Normal Lewis rat serum contains antibodies (IgM > IgG) that bind to hamster leukocytes and endothelial cells. Transplantation of either the heart or liver from hamster rat results in release of hamster hematolymphoid cells from the graft, which lodge in the recipient spleen (cell migration), where recipient T- and B-cell populations initiate DNA synthesis within one day. There is marked stimulation of splenic IgM++(bright)/IgD+(dull) B cells in the marginal zone and red pulp, which account for 48% of the total splenic blast cell population by 4 days after liver transplantation. CD4+ predominant T-cell proliferation in the splenic periarterial lymphatic sheath and paracortex of peripheral lymph nodes occurs almost simultaneously. The effector phase of rejection in cardiac recipients is dominated by complement-fixing IgM antibodies, which increase daily and result in graft destruction in 3 to 4 days, even in animals treated with FK506. In liver recipients, combined antibody and cellular rejection, associated with graft infiltration by OX8+ natural killer, and fewer W3/25+ (CD4) lymphocytes, are responsible for graft failure in untreated recipients at 6 to 7 days. FK506 inhibits the T-cell response in liver recipients and significantly prolongs graft survival, but does not prevent the rise or deposition of IgM antibodies in the graft. However, a single injection of cyclophosphamide 10 days before transplantation effectively depletes the splenic IgM++(bright)/Ig+(dull) cells and in combination with FK506, results in 100% survival of both cardiac and hepatic xenografts for more than 60 days. Although extrapolation of morphological findings to functional significance is fraught with potential problems, we propose the following mechanisms of xenografts rejection. The reaction initially appears to involve primitive host defense mechanisms, including an IgM-producing subpopulation of splenic B cells and natural killer cells. Based on the reaction and distribution of OX8+ and W3/25+ cells, antibody-dependent cell cytotoxicity and delayed-type hypersensitivity responses seem worthy of further investigation as possible effector mechanisms. Effective control of xenograft rejection is likely to require a dual pharmaceutical approach, one to contain T-cell immunity and another to blunt the primitive B-cell response.