■72 
IMPROVEMENTS  IN  THE  MANUFACTURE  OF  POTASSIUM. 
They  adapt  to  the  connecting  pipe,  a  kind  of  elongated  and 
flattened  box,  open  at  the  two  ends,  one  of  which  terminates  in  a 
round  collar,  to  permit  of  its  being  applied  to  the  intermediary 
or  retort  pipe.  This  box  is  formed  of  sheet  iron  of  the  thickness 
of  four  millimetres.    Its  length  is  thirty  centimetres  in  the  clear. 
Amongst  the  other  causes  which  exert  an  influence  on  the  pro- 
gress of  the  operation,  and  which  have  escaped  the  attention  of 
chemists,  MM.  Mareska  and  Donny  call  attention  to  the  rela- 
tive quantities  of  charcoal  and  carbonate  of  potash  in  the  mixture 
from  whence  the  potassium  is  obtained. 
The  most  suitable  proportions,  and  which  occasion  no  incon- 
veniences during  the  process,  are  those  indicated  by  the  theory 
of  equivalents.  Before  proceeding  with  the  extraction  of  the 
metal,  it  is  of  importance  that  an  analysis  of  the  charcoal  of  the 
tartar  should  be  made  ;  and  if  this  is  found  not  to  contain  the  ex- 
act quantity  of  carbon  required  for  the  reduction  of  the  metal, 
white  cream  of  tartar  must  be  mixed  with  the  crude  tartar,  and 
the  mixture  varied  until  the  calcined  product  obtained,  furnishes 
the  required  proportions. 
The  calcination  of  the  bitartrate  should  be  made  in  vessels 
quite  impermeable,  and  in  which  the  cooling  may  be  effected 
without  contact  of  air,  in  order  to  avoid  the  formation  of  cyanide 
of  potassium,  which  is  also  another  source  of  loss. 
All  who  are  employed  in  the  manufacture  of  potassium,  know 
the  inconveniences  attendant  on  the  employment  of  clay  luting. 
M.  Dumas  has  decidedly  rejected  its  use. 
Nothwithstanding  the  heat  necessary  for  the  reduction  of  the 
metal  is  such  that,  spite  of  the  most  minute  precautions,  an  ope- 
ration is  scarcely  ever  completed  without  the  breakage  of  the  re- 
tort, MM.  Mareska  and  Donny  have  succeeded  in  avoiding  the 
inconveniences  arising  from  the  employment  of  clay  luting,  and 
preventing  the  destruction  of  the  retorts,  by  running  on  them, 
when  they  are  at  a  dull  red  heat,  a  quantity  of  borax,  sufficient 
to  coat  them  with  a  vitrifiable  varnish. 
By  following  out  in  practice  the  method  thus  described,  MM. 
Mareska  and  Donny  have  succeeded  in  obtaining  a  considerable 
increase  in  the  quantity  of  metal  produced,  and  have  obtained  from 
150  to  200  grammes  of  potassium  at  each- operation.  They  possess 
a  specimen  of  400  grammes  of  pure  metal,  in  pieces  of  50  to  60 
grammes  each  ;  this  specimen  was  the  result  of  three  operations. 
