The splenic B-cell repertoire of unimmunized C57BL/6 mice can be examined for anti-(4-hydroxy-3-nitrophenyl)acetyl (NP) B cells of relatively high affinity by using a dual strategy. (14,15) and showed that it could prevent the generation of anti-HSA memory B cells in a dose-dependent manner. Adoptive transfer studies showed that the toleragen affected donor T and B cells, the former effect being more profound. Another way of studying the relative contributions of Tand B-lymphocyte populations to an antibody response is to make use of various hapten-carrier combinations (16-18). T-dependent anti-hapten antibody formation requires that anti-hapten B cells be "helped" by carrier-specific T cells. The anti-(4-hydroxy-3-nitrophenyl)acetyl (NP) response of C57BL/6 mice has been extensively investigated (19-21) because of the preferential appearance of B-cell clones utilizing the VH 186.2 V gene and the Al light chain gene, facilitating the study of V gene hypermutation. It is also a system in which clones making relatively high-versus relatively low-affinity antibody can be discriminated by the use of, respectively, relatively lowly versus relatively highly haptenated protein conjugates in an ELISA (22,23). In the present study, we immunized mice with highly haptenated NP-HSA, and attempted tolerance induction with lowly haptenated, freshly deaggregated, NP-HSA, with HSA carrier alone, or with NP attached to irrelevant carriers. The key findings were that NP-HSA .Jways worked better than carrier alone or NP attached i other carriers in preventing memory cell appearance and that tolerance could still be induced after immunogenic challenge, up to the time of the first appearance of V gene mutations. Apparently, the functional silencing of T and B cells is necessary for the full tolerance effect, and T-ceil help is still required for continued B-memory cell generation after the germinal center reaction is well under way.