482 
THE  ADMIEALTY  ASTEOXOMICAL  EXPEEBIEXT 
ing  through  the  whole  plane  of  the  stratum,  and  so  experiencing  the  maximum  of  its 
light-stopping  effect. 
Hence  the  occasional  deterioration  of  sunrise  and  sunset  were  infinitely  greater  than 
anything  that  occurred  at  noon;  and  on  some  days,  when  the  sky  was  perfectly  free 
fr’om  cloud,  and  the  sun  had  been  distressiugly  hot  and  bifight  when  high  in  the  sky. 
yet  it  had  almost  become  invisible  before  it  set.  It  was  seen,  though  made  out  with 
difficulty  on  such  occasions,  through  a darkling,  yet  luminous  haze  of  dull  lemon-yellow 
colour ; but  what  it  set  behind,  or  when  exactly  it  did  set,  there  was  no  ascertainiug. 
The  next  evening  perhaps  the  atmospheric  dust  had  removed,  and  the  change  in  the 
sunset  was  magical.  The  orb  radiated  hot  and  shone  bright  up  to  the  moment  of  going 
down,  sometimes  behind  Palma,  showing  hills  of  rich  dark  pur’ple ; sometimes  behind 
the  rollers  of  the  cloud  sea,  clearly  visible  to  the  extreme  r^erge  of  the  horizon.  Then, 
too,  in  place  of  the  uniform  yellow  colour  of  the  dusty  sunset,  the  most  gorgeous  scar- 
lets, yellows,  arrd  blues  took  its  place. 
To  eliminate  this  dusty  medium  would  be  of  the  utmost  importance  to  the  further 
improvement  of  astronomical  observation,  and  may  be  considered  the  greatest  and  most 
subtile  difficulty  which  the  observer  has  to  deal  with ; and  it  is  probably  general  over 
the  world,  as  on  the  South  African  mountains,  at  heights  of  5600  feet,  the  phenomenon 
was  almost  as  notable  as  on  Guajara.  From  Dr.  Mason’s  observations  of  solar  radiation 
in  Madeira,  and  from  the  relations  given  to  me  by  inhabitants  of  TerrerifFe  as  to  the 
periods  of  the  year  when  the  Peak  is  seerr  most  clearly,  I am  disposed  to  think  that 
there  is  least  of  this  dust  in  the  atmosphere  in  the  latter  errd  of  the  winter  and  the 
earlier  part  of  the  spring.  The  latter,  if  not  the  former  also,  would  probably  be  a 
practicable  time  for  mountain  observation ; for  orre  of  the  most  remarkable  results  of 
our  meteorological  inquiries,  is  the  indication  of  the  seasorrs  bemg  rreariy  two  or  three 
months  earlier  at  a height  of  10,000  feet  than  at  the  sea  level ; so  that  the  difference  of 
temperature  in  the  spring,  betweerr  the  two  zones,  is  by  no  nrearrs  so  great  as  in  autumn. 
CHAPTER  III. 
ASTEONOMICAL  OBSEEVATIONS  PEOCUEED. 
(1.)  Double  Stars. 
The  Astronomical  Journal  contains  many  notices  of  double  stars,  but  as  they  were 
merely  for  the  purpose  of  testing  the  performance  of  the  instruments,  the  nature  of  the 
objects  was  always  assumed  to  be  known  (the  “ Cycle”  being  adopted  as  authority),  and 
our  aim  was  rather  to  take  an  eye  recognition  of  as  many  objects  as  possible,  than  to 
apply  rigorous  measure  to  a few.  Several  cases  of  change  in  magnitude,  distance,  and 
position  were  however  encountered,  and  are  given ; many  stars  were  also  observed  tor 
colour,  to  compare  with  observations  below.  These  will  be  found  duly  entered  in  the 
