Postnatal skeletal stem cells are a subpopulation of the bone marrow stromal cell network. To date, the most straightforward way of assessing the activity of skeletal stem cells within the bone marrow stromal cell (BMSC) population is via analysis of the rapidly adherent, colony-forming unit-fibroblast (CFU-F), and their progeny, BMSCs. Several in vitro methods are employed to determine the differentiation capacity of BMSCs, using osteogenic and adipogenic "cocktails" and staining protocols, and pellet cell culture for chondrogenic differentiation. However, true differentiation potential is best determined by in vivo transplantation in either closed or open systems. By in vivo transplantation, approximately 10% of the clonal strains are able to form bone, stroma, and marrow adipocytes, and are true skeletal stem cells. Furthermore, when derived from patients or animal models with abnormalities in gene expression, they recapitulate the disease phenotype on in vivo transplantation. Although ex vivo expansion of BMSCs inevitably dilutes the skeletal stem cells, when used en masse, they are attractive candidates for reconstruction of segmental bone defects, and as targets for gene therapy.