364  The  Flash-Point  in  Mineral  Oils.  {Am^;^m' 
below  the  flash-point  (Abel).  The  experiments  also  prove  that? 
except  for  very  small  surfaces  of  oil,  the  flash-point  and  the  fire- 
point  are  the  same. 
The  following  happened  in  my  experience  : 
A  large  tank  of  very  high  flashing  oil  was  being  pumped  into,, 
and  the  oil  therefore  in  considerable  commotion.  The  temperature 
was  far  below  the  flash-point  in  Abel  cup ;  nevertheless,  vapors 
were  evolved  and  filled  the  top  of  the  tank,  and  expelled  by  the 
rising  oil,  overflowed  out  at  a  manhole  door  on  the  top,  which  was 
not  quite  close,  and  ignited  at  a  lamp  some  distance  below.  The 
fire  ran  up  the  stream  of  vapor  ;  there  was  an  explosion,  blowing  off 
the  top  of  the  tank,  and  the  oil  caught  fire  and  burned  uncon- 
trollably until  it  was  practically  all  consumed.  For  danger  in  oil 
works,  even  the  Abel  flash-point  is  deceptively  high.  Is  it  because 
this  has  not  been  realized  that  fires  in  oil  works  have  been  so  frequent  ? 
In  regard  to  danger  in  a  lamp,  in  1872,  before  a  Select  Parlia- 
mentary Committee,  a  chemical  expert  said  :  "  We  have  made  a 
great  number  of  experiments  to  ascertain  whether  oils  which  flash 
at  ioo°  (equivalent  to  73°  Abel  test)  or  even  a  little  below  100°, 
can  by  any  contrivance  be  exploded  in  a  lamp,  and  we  cannot  do 
it:  whether  by. electric  spark  or  flame  of  any  kind,  but  we  cannot 
fire  it ;  that  we  have  ascertained  to  be  a  fact."  Now,  the  real  fact 
is  that  a  lamp  filled  with  oil  of  100°  old  open  test,  if  shaken  up — 
as  by  carrying  the  lamp — can  be  exploded  with  electric  spark  quite 
easily,  even  at  73 °.  The  explosion  is  not  violent;  but  at  50  above 
the  flash-point  (Abel)  the  explosion  may  be  very  violent.  Such  an 
oil  can  explode  violently  at  any  temperature  between  780  and  120°  ; 
that  is,  at  all  temperatures  lamps  are  generally  exposed  to.  For 
lamps  burning  heat  up  the  oil,  more  with  large  lamps  than  with 
small,  more  with  metal  lamps  than  with  porcelain,  more  if  with  metal 
safety  tubes  than  without,  and  more  with  flat  burners  than  with 
central  draught.  As  all  ordinary  lamps  are  more  than  two  inches 
wide,  the  flash-point  in  them  will  be  the  Abel  flash-point,  or  a  little 
lower.  There  is  a  little  danger  even  at  the  Abel  flash-point,  for 
although  the  explosion  at  that  temperature  is  a  mere  puff,  yet  if  it 
happened  while  the  lamp  was  being  carried  it  might  cause  it  to  be 
thrown  down  on  combustible  material.  At  any  rate,  at  50  above 
the  Abel  flash-point  the  danger  is  very  great. 
I  have  tried  many  experiments  oversetting  cheap  gas  lamps  when 
