Prolonged kidney and liver allograft survival was produced in baboons by low cumulative doses (500 to 1200 rad) of total lymphoid irradiation (TLI). Continuing normal graft function for more than 1 yr after transplantation was seen in 10 animals in this series. The longest survivor (greater than 4 years) rejected a third party kidney allograft in typically acute fashion 1 yr after the original transplantation. The dose-response effect was parabolic, with a cumulative dose of 800 rad given as twice weekly fractions of 100 rad each being the most effective. With smaller cumulative doses (500 or 600 rad) results were improved if TLI was administered as a larger number of smaller fractions. These regimens are free of the radiation related mortality observed in earlier studies with larger cumulative doses. As in previous studies, the injection of donor bone marrow was not essential for prolonged and continuing graft survival. The field irradiated was extensive, and included a substantial part of the of the hemopoietic bone. Restricting the field irradiated to that approximating what is used clinically in patients with Hodgkin's disease resulted in markedly poorer graft survival. Alterations in T cell subpopulations were studied in these baboons with the use of the anti-human monoclonal antibodies OKT11 (anti-total T cells), OKT4 (anti-T helper cells; Th), and OKT8 (anti-T suppressor/cytotoxic cells, Tsc), which cross-react with baboon lymphocytes. After completion of TLI but before transplantation, the relative percentage of Tsc cells fell and the percentage of Th and the Th/Tsc ratio rose. Transplantation was followed by an inversion of this ratio due to reciprocal changes in the percentages of Th and Tsc cells. This pattern persisted and increased until 6 mo after transplantation, when it reverted to that seen in untreated control animals. This was maintained in long surviving animals up to 17 mo after transplantation. Transplantation was also followed by a marked early increase in the percentage of Ia+ (OKIa1 anti-human monoclonal antibody) total T, T4+, and T8+ cells. If findings in the human, in which activated suppressor but no cytotoxic T cells are Ia+, can be extrapolated to the baboon, a possible explanation for the serial changes observed in the present studies may be that transplantation after TLI results in a polyclonal expansion of nonspecific suppressor T cells. These disappear with time leaving specific suppressor T cells that may be responsible for continuing graft acceptance. Formal proof of this hypothesis will require appropriate functional studies.