IL-2 plays an important and complex role in the immune system, serving as a growth factor, a differentiation factor, and a regulator of cell death. It shares many of these functions with other cytokines such as IL-15, which complicates the interpretation of the IL-2-deficient phenotype. Nonetheless, it is clear that a major indispensable role of IL-2 signaling, at least in mice, is to limit the number of activated T cells in the periphery after exposure to self- or environmental antigens. Although we still do not understand the mechanism by which this occurs, there emerge several general conclusions that provide a foundation for future work on this issue. The autoimmune phenotype associated with IL-2 deficiency results from the dysregulated activity of thymus-derived TCR alpha beta CD4+ and/or CD8+ T cells. The process requires an initial antigenic stimulus (e.g., from intestinal flora or self-antigens), but can then spread to naive bystander T cells by an undefined mechanism. A key observation is that neighboring T cells with intact IL-2 signaling can suppress the dysregulated activity of IL-2- or IL-2R-deficient T cells in a dominant manner, indicating that IL-2-based immune regulation is a systemic rather than cell autonomous property. The IL-2R signal responsible for maintaining homeostasis is not dependent on either the Shc or Stat5 pathways, therefore AICD, which is Stat5-dependent in vitro, cannot be the critical mechanism. Finally, expression of the IL-2R in the thymus and not the periphery appears sufficient to suppress the autoimmune phenotype, suggesting that IL-2 signaling may be required for some aspect of thymocyte selection and/or differentiation that has so far gone undetected. These observations can be combined into a unified model wherein IL-2 signaling is required for the development of a regulatory T-cell subset that serves to terminate antigen-induced responses of TCR alpha beta T cells. On the other hand, the role of IL-2 in the maintenance of self-tolerance could be multifaceted, with no single model accounting for all experimental findings. Of note, most of our knowledge regarding IL-2 signaling and tolerance has come from studies of knockout mice. Conceptually, it is important to remember that the function of a molecule such as IL-2 is not simply the inverse of the phenotype seen when that molecule is absent, since necessity and sufficiency are not always coupled in physiological processes. In this regard, perhaps new insights into this important issue can be gained by better defining when and where IL-2 signaling normally occurs in wild-type animals and the functional status of the T cells involved.