VITALITY OF PUTREFACTIVE ORGANISMS. 223 



permanently clear. These latter only yielded the water- 

 hammer sound. The turbid flask, on the contrary, 

 ■when shaken yielded no such sound, and when its sealed 

 end was broken under water, its vacuum proved defec- 

 tive. Of the six cucumber-flasks, two became turbid, 

 the remaining four being perfectly clear. On breaking 

 their sealed ends under water, one-third of one of the 

 turbid flasks and one-fourth of the other remained un- 

 filled by the liquid. 



On the 6th of December eighteen retort-flasks were 

 charged with cucumber-infusion. They were boiled for 

 the usual time, that is three minutes, extreme care 

 being taken to seal them during the issue of the steam. 

 The water-hammer sound in all these flasks was parti- 

 cularly sharp and clear. Exposed to a temperature of 

 90° Fahr. for many weeks, seventeen of them remained 

 perfectly pellucid ; while the same infusion in a sealed 

 bulb with filtered air above it, and dissolved in it, 

 swarmed with life after boiling for sixty times the 

 interval here found efifectual. 



On the day subsequent to its preparation, one of 

 these well-exhausted flasks was unexpectedly found tur- 

 bid and covered with scum. But on examining the 

 sealed end it was found snipped off. The laboratory air 

 had thus entered the flask and given birth to the ob- 

 served organisms. Failures of this sort have a demon- 

 strative force greater even than successes ; they render so 

 obviously plain the external source of the contamination. 



It is worth saying here that the observation just 

 recorded was of frequent occurrence. The fineness of 

 .the sealed points of our retort-flasks renders them very 

 liable to be snipped off if they are not handled with 

 care. After preparation they are usually suspended on 

 a wire or on a wooden support ; and in frequent in- 

 stances, after such suspension, I have found a flask 

 11 



