But this is not all. Bodies sufficiently heated become luminous. According to the theory, this means that the molecules of matter, in their turn, communicate their vibrations to the ether; and here again we should find the influence of isochronism. The ether, it is true, is susceptible of vibrations of any velocity within certain very wide limits; but the molecules can give it none which are not isochronous with their own. Let us see what will result. Evidently, that the light which is emitted will, when developed into a spectrum, be concentrated in brilliant lines at those points where the velocities of undulation are the same as those of which the gas is capable; and, further, these lines should also evidently be in the same places, as the dark lines which this gas produces, as explained above, in a continuous spectrum, by absorption. This actually takes place in most cases, but some exceptions must be expected; because variations of temperature and pressure change the mutual connections of the gaseous molecules, and hence should also change the velocities of their oscillations. Thus, it is often found that the same gases change their systems of brilliant lines as their temperature or pressure changes; and Mr. Frankland has even obtained gases giving continuous spectra, sometimes attaining this result by pressure alone. The influence of heat also explains why solid or liquid bodies, when incandescent, give continuous spectra; while, at a low temperature, their interposition produces an elective absorption. For it is known that transparent solids or liquids become opaque when heated sufficiently to shine; the reason apparently being that, like the ether, they are capable of vibrations of any degree of rapidity within the usual limits, and hence allow no ethereal ones—or, in other words, no light—to pass through them, but absorb them all. Most flames or incandescent vapors, on the contrary, do not entirely lose their transparency. This property is of inestimable value in our investigations of nature.
Gases, by the combination of their elective absorption with their equally elective emission, produce results which at first sight might appear singular, but which can now readily be explained. Suppose that a flame is situated on the path of some rays which, without this interposition, would give a brilliant continuous spectrum. This flame only absorbs the ray having vibrations isochronous with its own; on the other hand, it emits rays similar to those which it absorbs. The resulting spectrum will vary according to the relative intensity of the emitted and absorbed rays. If these two intensities are equal, the spectrum will remain continuous; but if the absorption predominates, there will be dark lines in it; if the emission, brilliant ones. Similar phenomena of reversal have been often met with in the recent examinations of different parts of the sun.
The principles just explained have been known for several years, and were sufficient for astronomy as long as it restricted its investigations to the chemical analysis of the atmospheres of the heavenly bodies. But it was soon perceived that much greater use could be made of the spectroscope. Information is now beginning to be acquired by means of it which had previously appeared to be unattainable, regarding, for instance, the rapidity of the motion of stars the distance of which is still unknown; the great movements which are continually taking place in the great masses of gas in the solar photosphere, and the pressure of these masses at different depths; and it is even hoped that a direct determination of their temperature may be made. Let us speak first of the observations of stellar velocities. Their possibility may easily be shown by means of an acoustic phenomenon which the reader must frequently have noticed. Let us suppose two trains of cars to be moving rapidly in opposite directions, and that one of them whistles as it passes the other. If we are seated in the latter, we shall perceive that the pitch of the whistle suddenly falls as it passes us. The reason is manifest. A certain time is necessary for the sound to reach us; and while the train is approaching, this time is sensibly shorter for each succeeding vibration, so that the interval between the vibrations is apparently diminished, and the note is higher than it would be were the trains at rest. On the other hand, as the whistle recedes after passing, its pitch is lowered for a similar reason. Of course, no such effect is produced by that of our own train, which always remains at the same distance from us. By the amount of flattening of the sound, it is quite possible to calculate the velocity of the train, as compared with that of sound. [Footnote 198]
[Footnote 198: Suppose the sum of the velocities of the trains to be one-ninth of that of sound, and that the whistle is, at a given moment, 1140 feet (which is about the distance travelled by sound in a second) from our ear. The vibrations emitted at this instant will reach us in one second; and all those emitted in the nine seconds required for the train to arrive will be condensed into the remaining eight. Their frequency will then be nine-eighths of what it would be without the motion. It will be diminished in nearly the same ratio after the passage; since the vibration emitted nine seconds afterward will require an additional second to reach us; thus, the frequency will now be nine-tenths of what it would be without the motion, or four-fifths of what it was before meeting; corresponding to a flattening of two whole musical tones. This would require a relative velocity of 127 feet a second, or 87 miles an hour; which gives the rule, that, for every half-tone of flattening, the sum of the velocities, or the velocity of the moving train, if we are at rest, is 22 miles an hour.]
