There has been intermittent interest for many years in the possibility that immunological processes may play a part in the induction of neoplasms. Early in the history of cancer research it became evident that immunological responses could be provoked by transplantable tumours and, although the nominal aim of most research in the field has been to identify specific tumour antigens with a view to the eventual elaboration of methods of treating cancer, there has been an incidental yield of experimental results and theoretical ideas of great general interest.

It is now clear that some established tumour lines lack antigens which are present in normal cells (Zilber, 1958; Nairn, Richmond, McEntegart & Fothergill, 1960). This finding has obvious possible relevance to the rinding that chemical carcinogenesis is or may be associated with the deletion of a carcinogen-binding protein from the cells which have undergone malignant transformation (Miller & Miller 1953; Heidelberger & Moldenhauer, 1956). Green (1954) 1 suggested that the transformation took place by an immunological mechanism in which the carcinogen acted as a hapten, and by union with cell protein became a complete antigen. Antibody (or cell-bound antibody) induced by the antigen then became the active agent enforcing in one way or another the change to the deleted form. In addition to deletion of antigens there is equally extensive evidence that new antigens may appear with the development of malignancy, and again some interpretation in immunological terms is called for.