THE COLON GROUP OF BACILLI 115 



Clemesha confirmed these results by a long series 

 of examinations of naturally and artificially polluted 

 waters, all tending to show that with fresh pollution 

 most of the dextrose fermenters ferment lactose as 

 well, while with storage there is a relative increase in 

 the lactose-negative forms. Careful studies of the 

 history of faecal mixtures in water showed that the 

 resistant form was a particular type, called by Clemesha 

 Bacillus P. Houston (1911), reports similar results 

 for London waters. Of 12,744 specimens of raw river 

 water containing dextrose-fermenting organisms 81 

 per cent gave positive results in lactose and formed 

 indol as well, thus indicating the presence of the colon 

 group. Of 18,960 specimens of filtered water contain- 

 ing dextrose-fermenting organisms only 51 per cent 

 gave positive results in lactose and formed indol. 

 Clemesha (191 2 b ), in an analysis of Houston's results, 

 shows that the preponderance of dextrose-positive 

 lactose-negative forms is here not due to the Bacillus P, 

 which Clemesha found in India, but to two different 

 forms. 



From all these investigations it is clear that the dex- 

 trose broth test does not bear a constant relation to 

 the presence of the colon group, since another type of 

 organisms fermenting dextrose but not lactose is rela- 

 tively much more abundant in stored and relatively 

 pure water, particularly in warm weather. 



Phelps and Hammond (1909) have pointed out a 

 rather serious error in the routine isolation of B. coli, 

 as it used to be practised in this country, due to the 

 presence of this group of organisms which fail to form 



