45 
completely  precipitating  the  red  cells  by  centrifugalisation,  the  super¬ 
natant  liquid  was  removed  by  means  of  a  glass  tube  provided  with  a 
rubber  teat  and  drawn  out  to  a  capillary  end.  To  the  red  cells 
0'9  per  cent,  sodium  chloride  solution  was  added  in  such  amount  as 
to  produce  a  2-5  per  cent,  emulsion  of  the  red  cells.  A  sample  of 
this  emulsion  was  then  diluted  with  nineteen  parts  of  distilled  water 
and  a  haemoglobinometer  reading  made.  A  series  of  test  tubes  was 
next  prepared  to  which  the  respective  quantities  of  O’g  per  cent, 
sodium  chloride  solution  and  of  r92  per  cent,  quinine  bihydrochloride 
solution  given  in  Table  24  were  added.  The  requisite  amounts  of 
the  2' 5  per  cent,  emulsion  of  red  cells  were  added,  and  the  tubes 
placed  in  an  incubator  kept  at  37°  C.,  being  subsequently  stirred  up 
with  a  glass  rod  every  fifteen  minutes  until  the  conclusion  of  the 
experiment  at  the  end  of  three  hours. 
In  those  experiments  in  which  complete  haemolysis  did  not  occur 
the  degree  of  haemolysis  was  indicated  by  the  expressions  ‘  partial,' 
‘  slight,’  ‘  trace.’ 
In  carrying  out  these  e.xperiments  it  must  be  borne  in  mind  that 
the  emulsion  should  contain  a  definite  weight  of  red  blood  cells.  It 
cannot  be  made  up  with  a  definite  weight  of  blood,  for  such  a 
procedure  would  give  no  clue  as  to  its  real  composition  in  respect  of 
red  cells,  since  it  has  been  found  that  in  health  slight,  and  in  black- 
water  fever  and  malaria  considerable,  variations  in  the  relation  which 
the  volume  of  red  cells  bears  to  that  of  the  plasma  may  occur.  Nor 
is  it  of  any  practical  utility  for  our  purpose  to  determine  the  number 
of  red  cells  per  cubic  millimetre  of  blood,  since  this  also  gives  no 
measure  of  their  percentage  by  weight.  Nor  again  is  it  of  advantage 
to  know  the  red  cell  index  unless  the  percentage  by  volume  is  also 
known,  and  when  the  latter  has  been  determined  the  former  is  not 
required.  On  the  other  hand,  in  order  that  the  conditions  of  experi- 
jnent  may  be  more  fully  defined,  it  is  desirable  that  the  determination 
of  the  concentration  of  red  cells  by  weight  in  the  emulsion  employed 
should  be  supplemented  by  a  determination  of  the  percentage  of 
haemoglobin  given  by  the  reading  of  a  haemoglobinometer  stan¬ 
dardised  as  already  described.  When  the  two  determinations  are 
nearly  the  same,  no  difficulty  is  experienced  in  comparing  the  haemo¬ 
lysis  produced  by  quinine  with  that  obtaining  in  health.  When, 
however,  a  marked  variation  occurs,  as  in  Experiments  10,  14a  and 
