630 
Prof.  TV.  H.  Bragg  on  the  Ionization  of 
be  brought  into  play,  the  direction  o£  the  lines  of  force  being 
normal  to  the  plane  of  the  paper.    The  paths  of  the  particles  are 
curved  to   one   side,  and    the    curvature    is 
greater  the  nearer  the  particle  is  to  the  end 
of  its  course.       JJet  OA  and  OQ  represent 
two  such  paths.     Their  separation  from  each 
other   is    considerably    exaggerated    in    the 
figure.      If  all  the  paths   were    drawn    the 
locus    of    Q  would  be  seen  to   be  a  curve, 
whose  curvature  in  contrast  to  that  of  the 
path    of  any  one  particle  would  be  greater 
the  further  the  distance  from  A.     This  is  in 
agreement  with  M.  Becquerel's  experiments, 
as  I  have  previously  pointed  out*. 
The  width  of  the  trace  upon  the  paper  of  all 
the  paths  of  the  particles  is  very  small,  and 
is  almost  too  fine  to  be  shown  on  a  diagram. 
It  is  perhaps  well  to  point  out  that  there 
is  no  break  in  this  trace  at  the  critical  points 
1*8,  4*2,  and  3*5.  It  is  quite  smooth  from 
end  to  end.  These  points  mark  the  extreme 
distance  to  which  various  bundles  of  rays 
penetrate ;  but  the  deflexion  of  an  a  ray 
which  ends  its  course  at  a  given  point  is  independent  of  the 
particular  radioactive  material  from  which  it  has  come  :  the 
only  varying  characteristic  of  an  a  particle  is  its  velocity. 
We  must  now  take  into 
account  that  the  widths  of 
the  slit  and  the  groove  are 
not  negligible,  as  is  clearly 
to  be  seen  from  the  photo- 
graph under  consideration. 
There  is  consequently,  so  to 
speak,  a  large  penumbra. 
Thus  the  trace  upon  the 
plane  of  the  paper  of  all 
the  a  rays  is  such  as  is  re- 
presented in  fig.  3,  the 
deflexions  being  all  exagge- 
rated so  as  to  be  capable  of 
depiction. 
Now  suppose  an  alumi- 
nium plate  is  placed,  as  in 
M.  Becquerer's  experiment, 
over  the  slit  so  that  the  a 
Fi?:.  3. 
Th  o  to{f7ap/tic  plate 
Ha. 
*  Phil.  Mag.  Dec.  1904,  p.  737  j  Jahrbuch  d.  Bad.  u.  EleUr.  1905,  p.  14. 
