3,26 
NATORE 
[| August 6, 1885 
reflected image of this lighted aperture thrown by means 
of the magnet-mirror, already described, upon the scale. 
This image may be made to move over a large space of 
the scale for a comparatively small motion of the mirror. 
If the image be that of a round circle of light, with a wire 
in the centre, we shall be easily able to read the position 
on the scale of the image of the wire, and by this means 
measure the motions of the mirror with very great accuracy 
and delicacy. 
Having thus described in detail the various arrange- 
ments tending to make the thermopile and galvanometer 
a very delicate instrument for measuring radiant heat, let 
us proceed to discuss the reply which this combination 
gives to the question raised. 
Melloni, who it must be remembered did not work with 
the instrument in its most perfect form, soon began to find 
that very many of those substances which were transpa- 
rent for light, were, on the contrary, nearly opaque for 
dark heat. As he continued his labours he had, however, 
the satisfaction of finding a substance that was as nearly 
as possible equally transparent for both—this substance 
being crystallized rock salt. 
He next found that just as by placing together two 
screens of coloured glass, one of which absorbs the redder 
portion of white light, while the other absorbs all but the 
redder portion, we may virtually stop all the radiation, so 
by certain combinations of screens it was equally possible 
to stop all the radiation from a source of low temperature 
heat. In the one case the result was perceived by the eye, 
and in the other by the thermopile. By this means he 
found that green glass and alum formed a peculiarly 
opaque combination. He next tried the same combina- 
tion for the solar rays, and found that when they were first 
intercepted by a screen of green glass, they had a very 
feeble power of passing through a second screen of alum. 
The similarity in the behaviour of the rays from these two 
sources led him to imagine that heat accompanied with 
light and low temperature heat, are not physically 
dissimilar. 
Again, the discovery by Melloni of the dathermancy or 
transparency for heat of rock-salt, led him to construct 
prisms and lenses of this material. By these means he 
proved that dark heat is capable of refraction, thereby 
exhibiting another bond of similarity between it and light. 
The subject was afterwards taken up by Forbes, who 
showed that the refrangibility of dark heat is inferior to 
that of the luminous rays. Forbes likewise showed that 
dark heat is capable of polarization and depolarization, 
and more recently other observers have shown that all the 
various properties of light may be exhibited at will in 
similar experiments with dark heat. 
We have thus strong evidence for believing that dark 
heat is similar to light, the difference between them being 
physiological rather than physical, or, to speak more 
exactly, rays of dark heat may be presumed to differ 
from one another and from rays of light in no other respect 
than that in which the various rays of light differ from 
each other. 
length or refrangibility, this being of such a nature that 
dark heat is less refrangible and has greater wave-length 
than light. 
It is desirable at this stage to say a few words about 
the spectrum which is obtained from ‘a luminous source 
by means of a prism. 
Let us suppose, for the sake of simplicity, that the 
luminous source is a thread or slit of light, and that, by 
means of a lens, after the manner of a photographer, we 
wish to obtain an image of this slit of light and throw it 
upon a white screen. This image will appear as a white 
luminous slit of light. If, however, we interpose a prism 
betwcen the source of light and the screen we shall obtain 
a very different result. In the first place the rays will be 
much bent by the prism, so that the screen will have to be 
placed in a very different position in order to receive the 
In fine, the only “difference is one of wave | 
image of the slit. In the next place all the rays which 
go to constitute the light will not be bent to the same 
extent, so that the image of the slit given by one con- 
stituent ray will be thrown upon a different portion of the 
screen from that given by another. Thus the red rays 
will be least bent, then the orange, the yellow, the green, 
the blue, the indigo, and the violet, these last forming the 
most refrangible of the rays that enter into the composi- 
tion of white light. What we shall really have, therefore, 
will be a great number of images of the slit placed side 
by side without any interval between them, these images 
being red at the one extremity and violet at the other. 
We shall, in other words, be presented with a long parti- 
coloured ribbon instead of a single white slit. This ribbon 
forms what is known as the sfectrum of white light, and 
if before it is thrown upon the screen it be reflected from 
a plane mirror we may easily show, by making the mirror 
oscillate rapidly backwards and forwards, that this ribbon 
when in motion reconstitutes itself into a colourless white. 
The spectrum has various properties. Part of it can 
affect the eye—we say part of it, for there are dark rays at 
either extremity which the eye cannot perceive. 
Part of it can perform certain chemical changes. Here 
again we say part of it, because there are certain chemical 
changes which certain rays seem at first sight incapable 
of producing. 
All of it is, however, capable of heating a substance 
upon which it falls and by which it can beabsorbed. We 
have thus three effects—the luminous, the actinic, and the 
heating effects ; and certain portions of the spectrum are 
capable of exhibiting all the three. 
If we take the action of the rays in blackening chloride 
of silver as a type of actinic influence we shall find that 
the maximum of the action is near, if not beyond, the 
most refrangible extremity of the visible spectrum. 
If we take the effect upon the eye as our measure of 
light we shall find that the maximum is at the yellow, 
whilst if we take the heating effect of the spectrum 
under its usual circumstances of production we shall 
find that this has a maximum near the least refrangible 
extremity. 
Now these considerations give rise to the following 
question: Is there only one thing present at one part of 
the spectrum, or are there three things ? 
At first it was imagined by some of the physicists of a 
past generation that there was in reality more than one 
thing and that the light and heating effects were produced 
by different agents. It was, however, afterwards found 
that if you operate on any portion of the spectrum by 
reflexion, absorption, polarisation, or in any other way, 
all the various qualities of that region are affected in the 
same proportion, so that if the light effect is reduced by 
one-half the actinic and heating effects are reduced by 
one-half likewise. 
This decides the question, for we cannot imagine two 
or three separate agents existing at the same place and 
each possessing exactly the same physical qualities as the 
other ; in other words, things which are not physically 
different from each other must be the same. 
Thus we have now come to the conclusion that there 
is only one physical entity at any one part of the spec- 
trum, and we have likewise been driven to see that in 
order to compare one part of the spectrum with another 
we must not use the eye, which has its own peculiarity, or 
some particular chemical substance which has likewise a 
partiality for certain rays. 
What we have to do is to measure the amount of heat- 
energy possessed by the various parts of the spectrum, 
and this is done by allowing the rays in question to fall 
upon a suitable thermo-pile covered with lamp-black 
and then measuring the amount of heat to which 
they give rise by means of the indication of the 
galvanometer attached to the pile. A coating of lamp- 
black absorbs most of the rays, and if it is not absolutely 
