A reader may suppose that in speaking of the immense density or massiveness of ether, and the absurdly small density or specific gravity of gross matter by comparison, I intend to signify that matter is a rarefaction of the ether. That, however, is not my intention. The view I advocate is that the ether is a perfect continuum, an absolute plenum, and that therefore no rarefaction is possible. The ether inside matter is just as dense as the ether outside, and no denser. A material unit—say an electron—is only a peculiarity or singularity of some kind in the ether itself, which is of perfectly uniform density everywhere. What we "sense" as matter is an aggregate or grouping of an enormous number of such units.

How then can we say that matter is millions of times rarer or less substantial than the ether of which it is essentially composed? Those who feel any difficulty here, should bethink themselves of what they mean by the average or aggregate density of any discontinuous system, such as a powder, or a gas, or a precipitate, or a snowstorm, or a cloud, or a milky way.

If it be urged that it is unfair to compare an obviously discrete assemblage like the stars, with an apparently continuous substance like air or lead,—the answer is that it is entirely and accurately fair; since air, and every other known form of matter, is essentially an aggregate of particles, and since it is always their average density that we mean. We do not even know for certain their individual atomic density.

The phrase "specific gravity or density of a powder" is ambiguous. It may mean the specific gravity of the dry powder as it lies, like snow; or it may mean the specific gravity of the particles of which it is composed, like ice.

So also with regard to the density of matter, we might mean the density of the fundamental material of which its units are made—which would be ether; or we might, and in practice do, mean the density of the aggregate lump which we can see and handle; that is to say, of water or iron or lead, as the case may be.

In saying that the density of matter is small,—I mean, of course, in the last, the usual, sense. In saying that the density of ether is great,—I mean that the actual stuff of which these highly porous aggregates are composed is of immense, of wellnigh incredible, density. It is only another way of saying that the ultimate units of matter are few and far between—i.e. that they are excessively small as compared with the distances between them; just as the planets of the solar system, or worlds in the sky, are few and far between,—the intervening distances being enormous as compared with the portions of space actually occupied by lumps of matter.

It may be noted that it is not unreasonable to argue that the density of a continuum is necessarily greater than the density of any disconnected aggregate: certainly of any assemblage whose particles are actually composed of the material of the continuum. Because the former is "all there," everywhere, without break or intermittence of any kind; while the latter has gaps in it,—it is here, and there, but not everywhere.

Indeed, this very argument was used long ago by that notable genius Robert Hooke, and I quote a passage which Professor Poynting has discovered in his collected posthumous works and kindly copied out for me:—

"As for matter, that I conceive in its essence to be immutable, and its essence being expatiation determinate, it cannot be altered in its quantity, either by condensation or rarefaction; that is, there cannot be more or less of that power or reality, whatever it be, within the same expatiation or content; but every equal expatiation contains, is filled, or is an equal quantity of materia; and the densest or heaviest, or most powerful body in the world contains no more materia than that which we conceive to be the rarest, thinnest, lightest, or least powerful body of all; as gold for instance, and Æther, or the substance that fills the cavity of an exhausted vessel, or cavity of the glass of a barometer above the quicksilver. Nay, as I shall afterwards prove, this cavity is more full, or a more dense body of Æther, in the common sense or acceptation of the word, than gold is of gold, bulk for bulk; and that because the one, viz. the mass of Æther, is all Æther: but the mass of gold, which we conceive, is not all gold; but there is an intermixture, and that vastly more than is commonly supposed, of Æther with it; so that vacuity, as it is commonly thought, or erroneously supposed, is a more dense body than the gold as gold. But if we consider the whole content of the one with that of the other, within the same or equal quantity of expatiation, then are they both equally containing the materia or body."—[From the Posthumous Works of Robert Hooke, M.D., F.R.S., 1705, pp. 171-2 (as copied in Memoir of Dalton, by Angus Smith).]

Newton's contemporaries did not excel in power of clear expression, as he himself did; but Professor Poynting interprets this singular attempt at utterance thus:—"All space is filled with equally dense materia. Gold fills only a small fraction of the space assigned to it, and yet has a big mass. How much greater must be the total mass filling that space."

The tacit assumption here made is that the particles of the aggregate are all composed of one and the same continuous substance,—practically that matter is made of ether; and that assumption, in Hooke's day, must have been only a speculation. But it is the kind of speculation which time is justifying, it is the kind of truth which we all feel to be in process of establishment now.[6]

We do not depend on that sort of argument, however; what we depend on is experimental measure of the mass, and mathematical estimate of the volume, of the electron. For calculation shows that however the mass be accounted for—whether electrostatically or magnetically, or hydrodynamically—the estimate of ratio of mass to effective volume can differ only in a numerical coefficient, and cannot differ as regards order of magnitude. The only way out of this conclusion would be the discovery that the negative electron is not the real or the main matter-unit, but is only a subsidiary ingredient; whereas the main mass is the more bulky positive charge. That last hypothesis however is at present too vague to be useful. Moreover, the mass of such a charge would in that case be unexplained, and would need a further step; which would probably land us in much the same sort of etherial density as is involved in the estimate which I have based on the more familiar and tractable negative electron. (See Appendix 2.)

It may be said why assume any finite density for the ether at all? Why not assume that, as it is infinitely continuous, so it is infinitely dense—whatever that may mean—and that all its properties are infinite? This might be possible were it not for the velocity of light. By transmitting waves at a finite and measurable speed, the ether has given itself away, and has let in all the possibilities of calculation and numerical statement. Its properties are thereby exhibited as essentially finite—however infinite the whole extent of it may turn out to be. Parenthetically we may remark that "gravitation" has not yet exhibited any similar kind of finite property; and that is why we know so little about it.

