How Time Zones Were Born: Railways, Telegraphs and the Day of Two Noons

Two minute hands on one clock#

Look up at the clock on Bristol's old Corn Exchange and you will see two minute hands. One shows London time, the hour that the new railways ran by. The other shows Bristol time, set by the Sun, which ran about ten minutes behind. According to Bristol City Council, the second hand was added later so that people could read both.

The gap has a simple cause, the arithmetic behind the The Longitude Problem: How a Clock Taught Ships Where They Were. The Earth turns through 360 degrees in 24 hours, so the Sun appears to travel one degree every four minutes. One source puts Bristol about 2.6 degrees west of Greenwich, so the Sun reaches its highest point there roughly ten minutes after it does in London. Accounts quote the gap as ten or eleven minutes.

That oddity on a wall is this whole article in miniature. A town's own time worked well until a service arrived that ran to a timetable across many towns at once. Then someone had to decide whose minute counted. This is part four of our history of time. In part three, the trouble was keeping a home clock accurate at sea. Here the trouble comes ashore, and the cure is an agreement rather than a better clock.

When every town kept its own noon#

Before the 1840s Britain had no standard time. Each place set its clocks by the Sun, and Royal Museums Greenwich notes that there was no standard time in Britain before the railways and that local sun time was kept. With about 7.5 degrees of longitude between Lowestoft in the east and Land's End in the west, the gap across the country came to roughly half an hour.

Two terms help. A meridian is a line of longitude running from pole to pole, and every place on it has noon at the same moment. Mean time is clock time: days of equal length, each equal to the average length of a solar day. Greenwich Mean Time is that average measured at the Royal Observatory in Greenwich, and a town's local mean time is the same idea measured on its own meridian.

The clocks in the first parts of this series, from sundials and water clocks to the pendulum clock, were all set to the time of the place where they stood. A railway timetable is a different kind of promise: a train leaves at a given minute, and that minute has to mean the same thing at every station on the line.

Railway time: the Great Western goes first#

In November 1840 the Great Western Railway, Isambard Kingdom Brunel's line from London toward the south west of England, told all its stations to use London time, according to Network Rail. An 1841 timetable, which Network Rail quotes from the Science Museum, shows what that meant on the ground. London time ran about 4 minutes ahead of Reading time and 5½ minutes ahead of Steventon time, with bigger gaps further west. Sources disagree on the year: Network Rail and other accounts say 1840, while several Science Museum Group records say 1841.

Other lines followed. In 1847 the Railway Clearing House recommended that every company adopt railway time, meaning Greenwich time, as soon as the Post Office allowed it, according to the Science Museum Group. Accounts differ on how fast the companies obeyed; one has the London and North Western and Caledonian railways changing on December 1, 1847. The IANA time zone database also moves London from its own mean time to Greenwich Mean Time on that day.

Towns did not all fall into line. Exeter held out until November 1852, and Royal Museums Greenwich records concerted resistance in places from local people who refused to adjust public clocks. Bristol's two-handed clock is a surviving compromise. By 1855, an estimate quoted by the Science Museum Group says, 98 percent of public clocks in Great Britain and Ireland were on Greenwich time.

The law lagged. In 1858 the Court of Exchequer ruled in Curtis v March that the time for legal purposes was still local mean time. The Dorchester Civic Society says the judgment held that a railway company could not change a town's time by running a line through it. Parliament settled the matter with the Statutes (Definition of Time) Act, which received royal assent on August 2, 1880. Its preamble admits doubt about whether a time named in a statute or deed meant Greenwich time or local mean time. Section 1 says that, unless stated otherwise, it means Greenwich mean time in Great Britain and Dublin mean time in Ireland. That was 40 years after the Great Western's decision.

Time by wire: telegraphs and observatories#

Railway time needed a way to reach the stations. Greenwich had one tool already. In 1833 a time ball went up on Flamsteed House so that ships could set their clocks before a long voyage, as Royal Museums Greenwich records, but a ball on a roof was no use to a distant station. The Greenwich ball was not the first of its kind either: accounts of the invention credit the naval officer Robert Wauchope and mention an earlier trial ball at Portsmouth, though sources differ on the details.

