The thymus is the principal site of T-cell development and maturation. Failure to develop a functional thymus leads to severe immunodeficiency, while partially incorrect function of the organ can lead to a variety of autoimmune diseases as well as higher risk for infections and cancer. The thymus is organized into cortical and medullary regions, which are functionally distinct. The diverse array of thymic epithelial cells (TEC) are the key components of the thymic stroma, both the cortical and medullary TEC subsets are responsible for the establishment of a self-tolerant and self-restricted T-cell repertoire. The thymus is most active in young individuals, and undergoes a progressive naturally occurring involution from birth, which accelerates after puberty. Thymic involution is characterized by loss of thymus organization and function, including an overall reduction in the amount of functional thymic tissue. This results in decreased production of new naïve T-cells, and contributes to the diminished capacity of the aged immune system to adequately respond to new antigenic challenge. Involution of the thymus, both natural and in response to different therapies such as chemotherapy, raises interest in developing cell based treatment methods that will allow the restoration of the thymic architecture and so elevate immune reconstitution in vivo. The cellular mechanisms by which the postnatal thymus is maintained during homeostasis and involution are currently unknown. The earliest thymic progenitors in the thymus express Plet1; it has been established that from E12.5 to E15.5 these cells when purified are able to generate all thymic epithelial cell types and initiate thymus organogenesis. However, at least the latter capacity is reported to be lost from E18.5. A number of papers published provide evidence for the existence of both bipotent and unipotent TEC progenitors in the adult thymus. However the identity of these cells remains unknown, nor has the relationship between the mature and immature postnatal TEC compartments been established. The aim of my research was to investigate the cellular mechanism(s) that maintain the postnatal thymus. Specifically, I aimed to determine whether the thymus is maintained by a stem cell mechanism or by division of terminally differentiated thymic epithelial cells, and whether or not postnatal thymic epithelial stem/progenitor cells express functionally relevant levels of the transcription factor Foxn1. To address these aims, I used two approaches: in vivo genetically heritable lineage tracing and a novel grafting assay to assess the contribution of different lineages of TEC. This thesis describes the characterization of a novel mouse strain, the Foxn1CreERt2 line, which was predicted to allow conditional inducible manipulation of gene expression in TEC. I show that this deletor strain, while thymic epithelial cellspecific, could induce cre-mediated recombination in only in a low proportion of TEC and thus could not be used to address the initial aim of this work as described above. However, lineage tracing experiments using this line have provided evidence for a persistent cortical thymic epithelial progenitor/stem cell type, that was capable of rapid expansion within the cortical compartment over time. In parallel with characterisation of the Foxn1CreERt2 strain, I investigated the potential of various defined epithelial populations to contribute to the thymic environment in an assay of TEC potency. Using this technique I have established the potential of defined TEC subpopulations isolated from postnatal mice to generate cortical and medullary TEC. Among the populations analysed I have identified a minor TEC subset that can robustly contribute to both cortical and medullary TEC that coexpress Ly51 and Plet1. I have further shown, using a limiting dilution approach, that this population contains a postnatal common thymic epithelial stem/progenitor cells, present at a frequency of between 87.5 and 92.5 within this population. I have also produced evidence of a unipotent cortical progenitor population that is capable of long term expansion in vivo.