T he inflammatory bowel diseases (IBDs), Crohn's dis-ease and ulcerative colitis, are multifactorial disorders whose etiology remains unknown. Although the exact pathogenesis is poorly understood, there is evidence that it involves interactions among the immune system, genetic susceptibility, and the environment, most notably the bacterial flora. Mechanisms underlying initiation vs. progression and chronicity may well be distinct. Experimental animal models allow study of early events, dissection of the interactions among different components, and identification of immunologic processes and genes that determine susceptibility in ways that are not possible in humans. The complexity of these disorders is not reproducible in any simpler system, including reductionist tissue culture systems, although the latter can be effectively exploited to study selected, highly defined processes that are important to IBD. The results obtained with experimental models generate new paradigms to test in patients; they complement and expand studies in humans but do not replace them. This report is based in part on a workshop on experimental models of IBD at which both a variety of models in common use and some recently developed models were discussed. The premise of this review is that there is no right or wrong model of IBD but that the different models do differ in their relative use, and the purpose of this review is to define what the different models have told us and where they seem most useful for the study of IBD. Toward this end, the multiple factors contributing to IBD have been artificially divided into six major components, namely, genetic susceptibility, environmental factors, acquired immunity, innate immunity, nonspecific inflammation, and wound healing. Each of these major components can be subdivided into a variety of subcomponents as shown in Table 1. The first three components can be thought of as representing early events in the process and the last three as representing a final common pathway of inflammation and repair. The appropriate model for a given study depends entirely on the question being addressed. For example, many of the models have been used to test the use of new anti-inflammatory drugs that might be used in patients; simple, reproducible models involving only nonspecific inflammation may suffice for this purpose. Studies of genetic susceptibility or acquired immunity necessarily involve more complex models in which these components can be studied. As a guide, Figure 1 is provided to help investigators focus on which model might be most useful to study the various components of IBD shown in Table 1. The proposed usefulness for the different models is not meant to be dogmatic, and future progress will almost certainly modify it; however, it is presented as a convenient starting place for the interested investigator. In the text and tables that follow, there is a brief discussion of the available experimental models, including the method used to generate the model, its pathology, technical pitfalls or problems, advantages, disadvantages, and commentary on how it has been used. To facilitate this discussion, the models are arbitrarily clustered into categories and summarized in tables. This report is intended to provide a frame of reference rather than an exhaustive review.