It is very easy to apply what has just been said of the waves of sound to those of light. The motion of the sonorous body displaces its sounds on the acoustic scale; in the same way, the motion of the luminous body will displace its light on the optic, placing any particular line, dark or brilliant, in the spectrum nearer to the violet or rapid end, if the body is approaching; and nearer to the red, if it is receding. And we are not obliged to wait till the change has taken place in the character of the motion, as in the case of the train, since we can always obtain lines similar to those thus displaced, and having the same velocity of vibration, from some terrestrial substance, relatively at rest, and put the two side by side in the same field; and by this means we obtain at once the difference between the apparent number of vibrations in a second of the ray from the moving body, and the real number, and thus the velocity of the moving object. This observation has the advantage of being independent of the distance of the objects observed, being as accurate for the most distant stars as for the nearest. We may notice, in passing, also a singular consequence. If the motion were rapid enough, it would change the colors of objects; and, since outside the visible spectrum there are dark rays, it would even be possible for a luminous body to become invisible, by the mere effect of movement away from or to us. But the prodigious velocity of light places such a result among mere metaphysical possibilities. Indeed, it was thought, for a time, that the effect of motion on the spectral lines would never be perceptible. The first trials only gave negative results, either because the bodies observed were moving too slowly, or because the instruments used were not sensitive enough. This is no longer the case, as we shall soon see.
To conclude this explanation of principles, it only remains to say a few words on the spectroscopic observations of temperature and pressure. But here we shall indeed be obliged to be brief; since Messrs. Frankland and Lockyer, who have undertaken investigations on these important points, have not yet finished their labors; and what they have as yet communicated to the Royal Society of London, and the Academy of Sciences of Paris, is not sufficiently detailed. In 1864, Messrs. Plücker and Hittorf discovered that variations in temperature of some of the chemical elements, such as hydrogen, nitrogen, sulphur, and selenium, caused sudden changes in their spectra. At a certain degree of heat, their former lines instantly disappeared and were succeeded by new ones. This is evidently somewhat analogous to what takes place in a sonorous pipe when it is blown more forcibly. At first, the sound only becomes louder, then its pitch is suddenly raised. But here we know the relation of the new note to the old one; but the connection between the successive spectra has not yet been ascertained. As regards pressure, Messrs. Frankland and Lockyer inform us that one of the lines of hydrogen increases in breadth with increased compression of the gas. We have also already said that under very high pressures the gases have not only shown broader bright lines, but even continuous spectra. (It will be remembered that the usual spectrum given by a luminous gas consists of isolated bright lines.) Father Secchi, whose attention has lately been turned to composite rather than to simple substances, has observed, among other things, that the spectrum of benzine vapor is gradually modified with a gradual increase of density.
Let us pass to the recent applications which astronomers have made of these various principles. The eclipse of the 18th of August, 1868, and the beautiful discovery of M. Janssen, have naturally turned their attention to the sun, and some most interesting discoveries have been made. To study its various portions, an image of it is first produced in the focus of a large telescope, which image is afterward enlarged by a lens similar to those used for the objectives of microscopes; and its different parts are successively placed upon the slit of the spectroscope. (The slit is the small aperture of that shape through which the light enters before falling upon the analyzing prism.) This slit thus receives light from only a part of the sun's disc; for the light diffused in our atmosphere and falling upon it, although coming indeed from all parts of the sun, is too feeble to interfere with the observations. Suppose, then, that our eye is at the spectroscope, and that the slit is receiving rays from the centre of the sun. The movement of the heavens will bring all the points of the solar radius successively upon it, from the centre to the edge; and if the slit is placed perpendicular to this radius, it will come out, of course, tangent to the edge. Under these conditions, and if the atmosphere is steady, the phenomena will be as follows.
As long as we are upon the disc, we shall see nothing but the usual solar spectrum with its colors and its numerous dark lines. The region from which this light comes is called the photosphere; and its spectrum would be continuous were not its light absorbed by the interposed vapors of a great many substances. These vapors produce the dark lines; but where are they? It was for a long time supposed that they formed an immense atmosphere round the sun, only visible during total eclipses under the form of a brilliant aureola. This hypothesis seems now to have been abandoned, for reasons which will soon be given. It is generally thought that these absorbing vapors form the atmosphere in which the luminous clouds float, or, at least, that they are in immediate contact with the photosphere.
Secondly, when we have nearly arrived at the edge, the spectrum is covered with a number of bright lines. According to Messrs. Frankland and Lockyer, these probably indicate a very thin gaseous covering of the photosphere, the elective emission of which has no effect for want of sufficient thickness, except upon the borders of the sun, where it is seen very obliquely. Upon the rest of the surface it only acts by its elective absorption, and perhaps may be the only cause of the dark lines. This conjecture certainly agrees with the principles just developed.
Thirdly, at the moment of passing off the disc, the lines all disappear, and the spectrum becomes continuous. Father Secchi, who informs us of this fact, naturally ascribes it to a particular layer enveloping the photosphere. He adds that this layer is very thin, so that tremulousness in the air suffices to prevent its observation, on account of the mixture of lights. It is not found on the whole circumference of the disc; but we shall give an explanation of this. He supposes that it is the seat of the elective absorption which produces the dark lines; but how can this be reconciled with the continuity of the spectrum which it emits?