300 Profs. Percy Frankland and Marshall Ward. 



and placed in action within a few hours of its removal from the river. 

 The temperature of the room was 15 — 16° C, and remarkably con- 

 stant day and night. The flasks were exposed to diffused daylight, 

 and occasionally to the light of a Swan lamp. 



After the six days the two flasks showed the following results on 

 bacteriological analysis by means of gelatine plate cultures, using 1, 

 3, and 6 drops per plate, and diluting with 10 vols, sterile water to 

 1 vol. from flask. The numbers given are the calculated averages, 

 and it is possible that more extended observations would, perhaps, 

 alter them slightly. 



State 

 of water. 



Number 

 of 



drops in 

 culture. 



Total bacteria 

 per 1 c.c. 

 original. 



Proportion 

 of liquefyiug 

 bacteria. 



Remarks. 







1 

 1 



3 

 3 

 6 

 6 



35,640 

 17,695 

 33,000 

 15,840 

 uncountable 



5) 



26 : 64 

 24 : 23 

 26 : 64 

 1 : 3-5 



1 Liquefied too 

 J- rapidly for 

 J estimation. 



Non-aerated . . 



Non- aerated . . 



Non-aerated . . 



It must be remembered that in such experiments as these, the rapid 

 development of the liquefying organisms is always the chief trouble, 

 since we cannot allow the plates to incubate long enough to bring- 

 forward all the liquefying forms, on the one hand, while the lique- 

 faction overpowers the young non-liquefying colonies on the other : 

 consequently we lay no stress on the last column. 



As regards the direct effect of the aeration, we believe the third 

 column does express it more or less accurately, though of course the 

 reply is always possible that we have no absolute guarantee that no 

 aerial forms were filtered into the flask. 



We propose to extend this inquiry at a later period, and merely 

 put forward our results so far as tentative, and by no means devoid 

 of interest and suggestiveness. 



We next resolved to extend the comparison between aerated and 

 non-aerated cultures to flasks of Thames water treated exactly as in 

 the foregoing series except that we first infected both flasks with 

 anthrax. 



We employed a gelatine culture of the anthrax, selecting a tube of 

 rapid growth, and in which the gelatine was completely liquefied, 

 and used a relatively very large quantity (5 c.c. to the litre of water) 

 and of course a correspondingly large quantity of gelatine food- 

 material. This was done purposely in this first experiment, to en- 



