^45 
granules  towards  llie  distal  tubules,  both  in  blackwater  fever  and  in 
haernoglobinuria  due  to  potassium  chloi'aLc  and  other  haemolylic 
agents.  In  these  cases  it  may  be  questioned  if  the  small  and  large 
granules  in  the  proximal  and  distal  portions  of  the  renal  tubules 
respectively  are  of  contemporary  origin.  If  the  earlier  granules, 
which  as  secretion  of  urine  proceeds  pass  along  the  uriniferous 
tubules  towards  the  collecting  tubules,  are  large,  and  those  formed 
later,  when  haernoglobinuria  is  passing  off,  are  small,  then  the  latter, 
by  remaining  in  the  upper  portions  of  the  tubules,  might  suggest  an 
apparent  increase  in  size  of  the  granules  as  they  descended  towards 
the  renal  pelvis. 
The  presence  of  epithelial  nuclei  in  the  granular  casts  contained 
in  the  renal  tubules,  and  later  voided  in  the  urine,  supports  the 
hypothesis  that  haemoglobin  is  eliminated  by  the  tubular  epithelium, 
which  would  appear  sometimes  to  suffer  a  certain  amount  of 
degeneration  in  the  process. 
In  Case  ya  (Figs.  29  and  32)  the  individual  granules  were  of 
exceptionally  large  size.  Granules  nearly  as  large,  but  much  less 
densely  packed  together,  were  sometimes  met  with  in  rabbits  during 
experimental  haeinoglobinuria  (Figs.  22  and  33).  Obviously  sup¬ 
pression  would  be  most  likely  to  occur  when  the  granules  were  of 
large  size  and  numerous,  especially  if  at  the  same  time  the  amount 
of  water  in  the  urine  were  diminished  in  amount  as  this  fluid  passed 
along  the  uriniferous  tubules. 
3.  The  occurrence  of  a  primary  haemoglobinaemia  would  equally 
well  account  for  the  relapses  occurring  in  blackwater  fever  as  for  the 
original  attack  (cp.  Case  10,  Table  50).  In  reference  to  this,  it  may 
be  observed  that  the  relapses  of  short  duration,  such  as  occurred  in 
Case  10,  are  paralleled  by  the  haeinoglobinurias  of  short  duration 
induced  experimentally  (cp.  E.xp.  10,  1  able  35,  fig.  9.) 
4.  On  the  other  hand  the  presence  of  red  blood  cells  in  the  urine 
m  blackwater  fever  does  not  stand  in  any  direct  relation  to  haeino- 
globinaemia.  Putting  aside  Case  2,  which  will  be  referred  to 
separately  later,  the  appearance  of  red  cells  appears  to  be  due  to  the 
separation  from  the  basement  membrane  of  the  epithelial  cells  of  the 
uriniferous  tubules  which  is  observable  in  cases  of  suppression  and 
which  is  indicated  by  the  presence  of  epithelial  casts  in  the  urine  in 
K 
