294  Commercial  Oil  of  Citronella.  {AnuJSK'Safm' 
In  order  to  determine  on  what  the  difference  in  color  of  this  last 
and  the  change  from  yellow  to  green  which  takes  place  depend, 
eight  samples  of  citronella  oil  were  obtained  from  various  source's, 
and  a  small  quantity  of  each  exposed  to  direct  sunlight.  Of  this 
number  five  (A,  B,  C,  F,  G)  were  decidedly  green  before  exposure,  two 
(D  and  E)  were  yellow  at  first,  but  rapidly  became  green,  whilst  one 
(H)  was  yellow  originally  and  underwent  no  change.  The  fact  that 
the  presence  of  copper  has  been  shown  (Guibourt  and  Histed)  to  be 
reason  of  the  green  color  of  commercial  cajeput  oil,  led  me  to 
suspect  the  same  contamination  in  the  case  of  this  oil.  (Since 
writing  this  note  my  attention  has  been  called  to  the  fact  that 
Kremers1  mentions  incidentally  the  presence  of  copper  in  a  sam- 
ple of  this  oil  which  he  examined.) 
250  cc.  of  the  sample  a  was  shaken  with  a  dilute  solution  of 
ferrocyanide  of  potassium,  when  a  rapid  separation  of  a  red  pre- 
cipitate took  place,  which,  after  washing  with  spirit  to  free  it  from 
traces  of  oil  and  then  with  water  to  remove  any  excess  of  potas- 
sium ferrocyanide,  was  proved  to  be  ferrocyanide  of  copper. 
Examination  was  then  made  of  all  the  samples,  with  the  following 
results  : 
Sp.  gr. 
at  150  C. 
Color. 
Remarks. 
A 
•896 
B 
•895 
C 
•890 
D 
•887 
E 
•896 
F 
•896 
Cx 
•897 
H 
•870 
emerald  green, 
greenish. 
yellow,  becoming  green. 
emerald  green, 
greenish . 
brownish-vellow. 
copper 
copper  ent 
present. 
rely  absent. 
From  the  fact  that  only  those  samples  which  were  green,  or 
became  so  on  exposure,  contained  copper,  it  appeared  almost  cer- 
tain that  the  change  in  color  might  be  due  directly  to  the  presence 
of  that  metal,  which  was  readily  proved  by  precipitating  all  the 
copper  from  the  most  markedly  green  sample,  by  treatment  two  or 
three  times  with  solution  of  potassium  ferrocyanide,  when  the  oil 
became  pale  yellow  in  color.    One  portion  of  this  oil  was  then 
1  "Proceedings  American  Pharmaceutical  Association,"  1887,  p.  562. 
