ACTION AND UTILITY OF FILTERS. 275 



a bent platinum wira passed through the lateral arm, and 

 planted out on an agar-slope. The thin greyish growth which 

 resulted was then sub-cultured and tested for agglutination. 

 The B. typhosus used for the inoculation of the mantle was 

 also tested at the same time. The results are given in the 

 Table on the previous page. 



The organism which appeared in the Kitasato flask was evi- 

 dently the B. typhosus which had taken five days to grow 

 through the walls of the No. 12 filtering candle. This result is 

 in harmony with the results of Schofer and Sims Woodhead. 



Experiment II. The same form of apparatus as in Experi- 

 ment I. was employed, a new Berkefeld bougie, size No. 12, 

 being fixed in the mantle. Fifty c.c. of sterile broth having 

 been placed in the Kitasato flask, and all the joints made tight 

 with melted paraffin, the mantle was filled with a sample of 

 sterile barrack sewage. This sewage on chemical analysis gave 

 the following results, expressed as parts per 100,000 : Total 

 solids 610 parts, chlorine 52 parts, free ammonia 1*44 parts, 

 albuminoid ammonia 1*846 parts, oxygen absorbed (Tidy, four 

 hours) 8 parts. By releasing the pinch -cock 10 c.c. of the 

 sterile sewage were allowed to filter through into the broth ; 

 the apparatus was then incubated at 25 C. Next day the 

 broth in the flask being perfectly clear, the sewage in the 

 mantle was inoculated with a sub-culture of the B. typhosus 

 employed for the first experiment. Ten c.c. of the sewage were 

 then filtered into the broth every twenty -four hours, the amount 

 filtered being replaced by sterile sewage. The broth remained 

 perfectly clear until the fifth day, when a distinct turbidity 

 was noticed. A loopful of the turbid broth in the flask was 

 withdrawn through the lateral arm and planted out on agar. 

 The growth which appeared in twenty-four hours was tested 

 with Berne serum for agglutination and sub-cultured in the 

 usual media. The following results were obtained : 



The bacillus was completely agglutinated by the Berne serum 

 diluted 1-1000, and partially agglutinated by the serum diluted 

 1-10,000. On potato, there was a moist transparent growth. In 

 peptone solution, no indol was produced. Mil kwas unchanged. In 

 glucose-gelatine there was no gas formation. In litmus- whey, after 



seven days at 37 C. the acidity equalled 4 per cent. ^ alkali. 



