52 
brownish  in  colour  (cp.  Table  23).  Just  as  occurs  when  dissolved 
haemoglobin  is  added  to  urine,  a  precipitate  brown  in  colour  may 
make  its  appearance  at  the  end  of  twenty-four  hours  in  the  mixture 
of  urine  and  red  cells.  It  is  thus  seen  that  in  the  red  cell  the 
haemoglobin  is  broken  up,  by  urine  of  specific  gravity  exceeding 
roog,  in  situ,  and  does  not  leave  the  cell  until  it  has  become 
completely  converted  into  brown  pigment. 
In  order  that  some  idea  may  be  obtained  of  the  degree  to  which 
haemoglobin  is  likely  to  be  broken  up  in  the  urine  in  cases  of  black- 
water  fever,  and,  further,  in  order  to  determine  to  what  extent  small 
quantities  of  haemoglobin  passing  into  the  urine  at  the  time  of  its 
secretion  by  the  kidneys  may  be  recognisable  in  the  urine  when 
voided,  the  experiments  given  in  Table  25  were  carried  out.  These 
fall  into  two  series,  conducted  simultaneously  with  the  same  urine 
and  with  haemoglobin  derived  from  a  single  source. 
Method.  In  the  first  series  of  experiments  human  blood  cells 
were  added  to  the  urine  in  such  amounts  that  an  emulsion  containing 
from  5  per  cent,  to  0’33  per  cent,  of  red  blood  cells  was  obtained.  At 
the  end  of  four  hours  at  a  temperature  of  3;°  C.,  during  which  period 
the  red  blood  cells  were  distributed  in  the  urine  by  stirring  with  a 
glass  rod  every  fifteen  minutes,  the  mixture  was  centrifugalised,  the 
supernatant  fluid  poured  off  and  examined  spectroscopically,  while  the 
red  cells  were  mixed  with  a  measured  quantity  of  distilled  water  and, 
after  laking,  the  amount  of  haemoglobin  still  remaining  determined 
by  the  haemoglobinometer.  In  the  second  series  the  same  amounts 
of  haemoglobin  were  used  as  in  the  first,  but  the  haemoglobin  instead 
of  being  contained  in  red  cells  was  in  solution.  In  this  second  series, 
owing  to  the  formation,  especially  in  the  weaker  dilutions,  of  the 
brownish  pigment  already  referred  to,  accurate  haemoglobinometer 
readings  of  the  amounts  of  haemoglobin  in  the  urine  could  not  be 
made,  and  instead  the  oxyhaemoglobin  bands  present  were  matched 
by  the  aid  of  a  comparison  spectroscope  with  those  of  a  haemoglobin 
solution  of  known  concentration,  and  in  this  way  the  concentration  of 
haemoglobin  in  the  urine  was  determined. 
On  comparing  the  two  series,  it  is  seen  that  the  degree  of 
destruction  of  haemoglobin  proceeds  in  both  very  nearly  at  the  same 
rate,  being  somewhat  more  rapid  when  the  haemoglobin  is  in  solution 
than  when  it  is  contained  in  the  red  blood  cells,  but  the  difference  is 
