Marine Chronometer by Thomas Earnshaw

A one-day marine chronometer, serial number 509, made in London by Thomas Earnshaw around 1800 and issued by the Royal Observatory, Greenwich to a succession of Royal Navy ships over four decades, including HMS Beagle's second survey voyage of 1831 to 1836, the voyage that carried Charles Darwin. Captain Robert FitzRoy sailed with twenty-two chronometers rather than the usual three, a deliberate scientific redundancy intended to let him cross-check longitude readings against one another across a circumnavigation lasting five years.

Most of the twenty-two were sold, lost or scattered after the voyage, and tracing the survivors is itself active research rather than settled fact: how many are confirmed today depends on which published count is consulted. This instrument's own service record, by contrast, is unusually complete. The Greenwich ledger of chronometer issues tracks serial 509 by name through HMS Hecla, HMS Beagle, HMS Formidable and HMS Pembroke, then to shore duty at the Royal Society's Falmouth Observatory, and finally into the private collection of the engineer and horological collector Courtenay Adrian Ilbert, whose estate reached the British Museum in 1958.

The one genuine gap in that chain is exactly where an object usually loses its history: the interval between being declared beyond economical repair by the Navy's own watchmaker in 1886 and turning up, at an unrecorded date, in Ilbert's collection.

What it is

A one-day marine chronometer, 17.6 by 20.8 by 20.8 centimetres overall, in a rectangular mahogany box lined with green baize and fitted with brass carrying handles. The hinged lid has a glazed viewing aperture behind a sliding mahogany cover and a key-operated lock, with an ivory disc set into the front of the lid. The silvered brass dial carries Roman numerals I to XII around a minutes circle, a subsidiary seconds dial above VI, and plain blued-steel hands.

The movement sits in a lacquered brass bowl on gimbals, so the instrument stays level regardless of the ship's motion. It is spring-driven with a fusee and Harrison's maintaining power, jewelled bearings for the train wheels in screwed settings, and Earnshaw's own spring detent escapement with his split-bimetallic temperature-compensation balance and a free-sprung helical blued-steel balance spring. Both the movement and the dial are signed and numbered 509, the serial number by which the Royal Observatory, Greenwich tracked this specific instrument through every one of its subsequent postings.

Why this design mattered

A ship at sea can fix latitude from the sun or stars with comparative ease. Longitude requires knowing the time at a fixed reference meridian at the same moment as local noon, which in turn requires a clock accurate enough, and stable enough through heat, cold, damp and the motion of a hull, to keep that reference time across months at sea. John Harrison had proved in the eighteenth century that such a clock could be built at all. The problem after Harrison was that his solution was a bespoke masterpiece, not a manufacturable product.

Thomas Earnshaw's contribution was to make the same order of accuracy reproducible and affordable. His spring detent escapement and his own bimetallic compensation balance became, in the judgement of specialists in the instrument's own literature, essentially the standard design used in marine chronometers from his lifetime onward. That is the difference between an invention and an industry: Earnshaw's chronometers were, for the first time, cheap enough that a ship could carry several of them, which is precisely what let a captain like FitzRoy treat a set of chronometers as a cross-checking instrument rather than a single fragile authority.

Earnshaw against Arnold

Earnshaw did not work in isolation, and the record of how he was rewarded shows a real professional rivalry rather than a clean succession. He submitted chronometers to the Board of Longitude from 1791, competed again in 1793 with the support of the Astronomer Royal, Nevil Maskelyne, and in 1805 was awarded a share of the Board's prize money for improvements to timekeepers, alongside a matching award to John Arnold, the other leading maker of the period. Earnshaw considered the division unjust and said so in print, publishing a pamphlet in 1808 titled Longitude: an Appeal to the Public, stating T. E.'s Claim to the Original Invention of the Improvements in his Timekeepers, arguing that Arnold's innovations had followed rather than preceded his own.

The dispute was never definitively settled in either man's favour by the Board, and it is not settled by this record either; both makers' solutions to the spring detent escapement and temperature compensation are documented, and the priority argument between them belongs to specialist horological scholarship rather than to a single museum label. What is not in dispute is the outcome: Earnshaw's specific mechanical solutions, not a composite of the two men's work, are the ones that became standard.

Early Royal Navy service

This instrument's documented history begins with its return to Greenwich for servicing on 6 February 1826. Two general items appear in the secondary literature about Earnshaw chronometers of this period without being tied to serial 509 specifically: William Edward Parry's published account of his 1818-19 search for the Northwest Passage near Baffin Bay mentions an Earnshaw chronometer being used to check a longitude reading, without naming the ship it was aboard, and the Royal Observatory's system of issuing chronometers to Royal Navy ships is recorded as having made its first formal issue in 1823, to HMS Larne. Neither claim, in the sources consulted, is stated to be about this particular serial number, and this record treats them as background on Earnshaw chronometers of the period in general rather than as episodes in this object's own service history.

What is tied to serial 509 specifically, through the Greenwich ledger, begins in 1828: issued to Captain J. Bolder in command of HMS Hecla, a ship that had already carried two of Parry's own Arctic expeditions, then in survey service off the West African coast. The chronometer returned from Hecla in mid-1830, was serviced by the London watchmaker Robert Molyneux, and by March 1831 had been delivered to Devonport for its rate to be checked ahead of its next posting.

