American Scientist

In the early 1700s, European monarchies aspired to power by building world-spanning networks of colonies and commercial ventures. As a result, the merchant fleets and navies that connected and protected these assets were critically important. Eighteenth-century sailors led dangerous lives, not least because they seldom knew their exact location on the open ocean. Although navigators readily determined latitude, or north-south position, by estimating the height of certain stars at their zenith, they could not determine longitude (see Marginalia, September–October 2002). This failure caused shipwrecks that killed thousands of mariners and lost cargoes worth fortunes. Several countries offered immense financial rewards for a solution to the problem; Britain promised £20,000 (several million dollars in today's currency) for a way to establish longitude to within half a degree (30 nautical miles at the equator) after a journey from England to the West Indies. To judge proposed solutions, the crown established a Board of Longitude, made up of the Astronomer Royal, various admirals and mathematics professors, the Speaker of the House of Commons and 10 members of Parliament.

In effect, determining longitude depended on knowing the difference between local time and the time in Greenwich, site of the Royal Observatory. In principle, if a ship had a clock keeping Greenwich time, the navigator could measure the angle of the Sun to note local noon and compare it to the clock. If the clock read 2 p.m., his longitude was two hours, or 30 degrees, west of Greenwich. The problem lay in finding a clock reliable enough to keep time during the long voyages of that era. The best pendulum clocks of the day were accurate enough, but were useless on a heaving ship at sea. Alternately, a less reliable clock might be used if some means could be devised to correct it frequently. In practice this meant an astronomical method, the best of which became known as the method of lunar distances, in reference to the fact that the Moon's orbit causes it to continually change position in the sky. For example, a new moon, which appears close to the Sun, will have moved 180 degrees by the time it becomes a full moon two weeks later. The idea was for astronomers to provide tables of this angle between Moon and Sun (or Moon and selected stars in the night sky) as a function of Greenwich time. A measurement of this angle every few days would provide a correction to the mechanical clock. This scheme had two drawbacks: The first was that, at least initially, astronomers could not accurately predict the Moon's motion; the second was that the mathematical calculations required of the mariner were very complex—they took hours, and errors were common.

The chronometer and the method of lunar distances each had their proponents, and the solution of the longitude problem came down to a bitter battle between two Englishmen: John Harrison, a self-educated machinist who set out to make an accurate clock, and Nevil Maskelyne, astronomer and scion of the Church of England, who brought the method of lunar distances to fruition.

John Harrison was born in Yorkshire and followed his father's trade as a carpenter. This led to an interest in wooden (pendulum) clocks, which he became adept at making. His isolation from other clockmakers during this formative time was fortuitous—when most clockmakers learned that the prize required their clocks to maintain a rate constant to within three seconds per day for six weeks at sea, they regarded the task as impossible. Harrison saw it as a challenge, and developed the technical advances that justified his confidence, including the so-called gridiron pendulum (which used metals with different expansion coefficients to overcome temperature sensitivity), a nearly frictionless gear system and a ratcheted spring that kept the clock running while being rewound.

By 1728 the longitude prize had been open for 14 years without a serious contender. Young Harrison journeyed to London to learn about the award and met Edmond Halley, the Astronomer Royal. Halley treated him cordially and suggested that Harrison consult George Graham, one of London's leading clockmakers. Harrison and Graham took an immediate liking to each other—they talked for 10 hours that day, leaving Graham so impressed that he offered to lend Harrison money (without security or interest) to develop the young inventor's ideas.

Seven years passed before Harrison offered a solution to the problem: the clock later known as H1. It was cumbersome and heavy, with jutting arms and counterweights, but a small committee of the Royal Society, including Halley and Graham, declared it a masterpiece of ingenuity and urged submission to the Board of Longitude. This body, ever skeptical after years of spurious claims, proposed a short sea trial before attempting the expensive, official crossing of the Atlantic. So Harrison and H1 embarked on a run to Lisbon and back. The numerical result of the clock's performance has not survived, but the captain was very impressed. The ship's navigator admitted that on the return voyage, just before the English coast was sighted, his own calculations put them more than 90 miles offshore, while Harrison maintained that no, they were just about there.

Although the voyage didn't immediately lead to a trans-Atlantic test, it was productive for Harrison, who identified several potential improvements to H1. With these in mind, he withdrew his initial submission and set about building a second clock (H2). The Board of Longitude, impressed by H1, awarded Harrison £500 toward development expenses. He completed H2 three years later and spent two more years testing it on land. By 1741, testing was complete, but England was at war with France and her allies. No one was prepared to risk a sea trial of H2 and possible capture by the French. As it turned out, H2 was never tested at sea.

While waiting for the war to end, Harrison built a third clock, with the Board of Longitude contributing another £500 to meet his costs. But this time, Harrison's expertise failed him. H3 never worked to Harrison's satisfaction, and he abandoned it after five years. Taking a different tack, he proposed in 1746 to build two more clocks. Of these, H4 would be his masterpiece. It took another 13 years to complete, but was quite different from its predecessors. Instead of being all angles and arms, it appeared as a beautifully encased pocket watch, albeit five inches in diameter. After two years of testing the Board accepted it for the full trans-Atlantic test.

