626 
Climate  of  the  British  Islands 
have  found  from  many  experiments  the  minimum  of  heat  for  the 
twenty-four  hours  to  be  from  1 h.  30  m.  to  30  m.  before  sunrise, 
and  the  mean  to  occur  48  m.  before  the  rising  of  the  sun.  It 
would  appear  from  my  own  observations  that  the  greatest  cold 
is  found  nearer  to  the  time  of  sunrise  in  summer  than  in  winter. 
The  maximum  of  heat  is  found  upon  an  average  of  many  years  to 
occur  from  2 h.  30  m.  p.m.  to  2 h.  45  m.  p.m.  The  mean  tem- 
perature of  the  day  is  generally  approached  at  about  half  the 
time  from  sunrise  to  2 p.m.,  and  again  about  an  hour  after  sunset. 
The  variations  of  daily  temperature  have  a great  effect  on  vege- 
tation ; from  my  own  notes,  they  are,  in  the  midland  counties,  the 
least  in  November,  December,  and  January;  the  mean  difference 
betw'een  the  maximum  and  minimum  being  for  January,  7^;  De- 
cember, 6°;  and  November,  5°  30'.  The  mean  difference  is  the 
greatest  in  March,  April,  and  September;  the  mean  variation 
being  for  March,  20°  ; for  April,  18°;  and  for  September,  19°; 
the  remaining  months  being,  August,  16°  ; May,  15°;  June,  12°; 
October,  12°;  and  February,  9°.  The  decrease  in  heat  during 
the  night  being  evidently  owing  to  the  radiation  of  heat  from  the 
earth,  it  is  much  the  greatest  in  cloudless  and  windy  weather, 
these  being  favourable  to  the  radiation  of  heat.  In  cloudy  weather 
the  radiation  of  heat  is  checked,  and  the  heat  is  radiated  back  by 
the  clouds,  so  that  the  difference  between  the  maximum  and 
minimum  temperature  of  very  cloudy  days  and  nights  is  fre- 
quently not  more  than  3°  or  4°.  Taken  in  this  view,  the  distri- 
bution of  heat  through  the  day  is  of  a very  fluctuating  kind,  yet 
there  is  a surprising  regularity  in  its  general  laws.  When  I come 
to  speak  of  the  distribution  of  vapour,  it  will  be  made  clearly  to 
appear  that  February,  March,  April,  May,  and  June  are  the 
most  free  from  vapour,  clouds,  See.,  and  it  will  be  seen  at  once 
from  the  above  that  the  average  variation  of  temperature  is 
greatest  during  these  months  when  compared  with  others  of 
equal  temperature.  November  and  December  are  the  most 
foggy  months,  and  January  the  coldest ; and  we  find,  on  referring 
to  the  above  averages,  that  the  variation  of  temperature  is  least 
during  these  months.  Other  things  being  equal,  the  more 
aqueous  vapour  there  is  in  the  air,  the  less  will  be  the  difference 
between  the  temperature  of  the  day  and  night.  Westerly  winds 
also  cause  a less  variation  than  easterly,  hence  in  May,  1848,  with 
easterly  winds  and  clear  weather  the  difference  averaged  nearly 
36°  during  the  month.  On  the  western  parts  of  the  British 
Islands,  especially  south-west  of  the  line  of  41°  winter-tempera- 
ture, the  variation  between  the  temperature  of  the  day  and  night 
is  much  less  (in  many  places  by  one  half)  than  in  the  midland 
counties,  whilst  in  the  eastern  counties  the  average  difference  is 
somewhat  greater.  Thus  we  may  infer  another  general  law  in- 
