Newtonpedia ES
Home › Astronomy and space › Rømer's determination of the speed of light
Astronomy and space · 1676

Rømer's determination of the speed of light

Ole Rømer's 1676 demonstration that light travels at a finite, measurable speed, worked out from eclipses of Io, one of Jupiter's moons.

Year
1676
People involved
Ole Rømer
Section
Astronomy and space

History

Ole Rømer's determination, announced in 1676, showed that light does not travel instantaneously but at a finite speed that can be measured. The idea sounds simple, yet it changed how people understood light, since many astronomers had assumed it reached us at once from any point in space. His method relied on a regularly repeating celestial event, so that the time between events could be turned into a distance. Today that speed is a basic constant of physics and underpins space measurement, fibre-optic communications and satellite navigation.

Before Rømer, the most pressing practical problem was longitude, the east-west position of a place. Without it, navigation and mapmaking were highly unreliable. In 1598 Philip III of Spain offered a prize for a method of finding a ship's longitude out of sight of land. Galileo suggested using the eclipses of Jupiter's moons as a kind of celestial clock: if the exact time of an eclipse was known at a reference location, comparing it with local time would reveal the longitude. He proposed this to the Spanish crown around 1616 and 1617, but it proved impractical at sea, where eclipses were hard to observe from a deck. With refinements it could work on land. The Italian astronomer Giovanni Domenico Cassini pioneered this use and published tables predicting when each eclipse would be visible from a given place. Louis XIV invited him to France, and he became director of the Royal Observatory in Paris when it opened in 1671.

One of Cassini's first projects was to send the Frenchman Jean Picard to Uraniborg, on the island of Hven near Copenhagen, where Tycho Brahe had worked. Picard timed the eclipses there while Cassini recorded them in Paris. The difference between the two recorded times, just over 42 minutes, gave the difference in longitude between the two cities. Picard was helped by a young Dane who had recently finished his studies at the University of Copenhagen, Ole Rømer. Impressed by his skill, Picard arranged for him to join the Paris Observatory.

Io is the innermost of the four moons that Galileo discovered in January 1610. It completes one orbit of Jupiter every 42 and a half hours, and because its orbital plane is almost the same as Jupiter's, it often passes into the planet's shadow. From Earth, two events can be seen: an immersion, when Io vanishes as it enters the shadow, and an emergence, when it reappears as it leaves. Depending on the relative positions of Earth and Jupiter, one of the two may be hidden by the planet itself, and for several months each year Jupiter sits so close to the Sun that no eclipses can be observed at all.

Rømer's key observation was that the interval between eclipses was not constant but shifted slightly over the year. Because he was fairly confident that Io's orbital period did not really change, he concluded that the cause lay in the changing distance between Earth and Jupiter. When the two planets were moving apart, the interval grew; when they were moving closer, it shrank. This matched the idea that light takes longer to cover a greater distance. Rømer calculated that light would need about 22 minutes to cross a distance equal to the diameter of Earth's orbit, implying a speed of roughly 220,000 kilometres per second. That figure is noticeably below the true value of nearly 300,000 kilometres per second, but the order of magnitude was right.

His reasoning began with an order-of-magnitude argument: light had to be so fast that it crossed Earth's diameter in far less than a second. He considered an interval of about 42 and a half hours between two emergences of Io, during which Earth moved away from Jupiter by roughly 210 times its own diameter. If light travelled at one Earth diameter per second, that delay would add about three and a half minutes, which was not observed. He then applied the same reasoning to observations taken while Earth was approaching Jupiter. Finding no difference of that size, he concluded that the speed of light must be far greater.

Rømer also noticed that the effect built up over many observations. In spring 1672, for instance, Io's apparent orbital period came out about 32 seconds longer than the period derived from an earlier series. That seems tiny, but after thirty orbits the emergence on 29 April arrived roughly a quarter of an hour later than predicted. The only alternative was that the earlier observations were wrong by two minutes, which Rømer judged unlikely.

The announcement came in August 1676. On 22 August, Cassini told the Royal Academy of Sciences in Paris that he would change the basis of his eclipse tables, and that the new correction probably arose because light takes time to reach us from the satellite. Rømer predicted that the emergence of 16 November 1676 would be seen about ten minutes later than the old method gave. No observation of that particular emergence is recorded, but one was seen on 9 November. Rømer presented his method to the Academy around 21 or 22 November, according to the sources. The original minutes have been lost, but the Journal des sçavans published an anonymous report on 7 December, and an English translation appeared in the Philosophical Transactions of the Royal Society in July 1677. Rømer never published a formal account of his method, possibly because Cassini and Picard opposed his ideas.

Reception was uneven. Huygens backed the explanation from the start and wrote in its defence, and Newton also gave it his support. Cassini, however, never fully accepted it, and his objections fuelled lively debate at the Academy. He pointed out that the other Galilean moons did not seem to show the same effect and that some irregularities remained unexplained. Confirmation came nearly two decades after Rømer's death in 1710, when James Bradley explained stellar aberration in 1729, a small apparent shift in the positions of stars caused by Earth's motion combined with the finite speed of light.

Most of Rømer's papers were lost in the Copenhagen Fire of 1728, but a manuscript listing about sixty observations of Io's eclipses from 1668 to 1678 survived. Rømer went back to Denmark, where his work helped shape later efforts in timekeeping and mapping. Today measuring the speed of light is a pillar of physics and astronomy, used to calculate cosmic distances and to synchronise precise clocks. Rømer's method, with its limitations, showed that a repeating astronomical event could become a tool for measuring something as elusive as the travel of light.

Timeline

  • 1598Philip III of Spain offers a prize for a method of finding longitude at sea.
  • 1610Galileo discovers Jupiter's four moons, including Io.
  • 1671The Royal Observatory in Paris opens under Cassini's direction.
  • 1676In August Cassini announces a change to his Io eclipse tables at the Royal Academy of Sciences, and Rømer predicts a ten-minute delay for November.
  • 1677An English translation of the report on the method appears in London's Philosophical Transactions.
  • 1728The Copenhagen Fire destroys most of Rømer's papers.
  • 1729James Bradley explains stellar aberration, confirming that light has a finite speed.

Trivia and anecdotes

A quarter-hour delay

In 1672, a difference of just 32 seconds per orbit of Io built up until one emergence came about a quarter of an hour late. That cumulative effect was what Rømer announced.

A method never published

Rømer never published a formal description of his method. His explanation reached the public through an anonymous report in the Journal des sçavans.

Papers lost in a fire

The Copenhagen Fire of 1728 destroyed most of Rømer's papers. A manuscript with about sixty observations of Io's eclipses survived.

People involved

Other milestones in Astronomy and space