The Michelson-Morley Experiment, the Luminiferous Ether, and Precision Measurement

views updated

The Michelson-Morley Experiment, the Luminiferous Ether, and Precision Measurement

Overview

In 1887 Albert A. Michelson (1852-1931) and Edward W. Morley (1838-1923) performed what has become one of the most famous physics experiments in history. Using an extremely sensitive optical instrument—the interferometer—they attempted to measure Earth's velocity with respect to the luminiferous ether, a hypothetical substance that most nineteenth-century physicists believed necessary for the propagation of light. Against all expectations, their experiment yielded a negative result, indicating no motion of Earth relative to the ether. Ether theories were modified to account for this null-result, but no fully satisfactory solution presented itself until the introduction of Albert Einstein's special theory of relativity in 1905.

Background

The optical experiments of Thomas Young (1773-1829) and Augustin de Fresnel (1788-1827) at the beginning of the nineteenth century helped revived the wave theory of light. As with other wave phenomena—like sound waves in air and ocean waves in water—light waves were thought to require a medium of transmission. This medium was called the luminiferous (light bearing) ether.

An important nineteenth-century scientific question was the relationship between the ether and material bodies moving through it. Young believed that matter passed freely through the ether without in anyway disturbing it. This seemed necessary for ether-wave theories to explain stellar aberration. Discovered in the eighteenth century by James Bradley (1693-1762), aberration is the apparent displacement of a star from its actual position due to the combined velocity of Earth and starlight. It was thought that if Earth's motion disturbed the ether, then starlight would be deflected in a manner inconsistent with this well-known effect.

Dominique François Arago's 1810 failure to measure Earth's velocity relative to the ether challenged Young's conclusions. Arago (1786-1853) accepted that the velocity of light, c, was constant in the ether and could only be measured to be c if one were at rest relative to the ether. He reasoned that motion through the ether with velocity v in the same or opposite direction as a beam of light, such as Earth's motion away from certain stars and toward others in its solar orbit, should yield light velocity measurements smaller, c-v, or greater, c+v, respectively. Knowing that light beams with different velocities refract differently, Arago designed an experiment to observe this difference. However, no such effect revealed itself. This null-result suggested the existence of a stagnant layer of ether near Earth's surface. If true, this would have undermined Young's explanation of aberration.

In 1818 Fresnel introduced his hypothesis of partial ether drag, which reconciled Young's position with Arago's result. Fresnel argued that transparent bodies dragged a permanent amount of ether within them in proportion to the square of their refractive indices. This altered the velocity of light by just enough so that optical experiments like Arago's could not detect Earth's motion relative to the ether. Fresnel's hypothesis further entailed that the bulk of ether remained undisturbed by material bodies, which agreed with Young's aberration explanation.

According to Fresnel's theory, Earth's motion through the ether was in principle detectable. However, James Clerk Maxwell (1831-1879) noted in his 1878 Encyclopaedia Britannica article "Ether" that the expected effect was too small to observe with existing optical instruments. When Michelson learned of Maxwell's views in 1879, he took-up the challenge of designing a sufficiently precise instrument for which he was later to coin the name "interferometer."

As the name suggests, Michelson's instrument exploits optical interference. A coherent beam of light is produced and then split in two. Each beam is directed along one of two mutually perpendicular interferometer arms, then reflected by a mirror back along its path to recombine with the other beam. The recombined beam is then directed to an observing telescope, where a pattern of alternating light and dark regions—known as interference fringes—is produced. The mirrors are adjusted so a fringe center falls on the telescope's fiducial mark. Differences in the velocity of light between the interferometer arms means light beams arrive at the telescope slightly out of phase, causing the interference fringes to shift with respect to the fiducial mark. If Earth moves relative to the ether, such fringe-shifts should be observed while the interferometer rotates.

