Frank Watson Dyson: The Astronomer Royal Who Tested Einstein and Gave Britain the Pips

Sir Frank Watson Dyson (8 January 1868 – 25 May 1939) was an English astronomer who served as the ninth Astronomer Royal, heading the Royal Observatory at Greenwich from 1910 to 1933. Two achievements keep his name alive outside specialist circles. The first is the solar eclipse of 29 May 1919. Dyson saw, years in advance, that this eclipse offered an exceptional chance to test Albert Einstein's new theory of gravity, and he organised the British expeditions that made the measurement. When the results were announced in London on 6 November 1919, Einstein became world famous almost overnight. The second is the six short tones of the Greenwich Time Signal, known in Britain as "the pips". These were first broadcast by the BBC on 5 February 1924, using a clock at Greenwich, with Dyson in the studio as a guest.

Those two episodes make up only a small part of a long and busy career. Dyson was a mathematician by training who became one of the leading practical astronomers of his generation. He worked on star catalogues and the motions of stars, measured the spectrum of the Sun's outer atmosphere during eclipses, ran the national time service, modernised Greenwich's equipment, and led the main international astronomy organisation in the years after the First World War. This article covers his life, explains his main scientific work in plain terms, and looks at how his reputation has fared, including the later arguments over whether the 1919 eclipse results were handled fairly.

Quick facts

Early life: a minister's son with a gift for mathematics

Frank Watson Dyson was born on 8 January 1868 in Measham, Leicestershire, the son of the Reverend Watson Dyson, a Baptist minister, and Frances Dodwell. He was the oldest of seven children. Like many ministers' families, the Dysons moved with the father's postings. When Frank was an infant they lived at Wirksworth in Derbyshire, and they later moved to Yorkshire.

The family was not well off. The astronomy historian Joseph S. Tenn, writing a profile of Dyson for the series on Bruce Medal winners, pointed out that all seven Bruce medallists born in England during Queen Victoria's reign started life without money and rose through scholarships. Dyson was one of them. At thirteen he came first in a national mathematics competition, and the scholarship he won took him to a well-equipped and well-staffed school. He attended Heath Grammar School in Halifax and then Bradford Grammar School, the second on a scholarship.

Further scholarships took him to Trinity College, Cambridge, where he studied mathematics and astronomy. One of his daughters later recalled that he "loved mathematics for its own sake and was always happy puzzling over its problems."

What "Second Wrangler" means

In 1889 Dyson was placed Second Wrangler. To a modern reader the phrase needs explaining. At Cambridge in the nineteenth century, the students who earned first-class honours in the Mathematical Tripos (the final examinations in mathematics) were called "wranglers", and they were ranked in strict order. The Senior Wrangler came first, the Second Wrangler second, and so on. The examination was famously hard and the rankings were reported in the national press. Being Second Wrangler marked Dyson as one of the strongest young mathematicians in the country that year. Many second wranglers went on to distinguished careers, and the ranking is still used to group people in biographical reference works.

After graduating, Dyson stayed at Cambridge. He was elected a Fellow of Trinity in 1891 and held an Isaac Newton studentship for research in astronomy. Tenn notes that he won some renown for a piece of mathematical physics: calculating the gravitational potential of an "anchor ring", the doughnut-shaped solid that mathematicians call a torus. This is the sort of problem a gifted applied mathematician of the day would cut his teeth on. It has nothing to do with telescopes, and when Dyson moved into practical astronomy he admitted he knew little about instruments or observing.

Greenwich, the first time: Chief Assistant (1894–1905)

In 1894 Dyson was engaged to be married and needed a salaried job. He gladly accepted an offer to become Chief Assistant at the Royal Observatory, Greenwich, working under the eighth Astronomer Royal, William Christie. The same year he joined the Royal Astronomical Society and the British Astronomical Association, and married Caroline Bisset Best, the daughter of Palemon Best.

The couple settled near the Observatory. From 1894 to 1906 they lived at 6 Vanbrugh Hill in Blackheath, a house now marked with a blue plaque. Tenn describes a ten-room house on a hill near the Observatory and its park, which the couple "immediately began to fill with children." Dyson captained the Observatory's hockey team, and he and his wife regularly met friends to read Shakespeare aloud. He also began the habit of public service in astronomy that marked the rest of his life. He attended almost every monthly meeting of the Royal Astronomical Society and sat on its Council for thirty-seven years. He wrote popular articles and gave public lectures.

