A systematic study has been carried out to investigate the role of immunoglobulin isotype, epitope density, and antigen/antibody ratio on the capacity of immune complexes to activate the classical and alternative pathways of human complement and for the complexes subsequently to bind to erythrocyte C3b-C4b receptors (CRI). For this purpose, a series of chimaeric monoclonal anti-NIP antibodies was used, which all shared the same combining site but had different human constant domains. Antigen epitope density was varied by coupling different numbers of NIP hapten molecules to bovine serum albumin. All three parameters affect complement fixation. In general, complement activation is better in antibody excess and at equivalence than it is in antigen excess, and better at high epitope density than at low epitope density, although the effects are variable for different immunoglobulin isotypes and for the two pathways. It has been confirmed that IgG1 and IgG3 are good activators of the classical pathway and are tolerant to variations in both epitope density and antigen/antibody ratio. IgG4 and IgA do not activate the classical pathway in any circumstances. IgG2 activates the classical pathway only at high epitope density and at equivalence or antibody excess. IgM activates the classical pathway well only at the higher epitope densities and at equivalence or antibody excess but, in addition, shows an interesting and unexpected prozone phenomenon where immune complex in antibody excess inhibits complement activation by the classical pathway. The results of the alternative pathway activation are strikingly different. IgA is by far the best activator of the alternative pathway and is relatively tolerant to epitope density and to antigen/antibody ratio. IgM, IgG1 and IgG3 do not significantly activate the alternative pathway in any circumstances. IgG2 is the best IgG subclass for alternative pathway activation but requires high epitope density and equivalence or antibody excess. Binding to CR1 in general parallels the amount of complement fixed independent to the pathway by which it is fixed. However, IgG1 and IgG3 complexes in antigen excess activate complement well but bind poorly to CR1. Nascently formed complexes seem to bind complement in a way that is similar to that bound by preformed complexes, but are then less able to bind to red cell CR1. These observations help to explain the pathogenesis of complement activation in various autoimmune and immune complex diseases such as systemic lupus erythematosus, autoimmune thyroiditis and others.