Am.  Jour.  Pharm. 
July,  1903. 
Alkaloidal  Color  Tests. 
333 
designed  to  separate  are  wholly  insoluble  in  the  different  solvents 
employed  ;  and  unless  a  body  is  present  in  considerable  quantities,  it 
might  be  removed  completely  from  the  aqueous  solution  before 
the  best  conditions  for  its  recovery  were  reached.  For  example, 
morphine  and  its  sulphate  are  not  absolutely  insoluble  in  the  seven 
solvents  employed,  to  about  exhaustion,  before  treating  the  mixture 
with  the  solvent  in  which  the  alkaloid  is  most  soluble,  viz.,  amylic 
alcohol.  It  may  be  possible  that  sufficient  morphine  is  present  to 
carry  it  through  these  extractions,  yet  the  chances  are  decidedly 
unfavorable.  If  in  this  connection  we  consider  the  influence  of  one 
body  on  the  solubility  of  another,  it  can  readily  be  seen  how  incom- 
plete the  separations  must  be  even  with  this  excellent  method. 
That  such  influences  exist  in  the  organic  world  can  be  shown  by 
numerous  instances.  Take,  for  example,  a  simple  solution  contain- 
ing morphine  and  hydrastine  sulphate,  render  alkaline  and  extract 
with  ether,  evaporate  the  ethereal  solution  to  dryness,  and  it  will  be 
found,  contrary  to  expectations,  that  sufficient  of  the  morphine  is 
removed  with  the  hydrastine  so  that  the  color  reactions  of  the 
residue  are  quite  different  from  those  obtained  for  pure  hydrastine. 
Consequently,  it  seems  to  the  writer  that  the  toxicologist  should 
never  employ  this  method,  in  all  of  its  details  at  least,  unless  there 
is  absolutely  no  clue  whatever  as  to  the  general  nature  of  the  poison 
from  either  symptoms  or  circumstantial  evidence,  which  very  rarely 
is  the  case. 
T.  Graham1  introduced  a  method  in  which  he  utilized  the  remark- 
able property  that  moist  organic  membranes  possess,  of  allowing 
crystallizable  bodies  in  solution  to  pass  through  them,  whereas,  the 
non-crystallizable  mostly  fail  to  pass.  Graham  and  Hofmann2 
devised  a  process,  supposed  to  be  especially  adapted  for  the  isolation 
of  strychnine  from  certain  organic  substances  like  beer.  In  this 
method  advantage  is  taken  of  a  property  charcoal  possesses  of 
absorbing  strychnine  from  an  aqueous  solution  and  giving  it  up 
again  to  boiling  alcohol.  These  methods  are  seldom  used  at  present, 
because  small  quantities  of  strychnine  may  escape  detection,  or  will 
be  so  contaminated  with  foreign  matter  as  to  require  numerous 
extractions  with  ether  or  chloroform,  to  render  the  alkaloidal 
material  pure,  so  that  it  would  be  far  more  expedient  to  begin  the 
1  1862,  Jour.  Chem.  Soc,  15,  216. 
2  l%52»  Quart.  Jour.  Chem.  Soc.,  5,  173- 
