Measuring The Immeasurable

IN 1907 ALBERT ABRAHAM MICHELSON BECAME THE FIRST U.S. CITIZEN to win a Nobel Prize for science. He was honored as the laureate in physics “for his precision optical instruments and the spectroscopic and metrological investigations conducted therewith.” It was an honor well deserved. His ether-drift experiment of 1887, done in collaboration with the chemist Edward W. Morley, had put America in the forefront of experimental physics by challenging basic assumptions about the nature of the universe. His scientific accomplishment was also a techological one, for it grew from his work at assembling and operating instrumentation of unprecedented precision for measuring the seemingly immeasurable speed of light.

In the late nineteenth century scientists held the hope—as they do today—of finally unifying their understanding of physical forces in the universe. A chief challenge, if not the chief challenge, was to incorporate the newly expounded properties of electricity and magnetism into the Newtonian structure of mechanics. In the 1860s the great Scottish physicist James Clerk Maxwell had determined theoretically that light consists of electromagnetic waves, and scientists set about describing the mechanics of those waves. Every kind of wave previously encountered—such as waves in water or along a taut string or sound waves in air—had traveled in some medium. The postulated medium in which light traveled through space was given the name ether (or aether ), and understanding the mysterious and intangible ether became a fundamental goal of physics.

By the 1880s Michelson had developed a device for measuring with newfound accuracy the speed of light. He then undertook to calculate the earth’s speed through the ether, by comparing the speeds of two perpendicular beams of light sent in horizontal directions along the earth’s surface. In this he was far less successful. He repeatedly found that light traveled at the same speed in all directions, even though measured on a spinning planet in orbit around a sun hurtling through space. Michelson’s “failure,” as he considered it, was not explained until 1905, when Albert Einstein, with his special theory of relativity, postulated that the speed of light itself is a universal constant.

Michelson’s patient, ingenious, technically beautiful work made him such a paragon among experimental physicists that during his long life no one—except him—ever bettered his measurements of the speed of light. His interferometer is an instrument that should be listed with the telescope, microscope, barometer, and thermometer in the pantheon of scientific tools. In 1882, when he was working on an optical pathway along a railroad track in Cleveland, some reporters asked him what he was doing. He said he was measuring the speed of light. They asked him why. He said, “Because it is such fun.” Nearly fifty years later Einstein asked him a similar question. His answer was the same.

He was only the fourth or fifth person ever to measure the speed of light, but he devoted himself to the task wholeheartedly. Although trained as a naval officer, he never achieved the admiralcy of strategic scientific importance of some his protégés, like George Ellery Hale and Robert A. Millikan; he was held back from that by both his personal reserve and his insecurity about his mathematical abilities. He did grow to hold a commanding position in physical optics, bringing the talents of a practical navigator and an artful surveyor to his field of experimentation.

Michelson was born in 1852 in Prussia, about eighty miles from the birthplace of Copernicus. When he was two, his parents came to the United States, and they ended up in Virginia City, Nevada, where his father opened a dry goods store. In high school Michelson excelled in mathematics. His parents could not afford to send him to a private college, but he managed to be accepted by the U.S. Naval Academy. There he did well in math and the sciences—finishing first in physics and optics—but less well in warfare and seamanship. Upon graduation he spent two years at sea and then accepted a job as an instructor at the academy.

It was while teaching at Annapolis that Michelson became seriously interested in measuring the speed of light. Back in the seventeenth century a Danish astronomer, Ole Roemer, had calculated a probable finite speed of light based on the seemingly inconsistent period it took a moon to orbit Jupiter; he correctly concluded that the orbit seemed to take slightly longer as the earth and Jupiter moved apart in their orbits and then shorten as they neared, because of the time it took light from Jupiter and its moon to cross the earth’s orbit. But it was not possible before the mid-nineteenth century even to consider measuring the speed of light on earth. Suddenly, around 1850, however, magnificent demonstrations by Armand Fizeau and Jean Foucault in Paris showed how the big V—the velocity of light, today known as c—could be terrestrially measured, and in at least two different ways. Fizeau beamed light through a gap in a toothed-gear wheel to a distant mirror and back and then measured how fast the wheel had to spin to obstruct the returning light with a tooth. Foucault used a rotating mirror to measure similarly the deflection of a beam over a shorter course. After that all sorts of new possibilities were opened. M. Alfred Cornu, at the Sorbonne, and Simon Newcomb, at the U.S. Naval Observatory, were quick to improve big V measurements with minor improvements of these devices. Michelson followed in their wake but soon passed beyond their horizons.

