Figure 2-15. Trends in productivity per cow in US from 1998-2007. Source: USDA-NASS, 2008

population in Europe and desire for more meat consumption among North Americans. This drive for greater productivity accelerated in the 1950s as a response to the need to rebuild the food supply chain after World War II. Science-based live­stock breeding typically produced annual genetic changes of around 2% of the mean of a trait (or trait-related index), especially in species with high reproductive rates like pigs and poultry (Simm et al., 2005). Not only were yields from livestock varieties substantially increased, but standardized livestock systems were also developed, where in cattle (and to some extent sheep), the landscape was adapted to the system. In pigs and poultry, the whole enterprise was taken off the land and into intensive housing and feeding sys­tems. Varieties that maximized food conversion ratios were quickly developed, especially in pigs and poultry (Simm et al., 2005), with cattle breeding focused almost entirely on high milk and meat production. In N. America in the 1960s and 1970s, the so-called "British Breeds" of cattle were re­placed in much of the beef sector by "Continental Breeds" that introduced size and leanness, in response to consumer desire for leaner beef. The genetic techniques used to achieve these productivity gains include:
•     Better statistical methods of estimating the breeding value of animals
•     The use of artificial insemination that allowed produc­ers at any level to access superior genetics
•     Better techniques for measuring performance of new breeds
•     Selection focused on quantitative traits, such as weight gain and disease resistance

Despite the undoubted success of these science-based breed­ing programs, it is generally agreed that the maximum ge­netic potential of cattle, pigs and poultry has still not been

reached and intensive breeding programs are still maintained in Europe although the focus is now shifting away from continued productivity increases towards animal health and welfare traits (Garnsworthy, 2005).
     As a result of these breeding and husbandry techniques, the wide variety of landraces in 1945 was quickly replaced by a few high yielding varieties, such as Holstein/Friesian milking cattle (e.g., this breed comprised more than 85% of the Canadian dairy herd in 1999 [Kemp, 2001]) or white lines of pigs used in intensive production facilities. Most livestock landraces have survived in small numbers either by the activities of "rare breed societies" who try to maintain the genetic base of the "old" livestock breeds, or by being used to produce niche market high quality products, mainly meat and cheeses.
     The latest developments in animal breeding include genetic engineering. Its use is unpopular in Europe, but in North America, ancillary uses of GE technology, e.g., to in­crease milk production through the administration of re-combinant Bovine Somatotropin (rBST), has been widely adopted. Whether transgenic animals in the food supply are accepted by NAE consumers remains to be seen. In Europe, rBST use for milk production raised concerns about animal suffering and potential negative impacts on small farmers. In the context of surplus milk production in NAE, the ben­efits of this application continue to be debated.
     Simultaneously with breeding for improved produc­tivity, NAE scientists also focused on improving livestock feeding and management. For example, the weight gain for broilers at 56 days in 1957 was around 800g, compared to a 3900g weight gain in 2001 (Havenstein et al., 2003). Similar trends can be found for weight gain in pigs and for milk yields in cattle (Simm, 1998). Breeding and nutrition technologies for sheep and goats have not been subjected