Chapter 9

Michelson & Morley’s experiments, designed to detect the absolute[1] velocity of the Earth with respect to the stationary ether, were fraught with false assumptions.  These false assumptions included, inter alia, the existence of stationary ether, Maxwell’s assumptions of the greater distance and time interval for the to and fro passage of light in the direction of the absolute solar orbital velocity of the Earth, the existence of an “ether wind,” that light has lateral inertia, and the assumption that the mirrors in Michelson’s apparatus would displace from the stationary ether in the direction of the Earth’s solar orbital motion.  The null results of M & M’s experiments, in combination with such false assumptions, resulted in an absolute paradox.  This paradox totally baffled the scientific community and helped to trigger the invention of Special Relativity.  Nevertheless, such paradox is now fully explainable.

During early 1879, James Clark Maxwell was pondering possible experimental methods to detect the absolute velocity of the Earth with respect to the stationary ether (Goldberg, pp. 433 – 438).  In March 1879, he sent a letter to American astronomer D. P. Todd who had been assisting him in his endeavor.[2] In this letter Maxwell described a possible theoretical method to detect such absolute velocity of the Earth, based on the theoretical “effect of the earth’s motion on the there-and-back speed of light [from a light source to a mirror and back] as measured in the laboratory” (Hoffman, 1983, pp. 75-76; Figures 9.1A and Figure 9.2A).  A relevant portion of Maxwell’s letter states as follows:

“[I]n the terrestrial methods of determining the velocity of light, the light comes back along the same path again, so that the velocity of the earth with respect to the ether would alter the time of the double passage by a quantity depending on the square of the ratio of the earth’s velocity to that of light [(v/c)2] and this is quite too small to be observed”  (Nature , 1880, Vol. XXI, p. 315).

Stated a different way:  the round trip time interval for a terrestrial light ray “does in principle depend on the speed of the earth through the ether”  (French, p. 50).  It is important to realize and remember that these conjectures by Maxwell, French and others with respect to the velocity of light, the ether, and such greater time interval of light propagation were completely false assumptions (even though they may appear to be plausible).

In 1895, H. A. Lorentz interpreted Maxwell’s hypothesis to mean that “the time required for a ray of light to travel from a point A to a point B and back to A [on the Earth’s surface] must vary when the two points together undergo a displacement …”with respect to the stationary ether  (Figures 9.1B, Figures 9.2B and 9.2C).  Lorentz called this theoretical “in-tandem displacement” of points A and B “the Maxwell displacement”  (Lorentz, 1895 [Dover, 1952, p. 3]).  These displacements by Lorentz were also false assumptions, because stationary ether does not exist and something cannot physically displace from nothing.

According to Maxwell, if the time interval for light to travel from point A to point B and back to A on Earth varied in any absolute direction of the Earth’s velocity through the theoretical ether (Figures 9.1 and Figure 9.2), then this would be a way to detect the theoretical absolute velocity of the Earth through the stationary ether.  But what was the magnitude of the time interval for the double passage of light on the moving Earth varying from?  A priori, it was altering or varying from the theoretical absolute time interval for the double passage of light assuming that the Earth was at rest in the stationary ether  (Figure 9.1A).  This was yet another false assumption which we shall discuss in detail in Chapter 10A.

When Russian-American scientist Albert Michelson (1852 – 1931) read Maxwell’s letter in the scientific journal Nature , he asserted that this miniscule theoretical difference of time interval, if it actually existed, could easily be detected and measured by an interference of light method  (Michelson, 1881, pp. 120 – 121; Hoffman, 1983, p. 76).  Acting upon his intuition, Michelson in 1881 created an interference of light experiment to test for Maxwell’s theoretical time interval variation, by attempting to detect and measure the assumed absolute solar orbital velocity of the Earth (30 km/s) with respect to the stationary ether[3]  (Michelson, 1881, p 120).

Michelson assumed that “the actual distance that light travels in the [absolute direction of the Earth’s solar orbital motion in his experiment] is greater than in the [direction perpendicular to such motion]”[4]  (Michelson, 1881, p. 121).  Michelson further assumed that light which propagated from Earth perpendicularly to the absolute direction of the Earth’s absolute solar orbital motion “would be entirely unaffected” by such motion  (Ibid).  In Michelson’s words:

“Assuming then that the ether is at rest, the earth moving through it, the time required for light to pass from one point to another on the earth’s surface, would depend on the direction in which it [light] travels”[5](Ibid ).