The Astrographic Catalogue

Dyson's first big job at Greenwich was the Observatory's share of the Carte du Ciel ("Map of the Sky"), an international project launched in Paris in 1887. The aim was to photograph the whole sky and produce two products: a detailed photographic chart and a catalogue of precise star positions, the Astrographic Catalogue. The sky was divided into zones, and the directors of eighteen observatories each agreed to take responsibility for one.

The work was enormous. Tenn describes long exposures of forty minutes for the chart, intended to record stars down to the 14th magnitude, and shorter exposures of six minutes, three minutes and twenty seconds for the catalogue plates. Every star image on every glass plate had to be measured, and the measurements then had to be "reduced". Reduction means converting the raw positions measured on a plate into true positions on the sky, removing effects such as the plate's tilt and scale, flaws in the telescope's optics, the bending of light by the atmosphere, and the Earth's own motions.

Dyson improved Greenwich's methods for reducing photographic measurements and organised new determinations of the reference stars against which all other positions were measured. The Greenwich section of the Astrographic Catalogue was published in 1905, and Tenn notes that Greenwich was the first observatory to finish its share. One of Dyson's early papers, "On the Determination of Positions of Stars for the Astrographic Catalogue at the Royal Observatory", appeared in the Monthly Notices of the Royal Astronomical Society in the 1890s.

He also helped with a famous piece of measurement during the 1900–1901 close approach of the asteroid Eros. Eros comes unusually near the Earth at times, which made it a useful target for measuring distances in the solar system. Dyson organised the Greenwich observations and their reduction, and the work set new standards of accuracy.

Stars that move: Groombridge, Thackeray and proper motions

Stars are not truly fixed. Each one moves through space, and from Earth this shows up as a very slow drift across the sky called "proper motion", measured in seconds of arc per year. (A second of arc is 1/3,600 of a degree, roughly the width of a small coin seen from several kilometres away.) To measure such tiny drifts, astronomers compare a star's position at two dates as far apart as possible.

Dyson found a clever way to get a long baseline. Between 1806 and 1816 the English amateur astronomer Stephen Groombridge had measured, by eye, the positions of thousands of stars near the north celestial pole. Dyson re-reduced Groombridge's old observations with modern methods, greatly improving their precision. Then, with his colleague W. G. Thackeray, he re-observed the same 4,239 stars. Comparing the two sets of positions across an interval of about eighty years gave proper motions for all of them.

The results were useful in several ways. They gave improved values for the rate of the Earth's precession, the slow wobble of the Earth's axis that gradually shifts the sky's coordinate grid. They also gave a better value for the direction in which the solar system moves relative to the nearby stars. Dyson and Thackeray found that the stars' motions were related to their brightness and to their distance from the plane of the Milky Way.

Most important for the science of the time, the fastest-moving stars appeared to move in two great "streams". The Dutch astronomer Jacobus Kapteyn had announced this "two-stream" pattern a few years earlier, and Dyson's data confirmed it and extended it to fainter stars. Kapteyn, Karl Schwarzschild and Arthur Eddington were then trying to build models of how stars move within what we now know as our Galaxy, and Dyson's catalogue gave them important raw material. He later returned to the subject in a 1917 paper giving a statistical discussion of the proper motions in the Greenwich catalogue for 1910.

Chasing eclipses: the chromosphere and corona

The other strand of Dyson's research was the total solar eclipse. During totality, the Moon blocks the bright disc of the Sun and reveals two thin outer layers that are normally drowned out: the chromosphere, a reddish layer just above the visible surface, and the corona, the faint, pearly outer atmosphere that stretches far into space. Splitting their light into a spectrum shows bright lines that reveal which chemical elements are present and in what conditions.

Dyson went with Christie to observe an eclipse in Portugal in 1900. In 1901 he led his own expedition to Sumatra, with an amateur astronomer friend as helper and a good deal of assistance from the crew of a British naval vessel. Tenn describes how the party sailed through the Suez Canal and across the Indian Ocean, built concrete piers for the instruments, and rehearsed again and again so as not to waste a second of the few precious minutes of totality. The skies were clear. Dyson observed another eclipse in 1905. Over six eclipses he enjoyed good weather every time, and he called himself "a hundred percent eclipse observer."

