﻿460 
  Dr. 
  van 
  Rijckevorsel 
  on 
  the 
  Temperature 
  of 
  Europe. 
  

  

  compared 
  the 
  daily 
  temperatures 
  of 
  a 
  certain 
  number 
  of 
  places 
  

   mostly 
  in 
  Europe. 
  For 
  the 
  majority 
  of 
  these 
  the 
  figures 
  

   could 
  be 
  simply 
  taken 
  from 
  the 
  papers 
  published 
  by 
  different 
  

   institutions 
  or 
  private 
  observers. 
  For 
  some 
  stations, 
  however, 
  

   I 
  had 
  to 
  compute 
  the 
  daily 
  means 
  from 
  the 
  observations 
  them- 
  

   selves 
  as 
  published 
  periodically 
  by 
  the 
  observers. 
  

  

  However, 
  none 
  of 
  these 
  means 
  could 
  be 
  forthwith 
  used 
  in 
  

   the 
  state 
  they 
  were 
  in. 
  There 
  is 
  not 
  one 
  series 
  of 
  obser- 
  

   vations, 
  as 
  yet, 
  covering 
  such 
  a 
  number 
  of 
  years 
  that 
  we 
  may, 
  

   without 
  more 
  ado, 
  take 
  the 
  means 
  as 
  normal 
  temperatures. 
  

   Series 
  have 
  been 
  published 
  covering 
  a 
  century 
  and 
  more. 
  

   (St. 
  Petersburg, 
  Copenhagen, 
  among 
  others.) 
  Unfortu- 
  

   nately 
  no 
  series 
  of 
  even 
  nearly 
  that 
  length 
  seems 
  to 
  be 
  extant 
  

   for 
  a 
  more 
  southern 
  latitude. 
  And 
  for 
  a 
  station 
  in 
  the 
  north 
  

   of 
  Europe, 
  if 
  the 
  mean 
  temperatures 
  for 
  each 
  day 
  of 
  the 
  year 
  

   are 
  plotted 
  down, 
  the 
  resulting 
  curve 
  is, 
  even 
  in 
  cases 
  like 
  

   St. 
  Petersburg 
  and 
  Copenhagen, 
  so 
  hopelessly 
  distorted 
  that 
  

   not 
  much 
  can 
  be 
  done 
  with 
  it. 
  

  

  Therefore 
  it 
  is 
  necessary 
  to 
  manipulate 
  the 
  figures 
  to 
  a 
  

   certain 
  degree. 
  I 
  have 
  done 
  so 
  after 
  the 
  simplest 
  possible 
  

   method. 
  The 
  figures 
  were 
  smoothed 
  down 
  by 
  taking, 
  firstly, 
  

   the 
  arithmetical 
  means 
  of 
  every 
  set 
  of 
  nine 
  days 
  and 
  assigning 
  

   this 
  corrected 
  value 
  to 
  the 
  fifth 
  of 
  those 
  nine 
  days. 
  (Why 
  I 
  

   choose 
  the 
  number 
  nine 
  need 
  not 
  be 
  here 
  related. 
  It 
  has 
  

   been 
  printed 
  elsewhere*.) 
  

  

  In 
  most 
  cases 
  even 
  this 
  was 
  not 
  enough, 
  and 
  I 
  had 
  to 
  

   smooth 
  the 
  curves 
  once 
  more. 
  This 
  was 
  done 
  by 
  taking 
  the 
  

   mean 
  of 
  every 
  set 
  of 
  3, 
  5, 
  7, 
  or 
  even 
  9 
  days, 
  and 
  substituting 
  

   this 
  second 
  mean 
  for 
  the 
  figure 
  of 
  the 
  2nd, 
  3rd, 
  4th, 
  or 
  5th 
  

   day 
  of 
  the 
  set. 
  

  

  Of 
  course 
  this 
  may 
  seem 
  to 
  be 
  stretching 
  a 
  useful 
  method 
  

   to 
  a 
  rather 
  dangerous 
  extent. 
  I 
  think, 
  however, 
  that 
  the 
  

   results 
  sufficiently 
  show 
  that, 
  for 
  the 
  end 
  I 
  had 
  in 
  view, 
  the 
  

   expedient 
  was 
  a 
  good 
  one. 
  It 
  is 
  quite 
  possible 
  that 
  by 
  it 
  

   some 
  peculiarities 
  have 
  been 
  obliterated 
  which 
  may 
  be 
  inter- 
  

   esting. 
  But 
  this 
  does 
  not 
  in 
  any 
  way 
  invalidate 
  the 
  results 
  I 
  

   am 
  going 
  to 
  show 
  you. 
  

  

  Refore 
  proceeding 
  any 
  further 
  it 
  is 
  only 
  fair 
  to 
  add 
  that 
  

   not 
  all 
  my 
  figures 
  must 
  be 
  taken 
  as 
  real 
  normal, 
  or 
  even 
  mean, 
  

   temperatures. 
  For 
  some 
  stations 
  not 
  nearly 
  enough 
  data 
  

   were 
  at 
  hand. 
  In 
  some 
  cases 
  each 
  figure 
  is 
  the 
  mean 
  of 
  the 
  

   readings 
  on 
  three, 
  or 
  even 
  two, 
  chosen 
  hours. 
  In 
  some 
  others 
  

   it 
  is 
  the 
  mean 
  of 
  the 
  daily 
  maximum 
  and 
  minimum, 
  or 
  even 
  

   the 
  temperature 
  at 
  noon 
  only. 
  In 
  a 
  few 
  cases 
  also 
  the 
  series 
  

  

  * 
  Annuaire 
  Meteor 
  ologique 
  pour 
  1894, 
  publie 
  par 
  l'lnstitut 
  Royal 
  Me- 
  

   teoi 
  ologique 
  des 
  Pays-Bas, 
  quarante-sixierne 
  annee. 
  

  

  