Chemokine-driven accumulation of lymphocytes, mononuclear and polymorphonuclear proinflammatory cells in antigenic tissue sites is a key feature of several types of T-cell-dependent autoimmunity and transplant rejection pathology. It is now clear that the immune system expends considerable energy to control this process, exemplified by the sequential layers of regulatory cell input, both innate and adaptive, designed to prevent a classical Type IV or 'delayed-type' hypersensitivity (DTH) reaction from occurring in the visual field of the eye. Yet, despite an abundance of in vitro assays currently available to the human T-cell immunologist, none of them adequately models the human DTH response and its various control features. The theme of this article is that it is relatively easy to model the effector side of the human DTH response with xenogeneic adoptive transfer models. However, we show that in order to detect inhibition of a recall DTH in response to colocalized donor antigen (linked suppression)--a characteristic feature of peripheral tolerance to an organ transplant--both the challenge site and the immunocompetence of the mouse adoptive host are critical factors limiting the sensitivity of the trans-vivo DTH test.