AbstractThe adoptive secondary response of mice to conjugates of NIP (4‐hydroxy‐5‐iodo‐3‐nitro‐phenacetyl‐) and DNP (2,4‐dinitrophenyl‐) is here used to elucidate the mechanism of cellular cooperation. The framework into which the experiments fit can be formulated as follows. Priming immunization raises a crop not only of specific antibody‐forming‐cell‐precursors (AFCP) but also of specific helper cells. Upon secondary stimulation the helper cells serve a role as handlers or concentrators of antigen, thus enabling AFCP which would otherwise be incapable of reacting to initiate antibody synthesis. In this act of cooperation both cells recognise antigen; in the system examined here the helpers recognise carrier determinants and the AFCP recognise either the hapten or other carrier determinants.The first aim of the experiments was to raise populations of helpers and AFCP of distinguishable specificity. Mice were primed with NIP‐Ovalbumin (OA) mixed with chicken γ‐globulin (CGG) and bovine serum albumin (BSA); in comparison with controls primed with unmixed NIP‐OA, their cells after transfer were relatively more sensitive to secondary stimulation with NIP‐CGG or NIP‐BSA and similar findings were obtained in cross‐checks of these carriers. For reasons which are not entirely clear, non‐transferred cells did not show the same effect. In further experiments cells primed with one conjugate (e. g.NIP‐OA) were mixed with cells primed with another protein (e. g.BSA), transferred and challenged with the hapten conjugated to the second protein (i. e.NIP‐BSA). In comparison with controls lacking the protein‐primed cells, the mixture regularly showed greater sensitivity to stimulation. NIP and DNP were tested in many of the possible combinations with BSA, OA and CGG with the same result. The mixture system was used in the further analysis.Tests with allotype‐marked protein‐primed cells showed that these cells did not participate in the production of the anti‐hapten antibody and could therefore properly be regarded as helpers. Tests of specificity showed that physical union of the hapten and carrier were required: cells primed with BSA, for example, would not help NIP‐OA‐primed cells to make a response to NIP‐HSA even when stimulated at the same time with BSA. Transfer of less than one‐tenth of the spleen gives a maximum helper effect, whereas AFCP activity continues to rise as larger numbers of cells are transferred. Helper cells are therefore normally present in excess.Helper activity is more resistant than AFCP activity to irradiation, drugs and semi‐allogeneic cell transfer across an H‐2 barrier. This suggests that helper cells play a relatively passive role in the immune response.Several observations indicate that helper cells are thymus‐derived mediators of cellular immunity. Passively transferred antibody did not substitute for helper cells. After immunization helper activity developed faster than AFCP activity. Spleen cells obtained from lethally‐irradiated, thymocyte‐repopulated, immunized donors provided help. Cells from the thymus‐derived fraction of thymus/marrow chimeras also appear to provide help.Thus, the hapten‐carrier cooperative response maps onto the well‐establ