270 BACTERIOLOGICAL EXAMINATION OF WATER. 



of ammonia. The water was sterilised, and filtered through a 

 new bougie for thirteen days. B. typhosus cultures were added 

 five times daring this period, but the organism could not be 

 detected in the filtrate. Five c.c. of broth were then added 

 to the water in the glass cylinder, and two days later typhoid 

 bacilli were found in the filtrate. Schofer concluded from his 

 experiments that only those bacteria were able to grow through 

 the filter which found sufficient nutrient material in the water 

 to enable them to grow out and multiply. 



Sims Woodhead and Cartwright Wood in 1897 repeated and 

 extended the experiments made by Schofer. These observers 

 first tested the filtration of the New River Company water 

 supply through Pasteur-Chamberland and Berkefeld filters 

 attached to taps and used as pressure filters. In the case of the 

 Pasteur-Chamberland filter it was found that only one of the 

 four micro-organisms present in the water appeared in the 

 filtrate, and this microbe was so devitalised that its power of 

 liquefying gelatine and producing a yellow pigment was much 

 delayed ; it required several weeks cultivation before it recovered 

 its usual characteristics. In the case of the Berkefeld filter the 

 same organism appeared in the filtrate on the second or third 

 day of filtration, and was accompanied by the B. fluorescens 

 liquefaciens. These organisms, however, did not display any 

 evidence of the devitalising influence which was found in the 

 experiments with the Pasteur-Chamberland filter. Experiments 

 were then made with chromogenic organisms placed in the 

 reservoir of the filters before these were attached to the tap. 

 The Micrococcous prodigiosus, Staphylococcus pyogenes aureus 

 and Bacillus violaceus were employed for the tests ; but none of 

 these organisms could be discovered in the filtrate, although 

 the experiments were carried on for several weeks and large 

 quantities of the filtrate were added to concentrated broth and 

 incubated for some time. The test organisms at the end of the 

 experiments were easily recovered from the reservoir of the 

 filter, showing that their non-appearance in the filtrate was not 

 due to the death of the micro-organisms. 



The next tests were made with pathogenic organisms, the 

 filters being fed day by day with emulsions of cholera and typhoid 

 bacilli in sterilised water. After the lapse of six and eight weeks 



