IN spite of the immense amount of investigation that has been carried out on diphtheria, the literature still fails to provide a wholly satisfactory picture of the pathogenesis of this disease. In particular, there is confusion regarding the meaning of the term "virulence'' aa applied to C . diphtherice, and difficulty in understanding the failure of antitoxin to control the hypertoxic or malignant type of the disease in man. In this paper the t.hesis is developed that. the course and outcome of the infection in a fully susceptible host depend upon two factors only : the toxigenic power of the organism and the effectiveness of the primary Mammatory response of the host's tissues. It will be shown, furthermore, that when this concept is widened to include a consideration of the response of the completely or partly immune host, then every grade of severity, from subclinical infection to hypertoxic diphtheria, can be adequately explained.The meaning of the term "exotoxin" aa here employed must first be clearly defined. The work of Eaton (1936), Pappenheimer and Robinson (1937) and Pillemer et al. (1947) has led to the belief that the exotoxin of C. diphtherice is a single well-defined substance, which can be prepared in a state approaching chemical purity. It now appears that this belief may have to be modified in view of the work of Pope et al. (1951). By extending the earlier work of Oudin (1946), Ouchterlony (1948), Elek (1949) and others on the interaction of toxin and antitoxin in semi-solid media they demonstrated the presence of aa many 88 24 separate zones of precipitation, each of which they held to indicate the presence of a distinct antigen-antibody system. Moreover, even a highly purified toxin showed the presence of several zonw of precipitation. Much further work must be done before the significance of these findings can be f d y appreciated, but it is already evident from experiments recently reported by Bowen (1962) that the subsidiary antigens revealed by this delicate method of analysis probably account for only a small percentage of the total protein of 8 highly purified toxin.This question of the multiplicity of antigens does not materially affect the ooncept of the pathogenesis of diphtheria, here developed. For tne purposes of the present communication the term " exotoxin " will be used in its biological sense to signify the substance or complex of substances liberated by C. diphtherice that is lethal to the cells of a susceptible host and the thesis will be developed without prejudice to the question whether the toxigenicity of this organism is due to one or more than one chemical entity.