﻿and 
  Secondary 
  Gamma 
  Rays, 
  287 
  

  

  The 
  results 
  obtained 
  by 
  Bragg 
  "^ 
  were 
  found 
  with 
  an 
  

   arrangement 
  different 
  from 
  mine, 
  for 
  he 
  placed 
  the 
  radium 
  

   close 
  to 
  the 
  electroscope. 
  

  

  To 
  calculate 
  the 
  effect 
  of 
  reversing 
  plates 
  A 
  and 
  B 
  we 
  have 
  

   to 
  look 
  at 
  Table 
  III. 
  or 
  at 
  curve 
  II., 
  fig. 
  3, 
  and 
  thence 
  obtain 
  

   the 
  fact 
  that 
  the 
  reversal 
  from 
  Al, 
  Pb 
  to 
  Pb, 
  Al 
  decreases 
  the 
  

   ionization 
  from 
  100 
  to 
  98. 
  This 
  explains 
  the 
  change 
  from 
  

   I. 
  to 
  II., 
  and 
  also 
  from 
  III. 
  to 
  IV. 
  just 
  above. 
  

  

  To 
  calculate 
  the 
  effect 
  of 
  reversing 
  C 
  and 
  D 
  we 
  see 
  in 
  

   Table 
  I., 
  or 
  from 
  curve 
  I. 
  (fig. 
  2), 
  that 
  lead 
  at 
  gives 
  100 
  

   whereas 
  aluminium 
  at 
  C 
  gives 
  84. 
  

  

  Thus 
  II. 
  and 
  lY. 
  are 
  98 
  per 
  cent, 
  of 
  I. 
  and 
  III. 
  respectively, 
  

   but 
  III. 
  and 
  IV. 
  are 
  84 
  per 
  cent, 
  of 
  I. 
  and 
  IL 
  respectively. 
  

   Hence, 
  of 
  course, 
  the 
  product 
  of 
  the 
  extremes 
  I. 
  and 
  IV. 
  

   must 
  equal 
  the 
  product 
  of 
  the 
  means 
  II. 
  and 
  III. 
  The 
  

   calculated 
  and 
  observed 
  values 
  for 
  Al, 
  Pb 
  are 
  respectively 
  — 
  

  

  I. 
  IL 
  III. 
  

  

  IV 
  

  

  100 
  98 
  84 
  

  

  82 
  

  

  100 
  98-5 
  83-5 
  

  

  81-1 
  

  

  ', 
  again, 
  taking 
  aluminium 
  and 
  copper 
  plates,- 
  

  

  - 
  

  

  I. 
  II. 
  III. 
  

  

  IV. 
  

  

  Observed 
  100 
  103*5 
  94*3 
  

  

  96-8 
  

  

  Calculated 
  .... 
  100 
  103-4 
  93-4 
  

  

  96-6 
  

  

  In 
  the 
  following 
  Table 
  IV. 
  (p. 
  2%^) 
  are 
  given 
  the 
  results 
  

   for 
  many 
  sets 
  of 
  four 
  plates; 
  and 
  successive 
  columns 
  show, 
  — 
  

   reference 
  number, 
  nature 
  of 
  screen, 
  thickness, 
  and 
  distance 
  

   from 
  the 
  radium 
  to 
  the 
  near 
  face 
  of 
  the 
  electroscope. 
  In 
  the 
  

   fifth 
  column 
  the 
  substances 
  used 
  as 
  plates 
  are 
  set 
  forih, 
  and 
  

   then 
  follow, 
  — 
  the 
  results 
  for 
  the 
  arrangements 
  I., 
  II., 
  III., 
  IV, 
  

   as 
  explained 
  above 
  ; 
  the 
  products 
  of 
  the 
  extremes, 
  and 
  of 
  the 
  

   means, 
  also 
  their 
  ratio, 
  which 
  has 
  an 
  average 
  value 
  0'993. 
  

  

  From 
  No. 
  14 
  we 
  see 
  that 
  the 
  reversals 
  of 
  plates 
  of 
  nearly 
  

   equal 
  atomic 
  weights, 
  such 
  as 
  Pe, 
  Cu, 
  produce 
  very 
  slight 
  

   changes 
  within 
  the 
  electroscope. 
  

  

  The 
  hardening 
  effect 
  of 
  lead 
  is 
  well 
  seen 
  by 
  contrasting 
  

   No. 
  1 
  and 
  2 
  ; 
  for 
  on 
  increasing 
  the 
  thickness 
  from 
  0'6 
  to 
  

   2-0 
  cm., 
  IL 
  changes 
  from 
  98*5 
  to 
  106*5, 
  and 
  III. 
  from 
  83*5 
  

   to 
  78*0. 
  No 
  such 
  alterations 
  take 
  place 
  when 
  the 
  thickness 
  

   of 
  iron 
  is 
  increased 
  as 
  in 
  Nos. 
  3 
  and 
  4. 
  The 
  same 
  general 
  

   results 
  are 
  found 
  in 
  Nos. 
  5 
  to 
  9 
  when 
  carbon 
  and 
  lead 
  

   are 
  used. 
  

  

  * 
  Phil. 
  Mag-. 
  May 
  1908. 
  

  

  