Michelson then concluded that if it were possible to measure the time T required for light to pass from one point to another in “the direction of the earth’s motion [less the] time [T1] required for it to pass in the opposite [perpendicular] direction, [then] we could find v the velocity of the earth’s motion through the ether”[6]  (Ibid).  This theoretical difference between T and T1 was also the theoretical variation between T and T0.  According to Michelson, T0, was the “time required to perform the journey if the earth were at rest[7]  (Michelson, 1881, p. 120; Figure 9.1A).

How did Michelson propose to measure the theoretical difference between T and T1?  Prior experiments designed to determine this difference time intervals (or in velocity) to the first order or approximation v/c (one part in ten thousand), had failed to detect any motion or velocity of the Earth with regard to the ether.[8]  Pursuant to Maxwell’s suggestion, Michelson proposed to increase the sensitivity and precision of the measurement to an approximation of the second order:  v2/c2 (one part in one hundred million)[9]  (Goldberg, p. 90; Hoffmann, 1983, p. 76; Halliday, pp. 960, 961).

Similarly to the 1851 Experiment of Fizeau, Michelson built a very sensitive apparatus (an “interferometer”) with a 50% reflective glass mirror (a ‘beam splitter’) that would split a ray of light into two equal parts, called light ‘pencils’  (Figure 9.3A).  Light pencil 1 would propagate to and fro (between the beam splitter BS and mirror M1 on the apparatus) parallel (or longitudinal) to the assumed direction of the Earth’s absolute solar orbital motion.  Light pencil 2 would propagate to and fro in the direction of mirror M2, perpendicular (or transverse) to the Earth’s assumed absolute solar orbital motion.

Very importantly, using a micrometer screw, Michelson adjusted the light path for light pencil 1 so that it was slightly different in distance than the light path for light pencil 2  (Michelson, 1881, pp. 122 – 123; Figure 9.4).  This adjustment of distance caused the two light pencils to slightly change their relative wave phases by the time they propagated to and fro and joined again at the beam splitter.  As a result, Michelson created and observed an interference fringe pattern through the telescope  (Figure 7.4C).  Michelson’s above described adjustment of distance will be critical for our explanation of Michelson’s null result paradox in Chapter 12.

When Michelson’s apparatus was then rotated through 90°, the directions of the two light pencils were also rotated 90°.  As a result, Michelson assumed that first pencil 1 and then pencil 2 would take longer to propagate between their respective reflecting mirrors in the assumed direction of the Earth’s absolute solar orbital motion.  Also, as a result of this rotation, Michelson expected that the observed interference fringe would shift about 1/10th of a light wave[10]  (Ibid, pp. 122, 125).  Theoretically, this shift of interference fringes (Figures 7.4D-1 and 7.4D-2) would indicate the difference in time interval between T and T0 (or T1) in the apparatus.  In turn, Michelson assumed that such time interval difference would indicate the magnitude of the absolute solar orbital velocity of the Earth through the stationary ether (30 km/s).

As previously mentioned, Michelson based this theoretical magnitude of the distance/time interval difference for light propagation solely on “the velocity of the earth in its orbit” as compared to “the earth at rest” in stationary ether  (Michelson, 1881, p. 121).  However, Michelson did not realize that this theoretical difference in magnitudes of velocity was a completely meaningless assumption on its face.  Why was it meaningless?  First, because the false concepts of stationary ether, the Earth at rest in the ether, and the absolute velocity and absolute direction of the Earth’s solar orbital motion with respect to stationary ether as reference body, have no meaning  (Chapter 10A).  Secondly, because such theoretical difference in velocity did not take into account the infinite number of other relative velocities of the Earth, such as the seven times greater relative velocity of the Earth (together with the solar system) around the core of the Milky Way Galaxy at 225 km/s  (Figures 10.3 and Figure 10.4, and Chart 10.5).  Thirdly, since the Earth cannot be at rest in non-existent ether, such comparison of time intervals is impossible.

