Radiation  from  Ordinary  Materials.  225 
Note. 
Perhaps  it;  will  not  be  out  o£  place  to  remark  upon  two 
researches  the  results  of!  which  might  seem  to  be  contradictory 
to  mine. 
The  first  of  these  is  described  in  a  paper  by  McLennan 
and  Burton  *,  and  consists  of  the  measurement  of  the 
variation  of  the  ionization  with  the  pressure  in  a  cylinder  of 
zinced  iron,  of  height  125  cms.  and  diameter  25  cms.  The 
authors  point  out  that  the  resulting  curve  is  "  nearly  a 
straight  line,"  and  hence  conclude  that  the  ionization  is 
mainly  caused  by  penetrating  radiation. 
Now  it  is  clear  that  the  form  of  such  a  curve  ought  to  be 
of  the  same  nature  as  that  of  those  in  figs.  2-12  ;  for  large 
pressures,  when  the  radiation  from  the  walls  is  totally 
absorbed,  it  should  be  a  straight  line  which  should  cut  the 
axis  of  ionization  at  a  distance  from  the  origin  representing 
the  total  ionization  due  to  the  walls  when  the  rays  are  all 
absorbed.  The  pressure  will  have  to  be  considerable  before 
the  curve  becomes  approximately  straight,  for  the  density  of 
the  air  will  have  to  be  so  great  that  the  rays  from  any  point 
on  the  curved  surface  are  absorbed  before  they  strike 
another  point  on  the  same  surface.  McLennan  and  Burton's 
curve  is  not  plotted  for  pressures  higher  than  500  cms.,  at 
which  point  it  still  shows  noticeable  curvature  ;  but  we  may 
make  some  estimate  of  the  relation  between  surface  and 
volume-ionization  by  assuming  that  the  tangent  to  the  curve 
at  the  last  point  is  in  the  direction  of  the  final  straight  line. 
Taking  this  tangent,  we  find  10*7  for  the  total  surface 
ionization  and  12*8  for  the  volume  ionization  at  500  cms. 
(equivalent  to  1'94  at  760  mms.).  The  area  of  the  walls  is 
10780  sq.  cms.  and  the  volume  of  the  vessel  61500  c.c.  ; 
hence  the  ratio 
ionization  per  1  cm.2  of  surface        10-7  1*94   _o1..~ 
ionization  per  1  c.c.  of  volume  ~~  10780  '   61500 
From  Table  II.   we   find   the  same   ratio  for  zinc   to  be 
123 
— —  =  3*8,  for  an  unscreened  vessel.    Thus  the  ratio  of  the 
surface  to  the  volume  ionization  for  zinc  found  by  McLennan 
and  Burton  is  eight  times  that  found  by  me ;  the  difference 
between  our  experiments  is  in  the  opposite  direction  to  thai 
which  they  affirm.     It  is  very  probable  from  other  reasons 
*  Phys.  Review,  vol.  xvi.  p.  190. 
Phil.  Mag.  S.  6.  Vol.  11.  No.  62.  Feb.  1906.  Q 