His work began when a professor asked him to prepare a demonstration of the Foucault method of measuring the speed of light. In so doing, Michelson hit upon several simple changes in the apparatus that could better its accuracy by enlarging the distance over which the light was measured. He made this first determination of the speed of light in 1878, using his homemade but improved Foucault apparatus to achieve a result that was the most accurate thus far.

Michelson was fairly fluent in both German and French, and he read with delight the literature of European optical workers. Experiments by Foucault and Fizeau had corroborated wave over particle theory by measuring the velocity of light through a static column of water more than thirty feet long. Michelson made his own mark with his improvements to Foucault’s device for measuring the big V. In 1880 he took off for postgraduate studies at the laboratories and observatories of the European masters. The hybrid sail-and-steam Navy that he still served (in mufti, of course) paid his way, thanks to the patronage of his professor Simon Newcomb. Relative-motion calculations at sea or at anchor were second nature to Michelson by now; computing set and drift is an art vital to safe seamanship, so he was uniquely suited to think a new thought about light: What if I measured the velocity of light at right angles to itself? Would that allow me to compare the speeds of perpendicular transmissions? Might I not then have a meter like those used to measure currents of water, and maybe a speedometer for the earth itself? There is little doubt Michelson would have asked himself such questions in the first-person singular. He was proud of his thoughts, of his individuality, and his personal talents—not only as a scientist but also at boxing, billiards, tennis, watercolor painting, and music. There was not much room for collective responsibility in his makeup. He could and would delegate authority, but he jealously guarded his autonomy.

Even before Michelson left for Europe, Professor Newcomb was probably discussing his work with Alexander Graham Bell, recently made famous by his success in making the telegraph talk and now becoming a patron of science. Michelson was feeling increasingly challenged, especially by the theoretical and experimental ideas of James Clerk Maxwell, to measure the relative motion of the earth against the luminiferous ether. A good word from Bell to the Volta Fund, in Paris, ensured the funding (about five hundred dollars) that allowed Michelson to get the best optical-instrument makers in Berlin, Schmidt und Haensch, to construct his first interferometric apparatus.

His asymmetrical brass device, mounted to rotate on a small pedestal, had two optical-bench arms about a meter long. At its heart was a beam splitter, also called a “light divider” or “pencil separator.” This half-silvered mirror, carefully fashioned to reflect from its back face only half the incident light beam, allowed the other half of the yellow sodium light from the same source to pass through going and coming. Thus one-half of the beam could bounce off a mirror straight behind and be returned while the other half was reflected ninety degrees to one side and back. A second plate of glass behind the mirror, called a compensator, gave the straight beam of light two more passes through glass so that both optical pathways followed the same obstacle course (see illustration opposite). There seem to have been no very novel parts in the device. The light was provided by a standard Argand burner, widely used in lamps for vehicles and fashionable parlors as well as in spectroscopy. The mirror mounts, with their delicately machined screw threads for precise adjustment, were familiar in the Navy among navigators and were becoming ever more common as machine-tool technology advanced.

Through the eyepiece Michelson looked for changes in interference fringes—patterns of light and dark caused when overlapping waves of light amplified or canceled each other out. All waves—of sound, light, water, or anything else-have crests (high points) and troughs (low points). When waves coincide, they add together, so that two crests or two troughs together will make a crest twice as high or a trough twice as deep; a crest and a trough meeting will cancel out. When waves of refracted light coincide, the way they overlap and add together is visible in bands of dark called “interference fringes.” Thus Michelson could see if waves traveling the same distance and then meeting traveled the exact same speed by watching for changes in interference fringes where they met. Since simple mathematics shows that it is always faster to swim across a moving river and back than upstream the same distance and back, Michelson reasoned that he could rotate his instrument and, by determining from interference shifts when light traveled faster down one arm than the other, deduce which way—and maybe even how fast—the ether was moving relative to earth.

