﻿Calculation 
  of 
  the 
  Atomic 
  Weight 
  of 
  Radium. 
  411 
  

  

  but 
  it 
  does 
  appear 
  as 
  if 
  there 
  is 
  a 
  want 
  of 
  symmetry 
  in 
  the 
  

   lithium 
  dispersion 
  spectrum, 
  which 
  indicates 
  that 
  the 
  lithium 
  

   line 
  is 
  a 
  double 
  line 
  with 
  one 
  component 
  stronger 
  than 
  

   another. 
  

  

  Ebert 
  * 
  has 
  observed 
  the 
  anomalous 
  dispersion 
  of 
  potassium 
  

   vapour 
  by 
  a 
  method 
  in 
  which 
  the 
  vapour 
  was 
  kept 
  in 
  more 
  

   or 
  less 
  prismatic 
  form 
  by 
  means 
  of 
  two 
  currents 
  of 
  hydrogen 
  : 
  

   the 
  position 
  of 
  the 
  absorption 
  bands 
  given 
  in 
  the 
  diagram 
  

   in 
  Kayser's 
  Handhuch, 
  however, 
  does 
  not 
  seem 
  to 
  agree 
  

   with 
  that 
  shown 
  in 
  fig. 
  4. 
  The 
  method 
  does 
  not 
  seem 
  to 
  

   promise 
  such 
  steady 
  conditions 
  as 
  may 
  be 
  obtained 
  with 
  the 
  

   simple 
  method 
  of 
  Wood, 
  and 
  further, 
  there 
  must 
  always 
  be 
  

   some 
  moisture 
  carried 
  over 
  with 
  the 
  hydrogen 
  which 
  must 
  

   make 
  the 
  conditions 
  not 
  so 
  permanent 
  as 
  in 
  the 
  simpler 
  

   method. 
  The 
  author 
  hopes 
  to 
  continue 
  the 
  experiments 
  in 
  

   the 
  ultra-violet 
  region. 
  This 
  has 
  been 
  made 
  possible 
  by 
  a 
  

   generous 
  grant 
  from 
  the 
  Royal 
  Society. 
  

  

  XLYII. 
  On 
  the 
  Calculation 
  of 
  the 
  Atomic 
  Weight 
  of 
  Radium 
  

   from 
  Spectroscopic 
  Data. 
  By 
  W. 
  Marshall 
  Watts, 
  D.Sc.^ 
  

  

  THE 
  existence 
  of 
  exact 
  relationships 
  between 
  the 
  atomic 
  

   weights 
  of 
  allied 
  elements 
  and 
  their 
  spectra 
  does 
  not 
  

   admit 
  of 
  doubt. 
  If 
  we 
  knew 
  the 
  precise 
  law 
  of 
  this 
  relation- 
  

   ship, 
  the 
  measurement 
  of 
  the 
  wave-lengths 
  of 
  the 
  lines 
  in 
  the 
  

   spectrum 
  of 
  a 
  new 
  or 
  rare 
  element, 
  particularly 
  of 
  one 
  

   difficult 
  to 
  obtain 
  Id 
  a 
  state 
  of 
  purity, 
  would 
  afford 
  much 
  more 
  

   exact 
  data 
  for 
  the 
  calculation 
  of 
  the 
  atomic 
  weight, 
  than 
  the 
  

   chemical 
  method. 
  If 
  we 
  consider 
  that 
  even 
  in 
  the 
  case 
  of 
  an 
  

   element 
  whose 
  atomic 
  weight 
  has 
  been 
  determined 
  repeatedly 
  

   with 
  the 
  utmost 
  possible 
  care 
  — 
  say 
  silver 
  or 
  chlorine 
  — 
  the 
  

   result 
  is 
  not 
  certain 
  to 
  one 
  part 
  in 
  3000, 
  whereas 
  the 
  wave- 
  

   length 
  of 
  a 
  line 
  can 
  be 
  measured 
  with 
  an 
  accuracy 
  of 
  one 
  part 
  

   in 
  100,000, 
  we 
  see 
  how 
  important 
  the 
  spectroscopic 
  method 
  

   might 
  become. 
  

  

  In 
  the 
  Phil. 
  Mag. 
  for 
  Aug. 
  1904, 
  in 
  discussing 
  the 
  results 
  

   obtained 
  by 
  Runge 
  and 
  Precht, 
  I 
  have 
  pointed 
  out 
  that 
  the 
  

   number 
  257'8 
  obtained 
  by 
  them 
  for 
  Radium, 
  from 
  the 
  atomic 
  

   weights 
  of 
  Mg, 
  Ca, 
  Sr, 
  and 
  Ba, 
  must 
  be 
  too 
  high, 
  judging 
  

   from 
  the 
  similar 
  calculation 
  of 
  the 
  atomic 
  weight 
  of 
  mercury 
  

   from 
  the 
  atomic 
  weights 
  of 
  zinc 
  and 
  cadmium, 
  which 
  would 
  

   give 
  222 
  instead 
  of 
  200. 
  It 
  appears 
  that 
  Runge 
  and 
  Precht's 
  

   rule 
  — 
  that 
  the 
  separations 
  of 
  homologous 
  pairs 
  of 
  lines 
  are 
  

  

  * 
  Fhys. 
  Zeitschr. 
  4, 
  p. 
  473 
  ; 
  Kayser, 
  Handbuch 
  der 
  Spectroscopiey 
  

   vol. 
  iv. 
  p. 
  563. 
  

  

  t 
  Commimicated 
  bv 
  the 
  Author. 
  

  

  