Adult stem cells were once thought to produce only the cell lineages characteristic of the tissues in which they reside. Recent studies suggest that cells derived from one adult tissue can be reprogrammed to change into cellular phenotypes not normally found in that tissue. Bone marrow (BM) derived cells have been demonstrated to differentiate into multiple lineages, including glial cells and neurons, both in vivo and in vitro. This unexpected plasticity of BM cells occurs not only under experimental conditions, but also in humans following BM transplantation. As a result, BM transplantation has emerged as a novel approach to enhance neural regeneration and restore injured brain tissue. Several research teams have reported that transplanted BM cells can differentiate into neural derivatives; indeed, some of these cells were capable of integration into the host brain, where they promoted functional recovery after brain injury. Other researchers conducting similar studies were unable to find any evidence of neural differentiation, concluding that differentiation from marrow to brain is not a common phenomenon. More recently, two papers in Nature also cast doubt on the plasticity of adult stem cells, suggesting that the acquisition of different identities by grafted BM cells may merely reflect their fusion with host cells. Reasons for the wide discrepancies among findings in current BM stem cell research are unclear, making it difficult to understand the mechanisms by which transplanted marrow stem cells provide therapeutic benefit. Here, we summarize recent findings on this subject, and address some of the major controversies that have marked the evolution of adult stem cell research.