Twenty-two chronometers for one voyage

Standard Royal Navy practice through the 1820s and 1830s issued one chronometer per ship, rising to three only if the captain personally owned an instrument, on the reasoning that two chronometers were barely better than one: with only two, a discrepancy between them tells you that one is wrong but not which one. Three is the minimum number that lets a faulty instrument be identified by majority disagreement.

HMS Beagle's second voyage, under Captain Robert FitzRoy, departed in 1831 for a survey that would eventually take five years and circumnavigate the globe. FitzRoy carried twenty-two chronometers, a number so far outside normal practice that it required active justification, and one that historians of the voyage treat as a deliberate methodological choice rather than incidental equipment. The ship needed a running, mutually cross-checked determination of longitude that could survive years at sea, chronometers being carried ashore and up rivers in small boats to fix the coordinates of reference points as instructed by the Admiralty, with no guarantee that any single instrument would keep working, or survive the journey, at all. Six of the twenty-two were FitzRoy's own property; the rest were Admiralty-issued or on loan from their makers. Serial 509 was issued to the Beagle's book on 6 December 1831, and returned to Greenwich after the voyage on 7 November 1836.

The chronometers were kept in a dedicated cabin near the captain's own quarters, packed in sawdust for protection, with access restricted to the crew who needed to take readings or maintain the instruments, a level of physical security that itself indicates how seriously the exercise was treated. Their upkeep was the job of George James Stebbing, an instrument maker whose salary the Admiralty refused to fund; FitzRoy paid him personally, as he paid for a share of the ship's other equipment out of his own pocket. The Admiralty did concede that Stebbing could be fed from ship's rations, a concession it did not extend to the ship's naturalist: Charles Darwin paid 500 pounds for his own keep aboard the Beagle.

FitzRoy's own check on the result

FitzRoy had a way to test the whole exercise against itself. Local noon, measured by his chronometers, should shift by exactly twenty-four hours over a complete circumnavigation as the ship worked its way west around the globe. When Beagle returned to her home port after five years and tens of thousands of miles, FitzRoy's chronometers closed that circuit 33 seconds long, an error equivalent to about 8.25 nautical miles. Given the length and difficulty of the voyage, later writers have judged this an impressive result; FitzRoy himself, by his own published account, considered the discrepancy larger than he could explain. The voyage went on to establish, for the first time, a linked chain of reference points of known longitude circling the globe, a chain that subsequent expeditions could use to calibrate their own instruments rather than starting from nothing.

After the Beagle

Serial 509's working life continued for another three decades after it left FitzRoy's ship. Cleaned by Arnold and Dent, it was transported by HMS Lightning to Devonport or Portsmouth for reissue, went to HMS Formidable from January 1842 to November 1845, and then cycled through repeated servicing by Charles Frodsham and local issue from Portsmouth, being carried back to Greenwich for cleaning aboard HMS Rattlesnake in 1850 and HMS Odin in 1854. Its last ship posting was to HMS Pembroke, from April 1858 to February 1867.

From the Navy it passed to scientific shore duty: issued in November 1867 to the Royal Society's Meteorological Committee for observations at Falmouth Observatory, where it worked for nearly twenty years. It was withdrawn in November 1886, sent for assessment to the watchmaker J. Poole, and in December 1886 declared beyond economic repair. In July 1888 the Royal Observatory gave it to E. Dent and Co. in part exchange for a different chronometer. That is the last point at which the sources consulted for this record can place the instrument with a date attached.

The gap, and how it closed

Between 1888 and Courtenay Adrian Ilbert's death in 1956, nothing in the sources fetched for this record documents where this chronometer was or who owned it. Ilbert was a civil engineer who assembled what is described, without a stated rival candidate, as the most extensive private collection of watches, marine chronometers, clocks and sundials of its time, kept at his home in Chelsea. That serial 509 was part of it by 1956 is documented; how and when it got there is not, and this record states that plainly rather than inventing a dealer or a date to fill the space.

Ilbert's death triggered a scheduled public sale of the collection. That sale never happened. A private donation of funds, made by the businessman Gilbert Edgar, allowed the British Museum to purchase the entire Ilbert Collection outright in 1958 rather than see it broken up lot by lot at auction. The chronometer was accessioned as 1958,1006.1957, carrying forward its Ilbert Collection reference CAI.1957, L210, and has been in the museum's care since. It has been shown internationally, including venues in Shanghai, Beijing, Valenciennes and Hong Kong between 2017 and 2019, and remains on permanent display in Room 39, the museum's clocks and watches gallery.

Twenty-two chronometers, and how few are accounted for

What happened to the rest of FitzRoy's twenty-two is, by design, not this record's central subject, but it is the context that makes serial 509's own complete service record worth noticing. Most Admiralty-issued instruments of this kind were cycled indefinitely through Navy and scientific service, sold as surplus, or simply lost to time once no longer of practical use; FitzRoy's own six privately owned chronometers had a different and more exposed history, since they were his personal property rather than government stores.

Published sources do not agree on how many of the twenty-two survive today, and the number reported has grown as research and the market have turned up more of them rather than settled on a fixed figure. This is treated in this record's open questions rather than resolved by choosing whichever number is most convenient.