To study hematopoietic differentiation a variety of in vitro systems have been established using hematopoietic precursors derived from various explanted adult and fetal tissues. In this prospective we describe and discuss the potential of a novel system for studying the earliest stages of hematopoietic development. In addition, some of the applications of this system as a unique in vitro model for studying other developmental systems are discussed. Murine embryonic stem cells (ESC), which are totipotent and can be maintained undifferentiated indefinitely in vitro, have the capacity to differentiate in vitro into hematopoietic precursors of most, if not all, of the colony forming cells found in normal bone marrow. This potential can be exploited to study the control of the early stages of hematopoietic induction and differentiation. Recent results have indicated that there is a strong transcriptional activation, in a well defined temporal order, of many of the hematopoietically relevant genes. Examples of the genes expressed early during the induction of hematopoiesis include erythropoietin (Epo) and its receptor as well as the Steel (SI) factor (SLF) and its receptor (c-kit). Several other genes, including CSF-1, IL-I, and G-CSF were expressed during the later stages of hematopoietic differentiation. Contrasting with these observations, IL-3 and GM-CSF were not expressed during the first 24 days of ES cell differentiation suggesting that neither factor is necessary for the induction of hematopoietic precursors. Although these studies are just beginning, this system i s easily manipulated and gives us an approach to understanding the control of the induction and differentiation of the hematopoietic system in ways not previously possible. o 1992 WiIey-Liss, Inc.