Nevertheless, Michelson repeatedly pointed his 1881 apparatus in all possible directions over several periods of time as the Earth orbited the Sun.  However, the result of this experiment was only a very nominal shift of the interference fringes when the pencils of light were pointed in any different direction.[11]  The observed nominal fringe shift averaged only about 2% of that which Michelson expected, and he attributed these nominal fringe shifts to observational error  (Michelson, 1881, pp. 127, 128).

Very importantly, Michelson concluded from this basically null result[12] that:  “the hypothesis of a stationary ether…is erroneous…This conclusion directly contradicts [the hypothesis]…that the earth moves through the ether, the latter remaining at rest”  (Michelson, 1881, p. 128).

It was not realized at the time, but Michelson’s 1881 experiment, its empirical null result and Michelson’s above conclusions concerning non-existent ether, necessarily implied the most profound changes in the philosophy of physics.  If there was no ether absolutely at rest in space, then:  1) there was no absolute reference frame (body or material substance) at rest in space from which the absolute motions of celestial objects could be measured; and 2) there were no such things as absolute space, absolute time, absolute rest, absolute positions, absolute directions, and absolute motions or absolute velocities.  Thus, all positions, time intervals, states of rest, motions, velocities, directions of motion, and trajectories of co-moving celestial objects must be completely relative.  But the scientific community was not ready to consider or accept such “radical” and unorthodox conclusions.

It is one thing to realize and state (or infer) that there is no such thing as ether, absolute rest or absolute motion in the universe, as Einstein also did twenty-four years later in 1905 (Einstein, 1905d [Dover, 1952, pp. 37 – 38]).  It is quite another thing to make the ad hoc claim that relative motion and relative velocity (of itself and at a distance) alter the dimensions, mass and time intervals of physical objects.  But this is exactly what Einstein asserted in his 1905 Special Theory of Relativity  (Ibid, pp. 41 – 65).

It follows from the null results of Michelson’s 1881 experiment that, if a light ray emitted on the moving material Earth in all directions takes the same interval of time to travel from light source to mirror and back in all directions of the Earth’s motion, then light constantly propagates in all directions through empty space at the same constant transmission velocity, regardless of the assumed motions or velocities of its source body (the Earth).  It also follows that the assumed solar orbital motion of relatively stationary bodies (i.e. Michelson’s mirrors) toward which such light ray propagates are irrelevant to such transmission velocity of light and to the time interval of such propagation.[13] Likewise, Michelson’s 1881 experiment also implied that the inertial motions of any material body (the Earth, the mirrors or any distant other body) are irrelevant to such transmission velocity of light.  But none of these implications were either realized or articulated at the time, and the paradoxical null results of Michelson’s 1881 experiment completely baffled the scientific community of the late 19th century.

For many decades astronomers had observed, measured and documented the Earth’s orbital motion around the Sun.  So seemingly obvious and ingrained was the absoluteness of this observation, and that the Earth’s constant solar orbital motion would cause a specific absolute in-tandem displacement of the light source and longitudinal mirror from the light ray, that the great minds of Maxwell, Michelson, Morley, Fitzgerald, Lorentz, Einstein, and others never seemed to doubt its validity.[14]  Nor did they doubt that such absolute displacement and time interval comparison could be detected by Michelson’s interference method.  Thus they also reasoned that the time interval traveled by a light ray in the absolute direction of such constant solar orbital motion would be greater over the increasing distance caused by such absolute in-tandem displacement (Figures 9.1B, Figures 9.2B and 9.2C) than if the Earth were absolutely at rest in the stationary ether (Figures 9.1A and Figure 9.2A; Einstein, Relativity, pp. 58 – 59). But if the distance was greater, why had not Michelson detected a fringe shift which would indicate a greater time interval in the direction of motion?  What was the answer to this paradox, this missing time interval?

