The induction of donor-specific tolerance would eliminate the risk of long-term immunosuppression while ensuring allograft function and survival. Male Buffalo (RT1 b) rats were exposed to donor alloantigen by an intrathymic, intrasplenic, sc, or iv injection of 25±10 6 syngeneic Buffalo (RT1 b) or MHC fully mismatched Lewis (RT1 1), ACI (RT1 a), or UV-B irradiated Lewis (RT1 1) splenocytes. The Buffalo recipients were given 1 cc of rabbit antirat antilymphocyte serum (ALS) ip at the time of the donor antigen injection, and 21 days later received a heterotopic Lewis or ACI heart transplant. Only intrathymic alloantigen injection induced a donor-specific tolerance which allowed the cardiac allograft to survive indefinitely (mean survival time [MST]> 176.8 days) in> 86% of the recipients without the need for further immunosuppression, whereas groups receiving antigen injections at other sites rejected cardiac allografts in control time (MST±7.0 days). Histologic examination of long-term tolerated Lewis cardiac allografts revealed the presence of healthy cardiac myocytes without mononuclear infiltration. Buffalo rats with a long-term surviving Lewis cardiac allograft did not reject a second Lewis cardiac allograft (MST> 100.0 days), but rejected a heterotopic ACI cardiac allograft in normal time (MST±7.0 days). By limiting dilution analysis (LDA), maturation of donor-specific CTLs (pCTL) from long-term recipient splenocytes was markedly diminished, whereas third party pCTL was not altered, and T helper precursors were moderately decreased without alteration in the peripheral CD4+ and CD8+ phenotype frequencies. MLC responses of recipients with long-term surviving cardiac allografts to donor-specific and third party stimulation were not significantly different from naive controls. Microchimerism is unlikely because Lewis allograft survival was also prolonged (MST> 96.0 days) in rats receiving UV-B irradiated Lewis splenocytes which cannot proliferate. The absence of increased allograft survival after transfer of long-term recipient splenocytes into naive animals suggests that donor-specific suppressor cells are not present. Additionally, in vitro lymphocyte proliferative responses to mitogenic or allogeneic stimulation in MLC was not diminished by the addition of these longterm recipient splenocytes. This model emphasizes the importance of exposure of T cell precursors to foreign donor alloantigen in the thymic environment for the development of unresponsiveness to a donor-specific vascularized allograft.