Immunomagnetic cell separation uses binding of an antibody to its epitope to identify the target cell, which is then removed by attachment to an anti-immunoglobulin-coated paramagnetic bead, and passage through a magnetic field. This method has previously been shown to be less sensitive to the effects of low target antigen density than are other cell elimination methods, such as complement-mediated lysis. In this paper we demonstrate that, with certain antibody/target cell combinations, the efficiency of immunomagnetic depletion can be adversely affected by high expression of the target antigen. This can occur by two non-mutually exclusive mechanisms. These are (i) steric hindrance of bead binding due to crowding of monoclonal antibodies on the cell surface; and (ii) binding of the monoclonal antibody molecule in a configuration that is poorly-accessible to the anti-immunoglobulin immobilized on the microspheres. The predominant effect operating in any system can be determined by analysis of the cells remaining after the separation procedure. In both cases pre-attachment of the monoclonal to the beads results in improved separation efficiency. These results emphasize the necessity of optimizing experimental conditions in each system that is investigated.