64 Mr. E. W. A. Walker. 



but it was also incapable of absorbing appreciably ordinary typhoid 

 agglutinins. 



Table II. 



Culture 

 T. 36. 



Readings 



at: 



25 



50 



125 



250 



500 



10C0 



2500 



5000 



Controls. 



Standard 



2 hours 



t 



t 



t 



t- 



s — 



tr 











O 



serum 



24 „ 



t 



t 



t 



t- 



s + 



tr 















Serum (a) 



2 „ 



t 



8 + 



tr 



tr- 











o 







1 



24 „ 



t 



t 



tr + 



tr- 























Serum (b) 



2 „ 



t 



t 



' 



t- 



s 



tr 















24 „ 



t 



t 



t 



t 



t- 



s — 















Standard serum 100-unit standardised serum (Dreyer). 

 Serum (a) — Standard serum absorbed with T. 36. 

 Serum (£) — „ ,, „ T. dys. 



It is important to add that T. dys. was subsequently brought back to the 

 eu-agglutinating phase. After a long residence in bouillon with numerous 

 subcultures its lineal descendants were found to have reverted to the motile 

 phase, and gave good agglutination with standard typhoid serum. A very 

 similar result was subsequently obtained with a highly dys-agglutinable 

 culture of B. paratyphosus B. 



It is clear that observations of this character must profoundly modify our views 

 regarding the meaning and importance of serological differences, and the results 

 obtained by absorption methods, particularly when such differences are only 

 quantitative and not qualitative. They suggest that " serological strains " of 

 bacteria, even when apparently permanent, may represent no more than 

 particular phases of activity of the bacterial type concerned. As to how such 

 phasic differences may arise we have at present little evidence. How they 

 are maintained for long periods through successive generations is a problem 

 that urgently demands investigation. But it must certainly be admitted as 

 conceivable that changes which can be induced in test-tube experiments, may 

 on occasion also be produced in the body of an infected individual. If growth 

 in the presence of agglutinating serum under laboratory conditions can lead to 

 serological changes in one or other direction, it must be accepted as possible that in 

 the living body the agglutinin- containing fluids and agglutinin-producing tissues 

 of the animal may also under suitable conditions exert similar influences. In 

 this connection the occasional isolation of " inagglutinable " typhoid bacilli 

 from cases of typhoid fever, and the serological diversity of the strains of 

 dysentery (Flexner) isolated during recent epidemics of dysentery afford 

 suggestive evidence. 



