390 M. MELLONI ON THE IDENTITY OF THE AGENTS 
am about to submit to the consideration of the Academy, seem to me to 
leave no doubt whatever that light and radiant heat are the direct ef- 
fects of two different causes. 
If we decompose a bundle of solar rays by means of a rock-salt 
prism, and measure the degree of heat proper to each band of the 
spectrum (proceeding from that band in which the refraction is greatest 
to that in which it is least), we find that the temperature increases from 
the violet to the red, and that even in the dark space on the other side 
this increase continues until it has reached a point midway between the 
red and the yellow. . At this point a pretty rapid decrease of tempe- 
rature takes place, and a total cessation of calorific action is perceptible 
in the obscure band, the distance between which and the red is about 
one third of the length of the luminous spectrum. 
It is known that the refraction of solar undulations is greater in pro- 
portion as they are shorter. In the obscure part of the spectrum, we 
have none but calorific waves, which become shorter and shorter as 
we approach more nearly to the red limit. When we enter the lumi- 
nous part, the shortening of the undulations still continues from the 
red to the violet. But it is to be recollected that according to the 
theory of identity each simple colour is the effect of an undulation 
which produces, at the same time and by the same species of vibration, 
both heat and light. 
Now if all the parts of the spectrum are made to pass through a 
layer of water (from two to three millimetres in thickness) contained 
between two plates of glass, and the temperatures of the emergent rays 
are ascertained, we shall find that the maximum of temperature and 
the last obscure limit are both close to the red limit. These effects 
will be more decidedly marked if the layer of water be thicker. With 
a layer of about four millimetres, the maximum will be found in the 
red band. If we continue to increase the thickness of the interposed 
layer, we shall find the maximum continually take the same direction 
and pass successively over the different parts of the red, the orange, 
and the yellow. When the rays have passed through a layer of 300 
millimetres, the maximum becomes stationary at the commencement of 
the green. 
The obscure limit is found in that case much more close to the red 
than it is in the normal spectrum ; but there is still an appreciable in- 
terval between the two, and this interval is necessarily greater when 
the thickness of the layer of water is from eight to ten millimetres ; 
whence we conclude that a portion of the obscure heat emitted by the 
sun penetrates through a considerable depth of this liquid, and passes, 
no doubt, through the watery humour of the eye into the retina, with- 
out exciting the sensation of light. 
But (to continue the explanation of the changes produced in the 
