38 



schemes of classification if the authors happen to select different characters 

 for the initial subdivision. 



Statistical Classification of the Colon Group. To offset these 

 objections, Winslow suggested the utilization of the statistical method first 

 employed by Andrews and Horder and Winslow and Winslow in studies on 

 Coccaceae. Individual characters are not considered paramount and inde- 

 pendently but only in relation to each other. 



Howe first attempted the statistical method in 1912. This investigator 

 made a detailed study of acid and gas production and various other tests 

 on 630 strains freshly isolated from human intestinal contents. He con- 

 cluded that indol, nitrate reduction, motility, fermentation of dulcitol and 

 mannitol, and starch were not correlated with other characters and were 

 consequently not of classificatory value. He recognized only two groups 

 the sucrose positive, Bad. communior, and the sucrose negative, Bact. com- 

 munis. 



Rogers and his associates in 1914 and 1916 studied a large number of 

 colon strains from milk,, grains and bovine feces and on the basis of ac- 

 curately determined gas ratio from dextrose concluded that two distinct 

 groups may be distinguished. One, referred to as the low ratio group, 

 produced carbon dioxide and hydrogen in equal volumes and includes 

 about 52 percent of the strains from milk, 99.6 percent of the strains from 

 bovine feces, and only 4.8 percent of their grain strains; whereas the high 

 ratio group, which was characterized by the production of two or more 

 times as much carbon dioxide as hydrogen, included 47.5 percent of the 

 milk strains, only 0.5 percent of the bovine fecal strains, and about 95 

 percent of the strains obtained from grains. There is thus a very strong 

 correlation between these subdivisions and the source. 



Kligler (1915) suggested that salicin be substituted for dulcitol, in 

 subdividing coli-like bacteria, pointing out that salicin fermentation cor- 

 relates better with the Voges-Proskauer reaction than does dulcitol de- 

 composition. He recognizes a sucrose negative-salicin negative group (B. 

 acidi-lactici) ; sucrose negative-salicin positive group (B. communis) ; su- 

 crose positive-salicin negative group (B. communior) and sucrose positive- 

 salicin positive (B. aerogenes) . B. cloacae is differentiated from B. aero- 

 genes by its inability to ferment glycerol. 



The characterization of B. communior as salicin negative is probably 

 untenable. The term B. coli- communior was first employed by Durham 

 to describe members of the colon group which fermented sucrose and which 

 were motile. Later Ford recognized it as a species B. communior. 



Of 77 motile sucrose fermenting bacilli of the coli section, 56 (73%) 

 were found by the writer to be salicin fermenters. It is felt therefore that 

 Bact. communior should not be described as a salicin non-fermenter. 



Where the principle of correlation has been employed the best cor- 

 related character has apparently been picked out by inspection of the data. 

 Inspection is a tedious and difficult procedure, entirely inapplicable where 

 the number of characters considered is large, and it does not permit of a 

 concise statement of the degree of correlation which exists between differ- 



