Several groups, including our own, have independently demonstrated that effector memory T cells from non–alloantigen-primed donors do not cause graft-versus-host disease (GVHD). In the current study, we further investigated whether this approach could be extended to all memory T cells, and we studied the underlying mechanisms. Neither total memory T cells nor purified central memory T cells were able to induce GVHD. Memory T cells were at least 3-log less potent than bulk T cells in mediating GVHD. As expected, memory T cells failed to elicit cytotoxicity and proliferated poorly against alloantigens in standard 5-day mixed-lymphocyte cultures. However, the proliferative responses of memory T cells were more comparable with those of bulk and naive T cells when the culture time was shortened. Moreover, the frequencies of IL-2–secreting cells measured by 42-hour enzyme-linked immunosorbent spot (ELISPOT) assay were similar among naive, memory, and bulk T cells. These data indicated that memory T cells are able to respond to alloantigens initially but fail to develop to full potential. The abortive immune response, which was mediated by non–alloantigen-specific memory T cells in response to alloantigens, may explain why memory T cells from unprimed and non–alloantigen-primed donors could not induce GVHD.