A killing defect of natural killer (NK) cells in the absence of NK cytotoxic factors (NKCF) was first demonstrated in a child with Hodgkin's disease. The patient lacked detectable NK cell activity in every phase of the disease as measured by a four-hour 51Cr-release assay using K562 cells as a target. The percent lysis at a 40:1 effector:target ratio by the patient's lymphocytes was persistently below 0.3% as compared with the normal lymphocyte value of 46.2% +/- 5.8% (mean +/- SD). NK cell activity was not detectable at effector:target ratios of 10:1 to 80:1 and by prolongation of the incubation time, and the NK cell defect was not restored or improved by lymphocyte stimulation with polyinosinic-polycytidilic acid, interferon (IFN)-alpha, or interleukin 2 (IL 2). The numbers of Leu-7+ cells and Leu-11+ cells were normal as counted by flow cytometry. A single cell-in-agarose assay demonstrated normal numbers of target binding cells (TBCs), and they showed the morphology of "large granular lymphocytes." However, there were no TBCs with dead targets. These results indicated that the patient's lymphocytes contained normal numbers of NK cells that were capable of recognizing and binding to a target but were incapable of killing the bound target cell. The patient's lymphocytes were then studied for their release of NKCF upon interaction with K562 cells. The patient's cells did not release NKCF, and the NK cell defect was not restored or improved by stimulation of the cells with IFN or IL 2. It is suggested that the deficient release of NKCF may have been related to the killing defect of the NK cells in this patient.