Figure 1. The complex relationship between the thymus and the peripheral lymphoid pool can best be visualized as a tap filling a bath. During early life the tap is fully on, the thymus produces T cells each with a different antigen specificity, represented here by a different colour and these fill the peripheral T-cell pool represented by the bath. After this initial filling, the tap continues to be on and with no feedback mechanism affecting the outflow of new T cells from the thymus, we might expect that the total number of T cells within the pool would increase and eventually the bath would overflow, but this is not the case. In both humans and mice, there is usually no change in the total number of T cells in the peripheral pool with age. If the bath does not overflow, then there must be changes to either the inflow or outflow systems. Input certainly changes with time. With age, the thymus atrophies and its contribution of new T cells to the peripheral pool declines and this decline is not trivial, which raises the question about how, if the tap is gradually turned off, the level of water remains the same. If there are no changes in the rate of death within the lymphocytes in the T-cell pool then the only other means to maintain the pool is proliferation of constituent T cells, most likely the memory T cells. Consequently, there is potential for the T-cell repertoire to become limited as the input of naïve T cells declines with age and memory cells with specificity for previously seen antigens proliferate and come to dominate the T-cell pool.