The phenomenon of idiotypy was discovered almost thirty years ago, but it was only during the past decade that it attracted widespread interest and became the subject of numerous research investigations. From the outset, much of the interest in idiotypy was based on its implications with respect to the repertoire of antibodies. Kunkel showed, for example, that idiotypes associated with certain human myeloma or Bence-Jones proteins were present in normal human globulins at levels of less than one part per million. Also, Oudin's original definition of idiotypy implied that idiotypes could be uniquely associated with individual rabbits as well as with particular antigen-binding specificities. Such observations provided some of the earliest evidence for an extensive repertoire of immunoglobulin molecules. The implications of these findings have been amply confirmed by recent studies of protein struc ture and molecular genetics; many of these studies are reviewed in the present volume. It is known now that the diversity of antibodies is based on the presence of numerous V and L V H genes, on recombinatorial events involving D and] segments, on somatic mutations, and on processes involving deletion of DNA followed by repair with errors, including inser tions. Each of these parameters is capable of influencing the idiotype expressed by the final immunoglobulin product. Regulation of the immune response is another area in which idiotypy has significantly influenced modern immunology.