The answer was the telegraph. In 1852 Charles Shepherd installed a master clock at the Royal Observatory that sent electrical pulses to other clocks, including a public dial at the observatory gate. The museum calls that dial the first clock ever to show Greenwich Mean Time directly to the public, though other accounts say only that it was probably the first. The scheme was proposed by the astronomer George Airy, and railways were a major reason for it. From 1852 to 1893, the museum says, this clock was the heart of Britain's time system, and its time went out by telegraph to London, Edinburgh, Glasgow, Dublin, Belfast and many other cities.

In the United States, observatories found paying customers. Samuel Langley became director of the Allegheny Observatory in Pittsburgh in 1867, fixed the time from star observations and sold it by telegraph. According to the observatory's own building history, its time signals went out over more than 4,700 miles of telegraph line between 1869 and 1883, and the Pennsylvania Railroad was a major customer.

The telegraph solved how to send a time. It could not decide which time to send, and American railroads could no longer avoid that question.

A continent of railroad clocks and four campaigners#

Britain is small enough for one clock to serve every railway. North America stretches across several hours of sun time. Before 1883 each railroad kept its own standard, often the local time at its headquarters or its most important station, and a city served by several lines could have a clock for each. Estimates of how many railroad times were in use in the early 1880s run from about 50 to about 80.

Several people pushed for an answer, and the credit is shared and sometimes disputed.

Charles F. Dowd was a schoolmaster at Temple Grove Seminary in Saratoga Springs, New York. He presented a plan to railway superintendents in 1869 and met little enthusiasm. In 1870 he published a pamphlet proposing four zones, each 15 degrees wide, and later versions (one account dates the revision to 1872) centered the zones on the meridians at 75, 90, 105 and 120 degrees west of Greenwich, the layout eventually adopted. Sources also differ on what he called the zones. The Library of Congress's Today in History entry dates his first proposal to 1863, so sources differ on when it began.

Cleveland Abbe was a meteorologist. A committee on standard time was set up in 1879 at his urging, according to the Library of Congress, because he wanted weather observations to be taken at the same moment.

Sandford Fleming, who lived from 1827 to 1915, was a Scottish-born Canadian railway engineer. On February 8, 1879, he read a paper to the Canadian Institute in Toronto about a prime meridian common to all nations and a worldwide system of time-reckoning. The Virtual Museum of Canada says he recommended 24 global time zones.

William F. Allen edited a railway guide and was secretary of the General Time Convention, the railroads' managers' group. Asked to devise a plan, he drew on the ideas of Dowd, Fleming and Abbe, says the Library of Congress. Accounts, including Smithsonian Magazine, add that he toured the country lobbying local politicians, journalists and civic leaders, and was largely successful.

On October 11, 1883, a Thursday, railroad managers meeting in Chicago approved Allen's plan. It set zones centered on the 75th, 90th, 105th and 120th meridians west of Greenwich, one hour apart. Some accounts count five zones, adding one for the Canadian Maritimes. The switch was set for Sunday, November 18.

Sunday, November 18, 1883: the day of two noons#

On that Sunday, telegraph signals from observatories told the stations when the new noon had arrived. The best known came from the Allegheny Observatory, and accounts describe it as marking noon on the 90th meridian or noon Eastern time. The Library of Congress explains in a research guide that every railroad clock was reset, even those that had already passed noon, which gave the day its nickname.

The "two noons" followed from the four minutes per degree. Where the new standard ran behind the local sun time, clocks were set back and the minutes in between happened twice. New York lies a little under one degree east of the 75th meridian, so its clocks were set back by 3 minutes 58 seconds, and the new noon arrived that much later than the old local noon, according to Untapped New York. The IANA data records the same jump, from 12:03:58 back to 12:00. Chicago lies well east of the 90th meridian, and its clocks went back by about nine and a half minutes. Accounts give 9 minutes 32 or 33 seconds, while the IANA data implies 9 minutes 24 seconds, so the exact figure varies by source.

Not everyone welcomed it. The arrangement was voluntary and had no force of law, and some towns and cities resisted for a time.

• 4 minutesof sun time for each degree of longitude (24 hours divided by 360 degrees)
• 3 min 58 show far New York's clocks were set back on November 18, 1883, per Untapped New York
• 22 to 1the vote for Greenwich in Washington in 1884, with 2 abstentions
• 34 yearsfrom the day of two noons to the US Standard Time Act of 1918

Washington, 1884: a prime meridian but not time zones#

A year later, delegates gathered in Washington at the invitation of President Chester A. Arthur. A BBC-based account counts 41 delegates from 25 countries, though other sources say 26 or 27. Their job was narrower than this article's title suggests: to choose one prime meridian, the line from which longitude is counted, in place of the many then in use.

