The potential advantage of in utero HSC transplantation over a postnatal BMT is that early curative therapy could be given to an affected fetus, thus eliminating standard intensive immunosuppressive, marrow-ablative conditioning. It is apparent from studies in animals and humans that MHC-mismatched donor HSC of either fetal or adult origin can engraft in fetal recipients if the transplants are done sufficiently early in gestation. However, except for SCID, the percentage of donor pluripotent HSC that engraft is unacceptably low. We had hoped that for diseases such as thalassemia there would be a selective survival advantage for committed donor progenitor cells resulting in a high percentage of donor cell engraftment. At least based upon the experience in human fetuses with alpha- or beta-thalassemia, this has not been the case. Furthermore, for the majority of potential recipients of in utero HSC transplants, the marrow is non-defective, and the small percentage of pluripotent donor HSC that engraft would not be expected to selectively expand post-transplant. Our own results suggest that the non-defective fetal mouse and rhesus monkey are excellent models in which to study both stem cell engraftment, rejection, and tolerance induction. In our studies in non-defective mice with normal hematopoiesis, while the percentage of donor cells that are present is quite low, in only a small number of these animals were we able to induce permanent skin graft tolerance. Thus, while we found microchimerism in approximately 75% of recipients, less than 10% became tolerant. Even when we co-injected a large number of DC precursors, similar to what has been shown to induce tolerance to allogeneic liver, most of the animals failed to become tolerant to donor skin grafts. Interestingly, donor c-kit+ cells can be recruited with cytokines into the peripheral blood in engrafted mice, although these cells do not seem to be sufficient to induce tolerance to donor skin grafts, suggesting that the type (and location) of the engrafted donor cell plays a key role in tolerance induction. Our results in the fetal monkey model parallel those in the mouse, i.e., only a small number of donor cells engraft with limited tolerance induction. Interestingly, we found in our study of DC that GVHD was induced in those murine recipients of both allogeneic marrow and DC. It is likely that there were a sufficient number of mature DC in the preparation to facilitate a donor cytotoxic response towards the host. As a consequence there was also a significant increase in the percentage of donor cells that engrafted in the survivors. Future studies will focus on ways of blocking the graft vs host reaction while still maintaining the graft-promoting role of the donor T cell.