Michelson interferometer

A Michelson interferometer divides light into two beams, sends them along separate paths, and then brings them back together. The combined light forms an interference pattern that can change when the beams’ relative travel times change.

A partially reflecting beam splitter divides the incoming light: it transmits part of the light and reflects the rest. The two beams travel along perpendicular arms. A mirror at the end of each arm reflects its beam back towards the beam splitter, where parts of the returning beams recombine and travel towards a detector.

Phase difference and fringes

The returning beams can arrive at the detector at different stages of their light-wave cycles. This phase difference depends on the difference between their travel times along the two arms.

Where the waves arrive in phase, they reinforce each other and produce a bright region. Where they arrive in opposite phases, they cancel and produce a dark region. The resulting pattern of bright and dark regions is called a fringe pattern.

If the relative travel time changes, the phase difference changes, so the fringes can shift. For example, moving one mirror changes the distance its beam travels on both the outward and return journeys. This changes its travel time relative to the other beam and can shift the pattern.

Michelson-Morley experiment

The Michelson-Morley experiment tested whether Earth was moving through a hypothesised stationary medium called the luminiferous ether. The ether was thought to be the medium through which light travelled.

The experiment used an interferometer with two arms at right angles. A beam of light was split into two beams, which travelled along the separate arms, reflected back and recombined. Their waves formed a pattern of bright and dark interference fringes. A change in the difference between their travel times would change how the waves recombined, shifting the fringes.

The predicted fringe shift

If Earth moved through a stationary ether, the motion was expected to affect light’s travel time differently along the two perpendicular arms. Rotating the apparatus would change the arms’ orientations relative to that motion. The difference between the travel times should therefore change, producing a shift in the interference fringes.

The null result

No fringe shift of the expected size was detected. This null result failed to reveal the expected absolute motion of Earth through the ether and challenged the idea that light required a stationary ether as its medium.

The result is consistent with the invariance of the speed of light, but the experiment alone does not establish that principle in full.

Invariance of the speed of light

Light travelling through a vacuum has the same measured speed for every inertial observer: approximately $$3.00 \times 10^8\ \mathrm{m,s^{-1}}$$. An inertial reference frame is a frame that is not accelerating; inertial frames move at constant velocity relative to one another.

This measured speed is independent of the motion of both the light source and the observer. For example, if one inertial observer moves towards a light source while another remains at rest relative to it, both measure light travelling through a vacuum at the same speed.

This result differs from the classical expectation that an observer moving towards an approaching object measures a higher speed than an observer at rest relative to its source. For light in a vacuum, speeds do not combine in this simple way. The invariance of the speed of light is a postulate of special relativity.