important sites of surveillance for new pathogens that will emerge to jeopardize health in the future.

The potential for foods to be involved in the emergence or reemergence of microbial threats to humans is great, in large part because there are many points in the food chain at which food safety can be compromised. This chain of events begins wherever crops or animals are raised; it proceeds through a complex system of manufacturing, distribution, and retailing and ends with the use of a food product by the consumer. Changes in any of a number of aspects of the farm-to-consumer chain, or inattention to food safety in general, can result in outbreaks of food-borne illness.

Although food containing viruses or parasites can cause illness (as can chemical contamination), the majority of individual cases of food-borne disease of known etiology in the United States are caused by bacteria. However, in more than half of the outbreaks of food-borne illness, the exact cause is unknown (Bean and Griffin, 1990). Although in many cases the lack of an exact cause reflects an incomplete investigation, at least some proportion of those outbreaks are likely to be the result of as yet unidentified food-borne pathogens.

There has been a substantial increase in our knowledge of food-borne diseases during the past 20 years, as reflected in an approximate tripling of the list of known food-borne pathogens. An important component of this increase in understanding is a better scientific grasp of the factors that allow microorganisms, and bacteria in particular, to cause human disease. Because of better methods of identifying food-borne pathogens, it has become clear that only certain strains of a bacterial species may cause food-borne illness.

For example, Escherichia coli is part of the natural intestinal flora of humans; its presence in a water sample has been used as evidence of fecal contamination by other pathogenic microorganisms. The majority of isolates of E. coli pose no threat to humans as food-borne pathogens. Researchers, however, have identified five distinct groups of E. coli that cause enteric disease. Based on the mechanism of pathogenesis of each group, they are designated enteroinvasive, enterotoxigenic, enteropathogenic, enteroadherent, and enterohemorrhagic E. coli (Archer and Young, 1988). The ability to detect these pathogenic isolates has been greatly enhanced by diagnostic tests that identify specific virulence-related genes or gene products such as toxins, adhesins, and cell-surface markers.

Improved epidemiologic surveillance has also played an important role in identifying microorganisms that cause food-borne disease. This was the