One of the basic effects of antigens on lymphocytes is to stimulate DNA synthesis and cell division. This effect is readily demonstrable in vitro (1-3) and in vivo (4-7). Thymus "independent" antigens may stimulate DNA synthesis by B lymphocytes only (8, 9) or by both T and B lymphocytes (10, 11), while thymus-dependent antigens typically stimulate DNA synthesis by both T and B cells (12, 13), with the response of the T ceils generally preceding that of the B cells (5, 13). While some of this accelerated DNA synthesis reflects the activity of specific clones of lymphocytes with surface receptors for the antigen selected (14-17), much of it also seems to reflect the activity of cells with no intrinsic specificity for the antigen (15,16,(18)(19)(20)(21)(22)(23)(24)(25). These latter cells presumably are responding to mitogenic factor (18, 19), possibly to other T cell products (20-25) and possibly to other substances whose sources are not presently defined. This background of nonspecific stimulation has made it very difficult to study the antigen-induced DNA synthesis of specifically reacting cells directly.We have recently described a system whereby it is possible to concentrate recirculating long-lived lymphocytes of a particular specificity by secondary antigenic stimulation (15,16). Unfortunately this secondary stimulation also results in the nonspecific trapping of many other long-lived recirculating lymphocytes in the same lymph nodes (15, 16). The present studies, which were undertaken to try to refine this system, are based on the following two observations: first, most rapidly dividing lymphoid cells do not ordinarily recirculate (26) and therefore have little tendency to accumulate in antigenically stimulated lymph nodes (27, Emeson, Thursh, and Noble, unpublished observation), and, second, cells with specific reactivity for an antigen seem to rapidly synthesize DNA at very specific times after antigenic stimulation (28, 29). It was hoped that labeling cells for a very brief period after antigenic stimulation might make it possible to define conditions where specifically reactive cells were labeling much more rapidly than other recirculating