Hermann Helmholtz, Michelson’s mentor in Berlin at the time, was encouraging, yet he warned at the start that keeping a constant temperature would be imperative. Michelson soon discovered, in the spring of 1881, that this difficulty was indeed formidable. He tried the experiment first in Helmholtz’s laboratory in what is today East Berlin. Vibrations were too acute. He tried it in the wee hours of the morning; still too many disturbances. He tried “constant temperature” rooms in the basement; he tried covering the whole apparatus with various shields; he tried surrounding it all with melting ice, again to little avail. Yet the serendipitous surprise was that the instrument was amazingly sensitive to the slightest ambient disturbance. Human breathing or the heat from a human hand near the pathways could cause the fringes, once gained, to become lost. After several weeks Michelson got permission from Professor Hermann Vogel to use the extremely stable foundation for the equatorial telescope at the Potsdam Astrophysical Observatory, southwest of Berlin. Repetitions of his work there persuaded him to publish his null results. He could detect no differences or changes in the speed of light. He was disappointed but at the same time elated.

The operation had been a failure, but the patient had lived. Michelson’s interferometer, although unsuccessful in terms of its initial experimental raison d’être, had been born. It gave great promise of becoming useful in all sorts of extremely accurate precision measurements. His work with it was soon being discussed by Helmholtz, Vogel, and other physicists and opticians. André Potier, in Paris, and H. A. Lorentz, in Leiden, helped Michelson realize that he must greatly improve this instrument to realize its full potential. One obvious shortcoming was the small total path length for light. Maybe the device had to be enlarged by a factor of ten or so.

The U.S. Navy was near the nadir of its power at this time; Master Michelson was approaching the zenith of his. He was discovering his true vocation to be in physical science rather than in naval science or engineering. When offered a professorial job at the brand new Case School of Applied Science, in Cleveland, he jumped at the chance, resigned from the service, and collected equipment in Europe for his new job in the United States. Upon moving to Ohio in 1882, he first finished some velocity-of-light obligations to Simon Newcomb and the Naval Observatory. He set up his retests along the Nickel Plate Railroad right-of-way adjacent to his campus.

Edward Williams Morley (1838-1923), a fellow scientist next door at Western Reserve College, became interested in Michelson’s research, and the two soon became good friends. Morley was a chemist and a chaplain almost fifteen years Michelson’s senior and was deeply interested in precision analyses of the composition of water and air. Before long Michelson had convinced Morley of the need for and beauty of interferometry.

The two made a trip to Baltimore in 1884 to attend a series of advanced lectures in physics given by William Thomson (later Lord Kelvin) and John Strutt (Lord Rayleigh). These lectures could not have fallen on more fertile ears; part of their purpose was to try to stir American scientists to achieve feats in pure physics and chemistry equal to the best research in Europe. Talks with other scientists in Baltimore and correspondence with physicists elsewhere persuaded Michelson and Morley to do two things: first, to repeat another famous experiment by Fizeau that had evidently shown that flowing water would affect the velocity of light passing through the water by a certain minute fraction depending on the direction of the water’s flow; and second, to repeat Michelson’s 1881 experiment at a much more sophisticated level, to see, as Morley wrote in a letter to his father, “if light travels with the same velocity in all directions.”

Morley’s chemistry laboratory provided most of the equipment and ingredients to carry out the first test, sometimes called the “water-drift” or “ether-drag” experiment, between 1885 and 1886. A split beam from a single light source was made to traverse a racecourse with and against the flow of two streams of pure water, each about twenty feet long. Michelson did most of the optical instrumentation and observation, but Morley’s chemical apparatus was equally critical. When in 1886 they published their verification of Fizeau’s work, the two felt justified in concluding “that the luminiferous ether is entirely unaffected by the motion of the matter which it permeates.” That is, combining the speed of light and the speed of the moving water it traveled through slightly altered the observed speed of light, but to a degree that did not admit of any effect from ether, such as the water’s dragging on the ether.