ETHERIAL ENERGY.

Instead then of saying that the density of the ether is great, the clearest mode of expression is to say that the density of matter is small. Just as we can say that the density of the visible cosmos is small, although in individual lumps its density is comparable to that of iron or rock.

At the risk of repetition, I have explained this over again, because it is a matter on which confusion may easily arise. The really important thing about ether is not so much its density, considered in itself, as the energy which that density necessarily involves, on any kinetic theory of its elasticity. For it is not impossible—however hopeless it may seem now—that a modicum of that energy may some day be partially utilised.

Lord Kelvin's incipient kinetic theory of elasticity is a complicated matter, and I will only briefly enter upon it. But before doing so, I want to remove an objection which is sometimes felt, as to the fluid and easily permeable character of a medium of this great density,—that is to say, as to the absence of friction or viscosity—the absence of resistance to bodies moving through it. As a matter of fact there is no necessary connexion whatever between density and viscosity.

'Density' and 'Viscosity' are entirely different things; and, if there is no fluid friction, a fluid may have any density you please without interposing any obstacle to constant velocity. To acceleration it does indeed oppose an obstacle, but that appears as essentially a part of the inertia or massiveness of the moving body. It contributes to its momentum; and, if the fluid is everywhere present, it is impossible to discriminate between, or to treat separately, that part of the inertia which belongs to the fluid displaced, and that part which belongs to the body moving through it,—except by theory.

As for the elasticity of the ether, that is ascertainable at once from the speed at which it transmits waves. That speed—the velocity of light—is accurately known, 3 × 10¹⁰ centimetres per second. And the ratio of the elasticity or rigidity to the density is equal to the square of this speed;—that is to say, the elasticity must be 9 × 10²⁰ times the density; or, in other words, 10³³ C.G.S. units. That is an immediate consequence of the estimate of density and the fact of the velocity of light; and if the density is admitted, the other cannot be contested.

But we must go on to ask, To what is this rigidity due? If the ether does not consist of parts, and if it is fluid, how can it possess the rigidity appropriate to a solid, so as to transmit transverse waves? To answer this we must fall back upon Lord Kelvin's kinetic theory of elasticity:—that it must be due to rotational motion—intimate fine-grained motion throughout the whole etherial region—motion not of the nature of locomotion, but circulation in closed curves, returning upon itself,—vortex motion of a kind far more finely grained than any waves of light or any atomic or even electronic structure.

Now if the elasticity of any medium is to be thus explained kinetically, it follows, as a necessary consequence, that the speed of this internal motion must be comparable to the speed of wave propagation;—that is to say that the internal squirming circulation, to which every part of the ether is subject, must be carried on with a velocity of the same order of magnitude as the velocity of light.

This is the theory then,—this theory of elasticity as dependent on motion,—which, in combination with the estimate of density, makes the internal energy of the ether so gigantic. For in every cubic millimetre of space we have, according to this view, a mass equivalent to what, if it were matter, we should call a thousand tons, circulating internally, every part of it, with a velocity comparable to the velocity of light, and therefore containing—stored away in that small region of space—an amount of energy of the order 10²⁹ ergs, or, what is the samething, 3 × 10¹¹ kilowatt centuries; which is otherwise expressible as equal to the energy of a million horse-power station working continuously for forty million years.

Summarised Brief Statements concerning
the Ether

(As communicated by the author to the British Association
at Leicester, 1907).

1. The theory that an electric charge must possess the equivalent of inertia was clearly established by J.J. Thomson in the Philosophical Magazine for April, 1881.

2. The discovery of masses smaller than atoms was made experimentally by J.J. Thomson, and communicated to Section A at Dover in 1899.

3. The thesis that the corpuscles so discovered consisted wholly of electric charges was sustained by many people, and was clinched by the experiments of Kaufmann in 1902.

4. The concentration of the ionic charge, required to give the observed corpuscular inertia, can be easily calculated; and consequently the size of the electric nucleus, or electron, is known.

5. The old perception that a magnetic field is kinetic has been developed by Kelvin, Heaviside, FitzGerald, Hicks, and Larmor, most of whom have treated it as a flow along magnetic lines; though it may also, perhaps equally well, be regarded as a flow perpendicular to them and along the Poynting vector. The former doctrine is sustained by Larmor, as in accordance with the principle of Least Action, and with the absolutely stationary character of the ether as a whole; the latter view appears to be more consistent with the theories of J.J. Thomson.

6. A charge in motion is well known to be surrounded by a magnetic field; and the energy of the motion can be expressed in terms of the energy of this concomitant field,—which again must be accounted as the kinetic energy of ethereous flow.

7. Putting these things together, and considering the ether as essentially incompressible—on the strength of the Cavendish electric experiment, the facts of gravitation, and the general idea of a connecting continuous medium—the author reckons that to deal with the ether dynamically it must be treated as having a density of the order 10¹² grammes per cubic centimetre. (See Appendix 2.)

8. The existence of transverse waves in the interior of a fluid can only be explained on gyrostatic principles, i.e. by the kinetic or rotational elasticity of Lord Kelvin. And the internal circulatory speed of the intrinsic motion of such a fluid must be comparable with the velocity with which such waves are transmitted.

9. Putting these things together, it follows that the intrinsic or constitutional vortex energy of the ether must be of the order 10³³ ergs per cubic centimetre.

Conclusion.—Thus every cubic millimetre of the universal ether of space must possess the equivalent of a thousand tons, and every part of it must be squirming internally with the velocity of light.