Harrison, now 68, left H4 in the care of his son William for the long voyage in 1761. (Just as well, perhaps: Early in the voyage all hands were horrified when the captain announced that all the beer on board—more than a thousand gallons—had spoiled, and the crew would be reduced to drinking water!) The trial ended upon reaching Jamaica, where, amazingly, H4 reported the longitude with less than two miles of error. Back in England, the overjoyed Harrisons awaited their award of £20,000. Then everything started to go wrong. The Board of Longitude refused to believe the test results were not just a stroke of luck and demanded a second trial. The Harrisons protested vehemently. They had met the terms of the prize and now they wanted it. Uproar ensued, and to quell it, Parliament offered William Harrison £5,000 for results so far, in return for full disclosure of the construction details of H4. William refused this partial award and decided to attempt a second trial in 1764, this time to Barbados.

The Reverend Nevil Maskelyne now enters the story. He had no particular influence prior to this stage of Harrison's odyssey, but the Board of Longitude made a decision that quickly embroiled him in the controversy. Although Maskelyne was not then a member of the Board of Longitude, he was a Fellow of Trinity College, Cambridge, and a Fellow of the Royal Society. For some years he had maintained a strong interest in, and advocacy of, the method of lunar distances for determining longitude. Furthermore, as Tobias Mayer in Germany had recently resolved the problem of predicting the motion of the Moon, the method was ripe for testing. Thus, when the Board of Longitude asked Maskelyne to join the voyage to Barbados, primarily to establish the longitude of the capital, Bridgetown, by observation of Jupiter's satellites, he was also to test the method of lunar distances and its accuracy compared to Harrison's H4 clock.

The results were announced at a memorable meeting of the Board of Longitude in early 1765. H4 had done it again, producing the Bridgetown longitude with less than ten miles of error after a journey of more than 5,000 miles. The rival method of lunar distances fared slightly worse, yielding the result to "better than 30 miles." By way of explanation, four of the ship's officers at the meeting stated that their calculations were a product of Maskelyne's instructions, and, by implication, subject to their own inexperience. In any case, since the lunar distance method depended on tables that only Maskelyne could calculate, the method was not yet ready to claim the prize.

The Board of Longitude accepted the result of the Barbados trial of H4, but they remained unconvinced that the instrument was not just a fluke—a one-off that might never be replicated. Parliament passed a new Act that yielded £10,000 to Harrison but withheld the remaining half of the prize until he met a series of conditions. According to Parliament's terms, Harrison had to reveal in writing exactly how the watch had been made, including full drawings of each part and explanation of every detail to a select team of watchmakers. Then he needed to wait for these persons to manufacture similarly accurate watches themselves. Moreover, Harrison was compelled to relinquish H4 to the Astronomer Royal for long-term checking of its accuracy.

Here was the rub: Two Astronomers Royal had died in rapid succession, and no sooner had Maskelyne returned from Barbados than the King appointed him to the position; the appointment automatically put him on the Board of Longitude. As a result, Harrison's nemesis not only became an influential voice on the Board of Longitude but also took charge of checking his best clock. Not surprisingly, the resulting report on H4 was entirely negative. It seems Maskelyne prevaricated over the meaning of "accuracy" to condemn Harrison's creation. In this era, all clocks gained or lost time at some rate, but so long as that rate was constant and known, one could derive an accurate time. Maskelyne refused to allow these corrections.

While waiting for the copies of H4 to be completed, Harrison spent three precious years building H5—now approaching 80, he was becoming increasingly desperate. As the Board of Longitude debated sending his watch to the arctic and elsewhere for lengthy tests, Harrison appealed to King George III for help. The King had H5 installed at his own personal observatory at Kew, and he himself supervised its daily winding and checking. Under his care, H5 performed so admirably that "Farmer George" was outraged by what had gone before. "By God, Harrison," he roared, "I'll see you righted!" and threatened to appear in person before Parliament (under a lesser title, of course).

So the affair finally wound down. The Board of Longitude, under the continuing influence of Nevil Maskelyne, still managed to find reasons to withhold some of Harrison's prize money. Nevertheless, including interim advances, he received a total of £23,065 from the Board over four decades. A copy of H4 made by another watchmaker accompanied Captain James Cook on his second voyage to the Pacific, where it was hailed as having "an amazing degree of accuracy," although Cook also used the lunar distances method with good results. In a triumph of his own, Maskelyne had initiated the annual publication of The Nautical Almanac, which included the necessary tables for applying that method. Cook referred to it as "our faithful guide through all vicissitudes of climates." And since sufficiently accurate clocks (later called marine chronometers) remained very expensive for a long while, the method of lunar distances remained in use for more than a hundred years.

Harrison died in 1776, unreconciled with Maskelyne to the end. One cannot blame him, but because it turned out that posterity owed much to both of them, it would have been pleasing had their relationship ended otherwise.

© J. Donald Fernie