312 DR. AVTLLIAM ROBERTS. 



exposing them to the unfiltered air, decomposition would in- 

 fallibly follow. The filtration experiments supply a new 

 and telling argument on this point. Some of the liquids 

 become decomposed and full of bacteria while the filtration 

 was going ou, but the part which came over into the flasks 

 remained without further change, showing that decompo- 

 sition cannot go on without the actual contact of the living 

 organisms. 



We have next to ask ourselves what are the sources and 

 what is the nature of the fecundating iufluence which causes 

 organic liquids, when abandoned to themselves without pro- 

 tection, to become peopled Avith organisms. In regard to 

 their source, the answer is not doubtful. If I remove the 

 covering of cotton-wool from any of these preparations, and 

 admit unfiltered air, or a few drops of any ordinary water, 

 however pure, or anything that has been in contact with air 

 or water, organisms make their appearance infallibly in a 

 few hours. As to the nature of the infective agents, we can 

 say positively that they must consist of solid particles, other- 

 wise they could not be separated by filtration through cotton- 

 wool and porous earthenware. Is it not a most natural 

 inference that they are the parent germs of the brood which 

 springs up at their impact ? They are, however, so minute 

 that we cannot identify them as such under the microscope; 

 but Professor Tyndall has demonstrated that air which is 

 optically pure — that is, air which is free from particles — has 

 no fecundating power. 



It is contended in some quarters that these particles are 

 not living germs of any sort, but simply particles of albumi- 

 noid matter in a state of change which, when they fall into 

 an organic liquid, communicate to it their own molecular 

 movement, like particles of a soluble ferment, and so produce 

 decomposition, which, in its turn, provides the conditions 

 necessary for the abiogenic generation of bacteria. Filtration 

 through porous earthenware furnishes a complete answer to 

 this theory ; for I found on trial that the soluble ferments 

 passed with ease through the porous earthenware. If, there- 

 fore, this theory were true, the filtered liquids, if already 

 commencing to be decouiposed, would go on decomposing, 

 and would develope bacteria after filtration ; but instead of 

 that they remain unchanged and barren. We are absolutely 

 driven to the conclusion that these particles are living germs; 

 no other hypothesis squares in the least degree with the facts 

 of the case. 



We may formulate this conclusion in a third proposition 

 as follows : The organisms loliich appear as if spontaneously 



