Report on the Bacteriology of Water. 417 



From the above table it will be seen that the unsterilised Thames 

 water, which contained remarkably few bacteria (only 290 per c.c.) at 

 the time, was infected with a very large number of typhoid bacilli 

 (about 78,000 per c.c.). The total number of bacteria in the water 

 kept at the winter temperature of 6 8 C. underwent enormous 

 multiplication followed by decline, as in the case of the uninfected 

 unsterilised water. In the water kept at the summer temperature of 

 19 C., on the other hand, the numbers found exhibited almost con- 

 tinuous decline, and also closely resembled those found in the unin- 

 fected unsterilised water preserved under similar conditions. In 

 both cases, however, there must have been a great multiplication of 

 the water-bacteria, for whilst the gelatine plates prepared from these 

 waters during the first week after infection admitted of the ready 

 recognition of typhoid colonies, in the subsequent examinations this 

 was altogether impossible, so that the large number of colonies 

 present on these later plate cultivations must have been derived from 

 the extensive multiplication of the comparatively few water- bacteria 

 present in this unsterilised water at the outset of the experiments. 



I must, however, again emphasize what I have stated before, that 

 whilst the recognition of typhoid colonies on such plates containing 

 the colonies of numerous water-bacteria is often difficult and attended 

 with much uncertainty, any estimation of the number of typhoid 

 colonies on such plates, as has been attempted by some observers, is 

 altogether illusory and calculated to lead to the most erroneous con- 

 clusions. For whilst the surface colonies of the typhoid bacillus are 

 even liable to be confounded with the surface colonies of some other 

 bacteria, in the appearance of the depth colonies (and, of course, in 

 ordinary gelatine plates the majority of the colonies are beneath the 

 surface) there is nothing to distinguish them from an immense 

 number of other forms common in water. Thus, whilst in the above 

 series of examinations I have no hesitation in saying that on the 

 plates prepared on the llth, 16th, and 17th May, typhoid colonies 

 were present, I rely for the determination of their presence or absence 

 after those dates entirely on the results of the examinations by phenol 

 broth-culture which will be given below. Again, even in the case of 

 those plates which obviously contained typhoid colonies, I do not con- 

 sider that any estimate of their number could be justifiably made, as 

 such an estimate could only include the surface colonies which had 

 developed the characteristic expansions. 



Thus the examination by plate-culture of these unsterilised waters 

 does not enable us to ascertain whether the typhoid bacilli under- 

 went any numerical increase in these waters, but from the fact that 

 no such increase was observed in the case of the typhoid bacilli 

 similarly introduced into steam-sterilised Thames water (see p. 451), 

 and in which, therefore, the conditions for their multiplication were 



