208  Mr.  K  R.  Campbell  on  the 
(1.  Commercial  sheet),  Copper  (2.  Purest  electrolytic  copper 
specially  rolled) ;  Aluminium,  Zinc,  Iron,  Platinum*  (all  com- 
mercial sheet), Tin  (Tin-plate),  Silver  and  Gold*  ("  chemically 
pure").  The  surfaces  of  all  the  materials  except  the  gold, 
platinum  and  pure  copper  were  cleaned  with  sandpaper 
before  use,  in  order  to  remove  any  surface  radioactivity.  In 
the  three  cases  excepted  the  metals  were  newly  rolled  and 
might  be  expected  to  be  clean. 
§5.  A  word  should  be  said  as  to  the  variations  in  the 
ionization  which  were  described  on  p.  534  of  the  former 
paper. 
Much  less  trouble  than  formerly  arose  from  the  sudden 
discontinuous  changes ;  on  a  few  occasions  "  jumps "  were 
noticed,  but  I  think  it  is  probable  that  they  are  due  to  ex- 
perimental defects.  However,  I  am  unable  to  explain  their 
origin  or  to  prevent  their  occurrence ;  the  observations  were 
simply  suspended  until  the  normal  state  of  affairs  was  re- 
established. The  irregular  points,  often  more  than  50  per 
cent,  from  the  curve,  are  not  plotted. 
But  the  slow  continuous  variations  are  still  in  evidence. 
They  appear  to  be  periodic  in  some  multiple  of  24  hours,  but 
my  experiments  were,  of  course,  not  suitable  to  a  detailed 
investigation  of  the  phenomenon.  It  is  probable  that  they 
are  due  to  changes  in  the  intensity  of  the  external  radiation. 
That  their  magnitude  was  not  sufficient  to  incommode  the 
work  seriously  will  be  seen  from  the  nearness  of  the  experi- 
mental points  to  the  curve  drawn  through  them. 
It  may  be  mentioned  that  the  observations  on  Lead  (1) 
were  interrupted  for  a  period  of  eight  weeks.  On  resuming 
after  that  time,  the  observations  agreed  perfectly  with  those 
taken  before  the  interval. 
The  resulting  experimental  curves  are  shown  in  figs.  2-12 
(PI.  V.),  where  the  abscissas  give  the  distance  apart  of  the  sides 
A  and  C,  and  the  ordinates  the  currents  X  105. 
§  6.  The  curves  can  be  interpreted  on  the  same  principles 
as  those  applied  in  the  earlier  paper. 
Let  us  consider  first  the  case  where  the  external  radiation 
is  cut  off.     We  have  then  as  possible  ionizing  agents 
(1)  the  easily  absorbable  radiation  from  the  ends   of  the 
box; 
(2)  the  more  penetrating  radiation  from  the  same ; 
(3)  the  radiation  from  the  wire-gauze  cage. 
Let  the  dimensions  of  the  cage  be  a  x  b  x  x,  x  being  jjer- 
pendiculnr  to  the  sliding  side. 
*  I  have  to  acknowledge  again  the  generosity  of  Messrs.  Johnson  & 
Matthey  in  lending  me  sufficient  quantities  of  these  metals. 
