Natural killer (NK) activity, which has been implicated in the immune response against viral infections and neoplasias, is currently evaluated by means of a chromium (51Cr) release assay. However, criticisms have been raised with regard to the reliability and reproducibility of the test. We have developed a different in vitro method for measuring NK activity, based on the inhibition of the target clone growth in plasma clot semisolid medium. This method overcomes the limitations inherent to the 51Cr release test and more closely mimics the in vivo situation. The inhibitory activity revealed by the cloning assay was always greater than the lytic activity in the 51Cr release assay. Moreover, effector/target ratios of 3:1 and 1.5:1 still produced clonal inhibition. B-CLL cells, used as effectors, showed no inhibitory activity and the Raji cell line employed as target was resistant in both techniques. Thus, the clonogenic assay appears to be more sensitive for the evaluation of low levels of NK activity, for basic studies on the effector/target interactions, for the evaluation of LAK cell activity, and in diseases in which an involvement of the NK compartment has been hypothesized.