Human lymphocyte immunity to tumor-derived target cells was estimated by titrating lymphocyte concentration to achieve a 50% reduction of target cell survival. This lymphocyte titration assay gave estimates of cytotoxicity that were different from those obtained with the conventional cell-mediated cytotoxicity assays but were more proportional to lymphocyte activity. Estimates of cytotoxicity obtained using the lymphocyte titration assay were reproducible upon repeated testin over the course of several months and were relatively unaffected by two- to fourfold variations in target cell concentration. Target cell-specific cytotoxicity was reproducible but often did not appear to be tumor specific.