applets.

Longitude

← All apps

About Longitude

At sea in the 18th century a navigator could find his latitude from the height of the sun at noon, but not his longitude: for that he needed to know the time at Greenwich at the same moment as his own, and no clock kept time at sea. This page is his chart table, in three parts.

NOON SIGHT: a ship somewhere at sea. START WATCH runs the chronometer from 20 minutes before local noon, with the sun already brought down to the horizon. You keep it there: as the sun climbs its image lifts clear of the horizon, and you turn the drum up (drag it, scroll over the sextant, the +1′ −1′ buttons, or the ↑ ↓ keys, Shift for tenths) until its lower edge just touches again: a bigger reading brings the sun's image down. When turning it no longer brings the sun down and it starts to dip into the sea, press MARK NOON: the sight is what your sextant reads and the time is when you pressed, so your own errors carry into the latitude and longitude. The graph shows your readings while you watch and the sun's true height afterwards. HOLD IT FOR ME lets the page keep the sun on the horizon instead, so you only call noon. The log works the sight: the sextant reading, corrected for the height of eye (dip), refraction, the sun's half-width (semi-diameter) and parallax, gives the true altitude; 90° minus that is the zenith distance, and with the sun's declination it gives the latitude. The chronometer's Greenwich time at that moment, put into apparent time with the equation of time, is how far Greenwich is ahead of or behind the ship's noon: every 4 minutes is one degree of longitude. EQUAL ALTITUDES times noon better: mark the sun at some height in the morning, which clamps the sextant, then mark it again when it comes back down to touch the horizon after noon; noon is halfway between. NEW SHIP puts the ship somewhere else on another day.

THE CLOCK DRIFTS: a chronometer gains or loses a few seconds a day (its rate), and its rate wanders a little from day to day. If the rate was found before sailing (by comparing it for some days against an observatory's time) the navigator allows for it, and only the wander is left; if not, the whole rate piles up. Every second of time is 15″ of longitude, a quarter of a mile at the equator. The chart shows the error growing against the prize bands of the 1714 Act.

RECKONING: a passage from Madeira to Barbados, plotted by dead reckoning (the course steered and the distance by the log) with errors the navigator can't see: a current, a log that reads short or long, steering off the course. Choose how he checks his position each noon: DR ONLY, NOON LAT (latitude only, so he "runs down the latitude": gets on to Barbados's latitude well to the east, then sails due west along it), LAT + CLOCK (noon latitude and a chronometer longitude) or LAT + LUNARS (noon latitude and a longitude by lunar distance when the moon can be used). SAIL plays the passage; the red dotted line is where the ship really went, which no one aboard could see.

THE STORY

In 1714 Parliament passed the Longitude Act, offering £10,000 for a method that found longitude to within one degree of a great circle (60 geographical miles) on a voyage from Great Britain to the West Indies, £15,000 within two-thirds of that, and £20,000 within half a degree. Commissioners, later called the Board of Longitude, judged the claims; the Board lasted until 1828.

John Harrison (1693 to 1776), a carpenter and clockmaker who grew up in Lincolnshire, set out to do it with a clock. His first sea clock, H1, was finished in 1735 and tried on a voyage to Lisbon and back in 1736. H2 followed, then H3, on which he worked for about nineteen years and which brought in the bimetallic strip and the caged roller bearing. His answer turned out to be H4, finished in 1759: not a clock but a large watch about 13 cm across, with a fast-beating balance. On its trial his son William took it to Jamaica aboard HMS Deptford, sailing in November 1761 and arriving in January 1762, and after allowing for its rate it was found to be only about 5 seconds out, a little over a minute of longitude. The Board was not satisfied: among its objections, the rate had not been stated before the voyage, and the longitude of Jamaica itself was not well known. A second trial, to Barbados in 1764, also came well within the half degree. In 1765 the Board paid Harrison £10,000 once he had shown how H4 was made, and in 1773 Parliament awarded him a further £8,750. The full prize was never paid to anyone.

Larcum Kendall made a copy of H4, now called K1, finished in 1769. James Cook took it on his second voyage (1772 to 1775), and came to rely on it, praising it in his journal. Watchmakers such as John Arnold and Thomas Earnshaw then made simpler, cheaper chronometers, and by the early 19th century they were becoming common at sea.

The other answer was the moon. It moves against the stars by about its own width, half a degree, every hour, so its angle from the sun or a bright star tells the time at Greenwich, if someone has worked out those angles in advance. Tobias Mayer's tables of the moon made this accurate enough; Nevil Maskelyne, Astronomer Royal from 1765, published the first Nautical Almanac, for the year 1767, giving the "lunar distances" for every three hours of Greenwich time. A lunar needed a careful sight and a long sum, and every minute of arc wrong in the distance is about two minutes of Greenwich time, half a degree of longitude; but it needed no costly clock, and navigators used lunars to check their chronometers well into the 19th century.

HOW IT WORKS

The sun is placed by the NOAA solar calculator's method (after Jean Meeus): its declination, the equation of time (apparent minus mean time: about −14 minutes in mid-February, +16 in early November), its distance and so its semi-diameter. The noon sum is done in tenths of a minute so it adds up exactly as written. Dip is the Nautical Almanac's 1.76′ × √(height of eye in metres), refraction is Bennett's formula (the almanac's table), parallax 0.15′ × cos(altitude). Latitude = declination ± zenith distance: add when they have the same name, subtract when contrary. Longitude: the ship's apparent time at noon is 12h; Greenwich apparent time is the chronometer's mean time plus the equation of time; the difference times 15° an hour is the longitude, west when Greenwich is later ("longitude west, Greenwich time best").

The passage is worked hour by hour: the navigator steers the rhumb line (a constant course, a straight line on a Mercator chart) from where he believes he is, and the ship really goes on that heading plus any steering error, at the speed through the water, plus the current. Noon sights put his reckoning back to the truth plus a small random error: one standard deviation 1′ for latitude, 1.5′ of longitude for the chronometer (plus its rate times the days at sea), 15′ for a lunar. About one noon in seven is overcast, with no sights. Lunars can't be had within about three days of new moon. When the reckoning says she has arrived the ship holds her course; if nothing is in sight a day later, the passage ends there.

WHAT IT LEAVES OUT

The ship is stopped while you take the sight, and the sea is calm; the sight itself is perfect. The equal-altitudes noon is the plain halfway time, without the small correction for the sun's declination changing between the two sights (a few seconds of time). In practice a chronometer longitude was better found from a morning or afternoon "time sight", when the sun's height changes fast, with the noon latitude carried to it; timing noon itself is the weak part, as the page shows. The page reads the sextant to a tenth of a minute, as a modern micrometer sextant does, and draws its drum beside the sextant (one turn is one degree, so you can see it turn as the sun climbs and stand still at the top); the drum came in the 20th century, and a vernier of the 1770s read to about half a minute. The faint lines on the sextant are the light: the sun's ray comes down to the index mirror on the arm, across to the horizon mirror and into the telescope beside the horizon. A mirror turned by an angle turns its reflection by twice that, so the arm swings half the sun's height and the arc is marked double. Refraction is the standard value, with no allowance for temperature and pressure. The sun's place comes from modern formulae, not a 1772 almanac. The chronometer's wander is random day to day with no allowance for temperature, which was a real cause of changing rates. On the passage the speed is a steady 5 knots, the current is the same everywhere, the Earth is a sphere and the island is raised at 20 miles. The coasts are drawn roughly from a few points: this is not a chart for navigation.

Built from scratch for this site.

FURTHER WATCHING AND READING

Longitude west, Greenwich time best
Speed