The authors outline schematically the major histocompatibility complex as well as the relationship between tumor neo-antigen and pre-existing antigens at the cell surface. They note that these two antigenic systems are not independent and in particular, the major histocompatibility system (H-2) co-segregates with the tumour antigens. Considering the complexity of the H=2 system (a H-2 allele does not correspond to any single transplantation antigen but to a combination of several antigens units simultaneously present), the authors recall Boyse's hypothesis, revised by Haywood and Mc Khann, which propose that tumour antigens are a rearrangement of H-2 substructures. Moreover, it is possible that the relationship between H-2 and resistance to cancer may be attributed directly to the action of genes controlling the immune response (Ir region) which could also intervene in recognition of these tumour antigens. Finally, recent results obtained seem to show that some tumour antigens cross-react with H-2 fractions. This fact prevents the mice bearing these types of alleles from being immunized against the cross-reacting tumours. If this result could be transposed to the human situation, it would explain the frequency of certain types of tumours in defined H-LA groups and could allow the prediction of high-risk groups.