56 MASS. EXPERIMENT STATION BULLETIN 319 



SUMMARY AND CONCLUSIONS 



From the experimental evidence obtained in the investigation discussed here 

 the following conclusions seem justified. 



Variation of characters of 5. piillorum may be detected among freshly isolated 

 strains, as revealed by 13 of 163 strains studied. 



The behavior of the recently isolated variants in sodium chloride solution 

 of different concentrations and in the presence of positive and negative sera 

 differed greatly. 



Such properties as colonial and cellular morphology, Gram-staining, and 

 biochemical reactions revealed little or no variation in the majority of the de- 

 rived variants. 



Many strains of S. pullorum readily underwent variation when subjected to 

 frequent transfer and storage in meat infusion broth. 



Variants were isolated and stabilized on solid media by colony selection and 

 by frequent transfer. 



When variants were cultivated on both meat extract agar and liver infusion 

 agar the colonial and cellular features often appeared markedly different. Some 

 strains that appeared typical on meat extract agar exhibited pleomorphic ten- 

 dencies when placed on liver infusion agar. 



Variation of 5. pullorum was not observed when the organism was subjected 

 to alternate transfer in liquid and solid media, or cultivated in 5 per cent peptone 

 solution, in deep meat infusion broth, or in plain 3 per cent meat extract solution. 



5. pullorum variants may settle out in broth cultures, the degree of settling 

 varying with the different types. 



The ability to remain suspended in sodium chloride solution varies with the 

 different variants, and it may fluctuate even within the same type. 



The power of the variants to attack the different fermentable substances was 

 practically the same as that of the normal type. 



Passage of the variants through whole fresh eggs and through the natural host 

 (chicks and mature fowl) had little or no influence on the general features of the 

 different types. 



Most variants exhibited a lesser degree of virulence than their parent strains, 

 and in some instances appeared avirulent. 



Repeated subcutaneous inoculations of massive doses of variants may serve as 

 a means of detecting organisms that resemble the normal type in colonial and 

 cellular morphology, Gram stain and biochemical reactions, and in agglutinability. 



A menstruum which maintained the variants in suspension and which was also 

 satisfactory for conducting the agglutination tests was not discovered. 



The agglutinogen^ power and the absorptive capacity of the variants tested 

 were not equal to those of the normal type. 



