and  Composition  of  Uranium  Compounds. 
181 
the  first  method  of  calculating  the  total  activity  was  therefore 
excluded,  in  the  case  of  ores,  there  was,  on  the  other  hand,  no 
theoretical  objection  to  the  application  of  the  second  method. 
Only  the  purest  minerals  are  perfectly  homogeneous. 
Since  the  ordinary  method  of  powdering  might  involve  a  loss 
of  some  dust  of  greater  or  less  activity  than  the  average  for 
the.  ore,  the  crushed  samples  were  moistened  with  pure  redis- 
tilled alcohol  and  ground  in  an  agate  mortar  to  a  very  smooth 
paste,  The  alcohol  was  then  allowed  to  evaporate.  The 
powder  so  obtained  was  finer  and  more  uniform  than  could 
be  obtained  by  dry  grinding.  The  films  were  prepared  in 
the  manner  previously  described.  As  suspending  liquids  I 
used  for  pitchblende,  chloroform  ;  for  gummite,  alcohol ;  and 
for  carnotite,  water. 
The  uranium  in  pitchblende  and  gummite  was  determined 
as  previously  described  *,  by  treating  the  nitric-acid  solution 
of  the  ore  with  an  excess  of  sodium  carbonate  and  precipitating 
the  uranium  from  the  neutralized  filtrate  by  means  of  sodium 
hydroxide  f .  The  sulphuric  acid  solution  of  the  uranium 
precipitate  was  reduced  with  zinc  and  titrated  with  perman- 
ganate. In  the  analysis  of  carnotite,  vanadium  was  first 
removed  by  treating  the  dry  sample  repeatedly  with  dry 
hydrogen  chloride  gas  J.  Duplicate  analyses  were  made  in 
each  case.  Five  different  ores  have  been  studied  as 
follows : — 
Table  I. 
Mineral. 
Locality. 
Per  cent. 
Uranium. 
1.  Pitchblende. 
2.  Pitchblende. 
3.  Pitchblende. 
4.  Gummite. 
5.  Carnotite. 
Colorado. 
Bohemia. 
Unknown. 
North  Carolina. 
Colorado. 
511 
40-3 
611 
54-7 
39-9 
None  of  these  ores  contained  more  than  traces  of  thorium. 
*  Ber.  d.  ckem.  Ges.  xxxvii.  p.  2641  (1904). 
t  Brearley,  Analytical  Chemistry  of  Uranium,  1903,  states  that 
uranium  is  not  completely  precipitated  by  sodium  hydroxide.  This  is 
true  only  when  the  latter  contains  carbonate,  as  it  invariably  does  unless 
special  precautions  are  taken.  I  have  used  sodium  hydroxide  from  which 
the  small  amount  of  carbonate  originally  contained  was  removed  by  the 
addition  of  a  little  calcium  chloride  solution. 
%  Hillebrand,  Amer.  J.  Sci.  x.  p.  135  (1900;. 
