453 
FACTS  AND  THEORIES  OP  FERMENTATION. 
dergo  these  changes  if  common  air  was  substituted.  Thus,  meat, 
broth,  wort,  etc.,  were  preserved  for  weeks  in  flasks,  in  which 
they  were  boiled,  a  constant  current  of  filtered  air  being  drawn 
through  the  flasks.  No  change  of  any  kind  was  perceptible, 
even  in  summer  weather.  When  milk  was  tried  in  the  same 
manner,  however,  it  became  sour  nearly  as  soon  as  in  the 
open  air,  thus  indicating  an  essential  difference  in  the  principles 
involved  in  the  respective  decompositions."  The  author  has 
himself  repeated  the  experiment  of  preserving  boiled  meat  and 
water  in  a  flask,  having  an  aperture  of  at  least  one  inch  diame- 
ter, closed  merely  with  a  plug  of  raw  cotton,  part  of  the  cotton 
being  formed  into  a  ball,  surrounding  the  neck  of  the  flask,  and 
confined  with  a  thread,  to  prevent  the  passage  of  air  between 
the  sides  of  the  aperture  and  the  plug  of  cotton.  Meat  broth, 
thus  prepared,  was  found  to  be  perfectly  sweet  and  unchanged  in 
every  respect,  after  the  lapse  of  six  weeks,  in  the  months  of  June 
and  July ;  a  portion  of  the  same  broth  placed  in  a  bottle  with 
a  glass  stopper,  became  so  offensive  on  the  third  day  as  to  require 
its  removal. 
These  results  above  mentioned,  appearing  to  establish  the 
theory,  that  all  fermentations,  etc.,  are  induced  by  the  presence 
in  the  air  of  the  germs  of  organic  life,  led  the  author  to  make 
the  following  experiments,  with  the  two-fold  purpose  ;  1st,  of  de- 
ciding whether  this  property  possessed  by  cotton  was  peculiar 
to  it  alone  and  due  to  its  structural  arrangement,  or  whether  it 
was  common  to  it  and  to  all  other  finely  divided  substances  ;  and 
2d,  the  hope  of  detecting  in  the  air  these  invisible  germs,  or  at 
least  of  obtaining  satisfactory  proof  of  their  existence.  The  ap- 
paratus made  use  of  was  essentially  that  of  Schroder  and  Yon 
Dus(  h,  with  merely  such  alterations  as  the  purposes  in  view  re- 
quired. It  consisted  of  the  flask  a,  of  about  one  quart  capacity, 
in  which  was  placed  the  liquid  experimented  on,  closed  tightly 
with  a  cork,  through  which  passed  two  glass  tubes,  one  connect- 
ing with  the  five  gallon  tin  cannister  b,  the  communication  with 
which  could  be  intercepted  at  pleasure  by  a  stop-cock,  the  other 
leading  to  the  lower  end  of  the  filter  tube  c,  a  glass  tube, 
inches  in  diameter  and  18  inches  long,  closed  at  both  ends  by 
corks ;  a  diaphragm  of  fine  copper  wire  gauze  was  placed  a  little 
