• 
524 
CALCINATION  OF  MAGNESIA,  ETC. 
their  fragility,  and  the  length  of  time  required  to  heat  them, 
when  it  occurred  to  me  to  heat  a  fragment  of  sheet  iron  with 
magnesia,  in  a  platinum  crucible.  I  found  that  when  not  left  too 
long  in  contact  or  at  too  high  a  temperature,  the  product  was 
white,  and  not  perceptibly  more  ferruginous  than  the  carbonate 
which  furnished  it.  So  that  I  had  sheet  iron  cases  made  in 
which  to  put  the  carbonate  to  be  introduced  into  the  retorts. 
The  use  of  sheet  iron  is,  however,  not  without  its  inconveniences  ; 
it  becomes  covered  with  a  coat  of  oxide,  which  generally  de- 
taches itself  in  lamels,  which  require  to  be  sifted  from  the  mag- 
nesia. Occasionally  this  oxide  is  pulverulent,  but  it  does  not 
mingle  with  the  mass  if  care  is  taken  to  leave  a  •small  layer  of 
magnesia  in  the  case  to  protect  the  rest  in  the  following  opera, 
tions  ;  the  heat  puts  the  cases  out  of  shape,  and  thin  sheet  iron 
must  be  used,  that  there  may  be  no  difficulty  in  removing  the 
cases  in  and  out  of  the  furnace.  These  defects  are  compensated 
by  the  lightness  of  the  case,  the  little  space  it  occupies,  and  the 
facility  with  which  it  is  heated  and  cooled. 
To  produce  complete  decomposition  the  carbonate  should  re- 
main 45  minutes  in  each  retort.  Thus,  every  three  quarters  of 
an  hour  a  case  of  magnesia  is  taken  from  the  lower  retort,  and 
immediately  replaced  by  a  case  of  partially  decomposed  carbon- 
ate from  the  upper  retort,  into  which  a  fresh  case  of  carbonate 
is  then  introduced.  The  operation  is  continuous  and  the  con- 
sumption of  fuel  small,  because  it  is  never  necessary  to  cool  the 
furnace. 
Prepared  in  this  way,  magnesia  is  as  light  as  is  the  hydrocar- 
bonate.  It  combines  easily  with  water,  and  dissolves  immediate- 
ly in  d  iluted  acids. 
In  the  course  of  my  analyses,  I  have  had  occasion  to  examine 
a  great  number  of  samples  of  magnesia,  derived  from  various 
sources,  and  I  have  never  yet  found  a  single  specimen  of  mag. 
nesia  or  carbonate  completely  free  from  lime,  neither  have  I 
found  one  containing  a  sufficient  quantity  to  oifer  any  inconve- 
nience. I  cannot  say  that  the  admixture  with  a  considerable 
quantity  of  lime  can  never  be  the  case  ;  but  I  think  it  would  be 
quite  exceptional,  perhaps  because  attention  has  been  called  to 
this  point  by  recent  publications.  In  this  analysis,  I  have  always 
adopted  the  following  plan,  which  I  will  briefly  describe;  for 
