BIO-CHEMICAL CHARACTERISTICS OF BACILLI. 95 



ing epidemic. Arabinose was attacked by all of this series 

 except the sivine disease cultures, thus confirming the fact 

 noted before, that such strains do not affect this sugar, 

 and that this may be used as a means of differentiation. 

 The action on dulcit on certain of the Bureau strains, 

 however, showed wider variations, for in two swine fever 

 cultures and in one food poisoning culture, no action 

 occurred up to three weeks. 



The result of the above tests is in agreement with pre- 

 vious observers with two noteworthy exceptions. Whereas 

 I find arabinose is unaffected by old cultures from swine, 

 Bainbridge states that acid and gas are formed on this 

 sugar. I do not think there is any possibility of error of 

 observation on my part as the results have been repeated 

 more than once. The only alternatives are firstly that my 

 arabinose is not the same as Bainbridge's (mine is from 

 Merck), secondly that he has in the preparation of his 

 arabinose media somehow facilitated the breaking down of 

 it by the strains used. In the preparation of my sugar 

 media, the ordinary steam sterilisation (twice for half hour) 

 is used, and the media is neutral, so breaking up is out of 

 the question. Again, none of my Gaertner types affect 

 dextrin, while three observers above quoted, found that 

 acid and gas is produced. The above remarks re arabinose 

 apply equally well to dextrin. 



Notes on certain recently discovered Fseudo-Gaertner 

 type organisms recovered from various sources. 



(See Table III.) 

 Case I. Purcell. — This organism was recovered from the 

 blood of a septicaemic case, the history of which is as 

 follows: — The patient was operated upon for epithelioma 

 of lip and neck glands; these removed, he was progressing 

 to recovery when suddenly the temperature shot up and 

 the pulse became very quick. An erysipelatous condition 



