﻿226 
  Messrs. 
  H. 
  G. 
  J. 
  Moseley 
  'and 
  C. 
  G. 
  Darwin 
  on 
  

  

  efficiently 
  scatter 
  the 
  primary 
  radiation. 
  The 
  two 
  cases 
  are 
  

   shown 
  diagrarnmatically 
  in 
  fig. 
  3. 
  In 
  either 
  case 
  the 
  different 
  

   orders 
  will 
  "a 
  11 
  be 
  present, 
  hut 
  the 
  third 
  order 
  and 
  its 
  multiples 
  

  

  Fiff. 
  3. 
  

  

  A 
  B 
  

  

  may 
  well 
  be 
  much 
  stronger 
  than 
  the 
  others, 
  because 
  in 
  the 
  

   third 
  order 
  all 
  the 
  layers 
  cooperate, 
  while 
  in 
  the 
  first 
  order 
  

   only 
  every 
  third 
  layer 
  does 
  so. 
  If 
  all 
  the 
  layers 
  become 
  both 
  

   similar 
  and 
  similarly 
  situated, 
  the 
  previous 
  third 
  order 
  

   becomes 
  the 
  fundamental 
  and 
  the 
  others 
  disappear. 
  This 
  is 
  

   the 
  case 
  with 
  rocksalt, 
  for 
  which 
  no 
  subsidiary 
  orders 
  could 
  

   be 
  found. 
  In 
  this 
  case 
  d 
  is 
  presumably 
  the 
  distance 
  between 
  

   successive 
  layers, 
  and 
  this 
  happens 
  to 
  be 
  almost 
  exactly 
  

   the 
  same 
  as 
  the 
  average 
  distance 
  in 
  ferrocyanide. 
  The 
  be- 
  

   haviour 
  of 
  selenite 
  is 
  similar. 
  Here 
  although 
  the 
  intensity 
  of 
  

   the 
  first 
  order 
  was 
  more 
  than 
  twice 
  that 
  of 
  the 
  second, 
  the 
  

   latter 
  was 
  the 
  more 
  conspicuous. 
  The 
  fourth 
  order 
  also 
  was 
  

   much 
  more 
  conspicuous 
  than 
  the 
  third. 
  In 
  this 
  case 
  the 
  

   crystal 
  structure 
  evidently 
  repeats 
  itself 
  every 
  second 
  layer. 
  

   This 
  seems 
  to 
  offer 
  a 
  simple 
  method 
  of 
  investigating 
  the 
  

   structure 
  of 
  crystals. 
  Another 
  way 
  of 
  looking 
  at 
  the 
  matter, 
  

   which 
  is 
  really 
  equivalent, 
  is 
  to 
  regard 
  the 
  most 
  conspicuous 
  

   order 
  as 
  the 
  fundamental, 
  and 
  call 
  all 
  earlier 
  orders 
  "ghosts." 
  

   In 
  the 
  simple 
  case 
  of 
  rocksalt 
  the 
  intensity 
  falls 
  off 
  for 
  

   the 
  higher 
  orders 
  much 
  more 
  rapidly 
  than 
  theoretical 
  

   considerations 
  would 
  suggest. 
  This 
  is 
  doubtless 
  due 
  to 
  the 
  

   effect 
  of 
  the 
  temperature 
  oscillations 
  of 
  the 
  atoms 
  in 
  

   the 
  crystal. 
  These 
  oscillations 
  will 
  not 
  affect 
  the 
  resolving 
  

   power 
  of 
  the 
  crystal, 
  regarded 
  for 
  the 
  moment 
  as 
  a 
  grating, 
  

   but 
  will 
  greatly 
  reduce 
  the 
  intensity 
  of 
  the 
  reflected 
  beam. 
  

   For 
  suppose 
  an 
  atom 
  displaced 
  a 
  distance 
  b 
  out 
  of 
  its 
  plane. 
  

   The 
  wave 
  scattered 
  by 
  this 
  atom 
  will 
  have 
  a 
  phase 
  difference 
  

  

  from 
  those 
  of 
  the 
  other 
  atoms 
  in 
  the 
  plane 
  of 
  2b 
  sin 
  6 
  . 
  — 
  

  

  7 
  A, 
  

  

  or 
  ^2/Z7r, 
  where 
  d 
  is 
  the 
  distance 
  between 
  successive 
  planes 
  ; 
  

  

  (X 
  

  

  so 
  the 
  phase 
  deviations 
  are 
  more 
  marked 
  in 
  the 
  higher 
  

   orders 
  and 
  the 
  relative 
  intensity 
  greatly 
  reduced. 
  

  

  The 
  next 
  point 
  investigated 
  was 
  the 
  length 
  of 
  the 
  train 
  of 
  

  

  