Popular Science Monthly/Volume 77/August 1910/The Past and Present Status of the Ether




AUGUST, 1910




IN a recent letter to the New York Nation, Professor William James, in describing the philosophy of M. Emile Boutroux, makes the statement that "theories result from psychological variations, just as Roosevelts and Rockefellers result from biological variations." Of the entities of science he says:

The creative touch of human reason was needed in each case for the extrication; and that those particular creations resulted rather than a hundred others just as possible, is one of those selective interactions between living minds and their environment which can be "understood" when once it has occurred, but which no acquaintance with the previous conditions can show to an outsider that it was the sole thing possible.

Considering the prevalence of such philosophical views, and the fact that many persons believe that physics is now undergoing a sort of crisis, in which many of our most cherished ideas are about to be relegated to the scrap-heap, I believe it to be not without profit to consider the past and present condition of our views with regard to the luminiferous ether, and to cautiously forecast their future.

Certainly the postulate of the existence of the ether has been until very recently one of the fundamentals of physics (including astronomy). At the congresses of arts and sciences held at Si Louis in 1904, the subject of physics was, like all Gaul, divided into three parts, physics of matter, physics of ether, physics of the electron, and although this division was, I believe, not made by a physicist, this must have made little difference. In an interesting book published less than a year ago by Sir Oliver Lodge, entitled "The Ether of Space," the properties of the ether are set forth with a concreteness and dogmatic manner that is now becoming unfashionable, and relieves that writer from any suspicion of being called a pragmatist. For him the ether undoubtedly is a real thing. In a more ambitious treatise published ten years ago, by Sir Joseph Larmor, entitled "Ether and Matter," we have a thoroughgoing mathematical investigation of the properties of the ether, and, as the subtitle states, a development of the dynamical relations of the ether to material systems. And yet, since the publication of the latter work there have been voices heard with ever-growing distinctness, declaring in not dubious terms the lack of necessity of any such conception as that of the ether, and threatening the belief in its existence with relegation to the company of phlogiston in the morgue of dead theories. That we can not dismiss such voices with contempt is evident if among them are to be counted those of such leaders of physical science as Henri Poincaré, Sir J. J. Thomson and Professor Max Planck.

Before we can discuss the question of the existence of the ether, we must first determine what we mean by that term. This is undoubtedly the main difficulty with the whole matter. The article in the "Encyclopedia Britannica," written over thirty years ago by Maxwell, as competent an authority as could have been named at that time, begins with the definition, "a material substance of a more subtle kind than visible bodies, supposed to exist in those parts of space which are apparently empty," and ends with the statement, "Whatever difficulties we may have in forming a consistent idea of the constitution of the ether, there can be no doubt that the interplanetary and interstellar spaces are not empty, but are occupied by a material substance or body, which is certainly the largest, and probably the most uniform body of which we have any knowledge." This is certainly flat-footed enough, but how different from the conclusions of Lodge, one of the present survivors of the same school, we may see from his book above mentioned.

The need for the idea of an ether is well shown by the following quotation from Newton, who, after describing an experiment of two thermometers, one in a vessel filled with air and the other in vacuo, being carried from a cold place into a warm one, both rising at the same rate, says:

Is not the heat of the warm Room conveyed through the Vacuum by the Vibrations of a much subtiler Medium than Air, which after the Air was drawn out remained in the Vacuum? And is not this Medium the same with that Medium by which Light is transmitted, and by whose Vibrations Light communicates Heat to Bodies?

And yet Newton did not accept the wave theory, but by the influence of his great name bolstered up the emission theory for a hundred years. It was his contemporary Huygens, who must be credited with the invention of the ether in order to explain the propagation of light. Huygens's ideas of the properties of the ether were, however, very different from those that have now been held for a century. In order to cover all the different notions that have been held, without being so definite in making the ether a substance as was Maxwell, we need only ask the question, Since we know that light travels with a speed of about three hundred thousand kilometers per second, and takes about eight minutes to come from the sun, what is the state of the light after it has left the sun and before it has reached the earth? We reply, it is traveling through the ether. A similar definition was given by the late Lord Salisbury who said that the noun ether was the subject of the verb to undulate. But why undulations? The undulatory theory, as a successful explanation of optical phenomena, is just about a century old, and was propounded by Dr. Thomas Young, in two Bakerian lectures before the Royal Society in 1801 and 1803. The reason that convinced Young, and later the scientific world, of the undulatory nature of light, was the fact of interference, or the production of darkness by the simultaneous action of two beams of light, carefully investigated by Young. These views were savagely assailed by Lord Brougham, in a scurrilous article in the Edinburgh Review, in which he says that "it is a metaphysical absurdity, to assert that qualities can move in concentric surfaces." The violence of the attack may be seen from the quotation:

The long silence which he (Young) has since preserved on philosophical matters, led us to flatter ourselves, either that he had discontinued his fruitless chase after hypotheses, or that the Society had remitted his effusions to the more appropriate audience of both sexes which throngs around the chairs of the Royal Institution.

