﻿162 
  G. 
  E. 
  Hale 
  — 
  Photographic 
  Investigation 
  of 
  

  

  tor, 
  while 
  a 
  corresponding 
  slit 
  before 
  the 
  plate 
  moves 
  at 
  such 
  

   a 
  rate 
  that 
  the 
  line 
  in 
  use 
  passes 
  constantly 
  through 
  it. 
  

  

  Both 
  of 
  these 
  methods, 
  together 
  with 
  the 
  experiments 
  car- 
  

   ried 
  on 
  with 
  the 
  first 
  at 
  the 
  Harvard 
  Observatory 
  and 
  more 
  

   recently 
  at 
  the 
  Kenwood 
  Physical 
  Observatory, 
  have 
  been 
  

   already 
  described,* 
  and 
  in 
  the 
  present 
  paper 
  I 
  wish 
  to 
  consider 
  

   especially 
  the 
  results 
  obtained 
  in 
  Chicago 
  within 
  the 
  last 
  few 
  

   weeks. 
  

  

  In 
  my 
  earliest 
  attempts 
  at 
  photographing 
  the 
  prominence 
  

   spectrum 
  I 
  was 
  much 
  surprised 
  to 
  find 
  narrow, 
  sharp, 
  bright 
  

   lines 
  running 
  up 
  through 
  the 
  center 
  of 
  the 
  dark 
  shades 
  of 
  

   both 
  H 
  and 
  K, 
  apparently 
  to 
  the 
  very 
  top 
  of 
  every 
  prominence. 
  

   At 
  Mount 
  Sherman 
  in 
  1872 
  Professor 
  Young, 
  whose 
  eyes 
  are 
  

   exceptionally 
  sensitive 
  to 
  the 
  shorter 
  wave-lengths, 
  had 
  been 
  

   able 
  to 
  see 
  similar 
  reversals 
  of 
  H 
  and 
  K, 
  but 
  the 
  difficulties 
  of 
  

   observation 
  were 
  so 
  great 
  that 
  he 
  considered 
  it 
  probable 
  that 
  

   the 
  whole 
  width 
  of 
  each 
  dark 
  shade 
  at 
  H 
  and 
  K 
  was 
  reversed, 
  

   the 
  eye 
  being 
  able 
  to 
  perceive 
  only 
  the 
  maximum 
  of 
  intensity 
  

   at 
  the 
  center. 
  Once 
  or 
  twice 
  he 
  noticed 
  a 
  bright 
  line 
  esti- 
  

   mated 
  to 
  be 
  about 
  one 
  division 
  of 
  Angstrom's 
  scale 
  below 
  the 
  

   central 
  reversal 
  of 
  H, 
  but 
  with 
  the 
  utmost 
  precautions 
  the 
  eye 
  

   was 
  incapable 
  of 
  any 
  accurate 
  determinations 
  of 
  position 
  or 
  

   appearance 
  in 
  this 
  part 
  of 
  the 
  spectrum. 
  But 
  with 
  high 
  dis- 
  

   persion 
  and 
  care 
  in 
  manipulation 
  the 
  photographic 
  plate 
  meets 
  

   with 
  no 
  difficulties, 
  and 
  the 
  lines 
  are 
  obtained 
  with 
  ease. 
  Fig. 
  

   1 
  of 
  Plate 
  VIII 
  shows 
  the 
  reversals 
  photographed 
  with 
  a 
  radial 
  

   slit, 
  while 
  for 
  the 
  negative 
  used 
  in 
  making 
  fig. 
  2 
  the 
  slit 
  was 
  

   parallel 
  to 
  a 
  tangent 
  at 
  the 
  limb, 
  and 
  about 
  30" 
  from 
  it. 
  All 
  

   of 
  the 
  figures 
  were 
  made 
  directly 
  from 
  the 
  original 
  negatives 
  

   by 
  a 
  photographic 
  process, 
  and, 
  with 
  the 
  exception 
  of 
  fig. 
  3, 
  

   the 
  scale 
  is 
  the 
  same 
  as 
  that 
  at 
  the 
  focus 
  of 
  the 
  spectroscope, 
  

   the 
  fourth 
  order 
  spectrum 
  of 
  a 
  14,438 
  Rowland 
  grating 
  hav- 
  

   ing 
  been 
  employed. 
  Though 
  an 
  excellent 
  one 
  in 
  every 
  other 
  

   respect 
  the 
  grating 
  gives 
  two 
  orders 
  of 
  ghosts, 
  and 
  the 
  line 
  

   just 
  below 
  H 
  seems 
  to 
  coincide 
  with 
  one 
  of 
  these 
  ; 
  but 
  careful 
  

   measures 
  of 
  its 
  positions, 
  combined 
  with 
  its 
  appearance 
  as 
  com- 
  

   pared 
  with 
  the 
  corresponding 
  first 
  order 
  ghost 
  of 
  K, 
  makes 
  it 
  

   more 
  than 
  likely 
  that 
  it 
  is 
  an 
  independent 
  line. 
  A 
  set 
  of 
  pre- 
  

   liminary 
  measures 
  from 
  two 
  negatives 
  renders 
  it 
  extremely 
  

   probable 
  that 
  this 
  line 
  is 
  due 
  to 
  hydrogen, 
  as 
  the 
  wave-length 
  

   agrees 
  remarkably 
  with 
  that 
  obtained 
  by 
  Ames 
  for 
  a 
  hydrogen 
  

   line 
  at 
  this 
  point 
  (X3970*25)f 
  but 
  more 
  measures 
  from 
  a 
  num- 
  

   ber 
  of 
  negatives 
  already 
  in 
  my 
  possession 
  will 
  be 
  needed 
  to 
  

   settle 
  the 
  question. 
  There 
  seems 
  to 
  be 
  no 
  corresponding 
  line 
  

   in 
  the 
  solar 
  spectrum, 
  but 
  both 
  the 
  H 
  and 
  K 
  reversals 
  appear 
  

  

  * 
  Technology 
  Quarterly, 
  vol. 
  iii, 
  No. 
  4, 
  1890. 
  Astronomische 
  Nachrichten, 
  Nos. 
  

   3006 
  and 
  3037. 
  Sidereal 
  Messenger, 
  June, 
  1891. 
  Monthly 
  Notices 
  of 
  the 
  R. 
  A. 
  S., 
  

   July, 
  1891. 
  fPhil. 
  Mag., 
  July, 
  1890. 
  

  

  