454 CORYNEBACTERIUM 



Antigenic Structure. 



In considering the antigenic structure of C. diphthericB and of the other species 

 and types within this genus, we may note that those workers who have attacked 

 this problem have, in most instances, approached it with a view to the identification 

 of the type species, and its differentiation from related species, types and variants. 

 It is, therefore, only in regard to the type species that our knowledge is in any way 

 detailed or systematized. 



Almost all the studies on the antigenic structure of this group of organisms have 

 been carried out by the method of agglutination ; and the application of this 

 technique soon showed that C. diphtherice is an antigenically heterogeneous species. 

 Langer (1916) demonstrated the existence of two serological types, Durand (1918, 

 1920) six, Havens (1920) two. Smith (1923) seven, Eagleton and Baxter (1923) 

 ten, and Scott (1923) eight ; in all cases there remained strains that did not fall 

 into any of the groups. 



These conflicting results have been to some extent clarified by the more recent 

 work of Ewing (1933) and of Kobinson and Peeney (1936), who have shown that 

 the antigenic structure of diphtheria bacilli differs according to whether they 

 belong to the gravis, intermedius or mitis varieties. Ewing studied 106 gravis 

 strains derived from Great Britain, Berlin and Khartoum, and was able to divide 

 them by direct agglutination into four types, which she called A, B, C, and D. 

 Kobinson and Peeney studied 739 gravis strains derived from many different parts 

 of the world, and distinguished five antigenic types, which they called I, II, III, IV, 

 and V ; the first four of these corresponded to Ewing's types A, B, C, and D. 

 The antigens appear to be qualitatively distinct, so that absorption of agglutinins 

 is not required to separate the types. On the other hand there does seem to be 

 some group relationship between the gravis, intermedius, and mitis varieties, indi- 

 cating the existence of a species-specific antigen. Types I and II strains have 

 the characteristic cultural properties of gravis strains ; Types III, IV, and V 

 often differ from the classical description and may resemble the mitis variety. 

 Types I and III seem to be restricted mainly to Great Britain ; Type II has a 

 world-wide distribution, but is uncommon in Great Britain ; Type IV was the 

 only type found by Kobinson and Peeney in Egypt ; and Type V was found only 

 in the United States of America, No antigenic difference was detected between 

 virulent and avirulent strains of gravis, though too few observations have yet 

 been made to say whether loss of virulence is accompanied by change in antigenic 

 structure. 



Less attention has been paid so far to the intermedius and mitis types. Ewing 

 (1933), however, mentions the occurrence of at least two antigenic types in a small 

 series of intermedius, and of at least five antigenic types in a small series of mitis 

 strains that she studied. By means of complement fixation, using alcoholic 

 extracts of the organisms, Hoyle (1942) obtained evidence of the existence of two 

 different lipoid antigens in diphtheria bacilli : (a) a specific antigen characteristic 

 of mitis but present also in small amount in gravis and intermedius ; (6) a non- 

 specific or group antigen present in large amount in gravis and intermedius and 

 in small amount in mitis. C. hofmanni contained the group antigen in large 

 amount, as well as a specific antigen of its own. 



Apart from Hoyle' s observations, little antigenic relationship has been observed between 

 C. diphtherice and other species of corynebacteria. Scott (1923) included in the series 

 that he studied a number of strains of C. xerosis and C. hofmanni, but could detect no 