It is evident that Young had an excellent understanding of the analogy between sound and light waves, but he did not follow out the theory with the mathematical exactness bestowed upon it by Augustin Fresnel, whose superb researches, beginning in 1815, have made his name a classic of optical investigation. Both Young and Fresnel recognized, as Huygens had not, the fundamental difference in the nature of waves of light and sound, namely, that since by turning the proper apparatus traversed by light about the direction of the beam as an axis, the light is capable of alternate extinction and transmission, the undulations must be transverse to the direction of propagation. Fresnel introduced into his mathematical treatment certain mechanical principles, notably that one which we now call the conservation of energy, but he did not attempt to find a mechanical structure, in terms of properties of ordinary matter inertia and rigidity, which would explain the nature of the ether. This was done by George Green, who assimilated the ether to an elastic solid, which is capable of transmitting transverse waves in all directions with the same velocity. Unfortunately, such a solid transmits equally well longitudinal waves, like those of sound, but with a different velocity from that of transverse waves. But such longitudinal waves have no place in any optical phenomenon, and therefore constitute a difficulty for the theory. In order not to have them it was necessary for Green to suppose the ether incompressible. Thus the theory did very well for the propagation of light on free space. When light passes from free space to a transparent substance, however, it is partially reflected and partially refracted, travelling with a different velocity in the new medium. This change of velocity could be explained by a difference of either density or rigidity in the two media. Green chose one hypothesis, in fact the same as that of Fresnel, Neumann and McCullagh the other. This difference gave rise to a controversy over the direction of the vibration, as to whether it was in or perpendicular to the plane of polarization, a controversy vainly sought to be settled by experiment. Although reflection and refraction could thus be explained, there remained a very grave difficulty. The conditions to be satisfied at the surface between two different media are too many to be satisfied by a transverse wave alone, so that had there been originally only a transverse wave, it would give rise to a longitudinal wave on striking the surface limiting the media. To avoid this difficulty a mechanical theory was proposed by McCullagh, in which the elasticity was not like that found in any known substance, but was called into play when a portion of the medium was rotated, quite independently of whether neighboring portions were rotated or not. This theory gave a very satisfactory explanation of reflection and refraction, but long met with opposition on account of its postulating elastic properties not found in any substance.

Probably the person who took most seriously the view of the ether as having the properties of some familiar sort of matter was Lord Kelvin, who devoted a large portion of his life to the attempt to find a suitable mechanical representation of the ether. In fact he stated on the occasion of his jubilee that for forty years this question had not been absent from his mind for a single day. Lord Kelvin frequently uses the term "jelly" as typical of Green's elastic substance, and did finally, by a very ingenious assumption, succeed in assimilating the ether to such a substance. But in spite of all these attempts, we may agree with the opinion of Lord Rayleigh, who concludes that for many reasons "the elastic solid theory, valuable as a piece of purely dynamical reasoning, and probably not without mathematical analogy to the truth, can in optics be regarded only as an illustration."

Such was the condition of affairs at the close of what I may call the medieval period in optics, when, in 1864, Maxwell gave affairs an entirely new turn by the presentation of his famous paper on "A Dynamical Theory of the Electromagnetic Field." In this he was guided by the conjecture of Faraday that the same medium which is concerned in the propagation of light might also be the agent in electromagnetic phenomena. Faraday says:

For my own part, considering the relation of a vacuum to the magnetic force, and the general character of magnetic phenomena external to the magnet, I am much more inclined to the notion that in the transmission of the force there is such an action, external to the magnet, than that the effects are merely attraction and repulsion at a distance. Such an action may be a function of the ether, for if there be an ether, it should have other uses than simply the conveyance of radiation.

This expression of Faraday is the key-note of Maxwell's theory. In examining the properties of the medium necessary to transmit electric and magnetic forces, he concentrates his attention on two quantities having direction, namely, the magnetic and electric polarization of the medium at every point. He shows that these states of polarization are propagated in waves, and that these waves have all the properties of light-waves. They are transverse, no longitudinal wave occurs, and moreover for the first time the conditions at the surface of separation of two media are exactly sufficient to give the proper explanation of reflection and refraction. Everything accomplished by any undulatory theory was accomplished by the electromagnetic theory, with this in addition, so that it is perhaps surprising that it remained for the experimental production in 1888 by Hertz of undoubtedly electromagnetic waves having all the properties predicted by Maxwell to give this theory the overwhelming preponderance that it has since maintained.

