Yersinia enterocolitica is now the species of Yersinia most frequently isolated from human and animal infections. The species includes pathogens and ubiquitous strains. Among the human pathogens, those isolated in America are more virulent than those isolated elsewhere, especially in Europe and Japan, and these isolates differ biochemically and serologically. The relation between Y. enterocolitica and Y. pestis only became obvious in 1980 with the discovery that at 37°C Y. enterocolitica requires Ca++, a phenotype described in the 1960s for Y. pestis. This requirement as well as virulence is dependent on a 70-kilobase plasmid found later in Y. pseudotuberculosis and Y. pestis. Thus, many bacteriologists elected Y. enterocolitica as a model for bacterial invasiveness. However, studies with non-American strains were impeded by the lack of an inexpensive, simple animal test, a difficulty now circumvented by supplying an appropriate siderophore to the bacteria. Ca++ dependence can be viewed as a transition between free growth and protection against the immune system. In the latter phase, Y. enterocolitica synthesizes and releases large amounts of six plasmid-encoded outer membrane proteins. Most of these are under the control of the plasmid region governing Ca++ dependence. Mutants in this region either lose the Ca++ requirement at 37°C or become unable to grow at 37°C irrespective of the Ca++ concentration. The complex events leading to Ca++ dependence is still not understood. Virulence in Y. enterocolitica also depends on chromosomal genes: the endocytosis in intestinal epithelial cells seems not to be encoded by the pYV plasmid. Studies of Y. pseudotuberculosis suggest that this property depends on a single chromosomal locus, the study of which might be particularly important in the understanding of the first step in infection.