82 ELEMENTS OF WATER BACTERIOLOGY 



lo c.c. portions were withdrawn and placed in long, 

 narrow test tubes. To each of these tubes enough 

 typhoid serum of known potency was added to make 

 a regularly graded series, 1-50, i-ioo, 1-150, and 1-200. 

 The probable presence of the typhoid bacillus was 

 manifest by the formation of flocculi within a quarter 

 of an hour, and agglutination was complete in from 

 2 to 5 hours. 



The tube having the greatest dilution in which 

 agglutination was apparent was then examined by 

 breaking off the lower end, containing the precipitate, 

 washing the sediment two or three times with sterile 

 water after removing the clear supernatant liquid, 

 and allowing the bacteria to settle again. The organ- 

 isms remaining were plated upon various media, and 

 examined biochemically to determine the true character 

 of the suspected colonies. It was found that a dilution 

 of I to 60 was the highest which could be used with the 

 organisms examined, and it is therefore probable that 

 high dilutions (greater than 1-60) cannot be success- 

 fully used. 



Investigation of an organism isolated by this method 

 was made by Klotz (1904), who found the culture 

 to be not a typical B. typhi, but a form showing 

 certain points of similarity to both B. typhi and to 

 B. coli, and probably intermediate between them. 

 Frost (1910) isolated a bacillus of the B. proteus 

 group from filtered Potomac water which agglutinated 

 with typhoid serum in high dilutions. As Klotz points 

 out, therefore, it is evident that even when a positive 

 result is obtained with a relatively high dilution of 



