the  Sun  and  the  Lime-Light.  285 
This,  in  the  rock -salt  spectrum,  is  just  as  far  from  the  line  D  as 
the  latter  from  G. 
Of  the  four  different  maxima  of  thermal  effect  observed  by  us 
in  the  solar  spectrum,  in  most  of  our  experiments  the  first  alone 
kept  its  position,  while  the  other  three  moved  towards  the  red 
as  the  sun's  altitude  increased;  consequently  the  first  break  ap- 
peared narrower  in  the  experiments  towards  noon  than  in  the 
morning  ones. 
It  is  very  probable  that  the  displacement  of  these  maxima 
resulted  from  a  change  in  the  refractive  power  of  the  rock-salt 
prism,  which,  in  the  experiments  towards  noon,  may  have  taken 
place  through  a  stroDg  heating  of  the  prism.  In  order  to  test 
this  supposition,  I  determined  the  minimum  deflection  for  the 
line  D  at  apartment-temperatures  of  5°  and  16°  C. ;  and  I  found 
that,  for  a  rock-salt  prism  of  60°  18'  refracting  angle,  the  de- 
flection was  about  2'  greater  at  16°  than  at  5°  C.  The  displace- 
ment of  the  maxima  in  my  experiments  is  rather  less  than  ^  mil- 
ling, which  corresponds  to  an  angle  that  can  be  obtained  by 
heating  the  rock-salt  prism  10°  C;  and  such  a  heating  could 
easily  take  place  in  my  noon  experiments,  since  the  temperature 
of  the  room  at  that  time  was  always  very  high. 
In  all  the  experiments  (which  were  made  with  the  three  prisms 
above  mentioned,  at  different  hours  of  the  forenoon,  and  at  dif- 
ferent seasons  of  the  year)  it  was  distinctly  seen  that  the  heat- 
effect  of  the  solar  spectrum,  having  attained  its  last  maximum 
(after  the  last  break),  suddenly  sinks.  This  was  especially 
distinct  in  the  experiments  made  about  noon  with  rock-salt 
apparatus.  I  will  further  particularly  mention  that,  in  these 
experiments,  neither  the  slit  which  was  placed  in  the  focus  of  the 
first  lens,  nor  the  double  slit  in  front  of  the  thermo-apparatus 
had  more  than  £  millim.  breadth;  further,  a  single  shifting  of 
the  thermo-apparatus  amounted  to  rather  less  than  \  millim. ; 
and  yet,  under  these  circumstances,  the  deflections  diminished 
one  half  at  the  place  after  the  last  maximum,  when  only  two 
such  shiftings  had  taken  place.  We  are  therefore  entitled  to 
ask,  Is  not  the  limit  of  refraction  situated  at  the  place  where  the 
heat-effect  of  the  solar  spectrum  attains  its  last  maximum  ?  and 
does  not  the  thermal  effect  observed  beyond  that  maximum  arise 
from  diffused  reflected  heat  ? 
In  order  to  exclude  the  effect  of  diffused  heat  in  the  investi- 
gation of  the  individual  parts  of  the  spectrum,  I  made  use  of  the 
method  proposed  by  Geh.  Hath  Helmholtz,  the  one  in  which  two 
prisms  are  employed,  and  by  which  he  made  the  ultra-violet 
rays  immediately  visible  to  the  eye  by  excluding  the  diffused 
light.  I  adopted  this  method  several  times  in  these  investiga- 
tions, especially  in  those  cases  in  which  it  was  necessary  to  sepa- 
rate the  luminous  from  the  obscure  heat. 
