MIXED lymphocyte culture (MLC)1 has been used extensively as an in vitro model to analyse the reactivity of T cells to antigens coded by the major histocompatibility complex (MHC). When murine T responder cells are exposed in vitro to allogeneic lymphoid cells (stimulator cells) they proliferate and cytotoxic T lymphocytes (CTL) are generated2,3. Antigens coded by the central I region of the MHC are chiefly responsible for triggering proliferation4,5, whereas the target antigen of the CTL generated is either a H–2K or H–2D region or a I–A subregion gene product5–8. This dichotomy in the antigenic requirement of a MLC seems to be reflected at the level of the responding T lymphocytes. Two distinct subsets of T cells (T1/T2), which have been defined by both functional9,10 and physical11 criteria and possibly on the basis of their Ly phenotype12, seem to act synergistically during the in vitro generation of CTL9–13. Proliferating T1 helper cells are thought to respond mainly to I-region gene products2,5,12 whereas the precursors of CTL (T2 cells) are believed to react specifically to transplantation antigens coded by the H–2K, H–2D, or I–A region2,5,7. The proliferating T1 helper cell is believed to potentiate the development of CTL either by itself or through a secreted helper factor2,11,13,14. A serious limitation of this concept is, however, that so far the experimental evidence available is derived from work performed in vitro. We therefore aimed at verifying T–T collaboration during in vivo responses to antigens of the MHC.