AMA?gTOi,  mA3RM  }  Laws  which  Regulate  the  Distribution,  Etc.  355 
volume  of  the  other,  the  amount  of  substance  being  determined  in 
each  of  the  superposed  liquids  when  they  had  become  saturated. 
It  is  found  that  when  a  substance  is  simultaneously  in  presence  of 
two  solvents,  the  quantities  dissolved  by  equal  volumes  of  the  two 
liquids  have  a  constant  ratio,  which  is  called  the  co-efficient  of  distri- 
bution, and  is  independent  of  the  relative  volumes  of  the  two  solvents, 
but  varies  with  the  degree  of  concentration,  and  with  the  tempera- 
ture. In  the  case  of  succinic  acid,  a  decrease  of  temperature  causes 
a  diminution  of  the  co  efficient  of  distribution,  and  the  same  effect  is 
produced  by  dilution.  With  oxalic,  malic,  tartaric,  and  acetic  acids, 
on  the  contrary,  the  co-efficient  increases  with  the  dilution,  and  the 
same  with  ammonia.  The  co-efficient  for  iodine,  with  water  and  car- 
bon disulphide,  may  be  regarded  as  independent  of  the  degree  of 
concentration. 
The  fact  that  the  co-efficient  of  distribution  is  independent  of  the 
relative  volume  of  the  two  solvents,  may  be  readily  explained  in  the 
following  manner :  Imagine  the  superposed  liquids  to  be  saturated 
with  the  substance :  then  for  equilibrium  to  persist,  it  is  only  neces- 
sary that  there  should  be  equilibrium  at  the  surface  of  contact  of  the 
two  liquids,  and  this  would  be  undisturbed  by  the  addition  of  an  arbi- 
trary volume  of  the  same  liquid  saturated  to  the  same  degree,  to 
either  of  the  superposed  liquids.  From  the  consideration  of  the  in- 
fluence of  concentration,  it  is  evident  that  as  the  solutions  become 
more  dilute,  the  co-efficient  of  distribution  approaches  a  certain  limit, 
so  that  if  it  is  desired  to  remove  a  substance  from  a  solution  by  agi- 
tating it  with  another  liquid,  it  is  advisable  to  employ  the  latter  in 
successive  fractions  (ib.  [4],  xx,  422-425).  Moreover,  it  is  easy,  by 
successive  determinations  of  the  co-effiuients  of  distribution,  to  ascer- 
tain whether  the  substance  dissolved  is  homogeneous  or  a  mixture  (ib. 
[4],  xx,  425,  429  and  431).  As  there  is  a  limit  to  the  co-efficient  as 
"the  solutions  become  more  dilute,  there  will  likewise  be  one  as  they 
become  more  concentrated,  and  it  would  naturally  be  supposed  that 
this  limit  would  be  the  ratio  of  the  two  liquids  when  saturated  sepa- 
rately ;  but,  experimentally,  this  has  been  found  not  to  be  the  case, 
the  co-efficient  being  less  than  that  corresponding  to  the  ratio  of  the 
solubilities.  With  respect  to  the  relation  between  the  co-efficient  of 
distribution  and  the  chemical  composition  of  the  substance  dissolved, 
the  authors  find  that?  ether  removes  more  readily  from  their  aqueous 
solutions :  1.  The  more  highly  carburetted  of  two  homologous  acids ; 
