In a series of more than 6,000 cases of diphtheria gathered from many parts of the country and from Germany during the last five years, at least 95% of the strains have fallen within three principal types described in 1933 as Gravis, Intermediate, and Mitis. Of these the Gravis strains have been found to be associated with the highest case death-rate and the greatest incidence of paralysis. The Intermediate strains are more nearly related to Gravis than to Mitis in respect of their case death-rate, and at least equal to Gravis in tending to produce hæmorrhagic phenomena. The “Mitis” strains are the most likely of all to produce lesions extending to the larynx and lungs, but apart from such complications are rarely the cause of death. The suggested nomenclature for the types is further justified by observations on the incidence of diphtheria in the immunized and on diphtheria mortality in areas in which different types predominate. Also, it is justified though not so definitely by observations on animal pathogenicity. Typical Gravis strains are so constantly pathogenic to animals that virulence tests with such strains are superfluous. The stability of the types both in the animal and in the human body is marked although a number of observations are extant suggesting fluctuation of type in vitro. It seems most probable, although open to question, that the fluctuations of type observed in many areas over a period of years are due to the waxing and waning of virulence of a number of independent races, rather than to mutations from one to another. The discrepancy between the clinical severity of Gravis infections and their poor capacity to produce toxin in vitro has not yet received any adequate explanation. The nasal carrier is the most important factor in the spread of diphtheria. Clauberg's suggestion that the variants from the three well-defined types which are met may best be explained by rough to smooth variation within the types, the rougher strains representing the more virulent forms in each type, receives a good deal of support from a wide survey of the accumulated observations. A considerable group of freely growing and rough-non-starch fermenting forms which are non-pathogenic would form an exception to this rule as these would be classified as rough “Mitis.” These may, however, eventually prove with further investigation to be diphtheroid strains distinct from “Mitis.”
On the stability of the statistical classifier of states of human body systems to distortions in degenerate cases
