﻿18 
  

  

  ON 
  THE 
  DIURNAL 
  VARIATION 
  

  

  Number 
  of 
  Curve. 
  

  

  Hour 
  of 
  minimum 
  

   westerly 
  variation. 
  

  

  Remarks. 
  

  

  

  H. 
  Min. 
  

  

  

  No. 
  1. 
  

  

  8 
  52 
  

  

  Fog 
  clearing 
  away. 
  

  

  No. 
  2. 
  

  

  9 
  00 
  

  

  Fog 
  not 
  breaking 
  until 
  near 
  ten. 
  

  

  No. 
  3. 
  

  

  9 
  00 
  

  

  Cloudy. 
  

  

  No. 
  4. 
  

  

  f8 
  30 
  

   (_8 
  02 
  

  

  f 
  Clear. 
  The 
  curve 
  seems 
  to 
  indicate 
  that 
  the 
  true 
  

   \ 
  minimum 
  was 
  earlier 
  than 
  8 
  h. 
  30 
  m. 
  or 
  at 
  8 
  h. 
  2 
  m. 
  

  

  No. 
  5. 
  

  

  9 
  00 
  

  

  Cloudy. 
  _. 
  

  

  No. 
  6. 
  

  

  8 
  00 
  

  

  c 
  Nimbus. 
  There 
  is 
  a 
  doubt 
  if 
  the 
  true 
  minimum 
  was 
  

   £ 
  observed. 
  

  

  No. 
  7. 
  

  

  9 
  00 
  

  

  Cloudless. 
  

  

  No. 
  8. 
  

  

  9 
  00 
  

  

  Slight 
  haziness. 
  

  

  No. 
  9. 
  

  

  8 
  00 
  

  

  Dense 
  fog, 
  which 
  entirely 
  disappeared 
  before 
  9 
  o'clock. 
  

  

  It 
  cannot 
  be 
  affirmed 
  that 
  this 
  minimum 
  was 
  not 
  placed 
  at 
  this 
  particu- 
  

   lar 
  hour, 
  by 
  the 
  effect 
  of 
  the 
  weather, 
  for 
  there 
  are 
  but 
  two 
  observations 
  

   when 
  the 
  weather 
  was 
  clear, 
  and 
  but 
  three 
  when 
  it 
  was 
  steady, 
  in 
  one 
  

   of 
  which 
  (No. 
  6) 
  the 
  minimum 
  may 
  have 
  occurred 
  before 
  any 
  obser- 
  

   vation 
  was 
  made. 
  It 
  agrees, 
  however, 
  with 
  the 
  recorded 
  observations 
  

   of 
  others. 
  The 
  observed 
  minima 
  did 
  not 
  depart 
  in 
  any 
  two 
  observa- 
  

   tions 
  more 
  than 
  one 
  hour 
  from 
  each 
  other. 
  

  

  In 
  order 
  to 
  determine 
  these 
  points 
  still 
  more 
  unexceptionably, 
  and 
  

   with 
  the 
  further 
  view 
  of 
  observing 
  the 
  effect 
  of 
  meteorological 
  causes, 
  

   I 
  have 
  taken 
  the 
  mean 
  of 
  the 
  results 
  of 
  eight 
  of 
  the 
  days 
  of 
  observa- 
  

   tion, 
  including 
  No. 
  3, 
  for 
  each 
  hour. 
  They 
  are 
  given 
  in 
  the 
  following 
  

   table 
  ; 
  and 
  curve 
  No. 
  1 
  1 
  , 
  Plate 
  III.* 
  is 
  traced 
  from 
  them. 
  The 
  last 
  co- 
  

   lumn 
  of 
  the 
  table 
  shows 
  the 
  number 
  of 
  observations 
  from 
  which 
  each 
  

   mean 
  has 
  been 
  deduced, 
  giving, 
  therefore, 
  the 
  relative 
  authority 
  of 
  each. 
  

  

  When 
  the 
  observations 
  were 
  made 
  at 
  the 
  half 
  hours, 
  the 
  mean 
  of 
  

   half 
  an 
  hour 
  before 
  and 
  half 
  an 
  hour 
  after 
  was 
  taken, 
  as 
  the 
  number 
  

   for 
  the 
  variation 
  or 
  temperature 
  at 
  the 
  hour. 
  A 
  similar 
  calculation 
  

   was 
  made 
  for 
  the 
  variation 
  at 
  any 
  particular 
  hour, 
  when 
  no 
  observa- 
  

   tion 
  had 
  been 
  made, 
  whenever 
  the 
  interval 
  between 
  two 
  observations 
  

   did 
  not 
  exceed 
  an 
  hour 
  and 
  a 
  half. 
  The 
  mean 
  temperature 
  of 
  the 
  

   day 
  having 
  varied 
  considerably 
  in 
  the 
  changes 
  of 
  weather, 
  it 
  was 
  

   thought 
  more 
  correct 
  to 
  enter 
  into 
  the 
  columns 
  of 
  temperature, 
  at 
  the 
  

   several 
  hours 
  on 
  which 
  observations 
  were 
  not 
  made, 
  an 
  average 
  of 
  the 
  

   next 
  preceding 
  and 
  succeeding 
  observation, 
  whenever 
  the 
  intervals 
  

  

  