45 



by cultivation on media of variable constitution must undoubtedly 

 be very diverse. Then again it was possible to cultivate the bacillus 

 from the blood 17 hours after the injection in one of Bender's 

 rabbits which indicates that it possesses no trifling resistance against 

 the bactericidal action of the blood and thus far is in favour 

 of its identity with Cohen's. If we further compare this experi- 

 ment with corresponding experiments of Cohen and Prasek & 

 Zatelli (who Bender considered worked with Cohen's bacillus), 

 and also Wollstein, the only difference (apart from Cohen's inocu- 

 lations in series into fresh rabbits, which was not tried by Bender) 

 consists in the fact that these authors not only cultivated the 

 bacillus from the heart blood but also demonstrated it microscopi- 

 cally „in considerable numbers". How much weight may be attributed 

 to this difference, it is difficult to know. While Bender's rabbit 

 died 17 hours after the injection, Cohen's died 5 and 12 hours 

 after respectively, Prasek & Zatelli's 14 hours after, and Woll- 

 stein's 18—30 hours after. This does not speak well for the 

 view that in the first case there was a purely toxic action while 

 in the other cases there was multiplication without primary toxic 

 action., 



The reason for making this work the object of a detailed 

 review is merely to afford an example of the want of clearness 

 and certainly which besets all attempts to classify haemo- 

 globinophilic bacilli on a basis of virulence. It is of course 

 feasible that by means of very arduous and protracted experi- 

 ments with the use of a large number of animals more de- 

 finite boundaries could be established, but I consider it better 

 to agree with Henry that „the investigation of such a series 

 of haemophilic organisms from the standpoint of their patho- 

 genicity is not profitable, and it is probable as Ritchie has 

 suggested, that investigations into toxin production and into 

 the serum reactions of immunised animals hold out a much 

 better chance of differentiation." 



A qualification of this statement however, must be made. 

 As regards toxin formation or to use a more guarded expres- 

 sion, toxicity, it may certainly be claimed that this character 

 is often very prominent, but it is undoubtedly not justifiable 

 to propose it as a criterion of Pfeiffer's bacillus, and investi- 

 gations on this point have up to the present scarcely made 

 any appreciable contribution to the classification of this species 

 of bacteria. The problem assumes a different aspect with re- 

 gard to serum reactions as will later be explained. 



