ANTIGENIC STRUCTURE 689 



bacillus will agglutinate a Shiga bacillus to J or | titre ; but antigenically Schmitz's 

 bacillus and Shiga's bacillus are easily distinguishable ; a Shiga bacillus cannot 

 absorb the agglutinins from a serum prepared against Schmitz's bacillus, nor a 

 Schmitz bacillus from an anti-shigse serum. According to Schiitze (1944) Shiga 

 strains vary in their agglutinability ; hypoagglutinable strains may be rendered 

 more agglutinable by growth at 20-26° C, by heating in saline to 60° C. for 1 hour, 

 or by adding 0-5 per cent, phenol. 



Schmitz's bacillus used to be regarded as antigenically homogeneous, but the 

 observations of Boyd (1935) suggest that the freshly isolated organisms contain 

 two antigens which, for convenience, may be referred to as type and group. It is 

 believed that after some time in the laboratory the type antigen is lost and the 

 group alone remains. A serum, therefore, prepared against freshly isolated bacilli 

 will agglutinate both recent and old strains ; a serum prepared against old strains 

 will have little or no action on recent strains. The picture seems to be very similar 

 to that which will be described below for the 103 type of Flexner's bacillus, though 

 it is still doubtful whether the variation is to be regarded as of the type and group 

 or the smooth and rough order. 



Little is yet known of the antigenic structure of the para-Shiga group of 

 dysentery bacilli. Dudgeon and Urquhart (1919), who studied 11 strains, found 

 them to be antigenically alike. Similar findings were reported by Hazen (1938). 

 Ali (1938), however, was able by agglutination and absorption of agglutinins to 

 distinguish four serological groups among 8 strains ; and Sachs (1943) was able 

 to distinguish eight serological groups among 107 strains (see also Large and 

 Sankaran 1934, Christenson and Gowen 1944). It would appear that the para-Shiga 

 group of bacilli is far from being antigenically homogeneous. None of the members 

 appears to be related to either Sh. shigce or Sh. schmitzi. 



The mannitol-fermenting group has long been recognized as antigenically 

 heterogeneous. The work of Gettings (1919), Murray (1918), and Andrewes and 

 Inman (1919) afforded a picture of the Flexner bacillus as containing at least four 

 antigenic components. 



According to Andrewes and Inman, each of these components, which they 

 refer to as V, W, X and Z, is represented to some extent in every strain, but in 

 any given strain there is usually a preponderance of one antigen over the rest. 

 In certain races, V, W, and Z, there is so great a preponderance of a single antigenic 

 component, different in each instance, over the rest, as to make them behave 

 like distinct serological types ; each race requires its own antiserum for adequate 

 agglutination. The X race is peculiar in that it will not agglutinate with any 

 sera but its own ; yet it is able to give rise to a serum that will agglutinate not 

 only X races, but also Z, and, to a certain extent, V races. The agglutinins 

 corresponding to each of these four types cannot be more than partially absorbed 

 by the others. Andrewes and Inman found at least two sub-races, VZ and WX ; 

 these were members of the V and W races respectively, but contained so large 

 a proportion of a second antigenic constituent as to modify their serological 

 behaviour. One race, which is called Y, and which corresponds to the original 

 Y-strain of Hiss and Eussell (1903), contains a more evenly balanced mixture 

 of V, W, and Z components, with a small amount of X, For this reason a serum 

 prepared against a Y strain is more cosmopoUtan than the rest, having a wide 

 range of agglutination (Fig. 144). 



This conception has been challenged by the work of Boyd (1931, 1932, 1936, 



