The technetium-99m microcytotoxicity assay has been used to detect cell-mediated immunity in CS7BL/6 mice sensitized with A/J skin allografts. Our initial studies of the quantitative in vitro assessment of lymphocyte-mediated cytotoxicity in mice rejecting first-set skin allografts revealed a simple monophasic response peaking at 14 days postgrafting and declining to control levels by 21 days. Subsequent experiments in which the development and persistence of immunity was assessed at daily intervals from 9-21 days postgrafting revealed that the response was considerably more complex. A cyclic rise and fall in killer activity was evident. The first peak occurred 10-13 days after grafting and the second one 3-5 days later. A third peak of cytotoxic activity sometimes could be detected 16-19 days postgrafting. An attempt was made to characterize the phenomenon by studying the cytocidal effects resulting from the admixture of high- and low-responding lymphocyte populations. An intermediate effect generally was observed when lymphocytes with maximal killer activity were combined in equal numbers with those having decreased reactivity. Varying the ratio of high-and low-responding cells resulted in changes in the net killing effect which was consistent with dilution of more reactive lymphocytes with less reactive ones. Mixing lymphocytes from two peak periods produced a maximum killing effect at all effector to target cell multiplicities. Failure to demonstrate modulation of the reactivity of high-responding cells by low-responding ones suggests that these cyclic variations were not mediated by suppressor cells although a role for humoral factors cannot be excluded at the present time. Alternatively, the cyclic pattern may have been due to the specific depletion and subsequent regeneration of cytotoxic lymphocytes in the lymph nodes of sensitized animals. © 1974.