EXAMINATION OF rJLTEES 317 



and may be taken, therefore, to prevent organisms from 

 being directly washed through, the further examination is a 

 matter of some difficulty, and at the present time can only 

 be conducted inferentially, or by comparison with a standard. 

 The object of such examination is to discover whether 

 pathogenic organisms in water can grow through the walls 

 of the filter ; and the difficulty in making the examination 

 is that our information as to the circumstances which 

 favour the multiplication of organisms in water, and which 

 determine the maximum extent to which such multiplica- 

 tion may proceed in natural conditions, is quite incomplete. 

 It is impossible to state of any given water whether it offers 

 the maximum assistance to the growth of organisms that 

 may be found in natural water, or to say whether a speci- 

 men under examination is capable of multiplying to the 

 same extent as other specimens of the same organism 

 might multiply in a natural water. In many researches, 

 indeed, in which filters appeared to resist penetration of 

 organisms by growth, it was not even certain whether the 

 organisms under examination could grow in the water at 

 all. The method which must, therefore, be employed is 

 to take watei: containing known non-pathogenic organisms 

 known to multiply in it at suitable temperatures with 

 sufficient freedom to ultimately penetrate the Pasteur- 

 Chamberland tube, and to examine specimens of the filter 

 of which the efficiency is to be determined simultaneously 

 and with the same water-supply as specimens of the Pasteur 

 tubes themselves. The water must be kept at the optimum 

 temperature, and the filtrates examined periodically. If 

 the filter under examination retains the organisms for as 

 long a time, as the Pasteur, it must be considered as pos- 

 sessing the same efficiency. If, on the other hand, it 

 passes the test-organisms before the Pasteur tube will 

 do so, it is less efficient, and must for the present be 



