In summary, we currently hypothesize that systemic immunization within the lymphoid tissue is amplified by a sequence of specific and non-specific immunological interactions which lead to enrichment of specifically sensitized T-cells at the graft site. Nascent sensitized cells enter the graft via non-specific attractants generated by the healing response. The interaction of alloantigen with responder T-cells bearing specific alloantigen receptor results in production of lymphokines which increase blood flow and vascular permeability, and which are chemotactic for circulating lymphocytes. These events and other non-specific events recruit circulating cells regardless of the specificity of their sensitization. The state of activation of these cells may control their presence within the circulation. Cells which are more activated are released more readily from the lymphoid depots into the circulation so that relatively more specifically sensitized (though not necessarily mature cytotoxic cells) cells would enter the circulation and be recruited to the graft. At the allograft, there is an escalating sequence of lymphokine production and non-specific recruitment. II-1 has been demonstrated to be capable of mediating such recruitment. A second phase of amplication occurs at the graft site where precytotoxic (x-ray-sensitive) cells mature and expand into cytotoxic (x-ray-resistant) cells. These cells may be important effector cells of allograft rejection in concert with many unsensitized cells which can participate in non-specific attacks on graft viability. The essential point made from the studies summarized here is the consideration that trafficking patterns of lymphocytes and recruitment of lymphocytes and other inflammatory cells are in essential component to the local mechanism of graft rejection. The relative importance of specificity of response has not been defined.