The main finding of this paper is that CR1, the membrane receptor for C3b and C4b, together with C3b/C4b-inactivator (I), degrades C3b bound to immune complexes (C3b*). Two fragments are generated: C3c, which is released from the immune complexes, and C3d*. The C3c fragment released from the cell intermediate EAC1423b prepared with 125I-C3 was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and radioautography. It has a 135,000 mol wt and contains disulfide bonded labeled polypeptide chains of 75,000 and 31,000 mol wt, which presumably represent the beta and a fragment of the alpha-chain of C3b*. Silver staining of the SDS-PAGE gels revealed other C3-derived bands with 39-42,000 mol wt. Human erythrocytes + I also cleave C3b* into C3c and C3d*. The activity of the erythrocytes is CR1 mediated because it can be totally inhibited by monoclonal antibodies to CR1. In contrast with these results, I together with the serum protein beta 1H (H) transform EAC1423b into hemolytically inactive EAC1423bi and cleave the alpha' chain of C3b* into fragments of 70,000 and 40,000 mol wt. Small amounts of C3c are also released at relatively high concentrations of H. On a molar basis, the efficiency of CR1 in the generation of C3c and C3d is 10(4)-10(5) greater than H. An additional observation was that C3c could be released by treating EAC1423bi with CR1 + I and that this reaction was also inhibited by monoclonal antibodies to CR1. Therefore, it is likely that CR1 has binding affinity for iC3b and that the degradation of C3b* proceeds as follows: C3b (formula, see text) C3c + C3d*. Taken together, our findings argue that the processing of C3b* in vivo occurs in solid phase, that is, on the surface of cells bearing CR1.