On  the  Heat-Spectrum  of  the  Sun  and  the  Lime-Light.      283 
of  sulphide  of  carbon,  and  with  rock-salt  apparatus.  I  will  here 
mention  that  these  apparatus  were  of  the  same  dimensions  as 
the  above-mentioned  glass  ones,  were  of  perfectly  transparent 
Stassfurth  salt,  and  excellently  made  by  W.  Steeg,  optician,  of 
Homburg;  and  I  fresh-polished  them  before  each  experiment. 
My  experiments  have  reference  to  the  forenoons  in  the  summer 
and  autumn  of  the  past  and  the  present  year,  and  were  only 
made  with  a  cloudless  sky,  as  the  slightest  cloud  occasioned  a 
marked  difference  in  the  deflections. 
In  all  such  experiments,  if  our  observations,  commencing  at 
the  line  D,  advance  to  the  ultra-red  end  of  the  spectrum,  the  de- 
flections, answering  to  the  heat-effects,  become  gradually,  but 
not  proportionately,  stronger,  until  they  attain  a  certain  maxi- 
mum, and  then  diminish;  and  this  takes  place  four  times.  We 
therefore  see  here  a  discontinuous  distribution  of  heat  in  the 
solar  spectrum  ;  namely,  the  ultra-red  rays  are  interrupted  in 
three  places  by  breaks  or  bands. 
This  want  of  continuity  was  proved  by  Sir  John  Herschel  in 
the  following  manner  (Phil.  Trans.  1840).  By  means  of  a  flint- 
glass  prism  he  threw  a  spectrum  upon  paper  blackened  with  soot 
and  moistened  with  alcohol ;  and  by  the  time  of  drying  he  de- 
termined the  thermal  effect  of  the  spectrum.  He  at  the  same 
time  observed  that  the  moistened  surface  dried  in  a  series  of  four 
distinctly  separate  spots.  HerschePs  chief  concern,  however, 
was  merely  to  ascertain  the  conditions  under  which  these  four 
spots  made  their  appearance.  He  observed  that  in  the  spectrum 
of  a  crown-glass  prism  they  were  less  distinctly  separated  from 
each  other,  and  that  in  the  investigation  of  the  solar  spectrum 
with  a  water  prism  they  were  only  feebly  expressed. 
The  existence  of  such  bands  in  the  ultra-red  rays  was  after- 
wards noticed  by  Fizeau  andFoucault  (Comptes  Rendus, vol. xxy.), 
in  their  well-known  experiments  on  the  interference  of  the  heat- 
rays.  I  know  of  no  other  observations  on  these  bands  in  the 
solar  spectrum.  At  least,  all  the  philosophers  who  have  hitherto 
investigated  the  distribution  of  heat  in  the  solar  spectrum  with 
glass  prisms,  as  well  as  with  rock-salt  apparatus,  make  no  men- 
tion of  them,  and  still  as  previously  delineate  the  heat-curve  as 
continuous. 
These  bands  can  be  distinctly  observed  with  all  three  of  the 
above-mentioned  prisms  j  only  the  spectrum  must  be  perfectly 
pure. 
They  have  a  corresponding  position  in  the  spectra  of  all  three 
prisms,  and  only  differ  by  being  broader  when  the  prism  used 
has  greater  dispersive  power,  as  sulphide  of  carbon,  than  when 
it  has  less,  as  rock  salt. 
These  three  breaks  or  bands  are  not  of  equal  breadth ;  the 
