The  Osmotic  Pressures  of  Alcoholic  Solutions.        595 
4.  Most  penetrating;  rays,  probably  7  rays. 
\  =  2'1  to  2*0,  from  2'8  to  8*7  mm. 
(5)  A  kilogram   of   thorium  nitrate    sealed  in  a  thin   glass 
vessel  16  cms.  in  diameter,  placed  under  a  layer  of 
lead  1  cm.  thick,  might  be  adopted  as  a  convenient 
standard  for  measuring  the  quantity  of  radium  or 
thorium  contained  in  a  given  mass  of  ore. 
(6)  The  self-absorption  of  the  7  rays  from  1   kilogram    of 
thorium  nitrate  in  a  vessel  16  cms.  in  diameter  and 
about  3'1  cms.  deep,  is  such  that  about  18  per  cent, 
should  be  added  to  the  results  actually  obtained. 
When  used  as  a  standard  no  such  correction  should  be 
made. 
In  conclusion  I  take  pleasure  in  expressing  my  gratitude 
to  Professor  Rutherford  for  proposing  the  experiment,  and 
for  his  usual  kind  interest  and  ever  ready  assistance. 
McGill  University,  Montreal, 
December  19,  1905. 
L.    The  Osmotic  Pressures  of  Alcoholic  Solutions. 
By  P.  S.  Baklow,  B.A.,  St.  John's  College,  Cambridge*. 
rrVHE  earlier  experimental  work  of  this  paper  dealt  with  the 
_I_  use  of  the  ordinary  copper-ferrocyanide  membrane  for 
solutions  in  which  the  solvent  was  ethyl  alcohol.  Tammann  t 
showed  that  ethyl  alcohol  could  get  through  this  membrane 
against  the  osmotic  current,  but  did  not  itself  seem  capable 
of  setting  up  an  osmotic  current.  As  there  did  not  seem  to 
be  sufficient  evidence  on  this  point  to  justify  the  assumption 
that  all  copper-ferrocyanide  membranes,  however  prepared, 
could  not  be  used  with  alcoholic  solutions,  it  was  considered 
necessary  to  try  the  cells  which  had  already  been  in  use. 
The  substance  of  the  membrane  is  precipitated  in  the  colloidal 
state  ;  and  the  outward  passage  of  the  alcohol  would  appear 
to  point  to  the  absorption,  under  pressure,  of  the  alcohol  by 
the  membrane.  A  large  number  of  experiments  were  made. 
All  the  results  were  negative  and  in  agreement  with  Tain- 
mannas  work  ;  butthe  work  done  seemed  justified  by  the  limited 
knowledge  we  have  of  the  detailed  properties  of  colloids. 
The  inevitable  small  differences  of  preparation  of  the  mem- 
branes might  very  well  have  produced  some  change  in  the 
*  Communicated  by  Prof.  J.  J.  Thomson. 
f  Annal.  PJiys.  und  Cliem.  Neue  Folge,  xxxiv.  p.  309- 
