PREPARATION  OF  PYROGALLIC  ACID,  ETC. 
21 
Nothing  remains  in  the  retort,  or  only  a  residue  hardly  pon- 
derable. According  to  this  equation,  100  parts  of  dry  gallic 
acid  ought  to  give  74*1  of  pyrogallic  acid.  Now  the  processes 
in  actual  use  do  not  give  a  better  product  than  25  per  cent,  of 
the  gallic  acid.  They  are  therefore  very  far  from  satisfying 
theory. 
M.  Duuias,  in  his  lectures,  has  often  insisted  on  the  necessity 
of  weighing  bodies  which  are  being  operated  upon,  and  after- 
wards of  weighing  their  products  of  decomposition,  so  as  to  dis- 
cover the  discrepancies  which  exist  between  practice  and  theory. 
Now  the  separation  of  gallic  arid  into  pyrogallic  and  car- 
bonic acids  is  not  doubtful,  consequently  its  preparation  must 
be  defective. 
Yet  this  preparation  has  attracted  the  attention  of  eminent 
chemists.  In  1843,  M.  Stenhouse  indicated  the  process  gene- 
rally adopted  of  subliming  the  acid  in  cones  of  pasteboard. 
In  1847,  M.  Liebig  obtained  a  result  of  31  to  32  per  cent,  by 
mixing  gallic  acid  with  double  its  weight  of  pounded  glass  in  a 
glass  retort,  heated  by  an  oil  bath  and  by  effecting  the  sublima- 
tion in  a  current  of  carbonic  acid  gas. 
If  practice  conducts  to  results  so  far  from  those  which  theory 
indicates,  it  follows  that  substances  like  pyrogallic  acid,  orcine, 
and  all  analogous  compounds,  though  volatile  without  decompo- 
sition at  certain  temperatures,  decompose  at  these  same  tem- 
peratures when  they  are  exposed  during  a  long  time.  That 
which  renders  the  distillation  of  these  substances  impossible 
under  the  ordinary  pressure,  is  therefore  an  alfair  of  time  and 
mass  ;  and  a  stronger  reason  for  the  greater  destruction  of  the 
material  is  when  the  distillation  is  preceded  by  a  chemical  de- 
composition like  that  which  occurs  during  the  preparation  of 
pyrogallic  acid. 
Guided  by  these  considerations,  we  have  sought  to  separate 
gallic  acid  completely  into  pyrogallic  and  carbonic  acids,  by  sub- 
mitting it  to  the  action  of  bases  and  water  in  close  vessels,  as  one 
of  us  has  done  for  the  preparation  of  orcine  ;  the  reaction  worked 
well,  but  the  manipulations  necessary  to  separate  the  base  are 
too  complicated.  We  have  determined  that,  at  the  temperature 
of  392°  F.,  the  pyrogallic  acid  remains  combined  with  the  lime, 
