174  Prof.  E.  Butherford  on 
some 
Increase  of  the  Radius  of  Curvature  of  the  Path  of 
the  Rays  in  Air. 
"SXe  have  seen  that  M.  Becquerel  early  drew  attention  to 
the  fact  that  the  radius  of  curvature  of  the  trace  of  the 
a  rays  obtained  in  air  increased  with  increase  of  distance 
from  the  source.  Assuming  that  the  rays  were  homogeneous, 
he  suggested  as  an  explanation  that  the  mass  of  the  a  particle 
increased  as  it  passed  through  the  air. 
Proceeding  on  these  assumptions,  it  is  seen  that  the  a.  rays 
from  each  product  present  in  radium  must  increase  in 
curvature  with  the  distance  of  air  traversed.  In  order  to 
test  this  point  definitely,  I  made  the  following  experiment, 
with  an  active  wire  coated  with  radium  C  as  a  source  of  rays. 
Using  the  same  apparatus  employed  to  obtain  the  photograph 
in  fig.  1,  the  part  of  the  active  wire  on  one  side  of  the  mica 
screen  was  covered  with  a  sheet  of  metal  of  sufficient  thickness 
to  absorb  all  the  a.  rays.  The  apparatus  was  filled  with  air  at 
atmospheric  pressure,  and  the  trace  of  the  rays  obtained  in  a 
uniform  magnetic  field  is  seen  in  fig.  3  (PL  IV.).  After  40 
minutes,  the  metal  sheet  was  transferred  to  the  other  side  of  the 
screen  and  the  apparatus  was  then  exhausted.  The  trace  of 
the  rays  then  obtained  in  a  vacuum  is  shown  by  the  two 
narrower  lines  in  the  same  figure.  There  is  a  striking 
difference  between  the  two  pairs  of  lines  obtained  in  air  and 
in  vacuum. 
The  air-bands  are  more  deflected  than  the  latter  and  in 
addition  are  broader.  This  experiment  conclusively  shows 
that  the  a.  rays  from  the  active  wire  decrease  in  radius  of 
curvature  and  consequently  in  velocity  after  passing  through 
air.  The  greater  width  and  lack  of  definition  of  the  air-lines 
have  been  noticed  in  all  other  experiments,  and  show  evidence 
of  an  undoubted  scattering  of  the  rays  in  their  passage 
through  air.  The  slight  shift  of  the  vacuum-lines  in  the 
figure,  relative  to  the  air-lines,  was  due  to  an  accidental 
displacement  of  the  plate  in  transferring  the  metal  sheets  in 
the  middle  of  the  experiment.  Experiments  are  in  progress 
to  see  whether  this  scattering  also  occurs  in  the  passage  of  the 
rays  through  a  solid  substance.  This  scattering  of  the  rays 
probably  also  occurs  with  aluminium,  but  it  would  not  be 
evident  in  the  experiments  where  the  layers  of  foil  are  placed 
over  the  active  wire  between  the  source  and  the  slit. 
Since  the  rays  from  the  other  products  of  radium  in  all 
probability  behave  in  the  same  way  as  the  rays  from  radium 
C,  we  see  that  this  result  of  the  decrease  of  the  radius  of 
