By comparison with the rest of the nervous system, the olfactory epithelium has an unparalleled ability to renew and repair itself throughout life. However, the identity and capacity of the various types of progenitor cells that underlie that ability are not well established. We used selective isolation, transplantation, and engraftment of various types of marker-labeled cells into the epithelium of methyl bromide-lesioned, unmarked host mice to dissect progenitor cell capacity. Globose basal cells were purified from other potential progenitors using the monoclonal antibody GBC-2 (GBC, globose basal cell) and fluorescence activated cell sorting. Transplanted globose basal cells engraft and, in aggregate, give rise to globose basal cells, neurons, sustentacular cells, and several other kinds of non-neuronal cells. Individual clones, derived from single engrafted globose basal cells, can consist of a mixture of neurons and non-neuronal cells, only neurons, or only non-neuronal cells. Neurons that arise after transplantation mature to the point of expressing odorant receptors and olfactory marker protein and of projecting axons to the olfactory bulb. In contrast, other kinds of epithelial cells are neither neurogenic nor multipotent. For example, sustentacular and duct cells give rise only to themselves after transplantation. Furthermore, horizontal basal cells do not engraft in mice, in which the endogenous population is spared after lesion. Thus, some subtype(s) of GBC is a multipotent progenitor cell, whose multipotency is activated after destruction of both neurons and non-neuronal cells. The results suggest that progenitor cell transplantation may prove useful as a therapeutic modality as well as an analytical tool. J. Comp. Neurol. 469:457–474, 2004. © 2004 Wiley-Liss, Inc.