TO THE FLOATING-MATTER OF THE AIR. 



protected tea has remained clear, while the exposed tea 

 is turbid and covered with mould. Both the exposed 

 coffee and the protected coffee are, on the other hand, 

 turbid and covered with PenicUUum. 



The remarks already made with regard to the rabbit- 

 infusion apply here. The case is one, not for the hasty 

 admission of spontaneous generation, but for further 

 scrutiny. I examined the apparatus as it stood. Tlie 

 pipette used to introduce the coffee (and this one only 

 of the three employed in these experiments) rested 

 against the outer edge of the tube containing the infu- 

 sion. This had in part evaporated, had been in part 

 re-condensed, and had trickled down the pipette so as to 

 form a small drop at the point where pipette and tube 

 touched each other. The drop had virtually washed the 

 outer surface of the pipette, carrying with it, in part, 

 such matter as might have attached itself to that sur- 

 face. A portion of this washing-water reaching the 

 infusion was clearly the origin of the life observed. 

 The sure test, however, was the repetition of the experi- 

 ment under conditions which should exclude this source 

 of error. On the 27th of December accordingly two 

 tubes protected by lamp-glasses were prepared, two 

 other tubes of the infusion being exposed to the air. 

 The former remained clear for months, the latter in the 

 same number of days became turbid and covered with 

 Penicillium. 



§ 12. Infusions of Codfish, Turbot, Herring, and 

 Mullet. 

 With a view of causing these experiments on mote- 

 less and mote-laden air to run parallel with others made 

 with hermeticaUy-sealed tubes, to be described further 

 on, I added the fish named in the heading of this section 

 to the other substances examined. The muUet was in- 



