52 MASS. EXPERIMENT STATION BULLETIN 319 



Strain XIV exhibited only slight absorption, while Strain I, Experiment I, 

 caused no reduction in agglutinin titre. The variants derived from these strains 

 behaved in a like manner. Variants XH-a and Xll-b exhibited absorptive ability, 

 but not to the same degree; only the latter completely absorbed the agglutinin 

 content. No definite explanation is offered for this difference in action, except 

 that the one variant possessed a sufficient number of the cell type which was 

 capable of absorbing the agglutinins. The fact that this variant produced two types 

 of cells in its colony structure (See Figures 18, 19, and 20), makes it quite possible 

 that the more typical appearing cell was present in sufficient quantity to bring 

 about complete absorption. 



The four miscellaneous strains, which appeared typical, had been subjected 

 to treatment similar to that of the variants, but did not change in colonial and 

 cellular morphology. Likewise, their capacity to absorb agglutinins was as great 

 as that of the original parent strain. 



These results show that variants which yield different colonial and cellular 

 forms, which are different from the original strain, also possess little or no power 

 for absorbing specific agglutinins. 



DISCUSSION 



A review of the literature shows that bacterial variation of 5. pullorum has 

 been observed and studied by only a few investigators. Mallmann (1932) worked 

 principally with strains that had been maintained under laboratory conditions 

 for a period of time varying from four to nine years. The impression is given that 

 variants are seldom isolated from natural outbreaks of the disease. The criterion 

 used for the variant study seemed to include only a change in colony morphology. 

 On the other hand, Plastridge and Rettger (1930) reported a natural outbreak of 

 the disease among young and adult stock from which pleomorphic types were 

 isolated. These types appeared strikingly different from the normal type in 

 several respects. 



Since so few observations have been reported concerning the incidence of 

 5. pullorum variants among freshly isolated strains, the writer believed that an 

 examination of a large number of recently isolated strains might yield valuable 

 additional intormation. It was realized from the first that a change in colony 

 form was not the only feature that would classify a strain as a variant or normal 

 type. If the colony type had been selected as the only means of identifying 

 variants, all but two of the strains studied by the writer would have been des- 

 ignated as typical. Likewise, if these strains had been identified only by the 

 Gram-stain method and biochemical reactions, all would have been considered 

 typical. Hence, to classify an organism as typical, its various properties must be 

 studied. Furthermore, only slight deviations, which might be overlooked by one 

 not familiar with the characteristics of variants, may be observed at times. The 

 property of salt-sensitiveness would in all probability be overlooked when the 

 rapid serum agglutination method is employed to determine the agglutinability 

 of a freshly isolated strain. While this method may have its merits in the diagnosis 

 of pullorum disease, it should not be abused by being employed for the typing 

 of the organism unless every step is carefully controlled. 



The source of the strain apparently cannot be correlated with the incidence of 

 variants, since the different types were isolated from eggs, chicks, and adult fowl. 

 However, if a large number of cultures from one source were examined, it might 

 be possible to show such a correlation. Strange as it may seem, in the routine 

 testing of domestic fowl re-infected flocks generally show a very small number 