We may now touch upon the question, what is a mechanical theory. A mechanical theory is one that can be stated in terms of the principles of mechanics. The laws of mechanics, as they have been held since their exact statement by Newton, are all embraced in the single mathematical principle of least action, best comprised in the enunciation of Hamilton. In this enunciation occur two functions representing the two forms of energy, kinetic and potential. If these depend in a certain simple manner on two quantities having direction, or vectors, irrespective of their physical nature, the differential equations follow, which lead to wave propagation. Maxwell's field vectors have this property, and consequently Maxwell's theory is a mechanical theory. I will now define the properties of the ether, as they seem to me to be required by our present-day notions. The ether connotes those properties of space in virtue of which a change in either of two field vectors at any point gives rise to a field of the other sort, the lines of which tend to symmetrically surround the lines of the original and varying vector in circles. In addition the direction of these surrounding lines is contrary according to the field that we begin with. This is a qualitative statement in plain English of what is quantitatively stated in the six differential equations of Maxwell's theory, and it avoids the use of the electromagnetic terminology. It thus applies exactly to Fitzgerald's and Larmor's resuscitation of McCullagh's rotational elastic theory, which is found to be identical with the electromagnetic theory.

I believe that I have thus given that definition of the ether which best agrees with what Boltzmann calls the phenomenological view in physics which attempts to exactly describe phenomena, without any hypothesis, or any attempt at mechanical model to assist the imagination. This was the view of Kirchhoff, Helmholtz, Hertz and Boltzmann, and I believe it to be the most scientific. The English method, of which Lord Kelvin was the leading example, demands concrete models, which resemble the phenomena more or less, and which are frequently changed. In the words of an acute French critic, M. Duhem, for a geometer of the school of Laplace or Ampère, it would be absurd to give for the same law two theoretical explanations and to maintain that the two explanations hold simultaneously; for a physicist of the school of Kelvin or Maxwell, there is no contradiction in the same law being represented by two different models. I may also quote Fitzgerald's words:

I can not conclude without protesting strongly against Sir William Thomson's speaking of the ether as like a jelly. It is in some respects analogous to one, but we certainly know a great deal too little about it to say that it is like one. I also think that Sir William Thomson, notwithstanding his guarded statements on the subject, is lending his overwhelming authority to a view of the ether which is not justified by our present knowledge, and which may lead to the same unfortunate results in delaying the progress of science as arose from Sir Isaac Newton's equally guarded advocacy of the corpuscular theory of optics.

I feel that this protest is a very mild one, and that the attempt made by Kelvin to determine the density and elasticity of the ether, from very questionable assumptions, together with the recent attempts of Lodge, based on equally naive conceptions of the nature of the ether as a concrete substance, are greatly to be deplored.

