Data on a small number of allogeneic sibling donor marrow transplants have indicated that reconstituted hematopoiesis can be oligoclonal or even monoclonal, suggesting that hematopoiesis can be reconstituted from a very small number of stem cells. However, other studies have failed to confirm this observation. Animal data indicate that hematopoiesis is likely to be oligoclonal after transplantation using limiting numbers of hematopoietic stem cells, or in cases of poor engraftment. In this study we investigated clonality of engraftment after unrelated donor (URD) bone marrow (BM) and cord blood transplantation, clinical settings in which human leukocyte antigen (HLA) disparities have been known to impair engraftment and in which stem cell numbers have been limiting. Twenty-four URD transplant recipients (20, BM; 4, cord blood) together with 8 sibling donor marrow, 2 sibling donor peripheral blood stem cells (PBSC), and 1 sibling donor cord blood recipients were studied at 28 days, 100 days, 6 months, and 1 year post-bone marrow transplantation (BMT) using the polymerase chain reaction (PCR)-based human androgen receptor assay (HUMARA) as a marker of clonality. URD marrow and cord blood recipients show polyclonal reconstitution in all cases, and results are similar at early and late time points after BMT. Similar results were seen in sibling donor marrow and cord blood recipients. These data indicate that oligoclonal hematopoiesis is not a frequent event after URD stem cell transplantation, even for cases in which limited numbers of stem cells are infused.