CALCINATION  OF  MAGNESIA,  ETC.  523 
cept  by  preparing  very  small  quantities  at  a  time,  and  by  stir- 
ring the  carbonate  with  an  iron  spoon  during  calcination,  as 
advised  in  the  Codex.  These  plans  cannot  be  adopted  in 
wholesale  manufacture,  and  there  are  now  large  special  factories 
for  the  calcination  of  magnesia.  A  long  time  is  necessarily 
taken  up  in  calcining  large  quantities  of  carbonate,  because  the 
heat  penetrates  with  difficulty  to  the  centre  of  the  mass,  the 
periphery  of  which  during  the  operation  undergoes  the  action 
of  a  very  high  temperature  ;  the  heat,  in  fact,  must  be  very 
great,  for,  in  order  to  make  as  much  use  of  the  crucible  as  pos- 
sible, the  carbonate  is  heaped  up,  and  thus  forms  an  obstacle  to 
the  escape  of  «the  gases.  With  the  help  of  my  brother,  M.  L. 
Vee,  civil  engineer,  I  have  succeeded  in  overcoming  these  dif- 
ficulties, by  constructing  a  furnace  wherein  the  carbonate  of 
magnesia,  spread  in  a  layer,  not  thicker  than  seven  centimetres, 
is  submitted  to  the  action  of  the  degree  of  heat  strictly  neces- 
sary to  decompose  it,  I  will  briefly  describe  the  principal  features 
of  the  apparatus. 
It  consists  of  a  furnace  and  two  retorts.  There  is  no  pecu- 
liarity in  the  construction  of  the  furnace.  The  retorts  are  of 
cast  iron,  in  shape  rectangular ;  their  height  is  only  10  centi- 
metres, their  breadth  60,  their  depth  70.  The  opening,  occu- 
pying the  whole  of  one  of  their  smaller  sides,  is  shut  by  means 
of  a  cast  iron  slab,  fitting  in  a  groove,  and  held  in  place  by 
collars.  The  retorts  are  placed  one  above  another,  opening  on 
one  of  the  sides  of  the  furnace,  which  are  perpendicular  to  that 
in  which  the  furnace-door  is  situated. 
The  direct  action  of  the  fire  is  on  the  lower  retort;  the  pro- 
duct of  the  combustion  pastes  round  the  back  of  it  to  reach  and 
heat  its  upper  surface,  and  the  lower  part  of  the  other  retort, 
round  which  it  also  passes  before  escaping  up  the  chimney.  The 
result  of  this  arrangement  is,  that  the  lower  retort  becomes  red 
hot,  and  the  bottom  of  the  upper  retort  somewhat  less  so.  The 
decomposition  begins  in  the  upper  retort,  and  is  finished  in  the 
lower. 
The  carbonate  must  not  be  directly  introduced  into  the  re- 
torts ;  fearing,  but,  as  experience  has  shown  me,  without  reason, 
to  let  the  iron  come  in  contact  with  the  magnesia,  I  decided 
upon  employing  earthen  cases,  notwithstanding  their  weight, 
