180 ELEMENTS OF WATEE BACTEEIOLOGY 



observers in classifying colon bacilli is a long one. It 

 was pointed out by Smith (1893) long ago that saccharose 

 divides these organisms into two groups, and Winslow 

 and Walker (1907) have found that strains which 

 attack saccharose generally ferment rafJ5nose also. 

 MacConkey (1909) has made an exhaustive study of 

 the fermentative reactions of the group and suggests 

 an elaborate classification based on the use of a long 

 series of carbohydrate media. 



Howe (1912), from his recent statistical study of 630 

 strains of intestinal bacilli, concluded that fermenta- 

 tion tests in mannite, dulcite, and starch media are 

 of little value in the classification of colon bacilli, since 

 they are not closely correlated with other characters. 

 It'should be noted, however, that he worked only with 

 fresh intestinal strains and it is possible that types 

 characterized by definite reactions in these media, may 

 be rare in faeces but may develop so as to be important 

 in stored waters. 



Two noteworthy studies of the fermentative reactions 

 of the colon group have recently been published by 

 Kligler and by Rogers, Clark, and Davis, which will 

 be discussed in detail on pages, 200-200&. 



The Voges and Proskauer reaction has been extensively 

 used by MacConkey and his followers in England and 

 by Bergey and Deehan (1908) in this country. After 

 the carbon dioxide in the fermentation tube has been 

 absorbed by caustic soda, if the tube be allowed to stand, 

 an eosin-like color gradually develops in the open 

 arm, due to the presence of acetyl-methyl-carbinol. 

 West (1909) points out that the test used by Rivas, 



