The idea of an ether was introduced into science by Descartes in Principia philosophiae (1644). Until that time, forces between two bodies that are not in direct contact were assumed to act through space—by action at a distance. Descartes replaced this explanation by one based on an intermediate medium (ether) consisting of vortices that transmit forces between bodies at a distance.
The ether concept became especially predominant in the 19th century by the work of Young and Fresnel who revived Huygens' wave theory of light. They replaced Newton's light corpuscles by waves propagating through the ether. In order to explain stellar aberration, first observed in the 1720s and then shown to be caused by the velocity of Earth relative to the velocity of Newton's light corpuscles, Young (1804) assumed ether to be in a state of absolute rest. Maxwell showed in the 1860s that light waves are electromagnetic waves transverse (perpendicular) to the direction of the propagation of the waves. Following Young and Fresnel, Maxwell assumed that electromagnetic waves are vibrations of the ether.
In the 19th century it was known that transverse waves are not possible in a gas or a liquid, but only in a solid; hence ether was thought to have solid-like properties. Since light behaves in closed rooms the same as in open fields, and many materials are transparent to light, ether was assumed to fill up all of space and all of matter. Thus, at the end of the 19th century physicists had a picture of the ether as a quasi-rigid solid (not completely rigid because it can vibrate), luminiferous (light carrying) medium that is massless and transparent, at absolute rest, and present everywhere.
Today, the concept of ether does not play a role any longer in physics, but in daily life the word lives on in connection with radio and television signals, which commonly are said to be transmitted "through the ether".
It is not really possible to speak of "the" ether, because as a scientific concept it evolved through three centuries, from Descartes (1596 – 1650), who conceived it as a whirlpool of rotating chains of particles, to Hendrik Antoon Lorentz (1853 – 1928), who saw ether as a transparent massless solid at complete rest. Its only shared property, conserved through the centuries, is that it permeates all space and all matter, even the interstitial spaces between the atoms. Through the centuries ether served three different purposes:
- Transferral of forces; denial of Aristotelian action at a distance.
- Propagation medium for light; first in the form of particles (Newton) later as vibrations (Young and Fresnel).
- Absolute reference frame at rest; speeds of all (heavenly) bodies are with respect to ether.
The name ether comes from ancient Greek αἰθήρ (aithèr) where it means the upper, radiating, air. Aristotle introduced it as a fifth element (quinta essentia) next to Earth, Fire, Water, and (sea-level) Air. Aristotelian philosophy was introduced into Western Europe in the 13th century by scholastics as Albertus Magnus (ca. 1200–1280) and Thomas Aquinas (1225–1274). After Aristotelianism was accepted by the Church, Aristotle's views on the nature of motion were incorporated into medieval natural philosophy: a heavy object has as its natural place the center of the universe (which before 1600 was the center of the globe) and a light object has as its natural place the sphere of the Moon. Matter strives for its natural place in the universe, ergo a stone falls down and smoke rises up.
René Descartes considered the medieval views on motion occult and therefore superseded; he believed instead that all forces are transmitted by direct contact. With regard to the actions between bodies not touching each other, such as two magnets, or the influence of the Moon's position on the tides, he postulated that they must be in direct contact through intermediate contiguous matter. The force is transmitted through this matter—the ether—by two agencies, pressure and impact. Space, in Descartes' view, is a plenum occupied by an ether, which, imperceptible to the senses, is capable of transmitting forces on material bodies immersed in it. Descartes assumed that the ether particles are in constant motion, but, as there is no empty space for them to move to, he inferred that they move to places vacated by other ether particles. The particles participate then in the spinning motions of closed chains of particles (vortices). The Cartesian theory of light is—in the eyes of the modern beholder—rather convoluted. In the first place it is assumed that the speed of light is infinite and yet light is seen as a projectile whose velocity varies from one medium to another. The vehicle of light is "matter of the second kind", which is intermediate between vortex matter and ordinary, ponderable matter. This matter of the "second kind" forms globules and different rotational velocities of the globules give light of different colors.