Because of the numerous theoretical and technical problems that Michelson had encountered during his 1881 experiment,[15] he decided to repeat the experiment in 1887 with fellow American scientist Edward Morley (1838 – 1923).  As in the 1881 experiment, Michelson and Morley (M & M) assumed that an in-tandem physical displacement of the light source and the longitudinal mirror would occur relative to the light ray, because of the known solar orbital motion of the Earth at 30 km/s  (Figures 9.1B, Figure 9.2B and 9.2C).  M & M further assumed that this displacement would increase the distance interval and the time interval for the light ray to travel in the absolute direction of such solar motion from beam splitter to the longitudinal mirror and back, and that such increased time interval should be detectable by his interferometer.[16]

In 1887, the design of Michelson’s apparatus was improved, and the path that each pencil of light would propagate in each perpendicular direction was increased eleven fold to about twenty-two meters to and fro  (M & M, 1887, pp. 337, 341).  During this second experiment, only the solar orbital velocity of the Earth was again considered, because Michelson & Morley acknowledged that “little is known [of] the motion of the solar system” relative to the motion of the other celestial bodies  (Ibid, p. 341).  M & M also acknowledged that it would be hopeless to attempt to solve the question of the motion of the solar system, but if it were added in, “the result would have to be modified”  (Ibid).

The most significant theoretical modification that Michelson made for his 1887 experiment was to change his hypothesis concerning the path of the light pencil that was transverse to the solar orbital motion of the Earth.  In this regard he stated that in 1881, “[T]he effect of the motion of the earth through the ether on the path of the ray at right angles [transverse] to this motion was overlooked” [17]  (Ibid, p. 334).

Michelson then asserted that the paths of the two light pencils would only be perpendicular to each other if his apparatus was at rest in the ether  (M & M, 1887, p. 335).  According to Michelson, when the apparatus moves through the ether in the direction of the Earth’s solar orbital motion, the to and fro path of the transverse light pencil would be at an angle relative to the longitudinal light pencil, which angle would be in the direction of such solar orbital motion  (Ibid, pp. 335 – 336; Figure 9.5).  Michelson then concluded, that because this angle would increase the time interval which the transverse light pencil would have to propagate, the difference between T and T1 would actually only be 40% of the magnitude which he had assumed in his 1881 experiment.  This new magnitude of time interval difference would only be about 0.04 of a fringe, instead of the 0.10 of a fringe that Michelson assumed in 1881  (Ibid, pp. 335, 336).

Michelson’s new hypothesis was, of course, yet another major false assumption.  We now know that the photons that comprise a light ray do not have mass and therefore the light ray is not subject to the lateral inertial motion nor the momentum of its material light source (the Earth), which inertial motion would be necessary to create the angled path of light propagation envisioned by Michelson.  In reality, the transverse light pencil would always propagate perpendicularly to the longitudinal light pencil and rectilinearly toward the transverse reflecting mirror.  It would then reflect back toward the beam splitter in a perpendicular and rectilinear path[18]  (Figure 9.6).

The significance of these latest false assumptions by Michelson (for the purposes of this treatise) is two-fold:

1.  In 1895, Lorentz created his contraction of matter theory and his contraction factor

in order to explain why Michelson’s 1887 theoretical magnitude of a fringe shift (0.04% of a fringe) could not be detected by Michelson’s 1887 experiment (Lorentz, 1895 [Dover, 1952, pp. 3 – 7]).  Lorentz’s incorrect 1895 assumptions and contraction concepts were then incorporated into his 1904 transformation equations for distance and time (Ibid, p. 14), which were in turn ‘borrowed’ by Einstein in 1905 as the foundation for his Special Theory.[19]

2.  Michelson’s false assumption concerning the angular path of the transverse light pencil was later adopted by Einstein and his followers, incorporated into Einstein’s relativistic moving light clock theory, and asserted as an experimental confirmation for the validity of Special Relativity  (Cropper, pp. 209 – 213).  But as explained by the above analysis and in Chapter 26D, Einstein’s moving light clock hypothesis and its asserted confirmation of Special Relativity were both based on a false premise:  the lateral inertial and angular propagation of a perpendicular light ray.  Thus, what validity does Einstein’s relativistic moving light clock hypothesis have for any purpose?  None  (Chapters 26D, 28 and 36).

Very importantly, just before the 1887 experiment began, M & M adjusted the micrometer “screw altering the length of [the] path” for one light pencil to propagate  (M & M, 1887, p. 339).  As in the 1881 experiment, this adjustment of distance caused two slightly out-of-phase light pencils to result in an interference fringe pattern  (Ibid; Figures 9.4 and Figure 7.4C).  This seemingly insignificant change of distance for one light pencil to propagate to and fro will be exceedingly important for our analysis in Chapter 12.B of one empirical and technical reason for Michelson’s null results.

