Report on the Bacteriology of Water. 



263 



If we compare Table a with Tables b and c, there is a striking con- 

 trast as regards the maintenance of the anthrax colonies on the plates. 

 This is no doubt largely due to the removal of the water organisms, 

 enabling ns to count the anthrax colonies so much more readily ; but 

 I do not believe it is solely due to that cause. It seemed much more 

 likely — and a comparison of these tables with those of Series C ap- 

 peared to bear out the probability — that the competition of the water 

 organisms really affects the anthrax more directly, partly owing to the 

 former taking what organic food materials there are, and so starving 

 the anthrax, and partly owing to the rapid de-oxygenation of the water 

 by the competing forms. As will be shown later, these normal water 

 bacteria are aerobic in a very high degree, as we have convinced 

 ourselves by actual experiments ; and we have been surprised, there- 

 fore, at these results in the raw Thames water. As will be seen in 

 the sequel, however, the behaviour of the organisms towards one 

 another cannot be predicted (see pp. 290 — 298). 



There is one point in connection with the boiled Thames water 

 cultures (Table c) which seems worth further investigation : it is the 

 remarkable retardation of growth exhibited on some of the plates 

 after the first twenty-four hours. It seems by no means unlikely 

 that the explanation is due to two causes : — 



(1) The boiled water has been so far de-oxygenated that the living 

 bacilli fall off:, and only those which can pass into the spore condition 

 maintain themselves, and as it takes longer to get cultures from the 

 spores than from the actively vegetating bacilli, this might well 

 explain the retardation seen on the plates. 



(2) It may also be, however, that boiling the water renders many 

 of the organic food substances les3 available for the growth of 

 anthrax, and thus a partial starvation concurs in the fall. 



Or (3) it may be due simply to osmotic phenomena consequent on 

 immersion in the water. 



In any case it seems worth while to note the apparently more 

 rapid fall in the numbers in the boiled as contrasted with the filtered 

 water, in the first forty-eight hours, though we think the matter 

 would need a special inquiry to make certain of the phenomenon. 



Series G. 



This series of experiments was designed to secure answers to the 

 following questions : — Do the bacilli of the anthrax live and multiply 

 in Thames water at all, apart from any persistence of the spores ? If 

 so, is there any difference in their behaviour in the crude water, taken 

 fresh from the river, with all its normal bacteria flora and other im- 

 purities, and in the same water deprived of the aquatic microbes by 

 filtration through porcelain, or sterilised by boiling? And, further, 



x 2 



