Am*c&ur'i92farm '}      Theories  of  Blood-Coagulation  yn 
capable  of  reacting  at  once  with  cytozyme.  We  may  express  this 
fact  by  saying  that  plasma  contains  proserozyme  instead  of  active 
serozyme,  one  of  the  first  phenomena  of  the  whole  process  of  coagu- 
lation being  precisely  the  conversion  of  proserozyme,  unfit  until! 
transformed  to  unite  with  cytozyme,  into  serozyme  capable  of  this 
reaction. 
The  notion  that  in  original  plasma  or  in  circulating  blood, 
serozyme  does  not  exist  as  such,  that  is,  does  not  exhibit  affinities 
toward  cytozyme,  satisfactorily  explains  why  intravascular  injec- 
tions of  the  latter  substance  are,  as  we  found,  quite  harmless.  But 
the  blood  of  such  injected  animals  shows,  when  shed  within  about 
half  an  hour  after  the  injection,  a  strikingly  increased  tendency  to 
rapid  coagulation.  This  fact  as  we  have  pointed  out  may  probably 
be  available  for  therapeutic  purposes  in  cases  of  hemorrhage. 
Is  it  now  possible  to  investigate  under  what  influences  the  pro- 
serozyme is  converted  into  serozyme,  in  other  words,  under  what  in- 
fluences does  it  acquire  the  capacity  of  reacting  with  cytozyme? 
To  solve  this  problem,  we  have  at  our  disposal  a  very  adequate 
technic,  based  on  the  use  of  oxalated  salt-saturated  plasma. 
Some  minutes  ago  I  mentioned  the  fact  that  when  oxalated 
plasma  is  saturated  with  common  salt,  the  fibrinogen  is  entirely  pre- 
cipitated.  After  strong  centrifugalization,  pipetting  off  and  elimina- 
tion of  the  excess  of  salt  by  dialysis  in  presence  of  physiological 
oxalated  solution,  the  supernatant  fluid  represents  exactly  normal 
oxalated  plasma,  except  that,  having  lost  all  of  its  fibrinogen,  it  can- 
not coagulate.    Being  oxalated  it  does  not  contain  any  trace  of 
thrombin,  but  is  still  capable  of  producing  plenty  of  it  on  the  addi- 
tion of  calcium  salt  and  cytozyme.    Now  if  a  calcium  salt  and  cyto- 
zyme are  added  thrombin  appears  indeed,  but  only  after  a  rather 
important  delay.    Half  an  hour,  and  sometimes  more,  must  elapse 
before  the  mixture  becomes  capable  of  bringing  about  an  almost  in- 
stantaneous coagulation  of  a  fibrinogen  solution.    Thus  the  acquisi- 
tion of  the  ability  to  react  with  cytozyme,  that  is,  the  conversion  of 
proserozyme  into  serozyme,  requires  a  notable  length  of  time.  Now, 
on  the  other  hand,  if  the  aforesaid  fluid  without  fibrinogen  is  recal- 
cified  and  cytozyme  is  added  one  or  two  hours  later  the  thrombin 
appears  almost  instantaneously.    This  experiment  clearly  illustrates 
the  essential  assumption  that  the  whole  process  of  the  production  of 
thrombin,  the  first  stage  included,  which  is  the  conversion  of  pro- 
