THE SIGNIFICANCE OF B. CO LI IN WATER. 89 



Professor Jordan was himself somewhat sceptical as to 

 the value of the colon test, for he stated in 1890 (Jordan, 

 1890) that he had found, "in spring- water which was 

 beyond any suspicion of contamination, bacteria which 

 in form, size, growth on gelatin, potato, etc., were indis- 

 tinguishable from B. coli commune." In his recent 

 studies of self-purification (Jordan, 1901) the analyses 

 were made quantitative by the examination of numerous 

 measured samples, fractions of the cubic centimeter; and 

 the method employed was the enrichment either in dex- 

 trose-broth fermentation- tubes or in phenol broth, 

 with the subsequent making of litmus-lactose-agar plates. 

 The cultures isolated on these plates were tested as to 

 their behavior in dextrose broth, peptone solution, milk, 

 and gelatin; of the dextrose tubes made directly from 

 the water all were considered positive which gave more 

 than 20 per cent gas in the closed arm, with an appreciable 

 excess of hydrogen. The results were very significant. 

 In fresh sewage a positive result was obtained about 

 one-third of the time in one one-hundred-thousandth 

 of a cubic centimeter and almost constantly in one ten- 

 thousandth of a cubic centimeter. The Illinois and 

 Michigan canal proved almost as bad, giving positive 

 results on seven days out of twenty-eight in dilution of 

 one in one hundred thousand and on twenty-eight days out 

 of thirty-two in a dilution of one in ten thousand. At 

 Morris, twenty-seven miles below Lockport, where the 

 canal enters the bed of the Desplaines River, and nine 

 miles below the entrance of the Kankakee, the principal 



