AmDeca!im&cm'}  Separation  of  Resin  from  Fatty  Acids.  609 
which  do  not  admit  of  the  above-mentioned  methods  of  precipita- 
tion, e.  g.}  those  of  venomous  spiders,  a  more  complete  isolation  of 
the  poison  is  not  yet  practicable. 
If  the  second  experimental  animal  does  not  remain  in  health,  but 
is  taken  ill  like  the  first,  we  have  a  proof  that  the  poison  in  ques- 
tion does  not  belong  to  the  albumins — at  least  not  to  those  precipi- 
table  by  ordinary  means.  The  entire  filtrate  is  then  freed  from 
albumin  and  the  limpid  filtrate  is  examined  for  other  poisons  by 
DragendorfT's  method. —  Chemiker  Zeitung;  Chemical  News ,  Oct.  23, 
1891,  p.  206. 
THE  SEPARATION  OF  RESIN  FROM  FATTY  ACIDS. 
By  J.  Arthur  Wilson. 
The  detection  of  resin  and  its  separation  from  fats  and  oils  was 
at  one  time  a  very  difficult  problem  to  the  analyst,  and  in  many 
cases  was  impossible.  Amongst  many  processes  we  have  those  of 
Henderson,  Barfoed,  and  Gladding,  the  latter  of  which  has  found 
the  greatest  favor.  It  suffers  from  one  very  serious  defect,  how- 
ever, and  that  is  the  correction  for  solubility  of  the  fatty  acids 
and  silver  salts  in  ether.  This  is  a  very  variable  factor,  and,  of 
course,  is  dependent  on  the  nature  of  the  fatty  matter  used  in  the 
fabrication  of  the  soap.  With  small  quantities  of  resin  the  error 
thus  introduced  may  be  as  much  as  20-30  per  cent.,  but  of  course 
in  soaps  of  high  resin  content — say  20  per  cent. — the  error  dimin- 
ishes. 
Mr.  R.  Williams  {Analyst,  vol.  xv.,  p.  169)  obtained  very  good 
results  by  the  process;  thus,  in  a  soap  which  actually  contained 
30-0  per  cent,  of  resin,  he  obtained  307  per  cent.,  but  no  mention 
is  made  as  to  whether  the  amount  remaining  in  the  lyes,  after  pre- 
cipitation of  the  soap,  was  determined.  As  to  the  simple  detection 
of  resin,  no  doubt  the  best  test  is  to  boil  the  dry  fat  or  fatty  acids  (10 
drops)  with  5  cc.  acetic  anhydride,  cool,  and  add  2  drops  of  pure 
sulphuric  acid  (1-84  specific  gravity),  when  the  reddish  violet  color 
due  to  resin  is  easily  observed,  one  per  cent,  of  which  can  thus  be 
detected. 
It  seems  that  all  the  above  processes  must  make  way  for  the 
admirable  process  of  Twitchell  (Analyst,  Oct.,  1891),  who  found 
that  whereas  fatty  acids  are,    as  is  well-known,  converted  into 
eutral  ethers  on  passing  dry  hydrochloric  acid  gas  into  their  alco- 
