Graft failure in recipients of lymphocyte-depleted allogeneic bone marrow transplants is a major limitation to the success of this approach for preventing acute graft-versus-host disease. In a rat BMT model, we evaluated the effect of cyclosporine dose and schedule of administration on the engraftment of MHC-mismatched bone marrow. Lewis Brown-Norway rat recipients were prepared with myeloablative doses of busulfan (day -2) and cyclophosphamide (day -1) and then transplanted with MHC-mismatched ACI rat BM (day 0) that had been depleted of lymphocytes by counterflow centrifugal elutriation. Beginning on day -1, LBN rats received variable doses of CsA (i.e., 12.5, 10.0, 7.5, 5.0, 2.5, 0.0 mg/kg/day) for various lengths of time after BMT (i.e., 7, 14, 21, or 28 days). While all rats receiving high-dose CsA (i.e., 12.5 or 10.0 mg/kg/day) for 28 days had stable donor-derived hematopoietic reconstitution, rats receiving a daily dose of CsA less than or equal to 5.0 mg/kg/day or CsA for less than or equal to 21 days had a high incidence of graft failure. The data argue that (1) graft resistance can be suppressed by CsA and (2) the effectiveness of CsA for securing engraftment is dependent upon both dose and duration of treatment.