A major problem in understanding foodborne listeriosis from both the basic science and regulatory perspectives revolves around the role played by virulence factors of Listeria monocytogenes and how these interact with host susceptibility to result in the observed incidence of disease. From a mechanistic perspective, this problem has been well investigated, and many virulence components of L. monocytogenes have been discovered. Deletion of these genes results in large reductions in virulence functions in vitro and in vivo. The clonal bacteria and genetically identical hosts necessary to solve the riddles associated with virulence mechanisms are not likely to reflect the natural diversity found among wild populations of L. monocytogenes, including those associated with food. These factors contribute to a major dilemma in risk assessment and risk management of foodborne listeriosis: Although low-level L. monocytogenes contamination of certain foods is relatively common, suggesting widespread exposure, illness is overwhelmingly associated with only a relatively small subpopulation (3 of the 13 L. monocytogenes serotypes) and occurs in only a small proportion of susceptible individuals. Virulence testing based on DNA probes for virulence genes is confounded by the widespread distribution of these genes in food isolates. In terms of the distribution of virulence factors among food isolates of L. monocytogenes, only listeriolysin is well characterized, because beta-hemolysis is often used to confirm the presence of L. monocytogenes in foods. The presence of other virulence genes such as those involved in host cell invasion and cell-to-cell spread (inlA and actA) among food isolates has not been extensively investigated. How the presence of these components translates into functional virulence as measured in vivo and in vitro is also unknown. Animal studies and cell culture systems show a range of virulence among food isolates of L. monocytogenes. However, clinical isolates included in such studies are not consistently more virulent than food isolates with no known human disease association. Where multiple serotypes or ribotypes are compared, it has been difficult to demonstrate a consistent pattern of increased virulence associated with any subtype(s) in animal or in vitro studies. Development of model systems that adequately reflect the complexity of the host-pathogen relationship remains a challenge.