Two opposing immunologically-mediated phenomena currently limit the success of bone marrow transplantation (BMT) in HLA-identical situations and impede the application of this therapeutic modality across MHC barriers. These phenomena are: (1) the response of T cells within the donor marrow allograft to recipient alloantigen, resulting in graft-versus-host disease (GVHD) with its attendant morbidity and mortality; and (2) the response of recipient cells which have survived the ablative conditioning regimen against alloantigen borne by the donor marrow allograft, leading to failure of alloengraftment. While T cell depletion of donor marrow has successfully reduced the incidence of severe GVHD, this reduction has been associated with an increased incidence of failure of alloengraftment. In this communication we review several recent approaches being studied in animal models in our laboratory to avoid such undesirable effects of host-anti-donor and donor-anti-host alloaggression. The first approach is based on the observation that administration of T cell-depleted (TCD) syngeneic bone marrow (BM) appears to limit GVHD while still permitting engraftment by co-administered non-TCD allogeneic BM. This anti-GVH effect of TCD syngeneic marrow can be enhanced by delaying the administration of allogeneic BM by 8 days following whole body irradiation and syngeneic BMT. Evidence that natural suppressor cells derived from syngeneic marrow may be responsible for this phenomenon is reviewed. We also present evidence for the existence of a non-stem cell bone marrow subpopulation, distinct from those T cells which cause GVHD, which is capable of increasing levels of allogeneic chimerism.(ABSTRACT TRUNCATED AT 250 WORDS)