Sept. 3, 1885 | 
NATURE 
423 
Now in a compound gas these collisions sometimes 
cause dissociation of the compound molecule into more 
elementary constituents, which constituents will probably 
afterwards combine again, so that we may imagine that 
in such a gas (see “ Heat,” by Prof. Tait, page 203) equi- 
librium is maintained by a constant amount of dis- 
sociation, accompanied by an equal amount of recombi- 
nation, It is thus apparent that we have not here perfect 
simplicity and uniformity of molecular structure, and 
without discussing the question whether a simple mole- 
cule might or might not be expected to vibrate in only 
one way, we can readily imagine that the spectrum of 
such a gas should present us with more than one mode of 
vibration ; that is to say, more than one spectral line. 
Again, circumstances which conduce to proximity of 
molecules, and to the action of molecules upon each other, 
tend to bring about a state of things similar to that which 
we have in liquids and solids; that is to say, they will 
favour the emission of various kinds of rays, while on 
the other hand, the characteristics of a gaseous spectrum 
will be best shown by a perfect gas, that is to say, by 
a gas which is far removed from any tendency to con- 
densation. A rare gas at a high temperature will possess 
these properties. 
Having now defined the characteristics of the spectra 
of solids, liquids, and gases, let me say a few words about 
the methods by which we obtain gaseous particles heated 
to a high temperature. These are obtained in two ways. 
First, by means of flames, such as that of a Bunsen’s 
burner, into which the particles are introduced. In such 
flames we may imagine that we have before us a certain 
number of the particles of a certain gas all, or nearly all, 
heated to a temperature somewhat approaching that of 
the flame. The substance will probably have been intro- 
duced into the flame in a different chemical state from 
that in which it appears in giving out the light; for in- 
stance, we may introduce into a spirit-lamp a little 
chloride of sodium, or into a Bunsen’s burner a little bi- 
carbonate of soda. The flame becomes immediately of a 
yellow nature, giving us the double line D, or the yellow 
line of incandescent sodium vapour, and this affords us 
evidence that dissociation has taken place. In like manner 
the red line produced by salts of Lithium, the green line 
produced by those of Thallium, and so on, are indications 
that the compound saline molecules have become disso- 
ciated in the flame. 
The second way of producing gaseous spectra is by an 
application of electricity, as when a high tension spark is 
sent through a tube containing a small quantity of a 
given gas, or a vacuum tube, as this is sometimes called. 
We have then a momentary flash, consisting of the rays 
which characterise the spectrum of the gas through which 
the discharge has passed. 
case only a portion of the particles filling the tube have 
been brought to the high temperature which is denoted 
by the discharge. 
Before proceeding further, it may be well to mention 
that while from the title of our subject we must neces- 
sarily consider the spectrum to some extent, yet this is 
not to be regarded as a treatise on the spectroscope and 
its applications, which formed the subject of a previous 
set of essays in the NATURE Series by Mr. Lockyer. We 
shall discuss the subject in a somewhat different manner, 
and also give more especial attention to those branches 
which had not yet been developed when Mr. Lockyer 
wrote his work. With these preliminary remarks, we 
shall divide the subject before us into two sections. 
(1) Radiation and its consequences. 
(2) Absorption and its consequences. 
_In the first we shall discuss radiant spectra to a con- 
siderable extent, but shall not entirely confine our remarks 
to these phenomena; while in the second we shall dis- 
cuss absorption spectra to a considerable extent, but 
shall not entirely confine ourselves to spectral absorption. 
It is probable that in this | 
There is likewise another convenient way of dividing 
our subject, namely, in its application to terrestrial and 
celestial phenomena. 
Combining, therefore, these two principles of sub- 
division, we shall, in the first place, treat of terrestrial 
applications of the laws of radiation and absorption, and 
in the next place of their celestial applications; and, 
finally, we shall discuss the light which both of these 
branches together appear to throw upon the ultimate 
constitution of matter. 
With regard to our own Earth, it is abundantly evident 
that the great bulk of the heat which it receives is from 
the radiation of the sun, while, on the other hand, the 
great bulk of the heat which it loses is through radiation 
into space. 
There is a sort of balance kept up between the gain on 
the one hand and the loss on the other, in virtue of which 
we are placed under conditions in which life is endurable, 
and for the most part pleasant. The variations in these 
conditions in temperate latitudes may sometimes cause 
distress to the weak, but they are not less the source of 
enjoyment and vigour to the strong ; and, as a matter of 
fact, the most energetic races of mankind are they which 
dwell in those favoured regions that are neither too cold 
nor too hot. 
Inasmuch as the regions near the equator are hotter 
than those near the poles, it follows that there is greater 
radiation into space from the former of these than from 
the latter. If, therefore, we could imagine an observer to 
be placed many thousand miles above the earth, having 
an eye capable of distinguishing dark rays, and to regard 
that portion of the earth unilluminated by the sun, his 
eye would receive more rays from the equatorial than 
from the polar regions. 
On the other hand, the polar regions being manifestly 
colder than those of the equator, we have convection 
currents of hot air passing in the upper atmospheric 
regions from the equator to the poles, and currents 
of cold air passing in the lower atmospheric regions 
from the poles to the equator. These latter are known as 
the Trade Winds, and the former as the Anti-Trades. In 
like manner we have in all probability currents of hot 
water passing in the upper oceanic regions from the 
equator to the poles, and currents of cold water passing 
in the lower oceanic regions from the poles to the equator. 
It is not, however, our object to dwell on these phenomena 
here; suffice it to say, that our well-being depends on 
the balance between the radiant heat which we receive 
from the sun and that which we give out into empty 
space. 
The phenomena of dew form an exceedingly good illus- 
tration of the laws of radiation. This subject was first 
investigated by Dr. Wells, an English physician. When 
the sun has sunk beneath the horizon of any place, bodies 
of small mass and great radiating power for dark heat, 
such as the leaves of plants, become quickly cooled by 
their uncompensated radiation into space. They thus 
cool the air around them, until this air becomes so cold 
that it can no longer retain in the viewless state the 
aqueous vapour which it holds; part of this is conse- 
quently deposited in the form of dew, or of hoar-frost, if 
the temperature be sufficiently low. 
The following are the laws which regulate the deposition 
of dew :— 
(1) Dew is most copiously deposited under a clear sky. 
(2) And with a calm state of the atmosphere. 
(3) It is most copiously deposited on those substances 
which have a clear view of the sky. 
(4) And which are good radiators and of small mass. 
(5) And which are placed close to the earth. 
The first of these conditions is essential, because the 
cooling which precedes the deposition of dew is owing to 
radiation into free space. 
If there are clouds, these will radiate back to the body, 