The experiment was performed in 1881 at the Royal Astrophysical Observatory in Potsdam, Germany. No fringe-shifts of the expected magnitude were observed. Michelson suggested that this null-result refuted Fresnel's ether theory. However, this conclusion proved unwarranted. Alfred Potier, and later H. A. Lorentz (1853-1928), noticed an error in Michelson's theoretical calculations. Correct calculations indicated fringe-shift magnitudes smaller than originally predicted, in fact, smaller than could be observed with the 1881 interferometer.

In 1887 Michelson and Edward Morley repeated the Potsdam experiment with better experimental controls and a more sensitive instrument. The two most important changes were an increase in the optical paths lengths—making the interferometer ten times more sensitive—and the instrument being mounting on a massive stone floating in mercury, which insulted it from vibrations and allowed more accurate fringe-shift reading while the instrument rotated. Although Michelson and Morley expected a fringe-shift, they once again obtained a null-result.

Impact

Contrary to popular belief, the Michelson-Morley null-result was not considered a serious threat to ether theories, nor was it taken as proof that the velocity of light was absolutely constant. Though the null-result puzzled ether theorists, there was no sense of crisis in the physics community. Indeed, consensus had it that the result would eventually be explained within the ether framework.

One failed attempt to explain the null-result was a modified Stokes' ether-drag theory. George Stokes' (1819-1903) theory, requiring complete ether entertainment at Earth's surface, accounted for the null-result. However, in 1892 Lorentz pointed out inconsistencies in Stokes' explanation of aberration. Furthermore, Oliver Lodge's (1851-1940) 1892 experiments failed to detect any ether-drag near rapidly moving disks.

The most promising explanation within the ether paradigm was the Fitzgerald-Lorentz contraction hypothesis. Originally published in 1889 by G. F. Fitzgerald (1851-1901), the contraction hypothesis states that as an interferometer moves through the ether its arms shrink in the direction of motion by just the amount necessary to cause a null-result. Fitzgerald's proposal was motivated by Oliver Heaviside's (1850-1925) 1888 discovery that the electromagnetic field of a moving charge shrinks by exactly this amount. Although he had no electron theory of matter, Fitzgerald thought it reasonable to assume the intermolecular forces of the interferometer arms to be electromagnetic in nature. Thus, they could be expected to vary in accordance with Heaviside's result. Fitzgerald felt sure such a variation would cause the interferometer arms to shrink the required length.

The contraction hypothesis implied that the observed velocity of light would be constant. However, this constancy was not absolute. For Fitzgerald, Lorentz, and other ether theorists, the null-result did not mean light was always and everywhere traveling with the same velocity. It only appeared so because the interferometer arms had shrunk. In principle, an instrument either not susceptible to shrinking or vastly more sensitive would be able to detect Earth's motion. Attempts to measure an ether-drift continued into the late 1920s and beyond, always with null-results.

Lorentz independently proposed the contraction hypothesis in 1892 and attempted to justify it in terms of his electron theory. Its initial viability faded as Lorentz was forced to make more complex and implausible assumptions. This changed in 1905 with publication of Albert Einstein's (1879-1955) special theory of relativity. As Einstein himself noted, his theory provided "an amazingly simple summary and generalization of hypotheses which had previously been independent from one another."

The Michelson-Morley experiment had only a minor and indirect role in the genesis of Einstein's special theory of relativity. It did, however, play a significant role in convincing many physicists of the theory's validity, supporting as it did Einstein's postulate of the absolute constancy of the velocity of light.

Though the interferometer failed in the purpose for which it was created, it nevertheless remains one of the most sensitive and versatile instruments ever created. Michelson used the instrument to measure the gravitational constant, indices of refraction, soap film thickness, coefficients of expansion, and to test screw pitch uniformity, analysis of spectral lines, and stellar aberration, all with a precision never before achieved. Two of these measurement standout: his progressively better measurements of the velocity of light and his measurement of the world meter standard in terms of light waves. The latter was considered by his contemporaries vastly more significant than his ether null-result.

STEPHEN D. NORTON