The last decade has seen major advances in our knowledge of the molecular control of hematopoiesis, widespread access to cytokines, and the development of practical assays for quantitating highly primitive hematopoietic cells. This progress has now made feasible the predictable manipulation of hematopoietic stem cells (HSC) and progenitors for a variety of experimental and clinical applications. Nevertheless, our understanding of events that induce and/ or block the differentiation of primitive hematopoietic cells is still very limited. Therefore, it is not surprising that procedures for expanding HSC populations ex vivo are based largely on a small set of empirical observations. The incentive to improve this situation is provided by many clinical situations in which the number of stem cells available for particular types of transplants is inadequate, or where HSC amplification may be useful as part of a purging strategy to reduce the potential burden of malignant cells in an autograft. Cell division with retention of stem cell integrity could also facilitate the generation in vitro of many specific types of differentiated cells (e.g., dendritic cells) and is a requirement for retroviral-mediated gene therapy. Progress in each of these rapidly evolving areas has recently been reviewed in greater depth elsewhere (1 –5 ).