Why Greenwich? Royal Museums Greenwich says the reason was practical: Greenwich was already the standard for navigation, and by the 1880s the share of the world's ships using charts based on it is put at anything from nearly two thirds to 72 percent. That is the observatory at the centre of the longitude story.

The vote went 22 in favor, 1 against and 2 abstentions. San Domingo, now the Dominican Republic, voted no, and France and Brazil abstained. Most accounts date the vote to October 13, 1884, a Monday, while one gives October 22. The resolution proposed to the governments represented the adoption of the meridian through the centre of the transit instrument at the Observatory of Greenwich as the initial meridian for longitude, according to The Greenwich Meridian site, which treats the resolutions as proposals. Two accounts say France did not adopt Greenwich until 1911.

London still keeps Greenwich Mean Time in winter, and Royal Museums Greenwich says GMT was the international standard of civil time from 1884 until 1972, when UTC replaced it. The difference between the two is the subject of our guide to UTC vs GMT.

Zones become law, and why their borders wiggle#

Countries adopted zones on their own schedules. Japan chose 135 degrees east in an 1886 ordinance that took effect on January 1, 1888, and the IANA data moves Tokyo's clocks back about 19 minutes at that midnight. In Germany, mean solar time at 15 degrees east, one hour ahead of Greenwich, became legal time in 1893, according to the PTB, the national metrology institute. A US Bureau of Standards circular says nations accepted the plan gradually, with Germany, Italy, Denmark and Switzerland adopting central European time in 1893 and 1894.

India's standard time began on January 1, 1906, set from the meridian at 82.5 degrees east, which is exactly 5.5 hours from Greenwich (82.5 divided by 15), though Calcutta and Bombay kept their own times for decades afterwards. Nepal sits at UTC+5:45 today, and our guide, Why India Is UTC+5:30: Half-Hour and 45-Minute Time Zones, explains such offsets.

The American railroad zones had no force of law for 34 years. The Standard Time Act, approved on March 19, 1918, set the zones for the United States and left the boundaries to the Interstate Commerce Commission. The commission was to weigh the convenience of commerce and the existing junction points and division points of carriers, as the text of the law puts it. The same law introduced daylight saving, which Congress repealed in 1919 over President Wilson's veto while keeping the zones.

That wording explains the zigzags on a map of US time zones: the lines run through railroad towns instead of along straight meridians. Today the Department of Transportation, not the railroads, decides where the lines run. Our guide to US time zones has the current map.

Elsewhere the lines follow politics. China runs on one standard time, UTC+8, which Quartz traces to 1949. The exact timing is less settled: local switches are recorded from late 1949 into 1950, and the IANA time zone database long dated the change to 1980. Spain moved to central European time on the night of March 16 to 17, 1940 (the IANA record), though Madrid lies at about 3.7 degrees west of Greenwich. Madrid's mean solar time runs about 75 minutes behind its legal winter time. The popular story blames Franco courting Hitler, but a former observatory astronomer quoted in news reports, including Euronews, says the order aimed to line Spain up with neighbors that had already advanced their clocks, weeks before Germany did. The motive is debated. The offset is not.

1. 1833A time ball goes up at Greenwich so ships can set their clocks.
2. November 1840The Great Western Railway tells its stations to use London time (some records say 1841).
3. 1847The Railway Clearing House recommends Greenwich time for British railways.
4. 1852Shepherd's master clock at Greenwich begins sending time by telegraph.
5. 1858Curtis v March confirms that local mean time is still the legal time.
6. February 8, 1879Sandford Fleming proposes worldwide time-reckoning in Toronto.
7. August 2, 1880Greenwich Mean Time becomes the legal time of Great Britain.
8. November 18, 1883North American railroads switch to standard time: the day of two noons.
9. October 1884Delegates in Washington choose Greenwich as the prime meridian, 22 votes to 1.
10. January 1, 1888Japan's standard time at 135 degrees east takes effect.
11. January 1, 1906India's standard time begins.
12. April 30, 1916Germany puts clocks forward in the first nationwide daylight saving.
13. March 19, 1918The US Standard Time Act makes railroad zones federal law.

