Human embryonic stem (ES) cells were isolated from the inner cell mass (ICM) of blastocyst stage embryos (Thomson et al., 1998; Reubinoff et al., 2000). Under specific conditions, in vivo and in vitro, these cells are capable of differentiating into cell types of the three embryonic germ layers, namely ectoderm, mesoderm and endoderm. In vivo, following injection into severe combined immunodeficient (SCID) mice the cells develop into teratomas comprised of derivatives from many cell types (Thomson et al., 1998; Reubinoff et al., 2000). In vitro, when placed in non-adherent culture dishes, the cells form spherical structures termed embryoid bodies (EBs) and undergo spontaneous differentiation (Itskovitz-Eldor et al., 2000). Addition of growth factors during differentiation was shown to direct the commitment of the cells towards specific lineages (Schuldiner et al., 2000). Furthermore, protocols describing in vitro enrichment of specific cell types such as neurons (Carpenter et al., 2001; Reubinoff et al., 2001; Schuldiner et al., 2001; Zhang et al., 2001), pancreatic β cells (Assady et al., 2001), cardiomyocytes (Kehat et al., 2001; Kehat et al., 2002; Mummery et al., 2002; Xu et al., 2002a), trophoblasts (Xu et al., 2002b), endothelial (Levenberg et al., 2002) and hematopoietic cells (Kaufman et al., 2001) were published. However, these cultures still contain additional cell types and thus generating pure populations of specific differentiated cell types remains a challenge. Since human ES cells propagate indefinitely in culture without loosing pluripotency and their differentiation may be induced and directed rapidly, they were suggested to serve as an unlimited source of clinically transplantable cells. Yet, in order to achieve this ambitious task, improved differentiation protocols and purification procedures of particular human ES cell derivatives should be developed.