We come now to the most modern development of the ether theory. Maxwell had, as has been said, accurately described the propagation of the electromagnetic waves, and had given the differential equations governing their propagation. It remained to add to these equations terms expressing the genesis of the waves, to show how these resulted from the motion of charges of electricity. This was done in an important series of papers begun in 1892 and continued until the present by H. A. Lorentz, who may be characterized as the legitimate successor of Maxwell. Not only did Lorentz add terms shown to be necessary by the experiments of Rowland on the magnetic effect of moving electric charges, and later by the deflection of the cathode rays by a magnet, but he succeeded in showing for the first time how the potentials determining the field were propagated in time through the field, a result vainly sought by Gauss, Weber and Riemann, and almost reached by the latter. The object of Lorentz in his papers was to explain the transmission of waves in moving media, beginning with the explanation of astronomical aberration. Singularly enough this was the one phenomenon which was better explained on the emission than on the undulatory theory, and which had proved a stumbling-block for the latter. If the ether is a substance, the question arises whether it is carried along by the earth in its motion, or whether it remains fixed. Lorentz assumed that it remains fixed, and thus satisfactorily explained aberration. But if the earth moved through the ether, the velocity of light between terrestrial points should be affected in the same way that the velocity of sound is affected by the wind. To test this a celebrated experiment was made by Michelson in 1881, repeated by Michelson and Morley in 1887, and several times later, which showed the failure of the earth's motion to influence the velocity of light from a terrestrial source. This classical experiment may prove to be the beginning of the end of the ether. It is evident that if light is propagated through the ether in waves which have a velocity peculiar to the ether, and not influenced by the velocity of the source, then light will take longer to reach a point a given distance from it when both are moving in the direction of the line joining them when the second point is ahead than when it is behind, in the ratio of the sum of the velocities of the source and the waves to their difference. The time for the light to go to the forward point and come back is greater than it would be if the system stood still by an amount inversely proportional to where is the ratio of the speed of the source to that of light. In the case of the earth this is about one part in one hundred millions, and it was shown by Michelson that no such effect existed. Michelson assumed that this showed that the ether was fixed to the earth. For the contrary explanation, Lorentz adopted an hypothesis already proposed by Fitzgerald, namely, that all bodies in motion are thereby shortened in the direction of their motion, in precisely this ratio. This hypothesis, though startling, has now obtained great weight. In connection with it, Lorentz introduced the idea of local time, which is different for different points of the same system moving with a uniform velocity of translation. The modification, by the motion, of both distance and time leads to a most fundamental principle for all our physical notions, called the principle of relativity, which, though brought about by Lorentz, was most clearly expounded by Einstein, who is probably the high priest of the ultra-modern school. The principle of relativity assumes as a postulate that all phenomena are the same if observed with reference to a body moving with constant velocity with respect to the ether as if with respect to a body at rest. If this is so, and no experiments have contradicted it, we have as much right to suppose the ether at rest with respect to one body as another. It seems then unnatural to characterize one body as moving relative to a fixed ether. Hence Einstein abandons the ether, which he declares to be the totally unnecessary conception. Einstein makes two postulates which are sufficient to explain all phenomena now known. The first has been stated, the other is that the velocity of light is the same when measured in any system. By measures of this velocity, we can, therefore, not determine whether the system is moving or at rest. Clothed in a more mathematical form, such as has been given by Minkowski, we may state the principle as follows: If instead of the distance measured in the direction of the motion of the system, and of the time measured by a clock standing still, we substitute a quantity denoting a new length and a new time, then all the equations of electro-dynamics and presumably all those of physics admit of a so-called linear transformation of the variables and to the variables and . Under this transformation, the equations remain, therefore, absolutely unchanged. It is accordingly impossible by any observations to determine whether the time measured by the clock is or or whether the distance measured by the scale is or . As has already been said, this proposition is of the most startling nature and results in connecting the notions of time and space in a most unexpected manner. In fact we may briefly sum up by saying that we can not tell where a point is until we know when, and we can not tell the time when until we know the place where! If we accept this principle it may be necessary to totally abandon the hypothesis of the ether. Certain writers, such as Ritz in France, have established a system of electrodynamics in which the conceptions of the ether and of the magnetic and electric fields have totally disappeared. Ritz, for instance, bases his whole theory upon the so-called retarded potentials of Lorentz, by means of which the action of any electric charge, fixed or in motion, is calculated at any other time and place by means of definite integrals. This conception has been vigorously maintained: in England I may mention the name of Mr. Norman Campbell, who in a recent article in the Philosophical Magazine, as in his excellent modern treatise on electromagnetic phenomena, has vigorously assailed and even ridiculed the school of those whom he calls the "etherealists," as making use of a totally useless and hindering conception.

In 1900 Professor Poincaré had already asked the question, "does the ether exist?" This I may characterize as now the question of the hour. To sum up what I believe to be the state of the case, certain phenomena concerning radiation and the distribution of energy in the spectrum have led to the necessity of certain assumptions which seem difficultly explained on the ether hypothesis. Sir Joseph Thomson also, in order to explain certain phenomena connected with the emission of electrons from metals under the action of ultra-violet light and other phenomena with which he is particularly competent to deal, has propounded the hypothesis that a wave of light is not uniform but is somewhat of a fibrous nature. I find it difficult to see how such a hypothesis is to be reconciled with the hypothesis of the ether or the differential equations at all. In fact, the views of Sir Joseph are to me in many places incomprehensible. In his lectures recently delivered at the Royal Institution on the electromagnetic theory of light, however, Sir Joseph categorically expresses himself as of the opinion that the electromagnetic theory of light is one of the great achievements of modern science. To me this means that he approves of the ether. To take the extreme argument of Ritz, who employs as a fundamental necessity the retarded potential, seems to me to be exactly the same thing as to say that the ether exists, for since nothing whatever is propagated with finite velocity, this is the same to me as saying that it is propagated in the ether. In the first part of this paper, I have defined what I mean by the ether in very guarded form. This definition I see no reason to change. Whether we begin with the retarded potential and find that it satisfies a differential equation, or whether we begin with the differential equation and find that it is satisfied by a retarded potential is to me a matter of utter indifference and implies an ether. I admit that we still have to find a hypothesis for the ether which makes it give rise to this differential equation. The hypothesis of Maxwell seems to me the easiest one yet proposed. I will therefore close by stating my present opinion, that the ether is as good to-day as it ever was, but that apparently the notions of time and space have had to be modified in the method suggested by Lorentz and splendidly developed by Einstein and Minkowski. At the same time, we can not deny that there exists to-day what we may call la crise de l'éther, and we are far from being able to say with Lord Kelvin, "It is absolutely certain that there is a definite dynamical theory for waves of light, to be enriched, not abolished, by electromagnetic theory."

  1. Read at a meeting of the American Philosophical Society, April 22, 1910.