Daylight saving: an old joke, a late idea, a wartime law#

Daylight saving is often folded into the story of time zones, but it asks a different question: not which clock to follow, but whether the whole clock should move for part of the year.

The most quoted ancestor is Benjamin Franklin. In the spring of 1784, while serving as the American minister in Paris, he published an anonymous letter in the Journal de Paris. It began as a reply to the paper's praise of a new oil lamp. His narrator claims to have discovered that the Sun rises earlier than Parisians think, and proposes saving candles by waking everyone at dawn, with church bells and cannon fire to rouse the sluggards. It never proposes changing the clocks. Founders Online holds the letter, and the Independence Hall Association's ushistory.org calls the claim that Franklin proposed daylight saving false.

A serious proposal came from New Zealand. George Vernon Hudson, a Wellington entomologist, found that his shift work left him leisure time to collect insects and made him value daylight hours. In October 1895 he presented a paper on seasonal time adjustment to the Wellington Philosophical Society, which by most accounts proposed a two-hour shift. Te Ara says some members ridiculed it, and a one-hour summer time was trialed in New Zealand only in 1927.

In England the builder William Willett published The Waste of Daylight in 1907, urging that clocks advance at the start of spring and return to Greenwich Mean Time in autumn, as National Museums Scotland describes. Accounts of his scheme differ: one describes 80 minutes in 20-minute steps on April Sundays, another 90 minutes. A 1909 bill went to a parliamentary select committee, but nothing came of it. Willett died in 1915, a year before the law passed; one account gives influenza as the cause.

A small Canadian town tried it before any country did. Port Arthur, Ontario, now part of Thunder Bay, moved its clocks an hour in 1908, though timeanddate gives July 1 while local accounts describe a spring switch after council votes.

Germany is usually credited with the first nationwide change. The PTB's table shows clocks going forward on the night of April 30, 1916, and timeanddate calls Germany the first country with nationwide daylight saving. Austria-Hungary adopted summer time in the same period, so the "first" is a matter of days. The reason was the war: saving fuel and electricity. Britain was only weeks behind. The Summer Time Act passed on May 17, 1916, and clocks went forward on Sunday, May 21.

The arguments for daylight saving have shifted from decade to decade: wartime fuel first, then farming, energy and health. That is one reason the rules keep changing, and the European Union shows the deadlock. Its Parliament voted in 2019 to end the twice-yearly change, but the Council of the EU says on its own page that it has not yet agreed its position, so the clocks still change. How Daylight Saving Time Works, Why It Exists and Who Still Uses It covers the mechanics and the evidence, and When Do the Clocks Change? Daylight Saving Dates for 2026 and 2027 has the dates and the state of the EU proposal.

By 1918 the main pieces were in place: a prime meridian, zones born on the railways, and laws that made them official. Open the world clock and every row traces back to a decision made somewhere in this story, by a railway board, a ministry or a war. Time zones are agreements, which is why their edges follow rails and politics more than meridians, and why they can be changed.

Clocks were still tied to the turning of the Earth, though. The final part of the series, atomic clocks, leap seconds and the network that keeps your phone on time, picks up from here.

Frequently asked questions

Sources and further reading

1. St Nicholas Markets: history (Corn Exchange) — Bristol City Council
2. 180 years of railway time — Network Rail
3. 'Railway Time' clock — Science Museum Group
4. Shepherd Gate Clock — Royal Museums Greenwich
5. What is the Prime Meridian, and why is it in Greenwich? — Royal Museums Greenwich
6. History of legal time in Britain — polyomino.org.uk
7. When the Standardization of Time Arrived in America — Smithsonian Magazine
8. Whose Time is it Anyway? A Brief History of Standardized Time Zones in the United States — Library of Congress
9. Remembering Sandford Fleming and Standard Time — Toronto Public Library
10. Aux Auteurs du Journal de Paris, 1784 — Founders Online, U.S. National Archives
11. Hudson, George Vernon — Te Ara, Dictionary of New Zealand Biography
12. DST and midsummer DST in Germany until 1979 — Physikalisch-Technische Bundesanstalt
13. Seasonal clock changes in the EU — Council of the European Union