Failure to Detect the Ether
Michelson and Morley rotated their interferometer expecting to watch the fringe pattern shift by a small, precisely predicted amount. It never did — not that day, not at any other time of year, not at any orientation at all. That null result turned out to be far more important than a positive one would have been.
The null result
When Michelson and Morley rotated their interferometer, they found no fringe shift at all — not the shift the ether hypothesis predicted, nor any other detectable shift. This wasn't a borderline, ambiguous result: their apparatus was sensitive enough to have detected the predicted ether-wind effect many times over, and yet the fringe pattern stayed put. They repeated the experiment at different times of day and different times of year, in case the Earth's motion through the ether happened to cancel out at the moment of their first attempt — and still found nothing, at any orientation of the apparatus, at any time.
Purpose: to detect the Earth's absolute motion through the hypothesised luminiferous ether, by measuring the effect of the "ether wind" on the speed of light in different directions. Method: split a beam of light into two perpendicular paths of equal length using a half-silvered mirror, reflect each back along its own path with a mirror, recombine the beams to produce an interference pattern, and rotate the whole interferometer through 90° while watching for the predicted shift in that pattern as the arms' orientation relative to the Earth's motion changed. Result: no fringe shift was observed — none, at any orientation of the apparatus, at any time of day or year, to a precision easily good enough to have detected the expected effect many times over. Conclusion: the experiment's real significance lies in what this failure ruled out.
What the null result actually rules out
Faced with this result, there are really only two logical possibilities. Either the ether does not exist at all, or the Earth is somehow never moving relative to it. The second option can be ruled out directly: the Earth's orbital motion around the Sun continuously changes direction throughout the year, so if a stationary ether existed, the Earth's velocity relative to it would necessarily vary across the seasons — and yet the experiment found nothing, at any time of year, in any orientation.
The experiment's real significance, then, isn't really about the ether specifically — it's about what the failure implies more broadly. No experiment, however sensitive, was able to detect any absolute motion of the Earth through any proposed medium, in any direction, at any time. This undermined the entire concept of a fixed, universal frame of reference against which "true" or "absolute" motion could ever be measured — a concept classical physics had quietly assumed was available in principle, even if no one had measured it yet.
The single explanation that best accounts for the null result is that the speed of light is simply the same in every direction, for every observer, regardless of the Earth's motion — there is no ether wind to detect because light's speed doesn't depend on the observer's motion relative to any medium at all. This is the invariance of the speed of light: a genuinely strange experimental fact, confirmed by the total absence of any fringe shift, that classical physics had no mechanism to explain. Nothing in Newtonian mechanics predicts that a wave's speed should be independent of the observer's own motion — every other wave physicists knew of had a speed fixed relative to its medium, exactly the behaviour the Michelson-Morley experiment failed to find in light.
Two decades later, Einstein took this invariance not as a puzzle still waiting to be explained away by some undiscovered effect, but as a fundamental starting assumption — a fact of nature to be built on rather than solved. That decision, and the two postulates it led to, is the subject of the next two lessons.
Key takeaways
- The Michelson-Morley experiment found no fringe shift at any orientation of the apparatus, at any time of day or year — a genuine null result, well within the sensitivity of the instrument to detect the predicted effect.
- The Earth's changing orbital direction throughout the year rules out the possibility that the Earth simply happens to be at rest relative to the ether.
- The experiment's real significance is that no absolute motion through any medium could be detected — undermining the entire concept of a fixed, universal reference frame.
- The best explanation for the null result is that the speed of light is invariant: the same in every direction, for every observer, regardless of the Earth's motion.
- Classical physics had no mechanism to explain this invariance, since every other known wave had a speed fixed relative to its medium.
- Einstein later took this invariance as a fundamental starting assumption for special relativity, rather than an anomaly still needing explanation — covered in the next two lessons.