Professor Challis points this out very conclusively in the Phil. Mag. of 1859, where, writing on this point, he states: “It may also be remarked, that if the Law of Gravity be absolute, there is no security for the stability of a system of stars, whether the system be a Milky Way or a nebulous cluster. For, however small the mutual attraction between the constituent bodies may be, in the course of ages it must produce a general movement towards the centre or densest region. But the form of the Milky Way and of certain nebulae seems to present an utter contradiction to any such tendency.” With the conception, however, of a physical centrifugal force or motion due to the pressure of a physical medium, the stability of even the Milky Way may be physically conceived and understood.

Again, when we consider the sun as a star, we find that it has two motions of its own, one of rotation on an axis, and the other of translation in an orbit, such rotation being due to the fact that it is a magnet and has ever circulating round it electro-magnetic Aether currents ([Art. 91]). By inference, therefore, we arrive at the fact that every star is a magnet, as suggested by Professor Schuster, and possesses rotation on an axis, such rotation being due to exactly the same cause as produces the rotation of any other planetary or solar body ([Art. 92]). Not only has each star a rotation on its axis, but it must also possess translational motion in an orbit, and that orbital motion must be due to exactly a similar cause as that which produces the orbital motion of the sun. Are there any indications given by astronomical observations which lead us to the conclusion that stars do possess such orbital motions? The answer is unanimously in the affirmative; for, although all the stars and the constellations retain apparently the same relative position to each other, yet they are all in motion. The actual translational motion of the stars is termed proper motion, and has been calculated with more or less success in relation to many of the stars nearest to us. There are other motions of the stars known as apparent motions, which are easily noted by any observer. These apparent motions are due to the rotation of the earth on its axis, and its orbital motion round the sun.

Nothing is more certain, however, than that careful astronomical observations have revealed the fact that stars have actual orbital motions of their own through space. In many cases the orbital velocity has been approximately ascertained.

Halley discovered proper motions of certain stars as far back as 1715, when he found out, by comparing different observations, that Sirius, Arcturus, and Aldebaran had moved during the period which had elapsed since the respective observations were taken.

More recent observations tend to confirm the fact that stars have indeed proper motions, due to their actual translation through space. It has been ascertained, for example, that Arcturus is travelling at least 54 miles per second.

The proper motion of the stars, however, only gives us an indication of their relative motion through so-called space. It does not tell us whether the star is apparently receding from the earth, or approaching it.

Dr. Vogel has ascertained by a special system of photography in relation to the spectra of stars, that Rigel has a velocity away from the earth of nearly 39 miles per sec., Aldebaran of 30 miles per sec., and Capella of 15 miles per sec., while the Pole star is apparently approaching the earth at a rate of nearly 16 miles per sec.

Now if all the stars move through space with varying velocities, as spectroscopic and telescopic observations seem to suggest, the question naturally confronts us as to what is the particular kind of orbit which each star completes? Is the orbit that of an ellipse, or a circle, or a parabola?

That it must have some kind of orbit is obvious from the proper motions exhibited by the several stars. We have already learned from [Arts. 107] and [108] that the sun possesses an orbit, which orbit fulfils the first and second of Kepler's Laws.

If therefore the sun, as representing all stars, is subject to Kepler's Laws, then, according to our Second Rule of Philosophy by which we base our hypotheses on our experience, we are compelled to come to the conclusion that every star which possesses any motion at all through space must also be subject to Kepler's Laws, and therefore must each possess a controlling centre around which they severally revolve. Kepler himself was of the opinion that the stars were subject to the laws which go by his name, and this view of the subject was also accepted by Sir William Herschel.