434 
ON  THE  SUBLIMATION  OF  THE  ALKALOIDS. 
should  be,  if  we  would  obtain  satisfactory  results.  There  is  also 
some  risk  in  this  sudden  mode  of  applying  heat,  of  causing  the 
glass,  which  should  receive  the  sublimate,  to  break. 
The  method  of  procedure  to  which  I  should  give  the  prefer- 
ence is  the  following : — Provide  small  crucible  covers  or  slabs, 
or  fragments  of  white  porcelain,  a  few  microscopic  celhglasses, 
with  a  thickness  of  about  one-eighth  of  an  inch,  and  a  diameter 
of  circle  of  about  two-thirds  of  an  inch,  and  disks  of  window- 
glass  about  the  size  of  a  shilling.  Place  the  porcelain  slab  on 
the  ring  of  a  retort-holder  or  other  convenient  support,  then  the 
glass  cell,  and  upon  the  porcelain,  in  the  centre  of  the  cell,  a 
minute  portion  of  the  alkaloid  or  other  white  powder,  or  crystal 
reduced  to  powder.  Then  pass  the  clean  glass  disk  through  the 
flame  of  the  spirit-lamp  till  the  moisture  is  driven  off,  and  adjust 
it  with  the  forceps  over  the  glass  ring.  Now  apply  the  flame  of 
the  spirit-lamp  to  the  porcelain,  underneath  the  powder  or  crys- 
tal, aiid  continue  the  heat  till  the  powder  undergoes  its  character- 
istic change  and  gives  off  vapor.  Watch  the  deposit  of  this 
vapor  on  the  glass  disk,  and  remove  the  spirit-lamp  either 
directly  or  after  a  short  interval,  as  experience  may  determine. 
These  are  my  reasons  for  recommending  this  mode  of  pro- 
cedure in  preference  to  that  advocated  by  Br.  H  el  wig  : — By 
employing  a  flat  white  slab  of  porcelain,  the  heat  of  the  lamp  is 
applied  gradually,  and  every  change  of  consistence,  color  and 
position  which  the  powder  undergoes  is  easily  observed.  The 
ring  of  glass,  as  compared  with  a  ring  of  metal,  has  the  advan- 
tage of  conducting  the  heat  from  the  surface  of  the  porcelain  to 
the  glass  disk  so  slowly  as  to  guard  effectually  against  the  dan- 
ger of  breaking,  and  if  the  powder,  after  melting,  changes  its 
place,  the  glass  ring,  with  the  disk  upon  it,  is  easily  shifted. 
The  disk  of  window-glass  is  very  convenient  both  for  the  experi- 
ment itself  and  for  the  subsequent  application  of  liquid  tests- 
It  will  also  bear  a  moderate  heat,  if  required.  The  disks,  how- 
ever, are  not  essential ;  their  chief  recommendation  is  the  facil- 
ity they  afford  of  multiplying  experiments.  The  common  glass 
slide,  or  a  slip  of  window-glass,  (as  being  less  liable  to  scratch, 
and  bearing  heat  better  than  plate-glass,)  may  be  substituted 
