Jio  Theories  of  Blood-Coagulation  \A^S™v\9n*rm' 
appears,  the  activity  of  which  is  very  strong  at  the  outset,  but  de- 
creases as  we  know,  very  rapidly,  so  as  to  become  quite  attenuated 
by  the  following  day.  On  the  next  day,  lipoidic  cytozyme,  and, 
several  minutes  afterwards,  oxalated  plasma  are  added  to  both  tubes. 
Then  the  tube  which  has  received  cytozyme  on  the  preceding  day 
does  not  clot  at  all,  or  does  so  only  very  slowly,  whereas  the  tube, 
to  which  cytozyme  has  just  been  added  for  the  first  time,  clots  al- 
most instantaneously. 
It  is  clear  that  such  an  experiment  makes  it  possible  to  ascer- 
tain whether  the  same  cytozyme,  endowed  with  the  same  binding 
properties  as  the  pure  lipoid,  exists  in  fresh  or  heated  platelets,  or 
in  tissue  juice,  ground  muscle  for  example.  Adequate  experiments 
show  that  serum  which  has  already  reacted  with  any  one  of  such 
materials  does  not  generate  any  more  thrombin  when  subsequently 
brought  into  contact  with  any  one  of  them.  For  example,  serum  to 
which  lipoidic  cytozyme  has  been  added  will  not  react  any  more 
either  with  the  same  lipoid,  with  platelets,  or  with  muscle  juice,  and 
conversely. 
Without  entering  into  details,  I  may  add  that  the  manner  in 
which  the  two  substances  unite  closely  resembles  the  mode  of 
union  of  toxins  and  antitoxins,  in  that  the  process  is  not  governed  by 
the  law  of  strict  and  constant  equivalents,  but  takes  place  in  vary- 
ing proportions,  thus  seeming  to  result  not  from  true  chemical  af- 
finities but  from  contact  affinity  or  molecular  adhesion.  But  another 
fact,  more  noteworthy  for  its  bearing  upon  the  underlying  mechan- 
ism of  coagulation  is  disclosed  by  the  determination  of  the  lapse  of 
time  required  for  the  union  of  both  substances. 
Serozyme  being  found  in  serum  may  be  assumed  to  exist  also 
in  the  oxalated  plasma  from  which  this  serum  has  been  derived. 
Now  if  cytozyme  and  serum  are  mixed,  thrombin  appears  very 
quickly;  it  is  only  a  question  of  some  seconds.  But — and  this  fact 
is  undoubtedly  remarkable — if  cytozyme  be  added,  not  to  serum 
yielded  by  coagulation  of  -recalcified  oxalated  plasma,  but  to  an  iden- 
tical oxalated  plasma  recalcified  just  before,  that  is,  at  a  moment 
when  this  plasma  is  still  perfectly  fluid,  the  appearance  of  thrombin 
is  greatly  delayed.  In  other  words,  serozyme  reacts  with  cytozyme 
quickly  when  present  in  serum,  slowly  when  present  in  plasma.  We 
thus  reach  the  conclusion  that  the  serozyme  does  not  exist  in  the 
same  condition  in  plasma  as  in  serum,  that  in  plasma  it  is  not 
