342 FRAGMENTS OF SCIENCE 



vegetable. So long as the liquid contains living bacteria 

 a speck of it communicated either to the clear mineral 

 solution, or to the clear turnip infusion, produces in 

 twenty-four hours the effect here described. 



We now vary the experiment thus: Opening the back- 

 door of another closed chamber which has contained for 

 months the pure mineral solution and the pure turnip in- 

 fusion side by side, I drop into each of them a small pinch 

 of laboratory dust. The effect here is tardier than when 

 the speck of putrid liquid was employed. In three days, 

 however, after its infection with the dust, the turnip in- 

 fusion is muddy, and swarming as before with bacteria. 

 But what about the mineral solution which, in our first 

 experiment, behaved in a manner indistinguishable from 

 the turnip-juice? At the end of three days there is not 

 a bacterium to be found in it. At the end of three weeks 

 it is equally innocent of bacterial life. We may repeat 

 the experiment with the solution and the infusion a hun- 

 dred times with the same invariable result. Always in 

 the case of the latter the sowing of the atmospheric dust 

 yields a crop of bacteria never in the former does the 

 dry germinal matter kindle into active life. 1 What is the 

 inference which the reflecting mind must draw from this 

 experiment? Is it not as clear as day that while both 

 liquids are able to feed the bacteria and to enable them 

 to increase and multiply, after they have been once fully 

 developed, only one of the liquids is able to develop into 

 active bacteria the germinal dust of the air? 



I invite my friend to reflect upon this conclusion; he 



1 This is the deportment of the mineral solution as described by others. 

 My own experiments would lead me to say that the development of the bac- 

 teria, though exceedingly slow and difficult, is not impossible. 