M & M then pointed their new apparatus in all possible directions over a six-month period hoping to detect some fringe shift (time interval difference) in some direction.  But at the end of the day, the outcome of the M & M experiment in 1887 was substantially the same as Michelson’s 1881 experiment.  Again, only a very nominal shift of the interference fringes was observed (Ibid, pp. 340, 341); and basically it was considered to be yet another null result with the same conclusions as the 1881 experiment[20]  (Ibid; Figure 9.7).  The scientific community remained baffled.[21]

Why had not the absolute orbital velocity of the Earth and the displacement of Michelson’s mirrors relative to the light ray, produced the expected fringe shift and time interval difference?  How could light propagate atc over two theoretically different distances (longitudinal and transverse) during the same time interval?  Did this mean that the transmission velocity of light changed in different directions of motion?  Did it mean that the concepts of ether and ether at rest in space were invalid, as Michelson concluded in 1881?  Did it mean that Michelson’s interference fringe shift experiments and the assumptions that they were based upon were at fault?  Did it mean that Maxwell’s law of the constant transmission velocity of light at c needed to be revised (Einstein, Relativity , p. 23)? What could be the answers to these riddles?  These were some of the baffling questions that faced the scientific community after 1887.

The first attempted explanation for the null result was that the ether was completely carried along with the moving Earth at its surface, and thus the velocity of the Earth’s surface relative to the carried (dragged) along ether was zero (relatively at rest)  (Figure 9.7A).  But this theory, first suggested by English physicist George Stokes in 1845, was soon totally discredited by experiments, calculations and logic  (Hoffmann, 1983, p. 81).  Another attempted explanation of the M & M null results was the ballistic theory of light, which asserts that light consists of particles rather than waves.  This theory claims that the velocity of light “is uniquely defined with respect to the source…”  (French, p. 57).  In other words, that the transmission velocity of light varies depending upon the velocity of its source body.  But this explanation was also discredited by Willem de Sitter and others (Figure 7.2).

The empirical null results of Michelson’s experiments implied that the light source and longitudinal mirror rigidly attached to the Earth might not be physically displacing in-tandem away from the ray of light (propagating within the same apparatus), contrary to the assertion of such displacement in Maxwell’s 1879 letter.  But how could this be?  The Earth’s orbital velocity around the Sun was intuitive and obvious, so everyone assumed that the in-tandem displacement of the light source and longitudinal mirror from the propagating light ray must also exist.  They also assumed that this displacement would increase the distance interval and time interval that the light ray must propagate relative to the light source and longitudinal mirror in the direction of such motion.  Must not there be some plausible explanation for these paradoxes?  For many frustrated scientists, any explanation might do.

The quest for a plausible explanation led Fitzgerald, Lorentz, and ultimately Einstein to somewhat different radical ad hoc theories that the Earth and Michelson’s apparatus must be physically contracting in proportion to the solar orbital velocity of the Earth relative to the velocity of light  (Chapters 10A, 15 and 16).  “[T]he amount of [such] contraction being just sufficient to compensate for the difference in time”  (Einstein, Relativity, p. 59).  Miller characterized these radical contraction theories as “clearly a physics of desperation   (Miller, p. 28).

Why are we even concerned about the above paradox and the possible or real reasons for the M & M null result in the twenty-first century?  Because:  (1) the baffling theoretical paradoxes which such null results and the ether theory created ultimately led Einstein to his own radical ad hoc contraction explanation for such absolute paradoxes which he called the Special Theory of Relativity; and (2) the null result of the M & M experiment is now claimed by the followers of Einstein to be one of the major experimental confirmations for Special Relativity[22]  (Resnick, 1968, p. 37; French, p. 56 – 57; Chapter 36).

Despite such desperate and contrived contraction explanations, we now ask the questions:  Were the paradoxical results of Michelson’s experiments, and the contraction explanations of them, properly analyzed and adequately scrutinized? Are there other much more probable explanations for these puzzling null results?  In the next three chapters we shall discuss the many false assumptions that caused M & M’s null results to become a monumental paradox, and the much more probable real, empirical, and technical answers for this 120-year-old absolute mystery.