﻿198 
  Dr. 
  P. 
  Zeeman 
  on 
  Measurements 
  concerning 
  

  

  I 
  have 
  measured 
  negatives, 
  obtained 
  by 
  means 
  of 
  a 
  Rowland 
  

   grating, 
  using 
  electrodes 
  of 
  various 
  metals 
  and 
  a 
  very 
  strong 
  

   field. 
  The 
  measurements, 
  however, 
  only 
  refer 
  to 
  a 
  relatively 
  

   small 
  part 
  of 
  the 
  spectrum. 
  I 
  regret 
  that 
  it 
  is 
  not 
  greater. 
  

   Much 
  time 
  was 
  lost 
  because 
  the 
  grating 
  was 
  not 
  mounted 
  in 
  

   the 
  manner 
  of 
  Rowland, 
  the 
  various 
  parts 
  of 
  the 
  apparatus 
  

   not 
  being 
  ready 
  in 
  time. 
  Grating 
  and 
  camera-box 
  were 
  

   placed 
  upon 
  separate 
  stands 
  ; 
  it 
  was 
  now 
  necessary 
  to 
  find 
  

   the 
  position 
  of 
  slit, 
  grating, 
  and 
  camera-box 
  for 
  each 
  part 
  

   of 
  the 
  spectrum 
  by 
  a 
  laborious 
  process 
  of 
  error 
  and 
  trial. 
  The 
  

   dimensions 
  of 
  the 
  camera-box 
  permitted 
  only 
  the 
  nse 
  of 
  plates 
  

   13 
  or 
  18 
  centim. 
  long. 
  The 
  results 
  of 
  this 
  paper 
  are 
  hence 
  

   to 
  be 
  regarded 
  only 
  as 
  preliminary. 
  I 
  intend 
  to 
  go 
  over 
  the 
  

   measurements 
  once 
  more 
  and 
  to 
  make 
  negatives 
  of 
  the 
  whole 
  

   spectrum, 
  using 
  a 
  larger 
  grating 
  and 
  the 
  new 
  apparatus, 
  just 
  

   now 
  finished. 
  

  

  2. 
  Measurements 
  of 
  the 
  distance 
  between 
  the 
  outer 
  compo- 
  

   nents 
  of 
  the 
  triplet 
  will 
  also 
  put 
  to 
  test 
  the 
  possible 
  hypothesis, 
  

   that 
  it 
  is 
  only 
  one 
  kind 
  of 
  ions, 
  having 
  charges 
  proportional 
  to 
  

   their 
  effective 
  masses, 
  that 
  are 
  free 
  to 
  vibrate 
  in 
  the 
  atom 
  or 
  

   molecule. 
  On 
  such 
  an 
  hypothesis, 
  in 
  a 
  definite 
  field 
  the 
  inter- 
  

   vals 
  between 
  the 
  outside 
  components 
  of 
  the 
  triplets 
  should 
  be 
  

   (measured 
  in 
  change 
  of 
  wave-length) 
  proportional 
  to 
  the 
  

   square 
  of 
  the 
  wave-length 
  or 
  (measured 
  in 
  difference 
  of 
  fre- 
  

   quency) 
  the 
  same 
  for 
  all 
  lines 
  of 
  the 
  spectra 
  of 
  all 
  substances*. 
  

   The 
  measurements 
  communicated 
  prove 
  that 
  for 
  different 
  

   substances 
  the 
  magnetic 
  change 
  is 
  at 
  least 
  of 
  the 
  same 
  order 
  

   of 
  magnitude, 
  and 
  certainly 
  not 
  directly 
  dependent 
  upon 
  the 
  

   atomic 
  weight. 
  The 
  variation 
  of 
  the 
  phenomenon 
  with 
  wave- 
  

   length, 
  however, 
  seems 
  more 
  complicated 
  than 
  follows 
  from 
  

   the 
  hypothesis 
  mentioned, 
  and 
  in 
  fact 
  in 
  Lorentz's 
  theory 
  

   there 
  is 
  no 
  a 
  priori 
  reason 
  to 
  be 
  given 
  for 
  its 
  probability!. 
  

  

  3. 
  The 
  Rowland 
  grating 
  used 
  was 
  one 
  of 
  radius 
  6 
  feet 
  

   and 
  14,438 
  lines 
  to 
  the 
  inch. 
  The 
  second 
  spectrum 
  was 
  used 
  

   in 
  most 
  cases. 
  With 
  a 
  medium-sized 
  induction-coil 
  and 
  one 
  

   large 
  leyden-jar 
  twenty 
  minutes 
  exposure 
  was 
  sufficient 
  in 
  the 
  

   most 
  sensitive 
  part 
  of 
  the 
  spectrum. 
  Pure 
  metals 
  were 
  used 
  

   as 
  electrodes. 
  In 
  each 
  position 
  of 
  the 
  apparatus 
  always 
  two 
  

   (and 
  often 
  several) 
  negatives 
  were 
  taken, 
  one 
  with 
  and 
  one 
  

   without 
  the 
  magnetic 
  field. 
  The 
  plates 
  used 
  were 
  Cadett 
  and 
  

   Neal 
  and 
  Marion 
  instantaneous 
  ; 
  they 
  were 
  developed 
  with 
  

   hydroquinone. 
  In 
  some 
  photochemical 
  difficulties 
  Dr. 
  Ernst 
  

   Cohen 
  was 
  kind 
  enough 
  to 
  give 
  me 
  his 
  assistance. 
  

  

  * 
  Becquerel, 
  C. 
  R. 
  Nov. 
  8, 
  1897 
  ; 
  Laruior, 
  Phil. 
  Mag. 
  vol. 
  xliv. 
  

   p. 
  503, 
  § 
  4 
  (1897). 
  

  

  t 
  Lorentz, 
  Wied. 
  Ann. 
  Bd. 
  lxiii. 
  p. 
  278 
  (1897) 
  ; 
  Larmor, 
  I. 
  c. 
  p. 
  506. 
  

  

  