284  M.  S.  Lamansky  on  the  Heat-Spectrum  of 
first  is  much  more  sharply  separated  from  the  second  than  the 
second  from  the  third.  It  may  easily  happen,  if  the  movement 
of  the  thermo-apparatus  be  not  sufficiently  delicate,  that  the 
second  and  third  appear  as  one  common  broad  break. 
Nevertheless  we  can  only  obtain,  a  correct  notion  of  the  nature 
of  these  bands  from  experiments  with  rock-salt  apparatus,  be- 
cause the  ultra-red  rays  are  strongly  absorbed  by  glass.  At  the 
end  of  August  and  in  September  last,  when  we  here  had  very 
fine,  hot,  sunny  days,  I  made  many  experiments  on  this  point. 
At  that  time  I  could  examine  the  heat-spectrum  of  the  sun  be- 
tween 7a.m.  and  1  p.m.  ;  and  I  took  care  to  make  two  parallel 
experiments  in  one  forenoon — one  in  the  morning,  and  the  other 
about  noon.  In  each  such  experiment  I  traced  the  thermal 
effects  from  the  line  D  to  the  ultra-red  end,  to  where  the  heat- 
effect  entirely  ceased  or  had  become  very  feeble. 
These  parallel  experiments  showed  that  the  breaks  become 
somewhat  narrower  with  increasing  altitude  of  the  sun.  They 
were  rather  deeper  on  the  days  when  the  relative  moisture  of  the 
air  was  greater.  But  the  observations  cited  are  at  all  events  not 
sufficient  to  permit  us  at  once  to  explain  them  as  atmospheric 
lines,  especially  when  we  take  into  consideration  that  the  appa- 
rent magnitude  of  these  breaks  may  possibly  have  depended  on 
the  variations  of  intensity  of  the  ultra-red  rays  at  different  hours 
of  the  forenoon.  In  order,  therefore,  to  decide  the  question 
whether  they  have  their  origin  in  our  atmosphere  or  in  that  of 
the  sun,  it  would  be  very  desirable,  in  the  first  place,  to  make  the 
experiments  on  the  solar  heat-spectrum  shortly  before  sunset — 
when,  it  is  well  known,  the  atmospheric  lines  of  the  luminous 
part  of  the  spectrum- come  out  more  distinct, — and  also  on  high 
mountain-ridges,  where  the  disturbing  influence  of  the  humid 
atmosphere  of  our  regions  is  almost  entirely  excluded. 
From  the  above-mentioned  parallel  experiments  with  rock-salt 
apparatus  it  clearly  follows  that  the  ultra-red  rays  of  the  atmo- 
sphere are  strongly  absorbed.  In  this  absorption  of  the  ultra- 
red  rays  lies  the  reason  that  the  maximum  of  thermal  effect  in 
the  solar  spectrum  changes  its  position  with  increasing  altitude 
of  the  sun.  In  all  the  experiments  which  took  place  between  7 
and  10  a.m.  the  maximum  was  after  the  first  break,  and  was 
almost  exactly  as  far  from  the  line  D  as  this  latter  from  the 
line  F.  On  the  contrary,  in  many  of  the  experiments  made  to- 
wards noon  the  thermal  effect  after  the  last  break  was  quite  as 
great  as  that  maximum,  or  even  exceeded  it ;  and  this  was  ob- 
served with  peculiar  distinctness  in  an  experiment  which  took 
place  on  a  cool  day  in  October.  It  must  therefore  be  admitted 
that,  properly,  here  (after  the  last  break)  is  the  place  where  in 
the  solar  spectrum  the  thermal  effect  first  attains  its  maximum. 
