^-7 
It  will  be  seen  that  in  all  cases,  with  one  exception,  a  negative 
result  was  obtained.  In  the  cases  of  blackwater  fever  this  negative 
result  occurred,  not  only  when  the  patient’s  plasma  obtained  during 
haemoglobinuna  (twelve  observations  :  3,  7,  8,  14,  15,  16,  17,  20,  21,  22, 
28,  29)  was  used,  but  also  when  that  obtained  both  before  blackwater 
appeared  (three  observations;  i,  2,  6)  and  (with  one  exception) 
shortly  after  haemoglobinuria  had  disappeared  (nine  observations: 
4,  9,  10,  12,  13,  23,  25,  27,  30),  as  also  (with  one  exception)  some 
time  after  complete  recovery  had  taken  place  (six  observations . 
5,  II,  18,  19,  24,  26).  From  these  experiments  the  important 
conclusion  follows,  that  the  mechanism  of  production  of  blackwater 
stands  in  an  altogether  different  category  from  that  of  paroxysmal 
haemoglobinuria.  The  presence  of  haemolysin  or  of  defect  of 
antilysin  in  the  plasma,  which  is  present  in  the  latter  case,  is  absent 
in  blackwater  fever,  where,  therefore,  search  must  be  made  for  other 
factors. 
Table  30  shows  that,  as  might  be  expected  from  the  behaviour 
of  the  plasma  of  the  blackwater  fever  patients,  who  had,  in  all  the 
cases  we  examined,  been  the  subjects  of  malaria,  usually  having  had 
frequently  repeated  attacks,  a  negative  result  was  also  obtained  in 
all  but  one  of  the  cases  of  malaria  dealt  with  (nine  observations, 
18,  20,  21,  22,  23,  25,  26,  27,  28). 
In  three  cases  a  positive  result  was  obtained.  1  he  first  occurred 
in  a  patient  suffering  from  blackwater  fever,  the  plasma  being  taken 
four  days  after  the  cessation  of  haemoglobinuria  (18,  Table  29). 
When  the  plasma  mixed  with  the  patient’s  red  cells  was  placed,  after 
cooling  for  half  an  hour,  in  the  incubator,  slight  haemolysis  was 
visible  at  the  end  of  twenty  minutes,  and  complete  haemolysis  had 
occurred  at  the  end  of  an  hour  and  a  half.  When  an  examination  of 
the  same  patient’s  plasma  was  made  two  months  later,  when  he  was 
in  good  health,  though  still  troubled  with  occasional  malarial  attacks, 
a  negative  result  was  obtained  (19,  Table  29).  The  positive  result 
obtained  in  this  case  was  quite  exceptional,  and  does  not  affect  the 
general  conclusion  obtained  with  the  rest  of  the  blackwater  fever 
cases.  The  second  case  (25,  Table  29)  occurred  in  another  patient 
at  the  close  of  haemoglobinuria.  In  this  case  the  examination  was 
made  twenty-four  hours  after  collection  of  blood,  and  bacteria  were 