From his spectra of the 1900, 1901 and 1905 eclipses, Dyson measured the wavelengths of some 1,200 lines in the spectrum of the chromosphere and compared their strengths with laboratory data. He published the results in the Philosophical Transactions of the Royal Society in 1906. Tenn credits the Sumatra spectra with the first detection of the element europium in the Sun. Dyson's measurements also supported the view of Norman Lockyer and Alfred Fowler, which had been widely criticised, that there are real physical differences between the chromosphere and the photosphere (the visible surface). This work made Dyson a recognised authority on the spectra of the chromosphere and corona. In 1901 he was elected a Fellow of the Royal Society.

Edinburgh: Astronomer Royal for Scotland

By the mid-1900s Dyson had a reputation for being equally good at the old mathematical astronomy of positions and motions and the new astrophysics of spectra. When the post of Astronomer Royal for Scotland fell vacant in 1905, he was appointed. Sources differ slightly on dates: some give 1905 as the start of his tenure and others 1906, which probably reflects the gap between appointment and arrival. The post carried a professorship of astronomy at the University of Edinburgh and the use of an almost new observatory on Blackford Hill with 15-inch and 24-inch telescopes, plus a house large enough for his growing family. He was elected a Fellow of the Royal Society of Edinburgh in 1906.

Tenn lists a good deal of work from Dyson's five Scottish years. Having finished Greenwich's share of the Carte du Ciel first, he agreed to measure and reduce plates taken at Perth, Australia, whose astronomer had fallen far behind with his zones. He began a study of double stars too close to the north celestial pole for the Greenwich refractor to reach, a project that carried on long after he left. He obtained better equipment for Jacob Halm's spectroscopic measurement of the Sun's rotation period. He was a popular professor who often invited students to his home, and his introductory lectures became his first book, Astronomy, published in London by Dent in 1910.

In 1910 Dyson hosted a group of physicists who tried, without success, to detect electrical or magnetic effects from Halley's Comet as the Earth passed near its tail. He also travelled to Pasadena, California, for a meeting of the International Solar Union. On his return he learned that Christie was retiring and that he had been chosen as the ninth Astronomer Royal.

Astronomer Royal (1910–1933)

What the job involved

The title "Astronomer Royal" dates back to 1675, when King Charles II founded the Royal Observatory at Greenwich to improve navigation at sea. For most of its history the Astronomer Royal was the director of the Observatory, and Tenn notes that this remained true until 1971. The Observatory itself was renamed the Royal Greenwich Observatory after it left Greenwich following the Second World War.

In Dyson's day the job was part scientist, part civil servant and part manager. The Observatory was funded through the Admiralty, and it had duties as well as research: keeping the nation's time, publishing the positions of stars and planets used by navigators, and recording the Earth's magnetism and the weather. Soon after taking the post, Dyson wrote frankly: "It isn't all fun by any means. What between getting money from the Admiralty, satisfying the staff of various grades, and generally looking after 50 or 60 people, not to mention the genuine scientific work, there's plenty to do and a good deal of worry."

By most accounts he was very good at it. His chief assistant John Jackson later wrote that Dyson "never exaggerated his case or presumed on the ignorance of others," that when he asked for money "he never asked for more than what he himself was convinced was necessary," and that "he was transparently honest." Jackson thought that this honesty meant Dyson "probably sometimes got his way where a more forceful man might have failed." Dyson also tended to leave his own name off the many papers his staff produced. The Encyclopedia.com entry on him, drawn from the Dictionary of Scientific Biography, notes that most of his published work was jointly written.

Modernising Greenwich

Dyson upgraded the Observatory in many practical ways.

• Magnetism: Greenwich had long recorded variations in the Earth's magnetic field. When the electrification of local railways began to interfere with the delicate magnetic instruments, Dyson modernised the equipment and moved the magnetic department out to the countryside.
• Latitude: He reorganised the measurement of the Observatory's latitude, using a floating zenith telescope, an instrument that floats in mercury so that it points precisely straight up.
• Clocks: He brought in a new kind of precision clock, the Shortt-Synchronome free-pendulum clock, discussed below.
• The Earth's rotation: In 1929 he published a paper on the variability of the Earth's rotation, an early sign that the better clocks were beginning to show that the Earth itself is not a perfect timekeeper.

