ON  THE  EXTRACTION  OF  POTASSA  FROM  MARL.  237 
as  much  as  7  per  cent,  be  obtained.  (Report  N.  J.  State  Geo- 
logical Survey,  1854.) 
The  green  sand  with  which  the  following  experiments  were 
tried  contained  about  6  per  cent,  of  potassa.  It  is  from  the  se- 
cond or  middle  of  the  three  beds  into  which  this  deposit  in  New 
Jersey  has  been  divided,  and  was  dug  on  the  land  of  David 
Marshall,  near  Blackwoodtown,  Camden  County.  An  analysis, 
performed  by  the  general  process  for  the  analysis  of  soils,  has 
given  the  following  results,  viz. : 
Insoluble  Silica, 
Soluble  Silica, 
Protoxide  Iron, 
Alumina, 
Potassa, 
Soda, 
Lime, 
Magnesia, 
Phosphoric  Acid, 
Water, 
(1,.5 
48. 
22.74 
6.61 
5.01 
1.08 
1.975 
1.375 
4.821 
7.50 
(2) 
48. 
6.84 
1.47 
99.611 
The  phosphoric  acid  was  estimated  as  the  biphosphate  of 
magnesia.  This  specimen  is  in  the  usual  form  of  small,  dis- 
tinct, green  grains,  soft  enough  when  freshly  dug  to  be  readily 
crushed  by  the  nail,  but  becoming  harder  on  exposure  to  the 
air ;  and  exhibiting  under  the  microscope,  when  washed  from 
adhering  clay,  a  general  resemblance  in  shape  to  the  "casts" 
of  the  Rhizopods,  and  other  minute  marine  animals.  In  order 
to  separate  the  potassa  from  the  green  sand,  I  have  tried  seve- 
ral of  the  cheap  and  powerful  chemical  agents, — carbonic,  hy- 
drochloric and  sulphuric  acids,  quicklime  and  sulphate  of  lime. 
Carbonic  acid,  though  a  weak  acid,  has  been  stated  to  be  one 
of  the  readiest  agents  in  decomposing  green  sand.  But  I  have 
not  been  able  to  obtain  by  it  any  appreciable  amount  of  potassa, 
and  only  a  trace  of  iron  and  lime.  The  quantities  used  were 
100  grains  of  green  sand,  and  12  oz.  of  water  charged  with 
carbonic  acid. 
Hydrochloric  acid  decomposes  green  sand,  only  after  being 
boiled  with  it  for  several  hours?  or  after  contact  with  it  for  seve- 
