70 



b 



440 



THE BEGINNINGS OF LIFE. 



for four hours ^5 and it is almost needless to say that 

 every part of the sealed tubes and their contents was 

 exposed to this temperature during the whole time. 

 The glass tubes^ though of moderately thick glass only^ 

 ran no risk of fracture^ because the pressure inside them 

 was approximately counterbalanced by the pressure of 

 steam outside/ 



In all the subsequent experiments which I performed 

 alone^ an approximate vacuum was procured^ as in my 

 former experiments, by boiling the fluids and sealing 

 the flasks hermetically during ebullition. The vacuum 

 may have been somewhat less perfect in these cases 

 than when it was procured by means of the Sprengel 

 pump, though this circumstance does not in the least 

 diminish the value of the experiments. The vacuum 

 was not desired, because, by working under these con- 

 ditions, all atmospheric ^ germs ' might be abstracted 

 since in all cases the flasks were exposed to a 

 temperature which is acknowledged to be destructive 

 of living things whether in air or in fluids. In the 



^ This prolonged period of exposure was subsequently only resorted 

 to in some of the experiments. In others they were exposed, for shorter 

 periods, as will be seen from the different headings. 



'$ 



\l 



p 



(X^ 



fstd 



in^ 







■i]d 



// (13 



2 ' 



s< 



mint'i 



I 



'n(d 



•alt yif^' 



also expo 





f 



inint a 



Ai 



ine 



,0j alkali 



fibres of a ( 



taken from 

 •; assumed a pal( 

 jliflawarm plac 

 i sunlight. Th 

 tc partially pres 

 M two months 

 -reaction of the 



(differing j 



^'as sour, th 



^'^t consisting 

 ^klent 



experiments of Mantegazza, Wyman, and Cantoni, the '^^7 slightly tur 



portions of the closed flasks above the level of the 



fluids were filled with ordinary air. If, therefore, the 



vacuum may not have been quite so complete in some 



of my latter experiments as in those in which I had 



the benefit of Prof. Frankland's assistance, it is a matter 



depop: 



'5«ients u' 



re 



'ere ft 

 ^'^ilst the 



y^ lor "^ ^ 



^h 



tae 



Q'^'io 



Ur 



Tl 



% 





