264  Testing  Cocliineal.  {^j/neMm^"' 
into  pinother  flask  of  about  200  c.c,  and  the  measuring  vessel  rinsed 
with  a  definite  quantity  of  water,  say  10-15  c.c. 
A  weak  solution  of  permanganate  is  then  run  in  from  a  burette 
with  a  glass  cock,  the  flask  being  shaken  well  after  the  addition  of 
every  10  c.c. 
So  much  permanganate  solution  is  added  that  the  cochineal  extract 
shall  be  changed  from  its  original  color  to  a  pink  of  the  very  faintest 
shade,  almost  yellow,  in  fact,  but  never  reaching  a  full  yellow.  This 
pink  shade  should  be  persistent,  that  is,  it  should  not  turn  yellow 
after  standing  fifteen  minutes  ;  and  after  a  little  practice  it  will  be 
found  very  easy  to  obtain  the  tinge,  which  shows  that  the  coloring 
matter  is  almost  but  not  quite  destroyed. 
When  a  number  of  samples  are  to  be  compared  I  arrange  an  equal 
number  of  200  c.c.  flasks  and  test-tubes  on  the  table,  a  tube  standing 
in  its  rack  in  front  of  each  flask.  Then  the  number  of  c.c.  of 
the  permanganate  solution  (which  should  be  at  least  so  weak  that  bulk 
for  bulk  of  this  and  the  cochineal  solution  v/ill  be  required),  is  run 
into  each  flask,  taking  care  to  use  too  little  to  completely  destroy  the 
coloring  matter  in  all.  The  flasks  are  well  shaken  and  allowed  to 
stand  for  ten  minutes. 
"#Part  of  the  contents  of  eacli  is  then  poured  into  the  corresponding 
test-tube,  and  a  glance  at  the  tubes  as  they  stand  side  by  side  will 
show  which  is  the  least  affected  by  the  bleaching  liquid.  This  sample 
having  been  selected  to  serve  as  a  standard,  the  contents  of  the  test- 
tube  are  returned  to  this  flask,  and  more  permanganate  solution  is 
cautiously  jnided,  until  a  very  faint  pink  tinge,  which  a  fraction  of  a 
c.c.  will  turn  to  a  full  yellow,  is  obtained. 
The  number  of  c.c.  used  having  been  noted,  a  fresh  trial  is  made, 
in  which  the  c.c.  required,  minus  one,  are  used,  the  flask  agitated,  and 
the  last  c.c.  or  part  of  it,  as  the  whole  may  not  be  necessary,  added. 
If  the  two  results  agree,  the  next  sample  is  treated  in  the  same  way, 
and  so  on  until  all  are  tested. 
I  usually  make  a  final  trial  by  measuring  the  50  c.c.  of  each  solu- 
tion into  its  flask,  running  in  the  permanganate  in  the  ascertained 
amount  into  each  as  quickly  as  possible,  letting  the  flask  stand  10 
minutes,  and  tlien  making  a  comparison  of  all  in  the  test-tubes. 
If  the  shades  are  not  exactly  alike,  a  pretty  good  guess  can  gene- 
rally be  made  of  the  fractions  of  c.c.  required,  which  should  be  added, 
the  contents  of  the  tubes  being  joined  to  that  in  the  flasks,  and  a 
second  or  third  comparison  thus  made. 