Hooke, Huygens and Newton
The next event relevant to the history of ether is the publication (1665) of Micrographia by Robert Hooke (1635–1703). Hooke's description of the propagation of light is mechanical and in that sense it resembles that of Descartes. However, while the Cartesian hypothesis is a static pressure in the ether, Hooke's theory concerns a rapid vibrational motion of small amplitude. He introduces the idea of a wave front, which twelve years later (in 1679) was borrowed by Christiaan Huygens (1629–1695), who greatly improved and extended the wave theory of light. Huygens assumed that the ether, the medium in which light propagates, penetrates all matter and is even present in the vacuum. Huygens' ether was, like Descartes', constituted of particles. Huygens interpreted gravitation—a typical action without apparent direct contact—in terms of ether particles that are rapidly rotating in the space surrounding the planet. His rotating particles are reminiscent of the Cartesian vortices, which is not surprising as Descartes had had a strong influence on the young Huygens, whom he had known personally as a child.
Hooke's and Huygens' theories were obliterated (at least for over a century) by their contemporary, the scientific giant Isaac Newton (1642–1727). Newton started his career as a strict adherent of ether theory. He wrote in 1672 and 1675: (as summarized in Ref. )
All space is permeated by an elastic medium or aether, which is capable of propagating vibrations. This aether pervades the pores of all material bodies and is the cause of their cohesion; its density varies from one body to another, being greatest in the free interplanetary spaces.
Newton suggested three mechanisms by which light may proceed through the ether. His second suggestion that light consists of "multitudes of unimaginable small and swift corpuscles of various sizes springing from shining bodies" was generally selected by later scientists. In 1675 Newton submitted a memorandum to the Royal Society in which, among other things, he explained gravity. He wrote that ether condenses continually in bodies such as Earth and therefore there is a constant downward stream of it that impinges on gross bodies and carries them along. Further Newton suggested in this memorandum that the resulting movement of ether holds the planets in closed orbits.
Later, when writing the Principia (1687), Newton become more inclined toward considering gravity as an action at a distance. He realized that this would not be easily digested by his contemporaries, who had just freed themselves of the Aristotelian notion that an object falls downward because of its natural place in the universe. And, indeed, he was right. Huygens and Leibniz were very critical of the idea of attraction at a distance. In the second edition of the Principia (1713), Newton defended his point of view by adding a "General Scholium" in which he attacked the vortex theory of Descartes, pointed out that his gravitational law was mathematically correct, that he did not know the deeper reason for it, and pronounced Hypotheses non fingo (I do not make up hypotheses).
Because of Newton's enormous influence on 18th century science, action at a distance was no longer seen as a problem and generally accepted. This is witnessed by the resistance that Michael Faraday met in the early 19th century when he cast doubt on the action-at-a-distance concept for electric and magnetic forces.
The 18th century did not see much development in the theory of light, and Newton's corpuscular theory was universally accepted. It was forgotten that he had stated that light particles travel through ether; ether was not important to most 18th century natural philosophers.
Young and Fresnel
Ether re-entered the forefront of physics when Thomas Young revived the undulatory (wave) theory of light in 1800.. He noticed that Newton's emission theory: a light source emits corpuscles—had problems with the interference of light and with the simultaneous refraction and reflection of light falling under an angle on the surface of water. Wave theory can account elegantly for both effects, while corpuscular theory cannot. Young's theory was adapted and extended by his 15-year younger French contemporary Augustin-Jean Fresnel. Both workers recognized that stellar aberration needed to be explained by undulatory theory.
