100 



ves and incubated at 37° for 1—2 days, the first criterion of 

 the identity of this bacillus is that it shall be able to grow- 

 under these conditions and form well-developed colonies in the 

 time staled. Moreover it should be possible to grow a pure 

 culture in subcultures on the same medium. 



2. It is necessary that the bacterium shall be a Gram 1 - 

 negative rod. 



3. In pure culture it should not grow on agar or ascitic 

 agar (by which is meant, here and in what is to follow, 

 ordinary peptone broth agar, and the same medium with the 

 addition of blood-free ascitic fluid). 



4. On blood-agar plates (of suitable constitution) it should 

 not produce haemolysis. 



The significance of the conditions laid down under (1), 

 namely (a) rapid growth, and (b) power of growing on agar 

 containing dissolved haemoglobin, will be clear when the que- 

 stion of the boundary between Pfeiffer's bacillus and Bordet's 

 bacillus is dealt with later on. 



That (2) gives a sharp distinction has already been adequa- 

 tely explained on pp. 68—69. 



With regard to (3) it may be remarked in anticipation 

 that there has practically never been any difficulty in deciding 

 whether an organism should be termed haemoglobinophilic or 

 not. The bacteria examined fall naturally into two groups 

 according as they are markedly dependent on haemoglobin 

 or practically independent of it. 



In a number of cases however a pure culture on ordinary 

 agar has given a slight growth. But it is here exclusively a 

 question of bacteria which under various conditions, particu- 

 larly in giving haemolysis on blood-agar, naturally fall into 

 a group which it is reasonable to distinguish from Pfeiffer's 

 bacillus. If this group is eliminated, the distinction between 

 haemoglobinophilic bacteria (giving growth only on media con- 

 taining haemoglobin) and non-haemoglobinophilic bacteria 

 (giving growth on media with and without haemoglobin) is 

 very well marked. 



Point 4 is also a good distinctive mark, in that by using 

 a blood-containing medium of suitable constitution for those bac- 

 teria which possess other characters in common with Pfeif- 

 fer's bacillus, either there is no haemolysis even after several 



