A short history of measuring the speed of light

For centuries, many people assumed light moved instantly. The idea was reasonable. A lightning flash and the thunder from the same storm seem to arrive at different times, but the flash itself appears at the same moment the event happens. The first good proof that light has a finite speed came from astronomy, where distances are large enough for a delay to show up in timing data.

Roemer and Jupiter’s moon Io

In 1676, the Danish astronomer Ole Rømer studied eclipses of Io, one of Jupiter’s four large moons discovered by Galileo. Io goes into and out of Jupiter’s shadow on a regular schedule. If light travelled instantly, those eclipses would always be seen at the same times once Io’s orbit was known.

Rømer noticed a pattern. When Earth was moving away from Jupiter, the eclipses seemed to arrive late. When Earth was moving toward Jupiter, they seemed early. The reason is simple. The light from Io had to cross a slightly longer or shorter path depending on where Earth was in its orbit. Rømer used this effect to argue that light needs time to travel.

He did not get the modern value for the speed of light, and he did not know the exact size of Earth’s orbit. He estimated that light takes about 22 minutes to cross the diameter of Earth’s orbit. Christiaan Huygens combined that figure with the best estimate of the orbit’s size at the time and arrived at about 220,000 km/s. The true crossing time is closer to 16.6 minutes, but the order of magnitude was right. That was a major step. It turned a philosophical question into a measurement problem.

Bradley and the aberration of starlight

In 1728, the English astronomer James Bradley found another way to measure light’s speed. He was studying the positions of stars and noticed a small yearly shift in where they appear in the sky. The effect is called aberration of starlight.

The idea is like walking in the rain. Even if the rain falls straight down, you tilt your umbrella forward because you are moving. Earth moves around the Sun at about 30 km/s, and light from a star has to enter a moving Earth-based observer. That changes the apparent direction of the incoming light by a tiny angle.

Bradley measured this angle as about 20.5 arcseconds. From that and Earth’s orbital speed, he could estimate the speed of light. The result was close to 301,000 km/s, very near the modern value. This was a cleaner method than Rømer’s in one sense, because it relied on geometry rather than timing eclipses over long periods.

Bradley’s work also did something else. It gave strong evidence that Earth really moves around the Sun. The same observation that helped measure light’s speed also supported the heliocentric model.

Fizeau and the toothed wheel

By the 19th century, scientists wanted a direct laboratory measurement. In 1849, the French physicist Hippolyte Fizeau used a spinning toothed wheel. A beam of light passed through a gap in the wheel, travelled to a mirror about 8.6 km away, and returned. If the wheel turned at just the right rate, a tooth blocked the returning beam.

Fizeau’s light source was in Suresnes and the mirror on the hill of Montmartre in Paris, a round trip of about 17 km. From the wheel speed at which the returning light was blocked, he found a value near 313,000 km/s. That was not exact by modern standards, but it was a remarkable laboratory result for the time.

Other physicists soon improved the measurement. Léon Foucault used a rotating mirror in 1862 and got a value closer to the modern one. Later optical and electronic methods refined it further. Each new method brought smaller uncertainties and less dependence on large astronomical distances.

From measured quantity to exact definition

For a long time, scientists measured the speed of light because they needed its value to define other things. That changed in 1983. The General Conference on Weights and Measures fixed the speed of light in vacuum at exactly

\[c = 299{,}792{,}458\ \text{m/s}\]

and defined the metre as the distance light travels in vacuum in 1/299,792,458 of a second. So today the speed of light is no longer a quantity with measurement error in the SI system. The error moved to the realisation of the metre, while the value of c itself became exact by definition.

This change matters because it ties space and time together in a very practical way. If you know time extremely well, you know distance extremely well. Laser ranging, GPS, and many modern experiments depend on that link.

Why these old measurements still matter

Rømer, Bradley, and Fizeau used very different methods, but they all worked because light has a finite travel time. Rømer used the changing distance between Earth and Jupiter. Bradley used Earth’s motion around the Sun and the tiny shift it causes in the apparent positions of stars. Fizeau used a laboratory beam and a spinning wheel.

Each method fits its era. Rømer’s was the first clear proof. Bradley’s was elegant and precise. Fizeau’s brought the problem into the lab. Together they show how astronomy and physics often grow by the same process: careful timing, geometry, and patience.

The history also shows why the modern value of c is so reliable. It was measured many times with different instruments and different ideas, until the value became exact in the SI system. In that sense, the speed of light is one of the best-known numbers in science, but it started as a subtle delay in the eclipses of a moon around Jupiter.

This topic fits naturally into the Astronomy & Astrophysics Masterclass curriculum, where light, motion, and measurement are used to connect the sky with physics.