Stellar aberration was discovered by James Bradley in 1725–1726. A year later (1727) he explained his discovery in the framework of Newton's corpuscular theory. Bradley noted that the velocity of the stellar light observed on Earth, cE, is the resultant of the absolute velocity of light, c, expressed with respect to a frame attached to the fixed stars, minus the velocity of the planet v relative to the same absolute frame. The vectors cE≡ c−v and c make a small angle, the aberration angle. Thus, Bradley transformed the velocity c of the light-corpuscles from an absolute frame fixed to the stars to a frame attached to the moving globe. While doing this, he derived that the aberration angle is proportional to the ratio v/c, a ratio that was to become very important in 19th century physics. The speed of Earth, v, is about 3×104 m/s and c ≈ 3×108 m/s, so that v/c ≈ 10−4.
In 1804 Young made a first step in explaining stellar aberration by the wave theory when he assumed that the ether is at absolute rest, that is, ether offers the absolute reference frame used by Bradley. In the absolute frame light propagates with speed c (speed in vacuum).
It was known in the early 19th century that light waves travel through a transparent material—such as a block of glass—with a speed, cg, lower than c, while the corpuscular theory had reasons to suppose that cg was larger than c. Thus, the wave theory predicts the index of refraction n of the material to be larger than unity. The refractive properties of a prism depend, through Snell's law, on n and hence on the speed of light cg in the glass. François Arago was of the rather obvious opinion that the speed of light relative to the prism, which is fixed to Earth, should enter Snell's law. Inspired by Bradley's theory of stellar aberration, he performed in 1810 telescopic observations of the speed of stellar light. He mounted a prism on a telescope and observed stars situated at different angles in the sky, exhibiting different aberration angles. According to the Newton-Bradley theory, the light rays from the stars at different angles have different velocities relative to the prism; these should be observable in the refraction patterns of the prism. Arago got a null result, he did not observe any effects. This null effect was the first of several to come in the next hundred years or so.
The largest differences in speed of light on Earth can be expected when a fixed star is on the horizon, and the planet travels parallel to its light rays, toward or away from the star. The absolute speed of light (measured by an observer motionless in the ether, the velocities on one line) is,
where v is the speed of Earth; v is positive when the planet goes in the same direction as the stellar light and negative if it goes in opposite direction. Note that c=cg +v is a special case of the vector equation c=cE+v, introduced above.
In 1818 Fresnel gave some thoughts about incorporating Arago's observations in the wave theory of light. He made the assumption that ether is "dragged" along by the glass of the prism, so that the relative ether-glass speed is reduced. The ether in a transparent body is entrained with velocity v(1-1/n2) when the body itself moves with velocity v with respect to the absolute ether. The "ether drag factor" (1−1/n2) reduces the absolute speed of light, which becomes according to Fresnel,
Fresnel showed that inclusion of this drag factor into the theory gives contributions to Arago's results that start with (v/c)2, too small to be observable.
In 1851 Armand Hippolyte Louis Fizeau was able to confirm Fresnel's drag factor experimentally by guiding light through flowing water.
For a while wave theory had difficulties in explaining birefringence (double refraction) and the associated phenomenon of polarization of light. Around 1820 Fresnel was able to account for these effects by assuming that light in a crystal is a transverse wave (a vibration perpendicular to the propagation direction). In analogy he inferred that light propagating through the ether consists also of transverse waves. This was confirmed in 1861 when Clerk Maxwell showed that (visible) light is just a special kind of electromagnetic waves.
Ether in the second half of the 19th century
Michael Faraday, one of the fathers of electromagnetism, had a strong dislike of hypothetical entities for which no convincing experimental evidence existed. As a consequence, he was skeptical about the existence of the ether. But at the same time he was opposed to electric and magnetic action-at-a-distance theories, which he replaced by field theories. James Clerk Maxwell, who tread in Faraday's footsteps and accepted the physical reality of fields, formulated a mathematical theory of electromagnetic waves that propagate through the luminiferous ether. The difference between Faraday's conception of a field without an ether and Maxwell's conception of a field with an ether is subtle and not easy to understand for a modern physicist. The ether, in the view of Maxwell and almost all physicists at that time, permeates all space and has many of the characteristics of a polarizable dielectric. Further, Maxwell was of the opinion that terrestrial optical experiments aimed at determining Fresnel's ether drag, which is quadratic [i.e., of order (v/c)2], are not sensitive enough to detect the influence of the drag.
