150  Turpentine:  its  Nature  and  Adulterations.  {^^'mx^'^^^^' 
Method  of  Anali/sis.—The  terms  "petroleum  spirit  "and  "petroleum  oil " 
as  commercially  used  do  not  admit  of  very  precise  definition  ;  for  the  pur- 
pose of  this  paper,  I  would  therefore  define  petroleum  spirit  as  being  that 
portion  of  crude  petroleum  which  may  be  volatilized  by  means  of  steam 
from  water  boiling  at  atmospheric  pressure,  and  pretoleum  oil  as  being  the 
non-volatile  portion.  Judged  of  by  this  definition,  commercial  spirit  and 
oil  are,  as  a  rule,  more  or  less  mixtures ;  the  amount  of  spirit  in  the  best 
burning  oils  is,  however,  small. 
The  presence  of  petroleum  oil  in  turpentine  is  readily  detected  and  the 
amount  estimated  by  steam  distilling.  Unless  it  has  been  freely  exposed  to 
the  air  for  a  long  time,  but  a  mere  trace  of  viscid  matter  remains  on  steam- 
distilling  turpentine  ;  on  one  or  two  occasions  only  have  I  met  with  samples 
containing  a  small  amount  of  resin,  which  was  left  as  a  solid  on  distilling 
ofi"  the  turpentine  by  a  steam  current.  Should  more  than  a  few  tentlis  of  a 
per  cent,  of  non-volatile  matter  remain,  it  is  probable  that  petroleum  is  pre- 
sent. This  usually  betrays  itself  by  the  more  or  less  marked  blue  fiuorescence 
of  the  residue  ;  bnt  should  this  criterion  fail,  the  behavior  of  petroleum  and 
of  the  non-volatile  product  of  the  air  oxidation  of  turpentine  on  digestion 
with  dilute  nitric  acid  will  serve  to  differentiate  them.  The  latter  is  readily 
oxidized  and  dissolved;  the  former  does  not  alter  much  in  bulk,  but  appa- 
rently undergoes  more  or  less  complete  nitration.  I  have  never  yet  met 
with  a  sample  containing  resin  oil,  but  it  would  not  be  difficult  to  detect 
it,  as  it  is  oxidized  by  nitric  acid,  and  behaves  in  a  most  characteristic 
manner  when  triturated  with  a  paste  of  slaked  lime,  forming  the  well-  • 
known  grease. 
The  detection  and  estimation  of  petroleum  spirit  is  less  readily  eff'ected. 
The  method  which  I  employ  is  based  on  the  different  behavior  of  turpen- 
tine and  paraffins  with  sulphuric  acid.  The  paraffins,  it  is  well  known,  are 
almost  unaffected,  whereas  turpentine  is  polymerized  and  for  the  most  part 
converted  into  substances  of  high  boiling  point  which  do  not  volatilize  in 
a  current  of  steam.  I  say  for  the  most  part,  because,  as  I  have  elsewhere 
stated,  a  certain  amount  of  cymene  and  of  a  paraffinoid  hydrocarbon  is 
always  produced.  Inasmuch  as  the  amount  of  cymene  so  produced  varies 
with  the  strength  of  the  acid  and  the  temperature,  being  larger  the  more 
concentrated  the  acid  and  the  higher  the  temperature,  it  is  important 
always  to  work  under  uniform  conditions,  at  as  low  a  temperature  as  con- 
venient, and  to  use  diluted  acid.  I  employ  two  strengths  of  acid,  a  mix- 
ture of  2  vols,  acid  and  1  vol.  water  (2:1  acid)  and  a  mixture  of  4  vols,  acid 
and  1  vol.  water  (4  : 1  acid).  The  turpentine — 500  c.c.  is  a  convenient  quan- 
tity— is  placed  with  about  one-fourth  to  one-third  of  its  bulk  of  2 : 1  acid 
in  a  well-stopped  bottle,  and  the  mixture  is  somewhat  cautiously  agitated.. 
It  soon  becomes  more  or  less  heated,  and  as  it  is  important  to  effect  the 
polymerization  at  a  temperature  not  much  above  the  ordinary  atmospheric 
temperature,  the  bottle  is  placed  in  cold  water  for  a  short  time.  After 
repeated  agitation  with  the  acid,  the  turpentine  is  converted  into  a  viscid 
oil,  and  when  this  is  the  case,  and  no  more  heat  is  developed  on  continued 
agitation,  the  contents  of  the  bottle  is  transferred  to  a  separating  funnel, 
the  acid  layer  is  run  off  and  the  oil  poured  into  a  fiask  ;  the  latter  having- 
been  connected  with  a  condenser  and  a  steam-pot— an  ordinary  tin  cum 
