Canine radiation chimeras were used to investigate further mechanism(s) responsible for maintaining the stable chimeric state. Chimeras were studied 7 to 46 months after 1200 R total body irradiation and transplantation of marrow from a littermate donor matched at the major histocompatibility complex. An attempt was made to perturb the stable chimeric state by infusion of large numbers (0.6 to 13.7 x 10(8)/kg) of donor peripheral blood lymphocytes into each respective chimera. Two groups were studied: donors in Group A were normal; donors in Group B had been specifically sensitized against minor histocompatibility antigens of the chimera by repeated skin grafts. None of the nine chimeras in Group A developed significant clinical or histologic evidence of graft-vs-host disease (GVHD) after donor lymphocyte infusion. Eight of the 12 chimeras in Group B, however, developed GVHD which was transient in three and fatal in five. The results in Group A are not consistent with classical theories of tolerance, i.e., elimination or inactivation of potentially reactive cell clones, but suggest the presence of an active mechanism suppressing recognition of host antigens by the infused donor lymphocytes and development of GVHD. The results in Group B indicate that this mechanism can be overcome by infusion of sensitized donor cells. In an attempt to elucidate the nature of this postulated active mechanism, the cytotoxicity of donor lymphocytes for fibroblasts of the chimera and the presence or absence of serum-blocking factors were assessed in vitro by using a cellular inhibition (CI) assay. The presence of serum-blocking factors did not protect against the development of significant GVHD in two chimeras (fatal in one). GVHD did not occur in four other chimeras after infusion of cytotoxic donor lymphocytes despite the absence of serum-blocking factors. These and previous results suggest that serum-blocking factors are not the mechanisms suppressing the development of GVHD in canine radiation chimeras, and raise the possibility that a suppressor cell population may be responsible for preventing GVHD.