222  Mr.  N.  R.  Campbell  on  the 
less  than  1  per  cent,,  when  all  allowance  is  made  for  error, 
it  is  impossible  to  attribute  any  considerable  portion  of  the 
activity  of  lead  to  the  activity  of  radium  ;  and  radium  is  the 
only  impurity  to  which  anyone  has  ventured  to  attribute 
the  activity  of  ordinary  metals. 
Discussion  of  Results. 
§  15.  (1)  Comparison  of  Tables  I.  and  III. — It  is  clear  that 
in  the  main  features  the  new  series  agrees  with  the  old  ;  we 
may  instance  the  absence  of  secondary  absorbable  radiation 
from  lead  and  aluminium ;  the  large  ratio  of  secondary  to 
intrinsic  absorbable  radiation  for  platinum  ;  and  the  small 
volume  ionization  for  tin.  On  the  other  hand,  the  absolute 
values  differ  largely  :  this  may  be  partly  due  to  the  un- 
certainty regarding  the  value  of  the  unit  of  capacity  used  in 
the  older  series.  It  is  hard  to  compare  the  values  of  the 
penetration,  since  the  first  series  was  not  suitable  for  cal- 
culation, but  the  order  in  which  the  metals  would  be  arranged 
in  respect  of  this  quality  would  be  the  same  except  for  lead, 
in  which  case  the  penetration  appears  to  be  larger  in  the  new 
than  in  the  old  series.  It  should  be  noted,  that  the  inclination 
of  the  straight  portion  to  the  horizontal  axis  of  the  former 
curve  depended  on  the  intensity  of  the  absorbable  radiation 
from  the  walls :  it  would  be  large  for  lead,  and  tend  to 
make  the  transition  from  the  curved  to  the  straight  portion 
appear  too  abrupt. 
Since  the  later  experiments  are  more  reliable  in  principle 
and  more  accurate  in  execution  than  the  old,  in  any  case  of 
discrepancy  far  greater  weight  should  be  attached  to  the 
former. 
(2)  Value  of  s. — The  most  important  feature  of  this 
quantity  is  its  constancy  when  deduced  from  wholly  different 
specimens  of  the  same  metal  ;  the  emission  of  intrinsic 
absorbable  radiation  appears  to  be  an  inherent  property  of 
the  substance  itself.  I  am  unable  to  trace  any  relation 
between  the  value  of  s  for  a  metal  and  any  other  chemical  or 
physical  property. 
(3)  Value  of  s'. — It  might  be  expected  that  aluminium 
would  give  little  or  no  secondary  radiation,  but  the  absence 
of  such  radiation  from  lead  is  very  surprising.  Nevertheless 
it  isan  perfect  agreement  with  my  earlier  experiments  and 
those  of  Wood  *.  Perhaps  this  peculiarity  may  be  connected 
with  the  use  of  lead  as  the  screen  to  cut  off  external  radiation, 
but  the  connexion  is  not  clear.  Selective  absorption  would 
give  precisely  the  opposite  effect  :    moreover,  Wood  found 
*  Phil.  Mag.  April  1905. 
