554       DIFFERENCES  OF  ACTION  BETWEEN  POTASSA  AND  SODA. 
In  a  series  of  operations  made  with  the  view  of  knowing  what 
would  be  the  proportions  of  potassa  and  soda  capable  of  producing 
most  oxalic  acid,  I  observed: — 
1.  That  a  mixture  of  1  part  of  hydrate  of  soda  with  3  parts  of 
hydrate  of  potassa  is  capable  of  decomposing  one-tenth  more  fecula 
than  if  pure  potassa  were  employed,  and  that  the  quantity  of  oxalic 
acid  produced  is  increased  in  proportion  to  the  fecula  employed. 
2.  That  a  mixture  of  1  part  of  hydrate  of  soda  with  2  parts  of 
hydrate  of  potassa  is  capable  of  decomposing  one-eighth  more  fecula 
than  if  potassa  alone  were  employed,  and  that  the  quantity  of  oxalic 
acid  produced  is  increased  in  proportion  to  the  fecula  employed. 
3.  That  a  mixture  of  1  part  of  hydrate  of  soda  with  1  part  of 
hydrate  of  potassa  acts  nearly  like  pure  potassa. 
4.  That  a  mixture  of  2  parts  of  hydrate  of  soda  with  1  part  of 
hydrate  of  potassa  produces  one-tenth  less  than  pure  potassa. 
5.  That  a  mixture  of  3  parts  of  hydrate  of  soda  and  1  part  of 
hydrate  of  potassa  produces  one-fifth  less  than  pure  potassa. 
6.  That  below  this  proportion  of  soda  (No.  5)  the  production  of 
oxalic  acid  becomes  very  small  and  impossible  as  a  manufacture. 
Pure  soda  cannot  then  be  employed  as  a  substitute  for  potassa  in 
the  production  of  oxalic  acid,  as  former  works  on  this  subject  might 
lead  us  to  suppose ;  but  it  may  be  usefully  employed  mixed  with 
potassa. 
In  the  production  of  the  cyanides  by  means  of  animal  matters,  if 
we  endeavor  to  replace  potassa  by  soda,  we  find  that  pure  soda, 
whether  caustic  or  carbonated,  produces  much  less  cyanide  than 
potassa,  and  that  additions  of  soda  to  the  potassa  do  not  by  any 
means  increase  the  production  of  the  cyanides ;  on  the  contrary,  in 
proportion  as  the  quantity  of  soda  is  augmented,  the  production  of 
cyanide  is  diminished.  Thus,  whilst  in  the  ordinary  course  of  manu- 
facture, I  obtained  with  carbonate  and  sulphuret  of  potassa  25  parts 
of  ferrocyanide  with  100  parts  of  horn,  soda  in  the  same  conditions 
furnished  only  5  parts. 
This  smaller  production  of  the  cyanides  with  soda  seems  explicable 
from  the  fact  that  soda  is  less  readily  reduced  than  potassa  to  the 
metallic  state,  the  condition  in  which  the  combination  between  the 
nitrogen  and -the  carbon  can  alone  take  place. 
Indeed,  in  the  production  of  the  cyanides  by  free  gaseous  nitrogen 
