By combining a newly established single-cell cytotoxicity assay in agarose (16) with estimations of the maximum natural killer (NK) potential (Vmax) by 51Cr release that percentage of target-binding cells (TBC), the fraction of active killers among TBC, the kinetics of single-cell cytotoxicity, and the recycling of effector cells was studied. Using nylon wool-passed peripheral lymphocytes, approximately 10% of the cells will bind to NK- susceptible target cell lines. Most of these have receptors for IgG. Some 50% will go on to kill T cell targets and some 20% to kill the standard target cell K-562. As the individual NK cell is shown to have the capacity to recycle, i.e., to kill more than one target cell in the 3-h test period, and as recycling seems to vary between individuals, there is no consistent correlation between the number of TBC and 51Cr-release values. It seems as if the single-cell cytotoxicity assay, as presently performed in agarose, is a valuable complement to Vmax determinations by 51Cr-release to study the different steps involved in the cytolytic process: recognition, enzyme activation, and effector cell recycling. The discrimination between these steps will probably be necessary to define mechanisms influencing NK cells in different disease states as well as in learning more about the normal function and regulation of the human NK system.