The reports discussed above have increased our knowledge of idiotypes, mainly with respect to additional CRI on autoantibodies and to a relatively new aspect of 'pathogenic idiotypes'. It is obvious that much remains to be discovered about the normal role of idiotypes and how they might be involved in the pathogenesis of autoimmune disorders. The relationship of the idiotypic network to tolerance is a matter of speculation; tolerance implies the ability to distinguish between 'foreign' and self-antigens and it is important to remember that antibody V regions are also self-antigens. Several mechanisms have been proposed to explain immunological tolerance. Originally, it was envisaged that the repertoires of both T and B lymphocytes were in some way purged of potentially self-reactive clones. However, it is now evident that self-reactive lymphocytes do exist but are normally held under control. Finally, it may be that certain self-antigens are simply never exposed to immune surveillance. It seems that the control of self-reactive lymphocytes is central to the question of tolerance. In the absence of autoimmune disease, autoantibodies can be produced, for example, after many infectious diseases or vaccinations. However, this type of perturbation of the immune system is associated with the short-term production of low titres of low-affinity antibodies, generally of the IgM isotype, which are thought to represent germline gene products. Homeostasis is soon re-established, possibly by regulatory T cells interacting with autoreactive B cells through their idiotypic receptors.(ABSTRACT TRUNCATED AT 250 WORDS)