conservatively estimated at $133 M per year. On the other hand, the all-inclusive cost of the selection program (including losses by breeders due to the marketing of intact boars) was $8 M per year. Selection is a good investment.


During the last two years, national standards have been established for the recording of carcass and meat quality measurements, in support of the various programs developing in this area. Core measurements for meat quality include marbling score, structure score, colour score, fat hardness score and muscle pH. Other measurements for which standards are available are drip loss, loin colour, tenderness and iodine number.

Core measurements for carcass quality include estimated lean yield, depth of loin muscle, depth of loin fat, carcass length and carcass weight. Other measurements of carcass quality are weight of loin, loin area, weight of ham, belly thickness and complete carcass cut out.

There are a number of projects on-going in Canada in the meat and carcass quality area. The test station program entails the testing of boars along with 2 to 5 sibs (gilts and barrows), which are slaughtered for meat and carcass quality evaluation after the test. The boars can then be evaluated for meat and carcass quality on the basis of sibs and other relatives. The test station results have been encouraging, and meat and carcass evaluation is being extended to farm testing (i.e. the boars and sibs are tested on the home test program, but the sibs are later slaughtered and evaluated for meat and carcass quality in local meat packing plants).

The station and home test program are geared to the evaluation of purebred animals. However, some test stations are also used for the commercial product test program, which provides commercial producers with valuable data on the respective merit of different sources of genetic material. In this program, crossbred animals from different lines or herds are compared in the same station environment for growth performance

meat quality and carcass quality.

Pilot national genetic evaluations for meat and carcass quality were computed in 1996. The program will become official in 1997, and the corresponding EBVs will be incorporated into sire and dam line indices.

A number of promising technologies for live animal measurements are currently under investigation. They will be incorporated in the home and test station programs if they prove both practical and accurate. These technologies include the use of A-mode ultrasound and real-time ultrasound (ALOKA) for the measurement of loin eye depth, loin eye area and estimated lean yield on the live animal. The use of A-mode ultrasound to measure loin-eye depth has proven quite effective in the prediction of lean yield and loin-eye area and is now widely used in Quebec and Ontario. Research is also being carried out on the use of new probe sites (ham, belly) to improve live animal predictions of carcass quality.

A project which may be of special interest to breeders outside of Canada is the development by CCSI of a micro-computer package which will allow these breeders to identify and register breeding animals, record their performance and compute EBVs. The package works on a common PC (386 and above). It uses the same genetic evaluation method as for Canadian herds, and can link back to the Canadian system through the EBVs of Canadian parents, if any. The package will be available in January 1997.


The national program provides participants with third-party research and standards, nationally accredited technicians, public results and the ability to compare genetics from among 200 breeders and breeding companies. It has had a considerable impact on production efficiency and pork quality over the last 10 years.

Canadian swine breeders, with the help of the program, have produced breeding stock which enjoys a high reputation all over the world. Exports of swine genetic material from Canada have increased considerably during the last 10 years, from 2,941 heads in 1985 to 75,821 heads in 1995. The U.S., Mexico, Columbia, South Korea and the Philippines are some of the main importing countries.

The program is adapting to new trends in the industry: fewer but larger breeders and commercial producers; reduced government involvement; greater use of technology; and increased emphasis on product quality, feed efficiency and longevity. This adaptation is necessary if the program is to keep Canada a world leader in swine improvement.