172 THE FLOATING-MATTEE OF THE AIE. 



its centre. The portion of tubing above the pinchcock 

 ■was then filled with the infusion, and the end of the 

 separation-funnel introduced into the tube, all air being 

 thus excluded. On turning on the stopcock and releasing 

 the pinchcock, the liquid passed slowly down the shank 

 of the pipette, filling it wholly. The point of the shank 

 was then placed in succession over the test-tubes, the 

 infusion entering them without a single associated 

 bubble. The arrangement was not absolutely perfect, 

 but it was an improvement upon previous ones. As 

 before, the charged chambers were .placed in a room 

 the air of which was maintained at a temperature of 

 about 90°Fahr. 



The result was as follows : — Of the two turnip cham- 

 bers prepared as just described, on the 4th, one had 

 completely given way on the 6th. In the other cham^ 

 ber two out of the three tubes had given way, but the 

 third remained permanently brilUant. Previous to this 

 series of experiments I had never succeeded in saving 

 even a single tube of cucumber-infusion ; here, how- 

 ever, two out of the three chambers charged with it 

 remained perfectly clear for many days. Subsequently 

 one of these chambers yielded in part, through an 

 accident, but the other chamber is as brilliant at this 

 moment as it was on the day of its preparation many 

 months ago. 



Now, as regards inherent power to generate life, the 

 infusion of this chamber was in precisely the same 

 condition as its two neighbours. They, one and all, 

 contained the same infusion ; and there is no way of 

 accounting for the observed difference of deportment 

 save by reference to contamination from without. Here 

 we are certainly on the trace of the enemy which has 

 given us so much trouble. 



On the 5th of December two additional cases were 



