THE SIGNIFICANCE OF B. COLI IN WATER. 91 



More recently Hunnewell and one of ourselves (Wins- 

 low and Hunnewell, 1902 b ) examined a considerable 

 series of normal waters for B. coli, testing i c.c. 

 from each by the dextrose-broth method and a larger 

 portion of 100 c.c. by incubation with phenol broth as 

 described in Chapter VI. The samples were obtained 

 from the public supplies of Taunton, Boston, Cam- 

 bridge, Braintree, Brookline, Needham, and Lynn in 

 Massachusetts, and of Newport, R. L, from the Sudbury 

 River, from the ocean, from the waters of springs bottled 

 for the market, from ponds, pools of rain and melted 

 snow, springs, brooks, shallow wells, and driven wells in 

 various towns near the city of Boston. For comparison 

 50 samples of polluted waters from the Charles, Mystic, 

 Neponset, and North Rivers were examined. The colon 

 bacillus was defined as outlined in Chapter VI; and 

 organisms which lacked the power to reduce nitrates or to 

 form indol were classed in the "Paracolon group." The 

 results are summarized in the following table: 



PRESENCE OF B. COLI IN POLLUTED AND UNPOLLUTED WATERS. 



(Winslow and Hunnewell, igo2 b .) 

 UNPOLLUTED WATERS. 



I C.C. 100 C.C. 



Samples examined 157 153 



Dextrose broth positive 40 76 



Lactose plate positive 13 31 



Colon group 5 n 



Paracolon group 5 5 



B. cloacae group 5 



Streptococcus group 3 10 