The Michelson-Morley Experiment
Albert Abraham Michelson disagreed with Maxwell, judging that it was possible to observe quadratic effects on Earth. While Michelson was on leave in Berlin (1881), he built an interferometer that was sensitive enough to detect effects of the order (v/c)2 and tried to determine the speed v of Earth with respect to the ether. In other words, Michelson aimed to measure the speed of the "ether wind". He compared the times it takes for light to travel the same distance either parallel or transversely to Earth's motion relative to the ether. However, his conclusion was that the speed of the ether wind is zero. H. A. Lorentz found an error in Michelson's theory of the experiment and was dubious about his interpretation of the result. Lord Rayleigh (John William Strutt), who was a strict believer in ether, urged Michelson to repeat the experiment. So, Michelson, who in the meantime was appointed at Case School of Applied Science in Cleveland, Ohio, repeated the experiment in collaboration with Edwin Williams Morley, a chemist from Western Reserve University, also in Cleveland. They built an new interferometer and in August 1887 they measured again a null effect (a speed of 4.7 km/s or less was found, while 30 km/s was expected). 
Upon being informed of this null result, H. A. Lorentz and George FitzGerald, both adherents of ether, independently postulated that rods fixed to Earth show a dynamic (i.e., due to a change in molecular forces) length contraction,
where l′ is the length of the rod when it is oriented parallel to v; its length is l when it is perpendicular to v. When they applied this contraction formula to the arms of the Michelson-Morley interferometer, they could account for the null effect. Later Lorentz showed that the contraction is the first term in a series development of the exact formula,
Maxwell's electromagnetic theory was difficult for his contemporaries and hence did not receive much response. This changed in the mid-1880s when Heinrich Rudolf Hertz was able to generate waves of the kind predicted by Maxwell. (Hertz produced electromagnetic waves of wavelengths from a centimeter to a meter, much longer than the wavelengths of visible light that are on the order of 500 nm). As a 19th century physicist, who had studied under Hermann von Helmholtz, Hertz was committed to the ether idea throughout his life. His conviction of the ether’s importance developed throughout his career, intensified during his research on electromagnetic waves, and finally became his chief preoccupation during the final years of his life. After Hertz's empirical confirmation of the existence of Maxwell’s electromagnetic waves, it was universally assumed that the ether, as a carrier of these waves, was proved to exist, too. The experiments of Hertz had an overwhelming impact on physics. They occurred in the formative years of a generation of physicists who dedicated themselves to electromagnetism and who raised the ether to the status of one of the basic building blocks of nature. In a letter to Hertz dated August 14, 1889 from Oliver Heaviside, one of the founding fathers of electromagnetic theory, there is this paragraph:
Then there is the vexed question of the motion of the ether. Does it move when "bodies" move through it, or does it remain at rest? We know that there is an ether: the question is therefore a legitimate physical question that must be answered. I am in hopes that extensions of your researches will supply material for an answer. As for the structure of the ether itself, that is a far more difficult matter, and one that can never, it seems to me, be answered otherwise than speculatively.
Decline of the ether
At the end of the 19th century the ether served two purposes: first and foremost it was a transport medium for electromagnetic vibrations. Secondly it offered an absolute frame of reference. With regard to the latter it must be pointed out that in pre-Copernican times, when Earth was seen as the immobile center of the universe, the absolute frame of reference was so naturally fixed to our planet that the question of an absolute frame did not arise. When Copernicus posited that our planet is orbiting the immobile sun, then without much ado the absolute frame was shifted to the sun. But when it was observed that the sun, too, moves with respect to the fixed stars, the existence of a frame at absolute rest became problematic. The ether had solved this problem, or so was the communis opinio around 1890. Of course, there was still the remaining problem of the speed of the Earth and light on Earth with respect to the ether.
