18 MASS. EXPERIMENT STATION BULLETIN 319 



stimulating the production of agglutinins, as will be seen in the following data: — 



No significance can be attached directly to the variation in titres, since other 

 factors aside from the organism play a role in producing agglutinins. 



In a brief summary of the data presented in this section it may be stated that 

 variants among 5. pullorum may be isolated from naturally infected cases. The 

 incidence of variants as observed in this investigation amounted to 13 out of 

 163 cases, or 7.97 per cent. While this may be considered a significant percentage, 

 yet from the diagnostic point of view, this figure would have been reduced to zero, 

 since these strains were identified by tests for properties other than agglutinability 

 and colony morphology. Of course, one is unable to state to what extent such 

 organisms may vary further under natural conditions. The variants detected in 

 this investigation were isolated from naturally infected eggs, chicks, and adults. 

 The behavior of the variants in NaCl solution of various concentrations and in 

 the presence of positive and negative sera differed greatly. 



VARIATION INDUCED BY ARTIFICIAL METHODS 



According to Hadley, Arkwright, and others, many species of bacteria, if not 

 all, are subject to variation. Variability in one or more characters of an organism 

 may be the result of natural environment or artificial conditions set up in the 

 laboratory. In certain species which have been carefully studied the majority 

 of characters have been found liable to variation. Some investigators have 

 ventured so far as to claim the establishment of new species. These claims, how- 

 ever, are few in number and since such observations cannot be corroborated at 

 will, one feels reluctant to entertain a view of this nature. 



Different methods of studying variation have been described. In this investiga- 

 tion attention was directed chiefly towards deviations from the normal colony 

 type (See Figures 1 and 2). A change in colony form was selected as a starting 

 point to study the degree and type of variation. After having ascertained varia- 

 tion in one character, changes in other characters such as cellular morphology, 

 biochemical and tinctorial reactions, pathogenicity, antigenicity, and agglutinabil- 

 ity were studied. The author is cognizant of the fact that variation may occur 

 in certain properties such as antigenic structure and virulence, without the strain 

 undergoing macroscopic or even microscopic changes in colony form. To detect 

 strains that show a variation in antigenic structure and virulence would of neces- 

 sity involve a large number of experimental animals. Hence, it appeared expedient 

 to select any change in colony type as a starting point for investigation, since the 

 main objective was to determine to what extent 5. pullorum would lend itself 

 to variation in some of its properties under certain conditions. 



The isolation of variants was accomplished by repeated selection of single 

 colonies. While this method may have its shortcomings, success was attained in 

 isolating variants that appeared markedly, if not entirely, stable. 



In the historical resume it is shown that variation may be induced through 

 several means, such as nutriment, chemical agents, aging in broth, growth in 

 serum, bacteriophage action, etc. In this investigation repeated transfer and 



