Whether hematopoietic stem cells can proliferate without limit, or whether their regenerative capacity declines with repeated division, has been debated for decades. Prevailing opinion favours an intrinsic ‘decline', a view based on the finite degree to which murine bone marrow can be serially transplanted [1,2], the diminished self-renewal of spleen colony-forming cells (CFU-s) subjected to repeated passage [1], and the failure of stem cells to regenerate to normal levels after even a single transplantation [3,4]. However, serial transfer experiments did not specifically monitor input and output of long-lived stem cells (long-term reconstituting cells, LTRCs), leaving competing interpretations unresolved. We have re-examined the issue by quantitating 7–12 month LTRCs during sequential transplantations. Although these cells recovered to only 4% of normal levels after primary bone marrow transplantation, at each passage they increased around 10-fold relative to the amount transplanted, attaining an estimated cumulative expansion of 8400-fold over the original input after four transfers. Expansion was limited by transfer of increasing numbers of marrow cells and specifically of LRTCs, suggesting an extrinsically determined ceiling to stem cell growth. Conversely, expansion was enhanced in vivo by administration of stem cell factor (SCF, c-kit ligand) and interleukin-11. The results challenge the view that expansion of passaged stem cells is limited by exhaustion, and indicate that augmentation after transplant is limited by extrinsic mechanisms whose effects are reversible either by further transfer of the stem cells into irradiated hosts or by administration of exogenous cytokines.