﻿REPORT 
  OF 
  THE 
  SECRETARY 
  97 
  

  

  it 
  gave 
  excellent 
  agreement 
  with 
  previous 
  values. 
  Very 
  numerous 
  

   experiments 
  and 
  comparisons 
  were 
  made 
  at 
  Table 
  Mountain 
  in 
  the 
  

   effort 
  to 
  trace 
  the 
  cause 
  of 
  the 
  discrepancy. 
  These 
  were 
  without 
  

   result 
  until 
  September, 
  1928, 
  when 
  Doctor 
  Abbot 
  visited 
  the 
  station 
  

   and 
  observed 
  that 
  portions 
  of 
  the 
  vestibule 
  of 
  the 
  instrument 
  had 
  

   become 
  shiny 
  by 
  handling. 
  Hence 
  sunlight 
  in 
  addition 
  to 
  sky 
  light 
  

   was 
  reaching 
  the 
  sensitive 
  measuring 
  strip. 
  By 
  reblackening 
  the 
  

   limiting 
  diaphragm 
  nearly 
  all 
  of 
  tliis 
  error 
  was 
  removed. 
  

  

  It 
  was 
  now 
  necessary 
  to 
  perform 
  a 
  great 
  mass 
  of 
  statistical 
  com- 
  

   puting 
  in 
  order 
  to 
  determine 
  the 
  magnitude 
  of 
  the 
  pyranometer 
  

   error 
  at 
  different 
  dates. 
  Fortunately, 
  an 
  error 
  of 
  20 
  per 
  cent 
  in 
  

   pyranometry 
  makes 
  but 
  1 
  per 
  cent 
  error 
  in 
  the 
  solar 
  constant, 
  so 
  

   that 
  no 
  great 
  accuracy 
  of 
  determining 
  the 
  error 
  was 
  required. 
  Hence 
  

   it 
  appeared 
  sufficient 
  to 
  collect 
  all 
  the 
  pyranometer 
  values 
  of 
  each 
  

   month, 
  arranging 
  them 
  in 
  orders 
  of 
  atmospheric 
  humidity, 
  air-mass, 
  

   and 
  pyrheliometer 
  value, 
  and 
  to 
  compare 
  the 
  mean 
  pyranometer 
  

   values 
  of 
  corresponding 
  months 
  in 
  successive 
  years, 
  as 
  well 
  as 
  the 
  

   values 
  in 
  nearly 
  identical 
  sky 
  conditions 
  throughout 
  each 
  year. 
  

  

  It 
  soon 
  became 
  clear 
  that 
  no 
  change 
  in 
  the 
  instrument 
  had 
  occurred 
  

   prior 
  to 
  early 
  August, 
  1927, 
  At 
  that 
  time 
  there 
  had 
  been 
  many 
  ex- 
  

   perimental 
  comparisons 
  involving 
  handling 
  of 
  the 
  vestibule, 
  which 
  

   had 
  done 
  the 
  damage 
  and 
  led 
  to 
  the 
  sudden 
  change. 
  Afterwards 
  

   many 
  more 
  comparisons 
  were 
  made 
  to 
  find 
  the 
  trouble, 
  and 
  these 
  had 
  

   aggravated 
  it. 
  After 
  much 
  work 
  it 
  became 
  possible 
  to 
  determine 
  a 
  

   set 
  of 
  sufficiently 
  exact 
  corrections 
  to 
  the 
  pyranometry 
  of 
  1927 
  and 
  

   1928 
  suitable 
  to 
  each 
  of 
  the 
  13 
  months 
  during 
  which 
  they 
  were 
  needed. 
  

   These 
  studies 
  were 
  made 
  on 
  Table 
  Mountain 
  observations 
  exclu- 
  

   sively, 
  so 
  that 
  they 
  introduced 
  no 
  element 
  of 
  dependence 
  on 
  

   Montezuma. 
  

  

  To 
  prevent 
  a 
  future 
  mischance 
  of 
  this 
  kind, 
  imperative 
  orders 
  were 
  

   issued 
  to 
  all 
  stations 
  as 
  to 
  the 
  handling 
  of 
  instruments, 
  and 
  standard 
  

   instruments, 
  for 
  comparison 
  purposes 
  only, 
  were 
  added 
  to 
  the 
  equip- 
  

   ment, 
  with 
  instructions 
  to 
  make 
  fairly 
  frequent 
  comparisons 
  between 
  

   these 
  and 
  the 
  instruments 
  in 
  use. 
  

  

  (c) 
  Atmos'pheric 
  ozone.— 
  Mx. 
  Fowle, 
  having 
  become 
  impressed 
  

   that 
  the 
  variations 
  recently 
  investigated 
  by 
  Dobson 
  in 
  the 
  quantity 
  

   of 
  atmospheric 
  ozone 
  might 
  very 
  possibly 
  affect 
  the 
  observed 
  solar 
  

   constant, 
  made 
  a 
  fruitful 
  investigation 
  of 
  the 
  absorption 
  of 
  ozone 
  in 
  

   the 
  yellow 
  and 
  green 
  of 
  the 
  solar 
  spectrum.^ 
  He 
  found 
  that 
  this 
  

   absorption, 
  though 
  small, 
  is 
  clearly 
  and 
  quantitatively 
  indicated 
  by 
  

   means 
  of 
  the 
  atmospheric 
  transmission 
  coefficients 
  obtained 
  in 
  the 
  

   application 
  of 
  the 
  fundamental 
  long 
  method 
  of 
  solar 
  constant 
  de- 
  

   termination 
  invented 
  by 
  Langley. 
  As 
  we 
  frequently 
  employ 
  this 
  

  

  1 
  Published 
  in 
  Smithsonian 
  Misc. 
  Coll., 
  vol. 
  81. 
  No. 
  11, 
  1929. 
  

  

  