B-cell acute lymphoblastic leukaemia (B-ALL) is a clonal malignant disease originated in a single cell and characterized by the accumulation of blast cells that are phenotypically reminiscent of normal stages of B-cell differentiation. B-ALL origin has been a subject of continuing discussion, given the fact that human disease is diagnosed at late stages and cannot be monitored during its natural evolution from its cell of origin, although most B-ALLs probably start off with chromosomal changes in haematopoietic stem cells. However, the cells responsible for maintaining the disease appear to differ between the different types of B-ALLs and this remains an intriguing and exciting topic of research, since these cells have been posited to be responsible for resistance to conventional therapies, recurrence and dissemination. During the last years this problem has been addressed primarily by transplantation of purified subpopulations of human B-ALL cells into immunodeficient mice. The results from these different reconstitution experiments and their interpretations are compared in this review in the context of normal B-cell developmental plasticity. While the results from different research groups might appear mutually exclusive, we discuss how they could be reconciled with the biology of normal B-cells and propose research avenues for addressing these issues in the future.