Thus the stage was set to see—or to try to see with better equipment—if light itself was affected by the relative motion of the earth through the luminiferous ether that, presumably, permeated all space. Michelson had worried long and hard over problems of astronomical aberration and the elastic-solid idea of the ether. Ether had the wave-carrying properties of a rigid solid, yet it was less tangible than any gas and filled the space between the very atoms as well as galaxies of matter. He worried over George Stokes’s idea about wave motion, that the earth and other moving bodies must pull ether along with them; over Sir George Airy’s experimental observations with water-filled astronomical telescopes; over a critique by H. A. Lorentz of his original experimental design; and over various other comments and criticisms about his interferometric innovations. But he and Morley now felt ready to try again. They would try to measure the most obvious relative motion of the earth—that of its yearly orbital motion around the sun. Perhaps it had been too ambitious to seek the result of all the earth’s component motions. Perhaps the same sort of a turntable optical bench—only much more massive, stable, and with a tenfold increase in path length-might reveal the influence of the earth’s motion in orbit on the perceived velocity of light, and, reciprocally, the velocity of earth through space.

Michelson and Newcomb had often consulted two famous American optical-instrument makers and enthusiasts, John Brashear and Alvan Clark, for advice and aid in constructing critical components for their experiments. In all likelihood this consultation happened again in 1886 and 1887, but there was little if anything new enough or tricky enough to require their Yankee ingenuity. Morley was able to procure most of the parts off the shelf from suppliers right in Cleveland. The local high-tech firm of Warner and Swazey already had a reputation for helping support pure scientists in hope that profitable applications might result. Michelson and Morley’s main idea in 1887 was to iron out all the bugs from the 1881 experiment and redo it on a vastly increased scale, calculating the relative motion of the earth on both a daily and a seasonal basis throughout the year. They hoped to find out whether the ether moved or was stationary, and they still hoped to finally find the velocity of the solar system.

Michelson designed a symmetrical layout for his beam splitter and mirrors so that the pencils of light would be multiply reflected over a path ten times as long as at Potsdam before recombining to produce interference fringes. Morley developed a greatly improved mount for the apparatus. He built it on a stone slab about five feet square and one foot deep. To keep the stone rigorously horizontal and vibrationfree yet allow it to rotate slowly and smoothly, Morley ingeniously arranged to float it in mercury. The slab rested on a doughnut of wood sitting in mercury in a ring-shaped castiron trough. A pivot in the center kept the pieces concentric. A suitable sandstone slab was found, and Michelson and Morley had masons install the iron trough on a bed of cement atop a brick foundation pier sunk to the bedrock beneath their basement lab. The floating wooden raft became well coated with metallic-fluid lubricant as soon as the slab was in place, centered, and leveled on it. The whole basic device worked so well that it would rotate very slowly for half an hour from the momentum of one gentle push of the hand.

Michelson’s arrangement of the optical parts fixed them all at comfortable working height. The light came as before from the bright flame of an ordinary Argand oil burner, passed through a small, narrow slit and rendered parallel by a lens. Then the beam shot to the surface of the beam splitter. That half-silvered, plane-parallel glass, set at forty-five degrees near the center of the apparatus, divided the beam into two equal pencils of light. Each moved over paths at right angles to the other, being reflected among a set of four speculum mirrors clustered at the four corners of the square stone. Again, a plane-parallel compensator glass equalized the pathway of that half-beam or pencil that had traversed the separating plate only once. Both the observer’s small telescope and the micrometer adjustment screw for one mirror were located at the same corner. Sixteen marks for compass points around the base ring made easy the uniform recording of fringe displacements as the observer followed the device’s slow rotation, completing a full circuit in about five minutes.

On April 17, 1887, Morley wrote his father that he and Michelson were progressing well with the new experiment. They expected “decisive results” to begin to accumulate in a month or two, when after many delicate adjustments the apparatus should become operational. They also expected to make observations “every month for a year” before finishing. By the end of classes in June, Michelson and Morley had taken many preliminary observations. It was tedious work. The observer had to walk slowly around with the rotating apparatus and neither let a moment lapse when he wasn’t peering into the quarter-inch eyepiece nor ever actually touch the eyepiece or any other part of the equipment. Morley remarked, “Patience is a possession without which no one is likely to begin observations of this kind.”