Around the change of the century physicists started to realize that ether was not indispensable as a transport medium. Hertz had quipped: "Maxwell's theory is Maxwell's system of equations", by which he meant that Maxwell's equations for the propagation of electromagnetic waves are valid irrespective of the model for the underlying ether. Paul Drude wrote in 1900: "The conception of an ether absolutely at rest is the most simple and the most natural—at least if the ether is conceived to be not a substance but merely space endowed with certain physical properties."
The death knell of the ether tolled in 1905, when Einstein published his special theory of relativity.  Einstein assumed that an absolute reference frame does not exist and, even more than that, he showed that physics is not in need of such a frame. He declared that all inertial frames are equivalent, one cannot prefer one over another (see special relativity). At one stroke he solved the problem of the speed of light, too: this speed is the same in any inertial frame, and is independent of the velocity that a frame may have with respect to any of the infinitely many other inertial frames. In his 1905 paper Einstein refers to the ether only once:
The introduction of a "luminiferous aether" (Lichtäther) will prove to be superfluous inasmuch as the view here to be developed will not require an "absolutely stationary space" provided with special properties, nor assign a velocity vector to a point of the empty space in which electromagnetic processes take place.
Defenders of the ether still remained after 1905, however, since a number of physicists failed to read or understand Einstein, and those who did, came only gradually to a full appreciation of its impact on the ether. Not until about 1910 did the general opinion of physicists shift away from the ether. Among the more stubborn hold-outs were Oliver Lodge, A. A. Michelson, and Joseph John Thomson. This older group of physicists gradually died out and was replaced by a new generation that had grown up with Einstein’s theory and for which ether was an esoteric historical concept like phlogiston.
- Generally speaking the spelling "ether" is more modern than "aether". Note, however, that two Americans, A. A. Michelson and E.W. Morley, spelled it as "ether" as early as 1887.
- E. Whittaker, A History of the Theories of Aether and Electricity 2n edition p.19. First edition is online: Whittaker, History of ... )
- R. S. Westfall, Never at Reʃt; A Biography of Isaac Newton, Cambridge University Press, (1980), p. 271
- T. Young, Philosophical Transactions, vol. 90, p. 106 (1800)
- The index of refraction is the ratio of two speeds: n ≡ c/cg.
- Ronald Newburgh, Fresnel Drag and the Principle of Relativity, Isis, Vol. '65, (1974) pp. 379-386
- R. S. Shankland, Michelson-Morley Experiment, American Journal of Physics, Vol. 32, pp. 16-35 (1964) DOI
- A. A. Michelson and E. W. Morley, On the Relative Motion of the Earth and the Luminiferous Ether, Am. J. Sci., vol. 34, p. 333-345 (1887) online.
- H. A. Lorentz, The Relative Motion of the Earth and the Ether, Verslagen Koninklijke Akademie der Wetenschappen, Amsterdam, vol. 1, p. 74 (1892)
- H. A. Lorentz, Versuch einer Theorie der electrischen und optischen Erscheinungen in bewegten Körpern [Attempt of a theory of electrical and optical phenomena in moving bodies], E. J. Brill, Leiden (1895) online
- Reprinted in: James G. O’Hara and W. Pricha, Hertz and the Maxwellians Peter Peregrinus Ltd., 1987 (London), pp. 73–74.
- P. Drude, The Theory of Optics, (Transl by C. R. Mann and R.A. Millikan) Dover, New York (1959). p. 457
- A. Einstein, Zur Elektrodynamik bewegter Körper, Annalen der Physik, vol. 17, pp. 891-921 (1905) Online pdf Original (German) version
- A. Einstein, On the Electrodynamics of Moving Bodies, Online pdf English translation by W. Perrett and G.B. Jeffery