A chronometer is not a type of watch. It is a verdict: an independent body tested a timekeeper and found its rate good enough to earn the title. The idea began with marine instruments, grew into the observatory trials of Neuchâtel, Geneva and Kew, and survives in COSC and the other certification schemes, and its logic has not changed since the first trial. A precision claim counts only when someone other than the seller has checked it.

A marine chronometer, the instrument that gave the word its meaning. The first chronometers kept time at sea, where a small error meant a large error in position. Photo: Wikimedia Commons, public domain.
Marine chronometers set the template
The London maker John Arnold coined the term “chronometer” together with the hydrographer Alexander Dalrymple, and the word at first meant only the marine timekeeper, a spring-driven box kept in gimbals so it stayed level at sea. A ship that knew the time at a reference meridian could work out its longitude, so the instrument’s accuracy was a matter of navigation, not of pride. The marine chronometer’s job was to hold its rate over weeks and months in changing temperatures and under motion, and that standard of long, steady performance is still the one a modern chronometer certificate is judged against.
The marine service also created the habit of testing. Before a chronometer went to sea it was rated against an observatory clock, and its daily gain or loss was recorded so a navigator could correct for it. That habit, of an outside body checking a timekeeper against a reference and publishing the result, is the direct ancestor of every modern chronometer test.
The observatory trials
Neuchâtel’s observatory, founded in 1858, hosted what Worn & Wound calls the most stringent accuracy competition in watchmaking history. Its trials ran for weeks, tested movements in several positions and at several temperatures, and scored them on the variation between those readings rather than on a single day’s rate. Geneva and Kew ran similar contests, and the annual results became a public ranking of makers. Winning at Neuchâtel was a marketing prize and a technical one, because the trials exposed weaknesses that a maker’s own testing had missed.
Wristwatches entered the system early. Rolex’s small ladies’ wristwatch received a Class A certificate at Kew in 1914, the first time a wristwatch had earned the top grade, and Omega’s 1936 result at Kew, with a score of 97.4 out of 100, stood as the best ever recorded there. The trials rewarded thin, well-finished movements that held their rate across positions, and the winners were often the same calibres that later became the basis of production watches.

Kew Observatory in London, home of some of the most famous chronometer trials in watchmaking history. Photo: Wikimedia Commons, CC BY-SA 4.0.
Zenith won no fewer than 2,333 prizes in its first century, according to Europa Star, including five consecutive first prizes at Neuchâtel in the 1960s. The prizes were not for a single watch but for a series of movements submitted over the year, so the count reflects a factory’s sustained accuracy rather than one lucky entry.
Girard-Perregaux made 662 wristwatches with the 36,000 bph Calibre 32A in 1966 and 1967. Neuchâtel tested 662 of them and awarded the observatory’s highest distinction to every one, a result that showed how far production accuracy had come since the trials began.
Quartz empties the scoreboard
In 1968 Geneva ran its own trials, and experimental quartz movements took the top three places overall, beating every mechanical entry by a wide margin. The following year the observatories stopped the contests altogether, because a quartz movement’s rate was so much better than a mechanical movement’s that the comparison had become meaningless. The trials had done their work: they had pushed mechanical accuracy to its limit, and a new technology had passed it.
COSC replaces the contest with a standard
The COSC (Contrôle Officiel Suisse des Chronomètres) was founded at the Château de Neuchâtel, and its statutes were adopted at a founding assembly in La Chaux-de-Fonds on 7 September 1973. It replaced the contests with a pass-or-fail test: a movement is tested for 15 days in five positions and at three temperatures, and its average daily rate must fall between -4 and +6 seconds. A movement that passes earns the right to be called a chronometer, and the certificate is issued for the movement, not for the finished watch. The full test is covered in the COSC article.
The criteria correspond to ISO 3159. In 2009 the Watch Museum of Le Locle launched a contest based on the old observatory format, testing finished watches rather than bare movements, but it remains a niche event. The mainstream chronometer label is now the COSC certificate, and its rivals are the stricter in-house standards that some makers apply on top of it.
Outside Switzerland
Germany’s Glashütte Observatory tests to DIN 8319, which is nearly identical to COSC’s regimen. The tests differ in one respect: DIN 8319 is applied to the finished, cased watch instead of the bare movement. The procedure is 15 days in five positions at 8, 23 and 38 °C against a rate of -4 to +6 seconds per day. Only movements with a hacking function are accepted, and the observatory certificate also requires final assembly in Glashütte with at least 55 percent of the movement components made in the town.
Sources
- “John Arnold,” Grace’s Guide
- “History of Chronometers Pt. 2: Observatory Trials,” Worn & Wound
- “History of Chronometers Pt. 3: Enter COSC,” Worn & Wound
- “History of Chronometers Pt. 4: In-House and Beyond COSC,” Worn & Wound
- “Observatory chronometer,” Wikipedia
- “1914: A Milestone in Watchmaking History,” Watchonista
- “Rolex, the Kew Observatory, and the ‘A’ Certificate,” Bob’s Watches
- “1936: In Reach of Perfection,” Omega
- “Zenith’s Two Chronos,” Europa Star
- “The COSC Is 50 Years Old,” FHS
- “Glashütte Observatory,” Horopedia
Recommended articles
Thanks for reading this.