Their data reduction quickly proved sorely disappointing. The results were turning out uniformly null. Yet they drafted a report, “On the Relative Motion of the Earth and the Luminiferous Ether,” and published it forthwith. After that summer of 1887 the two abandoned all further observations with their classic interferometer, although they later continued working in interferometry separately and with others. Their paper offered at least seven ideas (four for laboratories and three for observatories) for attacking all over again the problem of the motion of the whole solar system through space. Michelson concluded that nearby ether must be spinning with the earth like an atmosphere. Physicists were bewildered by the test’s results but accepted them with complete confidence nonetheless.

Twelve years later, in his final lecture in a series at Harvard, Michelson confessed: “With good reason it is supposed that the sun and all the planets as well are moving through space at a rate of perhaps twenty miles per second in a certain particular direction. The velocity is not very well determined, and it was hoped that with this experiment, we could measure this velocity of the whole solar system through space. Since the result of the experiment was negative, this problem is still demanding a solution.”

By 1903, when his Harvard lectures were published as his first book, Light Waves and Their Uses , Michelson’s interferometer had evolved into numerous species of his own devising, many of which, such as his stellar interferometer and his solid-earth-tides interferometer, were still being thought out. He had constructed and used for momentous investigations an interferential comparator (to standardize the meter as a number of light wavelengths), a large-scale vertical interferometer, an echelon spectroscope (using a series of glass plates to observe spectra of light), a mechanical harmonic analyzer (to analyze harmonic motions in interference fringes), and some two dozen other interferometric instruments for specific applications. The British firm of Hilger and Watts was one of several companies that regularly manufactured a standard bench model of Michelson’s interferometer by the turn of the century. By then there were also a host of competitors. They all owed much to the fertile, unpatented ideas and investigations of Michelson.

He had gone on to Clark University for a few years and thence to Paris to measure the official platinum-irridium onemeter bar in terms of light waves from the homogeneous red radiation of cadmium vapor (1 meter, he found, equals 1,553,163.5 red cadmium light waves in air at fifteen degrees Celsius and normal atmospheric pressure). He moved to the new University of Chicago, built on the remains of the Columbian Exposition in 1894, and created its physics department. Most important, by the turn of the century he had studied and explored the principles of spectroscopy and interferometry enough to understand fully the amazing versatility of interference methods for practical applications. Any microscope or telescope could be converted into an interferometer. The accuracy of linear and angular measurements was increased dramatically.

Albert Einstein in 1905 declared the ether to be superfluous. This came out of his critique of the idea of simultaneity, in his analysis of the electrodynamics of moving bodies, wherein he postulated the invariance of the velocity of light and the ubiquity of relativity. Two years later Albert Michelson was awarded the Nobel Prize. He continued his contributions to experimental science with hardly a pause until his death in Pasadena in 1931, at a time when he was trying to perfect his measurement of the velocity of light in a mile-long vacuum tube. By then Einstein had become the world’s most famous scientist, but Michelson and Morley went to their deaths continuing to worry about the skeleton in the closet that their interferometry had seemed to uncover and that relativity and quantum theories seemed to cover up. Max Planck wrote that old theories never die, old theorists just die off. He must have been thinking in part about Michelson and Morley, although they were experimentalists first, musicians second (Morley played the organ, Michelson the violin), and theorists hardly at all.

At the end of his second and last book, Studies in Optics , Michelson reluctantly urged a “generous acceptance” of Einstein’s general theory of relativity. But characteristically he withheld acceptance of the annihilation of the ethereal ether. How can we ever explain the propagation of light if there is no medium for waves to wave in? he asked. And we may ask, If the velocity of light is evidently as well as theoretically a universal constant, or a finite limit with infinite implications, then whose inventions and measurements made it so? Michelson and Morley stumbled into misplaced glory by association with Einstein’s name after 1920, when it became a common but wrong assumption that their 1887 experiments had directly inspired Einstein to theorize relativity. They were embarrassed ever after; their relation to Einstein’s work was complex and indirect.

In 1903 Michelson acknowledged that although the etherdrift experiment had failed in its stated purpose, it had certainly borne fruit. He wrote, “I think it will be admitted that the problem, by leading to the invention of the interferometer, more than compensated for the fact that this particular experiment gave a negative result.” At another time he described why his work had pleased him personally: “It is the pitting of one’s brain against bits of iron, metals, and crystals and making them do what you want them to do. When you are successful that is all the reward you want.”