The cytolytic efficiency of the terminal complement complex (C5b-9) against erythrocytes of different species is, in part, dependent on the species of C9 origin. In the present study, we have examined the interaction of C9 with erythrocytes in terms of the binding, dimerization, and insertion of C9 into the membranes of sheep and human erythrocytes lysed by human complement (C). The membranes of these C-lysed erythrocytes were analyzed for bound, dimerized, and inserted C9 by a combination of photolabeling, SDS-PAGE, electroblotting, and immunostaining techniques. We found that neither binding nor dimerization of C9 could be correlated with the relative hemolytic efficiency of human C on these erythrocytes, but that C9 insertion into the membranes of these cells varied in direct relation to the extent of lysis. Interestingly, the binding of C3 to these cells under conditions of equivalent C1 fixation also correlated with lytic efficiency. These data indicate that the C9-related differences in the cytolytic efficiency of C against erythrocytes from different species is primarily due to the efficiency of C9 insertion into these cells. Moreover, these data emphasize that neither the binding of C9 to a target membrane nor the formation of C9 dimers necessarily leads to the insertion of C9 into the membrane, suggesting the presence of membrane-bound but inactive C5b-9 complexes. Because the extent of C3 binding also correlated with the relative degree of lysis of sheep vs human erythrocytes, the possibility exists that surface-bound C3 may regulate hemolysis by directing the insertion of C9 in terminal complexes into cells.