BACTERIOLOGICAL STUDY OF WATER 649 



by passing very large quantities of water through an ordinary 

 Pasteur or Berkefeld filter, brushing off the matters collected 

 on the filter into a sterilized vessel and examining this by 

 plate methods. 



It has occurred to us that possibly the employment of 

 chemical coagulants, such as alum and iron, might prove 

 serviceable for this purpose. Their action would be to 

 mechanically drag down, in precipitating as hydroxides, 

 the suspended bacteria contained in the fluid. This preci- 

 pitate could then be examined bacteriologically, * instead 

 of the water, and the recent experiments of Ficker (loc. tit.) 

 appear to demonstrate' the value of such a procedure. 



The difficulties in this field of work are obviously due to 

 the suspension of a very small number of the disease-pro- 

 ducing organisms sought for in large volumes of fluid, and 

 the association with them of large numbers of other species 

 that offer a very great obstacle to the successful search for 

 the pathogenic varieties. 



If by either of the above procedures bacilli that bear 

 any resemblance to bacillus typhosus be isolated, recourse 

 must then be had to all the differential tests detailed in the 

 chapter on that organism. 



The Quantitative Estimation of Bacteria in Water. Quan- 

 titative analysis requires more care in the measurement of 

 the exact volume of water employed, for the results are to 

 be expressed in terms of the number of individual organisms 

 to a definite volume. The necessity for making the plates 

 at the place at which the sample is collected is to be particu- 

 larly accentuated in this analysis, for multiplication of the 

 organisms during transit is so great that the results of 

 analyses made after the water has been in a vessel for a 

 day or two are often very different from those that would 

 have been obtained on the spot. 



