The treatment of guinea pigs with 2,4-dinitrophenyl (DNP) t derivatives of the copolymer of D-glutamic acid and D-lysine (D-GL) induces a state of specific immunologic tolerance in the recipient (1). The type of tolerance induced is unusual in that it is DNP specific. That is, challenge of such guinea pigs with a DNP conjugate of ovalbumin (DNP-OVA) elicits anti-DNP responses markedly smaller than those of animals which have not been pretreated with DNP-B-GL. The degree of depression in responsiveness is related to the dose of tolerogen, to the mode of immunization with DNP-OVA, and to the interval between the two events. A similar type of haptenspecific tolerance, in mice, has recently been described by both Hraba et al. (2) and by Golan and Borel (3).The tolerant state, in this model, appears to be largely expressed in the precursors of antibody-secreting cells. The evidence for this view is the following:(a) The number of DNP-specific antigen-binding lymphocytes in the lymph nodes of tolerized animals is markedly reduced (1). These antigen-binding lymphocytes are members of the bone marrow-derived (B) lymphocyte pool and include the precursors of antibody-secreting cells (4).(b) The tolerance is hapten specific (1). Thymus-derived (T) lymphocytes acting as "helpers" in the activation of precursor cells are almost exclusively carrier or conjugate specific (5, 6); on the other hand, hapten-specific precursors of antibodysecreting cells are relatively common (4, 7). Hapten-specific tolerance thus implies precursor cell tolerance.(c) Even when an excess of helper cells are provided through an immunization with carrier after tolerance induction with DNP-D-GL but before challenge with DNPcarrier conjugate, the subsequent anti-DNP response is markedly diminished (1). Thus, in a situation in which helper T cells are not limiting, the hapten-specific tolerance is easily demonstrable, again implying a defect in the population of precursors of antibody-secreting cells.