of aluminium. The fluorescence in the walls of the vessel, when it was exhausted, showed that the cathode rays went out from every element of the cathode at right angles to it. By bending the cathode into an arc of a circle the cathode beams travelled over the surface of the vessel, forming zones of light the centres of which were in the bent wire. Is it not possible that by the electrostatic action the few molecules of air left in the high vacua are shot off with great velocity and bombard the walls of the vessel, and thus give rise to the fluorescent light, and also to an agitation of the molecules of matter outside the vessel? This may be called the molecular view of the phenomenon. I confess it is difficult to see why the molecular agitation is stopped by a thin sheet of glass and not by an inch of wood. It is certain that a few molecules must be left in the high vacua, for the cathode rays can not be formed in a perfect vacuum.

It is also true that it is useless to attempt to obtain photographs in any reasonable time from tubes which do not show a strongly marked cathode beam, or from tubes which on reversing the electric current through them do not show a marked difference between the light at the cathode and that at the anode. In poorly exhausted tubes one can perceive a faint appearance of a cathode beam, which is lost at a short distance from the cathode, as if the molecules which are shot off meet with such a crowd of more slowly moving ones that their energy is soon lost, and the cathode beam is quickly diffused like a beam of sunlight passing into milk and water. Thus the beam of cathode or X rays emerging from the glass vessel into the air is soon no longer conical in form. The sides of the cone of rays are no longer straight; they are curved, as if the gen

eratrix of the cone were a curved line instead of a straight line, and the beam is soon lost in a turbid medium. One can imagine a stream of projectiles being similarly dispersed in striving to pass into a region of sluggishly moving shot. This molecular view of the phenomenon seems at first sight to be a more tangible one than the longitudinal wave theory. Yet the amount of energy required by any corpuscular theory would seem to be enormous. It is possible, too, that the impact of the molecules on the aluminium window of Lenard, or on the glass sides of the vessel, may serve to start ripples, so to speak, in the ether, which are propagated with the velocity of light.

The Röntgen phenomenon seems to be a manifestation of cathode rays brought to light and endowed with great practical interest by its application to dryplate photography. When we return to the classical investigation of Lenard mentioned in the last chapter, we are impressed by his apparently crucial experiment which he describes in regard to the existence of an ether or medium. Energy passed into the vacuum he formed, and could be detected from point to point. We can conceive of its passing through the ether in the tube by a wave motion, but not by a molecular movement, for there were no molecules to move. molecular bombardment must have stopped at the aluminium window, and the resulting energy may have been propagated by ripples in the ether. This experiment of Lenard seems to me the most interesting one in the subject of cathode rays. The greatest mystery, however, which envelops the subject is the action of the X rays on bodies charged with electricity. When the rays fall on, for instance, a charged pith ball, the charge disappears. Prof. J. J. Thomson and Prof.

The

Rhigi have found that a positive as well as a negative charge is dispelled by the X rays. The energy of the medium about the pith ball is changed to a marked degree, and in this phenomenon we seem to be brought closer to a wave theory in a medium than to a molecular theory of movement of matter.

The tendency at present is to believe that the X rays are waves of ultra-violet light of much smaller dimensions than any that have been hitherto detected. D. A. Goldhammer* strongly advocates this view. Prof. Röntgen's reasons for believing that the new radiations discovered by him were not those of ultraviolet light were as follows:

a. The X rays suffer no refraction in passing from air to water, bisulphide of carbon, aluminium, rock salt, glass, zinc, etc.

b. They are not regularly reflected by known bodies.

c. They can not be polarized by known means.

d. The density of a body apparently influences their absorption more than that of any other factor.

If the X rays are very short transverse waves of light which are too small in comparison with unevenness of highly polished substances to be regularly reflected or polarized, b and c can be explained.

When we consider also the phenomenon of anomalous refraction and dispersion the behaviour of the socalled X rays is not so remarkable. Certain substances, like fuchsin and aniline, exhibit anomalous refractionthat is, a ray of blue light may be more refracted in passing through a solution of these substances than a ray of violet light; while with substances like glass,

* Annalen der Physik und Chemie, No. 4, 1896.

which exhibit normal refraction, the violet rays are more refracted than the blue

rays.

In certain cases of anomalous refraction and dispersion the amount of refraction (index of refraction) diminishes as the length of wave grows shorter. Goldhammer therefore concludes that a and c can be thus explained by anomalous refraction and dispersion, together with the hypothesis that the X rays are ordinary transverse vibrations of the ether, such as constitute ordinary ultra-violet light. The wave lengths of the X rays is, however, much smaller than those of any hitherto observed ultra-violet rays.

In 1867 M. Boussinesq presented a paper to the French Academy on the Théorie nouvelle des ondes lumineuses,* in which the effects of the momentum communicated to the molecules of matter by the ether are considered to be the cause of reflection, refraction, polarization, dispersion, etc. The ether is supposed to be homogeneous and of the same density and rigidity in all bodies, and that when light enters a transparent medium the molecules of that medium may be set in motion isochronously with the motion of the ether. Sellmeyer also, in 1872, adopted the hypothesis that the ponderable atoms vibrate, but with much smaller amplitudes than the ether particles. The theories of Boussinesq and Sellmeyer lead to expressions for indices of refraction in cases of anomalous refraction which bear upon the Xray phenomena. The electrical stress acting on the ether may probably serve to set the molecules of the fluorescent substances into their peculiar rates of vibra

tion.

* Glazebrook, Optical Theories, British Association Report, 1885.

CHAPTER XXI.

THE SUN.

In asking ourselves What is electricity? we are naturally led to inquire into the constitution of the sun, to which we owe our electrical energy. What, therefore, is the constitution of the sun? It seems strangely analogous to an enormous electrical furnace. If one gazes into an electrical furnace in which there is a mass of molten metal-silver, for instance-one perceives vapours shifting over the glistening mass. In this furnace carbon becomes freed from its impurities; the iron and sodium, for instance, are driven off in vapour, and the pure carbon lies in the heart of the furnace, surrounded by clouds of what once existed throughout its mass.

When one gazes at the spectrum of the sun one marvels at the mysterious arrangement of the dark lines which indicate the absorption of the vapour of some metal. According to the electro-magnetic theory of light, these dark lines represent an absorption of electric energy also. To what is due the electric energy which reaches us in electro-magnetic waves propagated through the infinite space between us and the farthest star?

The spectrum of the sun is like some ancient palimpsest, with inscription upon inscription laid upon each other. A photograph of it is a composite photo