The Natural Killer (NK) activity in the thymus and NK-sensitive thymocyte targets of dystrophic mice was investigated. Dystrophic and normal mouse thymocytes or spleen cells were layered on discontinuous Percoll gradients (5 or 10% increments, respectively) between 40 and 70% and centrifuged at 1700 g for 30 min. All fractions were tested for either NK activity or used a 51Cr-labeled NK-sensitive targets in a 6-hr 51Cr release assay. The density interface between the 50% (1.060 g/ml) and 60% (1.075 g/ml) Percoll fractions of either dystrophic or normal mouse spleen cells and the 40% (1.050 g/ml) and 50% (1.060 g/ml) Percoll fractions of either dystrophic or normal mouse thymocytes were found to contain the largest proportion of NK activity using YAC-1 lymphoma tumor cells as targets. In addition, the NK activity in dystrophic mouse spleen cells and thymocytes was significantly greater when compared with normal mouse controls. Target binding cell studies revealed that these Percoll fractions of dystrophic mouse spleen cells and thymocytes had greater numbers of conjugate-forming cells when compared with normal control groups. Cell depletion experiments using either anti-Thy 1.2, anti-asialo-GM 1 or anti-NK-1 plus complement treatment revealed that the cell responsible for NK activity in the 50% Percoll fraction interface of dystrophic mouse spleen cells was asialo-GM 1 positive. NK-1 positive, and partially Thy 1.2 positive. However, the cells displaying NK-activity in the thymus of normal or dystrophic mice were found to be highly Thy-1.2 positive and peanut agglutinin (PNA) negative. The density interface between the 60% (1.075 g/ml) and 65% (1.081 g/ml) Percoll fractions of either normal or dystrophic mouse thymocytes contained the largest proportion of NK-sensitive target cells. Interestingly, the 60% Percoll fraction of dystrophic mouse thymocyte targets was significantly more susceptible to NK-mediated lysis than that of the normal mouse thymocyte population. Cell depletion experiments revealed that the NK-sensitive thymocyte population was similar in both mice, that is, Thy-1.2 positive, cortisone sensitive, PNA positive, Dolichos biflorus (DBA) negative and asialo GM-1 negative. The results indicate that there are density differences between splenic and thymic NK cells. In addition, there are density and phenotypic differences between thymic NK cells and thymic NK-sensitive target cells. The findings support the hypothesis that there are different populations of NK cells.