﻿480 
  G. 
  C. 
  ComstocJc 
  — 
  The 
  Secular 
  Variation 
  of 
  Latitudes. 
  

  

  more 
  closely 
  investigated 
  than 
  has 
  yet 
  been 
  done. 
  For 
  the 
  

   present 
  the 
  only 
  available 
  data 
  seems 
  to 
  be 
  contained 
  in 
  the 
  

   following 
  table 
  : 
  

  

  No. 
  of 
  Comput'd 
  

   Station. 
  Longitude. 
  Ann. 
  Var. 
  of 
  <p. 
  Weight 
  Determin's. 
  Ann. 
  Var. 
  

  

  Pulkowa 
  — 
  30°-3 
  — 
  0"-006 
  4 
  3 
  — 
  0"*007 
  

  

  Konigsberg 
  —20-5 
  —0-003 
  1 
  1 
  —0*000 
  

  

  Washington 
  + 
  77*0 
  +0-042 
  4 
  1 
  +0-044 
  

  

  Madison 
  +89'4 
  +0*043 
  4 
  2 
  +0'041 
  

  

  The 
  longitudes 
  are 
  reckoned 
  from 
  Greenwich. 
  

  

  I 
  have 
  made 
  a 
  least 
  square 
  solution 
  of 
  these 
  data 
  to 
  determine 
  

   the 
  most 
  probable 
  direction 
  and 
  amount 
  of 
  motion 
  of 
  the 
  pole 
  

   and 
  find 
  a 
  motion 
  of 
  0"O44 
  along 
  the 
  meridian 
  69° 
  west 
  of 
  

   Greenwich. 
  The 
  last 
  column 
  of 
  the 
  table 
  above 
  contains 
  the 
  

   values 
  of 
  the 
  annual 
  variation 
  at 
  the 
  several 
  stations 
  computed 
  

   from 
  these 
  elements. 
  

  

  If 
  the 
  elements 
  of 
  the 
  motion 
  of 
  the 
  pole 
  thus 
  derived 
  are 
  

   even 
  a 
  rough 
  approximation 
  to 
  the 
  truth 
  they 
  furnish 
  valuable 
  

   indications 
  of 
  the 
  methods 
  by 
  which 
  our 
  knowledge 
  may 
  be 
  

   extended. 
  In 
  the 
  first 
  place 
  European 
  observatories 
  cannot 
  be 
  

   expected 
  to 
  show 
  any 
  considerable 
  change 
  of 
  latitude. 
  Obser- 
  

   vations 
  made 
  there 
  will 
  be 
  chiefly 
  valuable 
  for 
  determining 
  

   the 
  direction 
  of 
  motion 
  of 
  the 
  pole 
  and 
  for 
  this 
  purpose 
  a 
  care- 
  

   ful 
  comparison 
  of 
  the 
  older 
  latitudes 
  with 
  modern 
  determina- 
  

   tions 
  is 
  much 
  to 
  be 
  desired. 
  In 
  particular 
  the 
  latitude 
  deter- 
  

   minations 
  made 
  at 
  Dorpat 
  by 
  W. 
  Struve 
  in 
  1824 
  and 
  1827 
  

   with 
  the 
  meridian 
  circle 
  and 
  prime 
  vertical 
  transit 
  are 
  for 
  

   this 
  purpose 
  probably 
  the 
  most 
  valuable 
  data 
  not 
  yet 
  utilized. 
  

   I 
  have 
  endeavored 
  to 
  compare 
  these 
  with 
  similar 
  modern 
  

   determinations 
  by 
  Schwarz 
  and 
  Renz 
  but 
  the 
  printed 
  results 
  of 
  

   the 
  later 
  determinations, 
  at 
  least 
  so 
  far 
  as 
  I 
  have 
  access 
  to 
  them, 
  

   do 
  not 
  furnish 
  sufficient 
  data 
  for 
  the 
  purpose. 
  In 
  America 
  the 
  

   older 
  latitudes 
  of 
  the 
  Coast 
  Survey 
  could 
  very 
  profitably 
  be 
  

   rediscussed 
  and 
  compared 
  with 
  redeterminations 
  at 
  such 
  sta- 
  

   tions 
  as 
  can 
  now 
  be 
  identified. 
  A 
  redetermination 
  of 
  the 
  lati- 
  

   tude 
  of 
  Cambridge 
  with 
  both 
  the 
  prime 
  vertical 
  transit 
  and 
  

   the 
  zenith 
  telescope 
  seems 
  especially 
  desirable 
  and 
  the 
  Asiatic 
  

   stations 
  occupied 
  by 
  the 
  American 
  Transit 
  of 
  Yenus 
  parties 
  in 
  

   1874 
  can 
  be 
  made 
  to 
  furnish 
  most 
  valuable 
  data, 
  since 
  their 
  lati- 
  

   tudes 
  should 
  now 
  be 
  three 
  quarters 
  of 
  a 
  second 
  less 
  than 
  in 
  

   1874: 
  

  

  I 
  wish 
  now 
  to 
  consider 
  briefly 
  a 
  plan 
  for 
  the 
  systematic 
  

   investigation 
  of 
  the 
  motion 
  of 
  the 
  pole. 
  For 
  the 
  present 
  it 
  

   seems 
  best 
  not 
  to 
  attempt 
  the 
  absolute 
  determination 
  of 
  lati- 
  

   tudes 
  for 
  this 
  purpose 
  on 
  account 
  of 
  their 
  great 
  liability 
  to 
  

   systematic 
  error 
  but 
  rather 
  to 
  rely 
  upon 
  differential 
  methods. 
  

  

  