51 
observations  have  been  made.  In  the  meantime,  we  have  thought 
it  worth  while  to  refer  to  the  line  of  experiment  followed. 
(d )  The  action  of  urine  on  red  blood  cells  and  on  dissolved 
haernoglubin. 
In  studying  the  mechanism  of  blackwater  fever,  it  is  essential  to 
ascertain  how  far  the  condition  of  the  urine  when  voided  represents 
its  original  condition  when  issuing  From  the  collecting  tubules  of  the 
kidney,  and  how  far  it  has  been  changed  while  retained  in  the  bladder. 
This  enquiry  resolves  itself  into  a  two-fold  one  ;  into  the  action  of 
urine  upon  haemoglobin  contained  in  the  red  blood  cells ;  and  into 
the  action  of  urine  upon  laked  haemoglobin. 
When  a  solution  of  haemoglobin  is  added  to  urine  in  amount 
sufficient  to  produce  a  light  red  coloration,  it  is  found  that  the  red 
tint  soon  becomes  fainter,  the  urine  turning  browm  in  colour,  and  that 
ultimately  the  w'hole  of  the  haemoglobin  is  broken  up,  a  brown  soluble 
pigment  remaining,  w'hich  gives  no  absorption  bands  on  spectroscopic 
examination.  Sometimes  a  variable  amount  of  a  brow-n  precipitate 
appears  at  the  end  of  twenty-four  to  forty-eight  hours,  and  the  urine 
may  then  become  less  browm  in  colour. 
If  an  emulsion  of  red  blood  cells  in  o-g  per  cent,  solution  of  sodium 
chloride  is  added  to  urine,  it  is  seen  that  the  latter,  when  the  changes 
in  the  red  cells  described  below',  become  advanced,  gradually  turns 
brown  in  colour,  but  no  reddish  tint  is  noticeable.  Urine  does  not 
produce  visible  laking  of  red  blood  cells  unless  its  specific  gravity  is 
less  than  I'oog.  Urine  of  specific  gravity  r002  lakes  red  blood  cells 
as  rapidly  as  distilled  water.  Red  blood  cells  in  urine  whose  specific 
gravity  e.xceeds  roog  become,  on  standing,  poorer  in  haemoglobin 
and  assume  a  darker  somewhat  browmish  aspect,  but  are  ultimately 
decolourised  and  form  a  brownish-w'hite  precipitate  at  the  bottom  of 
the  tube  in  which  the  urine  is  contained.  If  such  urine  at  any  time 
after  adding  the  red  blood  cells  is  centrifugalised  and  the  supernatant 
fluid  examined  spectroscopically,  it  is  found  that  no  oxyhaemoglobin 
bands  are  at  first  seen,  the  red  blood  cells  becoming  decolourised 
without  any  laking  of  their  haemoglobin.  When,  however,  the  red 
blood  cells  have  lost  75  per  cent,  or  more  of  their  haemoglobin,  oxy¬ 
haemoglobin  bands  make  their  appearance  in  the  urine,  w'hich  is  now 
