Report on the Bacteriology of Water. 



217 



bolites of the nature of ptomaines and the like must be altered or 

 destroyed. 



Filtration, through porous films of porcelain, certainly acts less 

 violently on the water ; but it must by no means be concluded that 

 either the chemical constitution or the physical character of such 

 filtered water is absolutely unaltered. In the case of ordinary filtra- 

 tion one series of changes alone, viz., the alteration of the propor- 

 tions of the gaseous contents owing to the difference of pressure on 

 the two sides of the filtering film, will illustrate this.* 



On the whole, however, we may conclude that in cases where it is 

 necessary to eliminate the living bacteria of a natural water, the 

 process of filtration through porous porcelain is a better method than 

 that of sterilisation by heat. 



There can be little doubt that some of the discrepancies between 

 the results of the various observers, referred to on pp. 204 — 214, are 

 chiefly due to the sources of difference here indicated. 



It may be concluded, with some show of certainty : (1) That the 

 numerical results obtained by the gelatine-plate method, are, on the 

 average, too low. 



(2.) That several workers employed temperatures too high for 

 comparison with what occurs in natural waters in this country. 



(3.) That many of the results are vitiated by small quantities of 

 very concentrated food materials having been introduced into the 

 waters with the pathogenic germs employed for infection. 



(4.) That the conclusions drawn from experiments with distilled 

 water must be received with great caution, and are of little practical 

 value. On the whole we may regard "pure" water as a worse 

 medium for the life of pathogenic bacteria,f in spite of apparently 

 contradictory results in the hands of some of the investigators. 



(5.) That the conclusions drawn from cultures in sterile waters 

 must also be received with due regard to all the facts, and especially 

 those where the water was sterilised by heat. 



On the other hand, the enormously greater experimental difficulties 

 attending the investigations in which unsterilised waters are used 

 necessitate that the results should also be very carefully scrutinised, 

 and should not be finally accepted until confirmed by numerous 

 investigators attacking the question from different points of view, and 

 using different methods of research. 



(6.) That it is not safe to regard mineral waters as necessarily 



* For the effect of filtration through porous porcelain on the chemical composi- 

 tion of water, see Percy F. Frankland (" The Removal of Micro-organisms from 

 Water," ' Eoy. Soc. Proc...' 1885). 



t It may be remarked that the pathogenic bacteria, from the nature of the case, 

 are less adapted for life in media poor in organic materials than are the sapro- 

 phytic forms, and especially those known as " aquatic.'' 



Q * 