Leading through the First World War

The war of 1914–1918 hit the Observatory hard. Tenn records that Dyson lost thirty-six members of staff to the armed forces, and that reduction work fell behind even though he hired retired staff, conscientious objectors, Belgian refugees and women to fill the gaps. The Observatory did its own bit for the war: Dyson and his staff collected, tested and cleaned binoculars for the British forces, and the Dysons held garden parties at the Observatory to raise money for the Red Cross. He was knighted in 1915, and served as vice-president of the Royal Society from 1913 to 1915 and president of the British Astronomical Association from 1916 to 1918.

The 1919 eclipse: testing Einstein

The question

In November 1915, in Berlin, Albert Einstein published his general theory of relativity, a new theory of gravity. In Isaac Newton's picture, gravity is a force pulling masses toward each other. In Einstein's, mass curves space and time, and objects, including light, follow the curves. For most everyday purposes the two theories give almost identical answers. They differ measurably only where gravity is strong.

Einstein proposed three tests. One was the slow shift in the orbit of Mercury, which Newton's theory could not fully explain and Einstein's could. Another was the bending of starlight as it passes the Sun. General relativity predicted that a ray of light grazing the edge of the Sun would be bent by 1.75 seconds of arc. A theory in which light simply responds to gravity like a falling object, but space is not curved, predicts exactly half that, about 0.87 seconds of arc. This half value was referred to at the time as the "Newtonian" prediction. A third possibility was no bending at all. Einstein's own earlier version of the theory, published in 1911, had predicted the smaller value. His final 1915 theory doubled it.

The bending would make stars near the Sun appear pushed slightly outward, away from the Sun. The shift is tiny, and the stars can only be seen near the Sun when the Sun's glare is blocked, which means during a total eclipse. The method is to photograph the star field around the eclipsed Sun, then photograph the same field months later at night, when the Sun has moved elsewhere in the sky, using the same telescope. Comparing the two sets of photographs shows whether the stars near the Sun have shifted.

Earlier attempts had failed. The German astronomer Erwin Freundlich tried to measure the effect at a 1914 eclipse in the Crimea, but the war broke out and he was arrested and expelled from Russia. According to Gilmore and Tausch-Pebody, several observatories examined old photographs or sent expeditions to eclipses in 1912, 1914 and 1918, but none produced a published result.

Dyson's insight

Because Britain and Germany were at war, scientific journals from Berlin reached Britain only indirectly. Arthur Eddington, by then a professor at Cambridge and Dyson's former chief assistant at Greenwich, received Einstein's papers through neutral Holland and became the theory's leading champion in Britain.

Dyson's contribution was to spot the opportunity. In 1917 he published a short paper in the Monthly Notices of the Royal Astronomical Society titled "On the opportunity afforded by the eclipse of 1919 May 29 of verifying Einstein's Theory of Gravitation". On that date, the eclipsed Sun would sit in front of the Hyades, a cluster of bright stars in the constellation Taurus. Most eclipses have few bright stars near the Sun. This one would have many, spread around the Sun at a useful range of distances, and the totality would also be unusually long. Tenn reports that the eclipse offered thirteen stars close enough and bright enough to photograph, and calls it the best eclipse in a thousand years for measuring the effect. Gilmore and Tausch-Pebody note that the combination of bright, well-spread stars and long totality made it possible to test both the size of the bending and the way it falls off with distance from the Sun.

Planning in wartime

The eclipse's path of totality crossed the South Atlantic, from Brazil to the west coast of Africa. There was no prospect of getting a ship while the war continued. Dyson planned anyway. He persuaded the Admiralty to let him prepare one expedition, and Eddington's deferment from military service allowed him to plan another. (Eddington was a Quaker and intended to declare himself a conscientious objector, which his Cambridge colleagues feared would embarrass the university.) The physicist James Jeans later summed up the situation: "In 1918, in the darkest days of the war, two expeditions were planned, one by Greenwich Observatory and one by Cambridge, to observe, if the state of civilisation should permit when the time came, the eclipse of May 1919 with a view to a crucial test of Einstein's generalised relativity."

The Armistice was signed on 11 November 1918, and the expeditions went ahead. Dyson himself stayed at Greenwich and directed from there.

The two expeditions

Gilmore and Tausch-Pebody point out that no Cambridge astronomy staff were available to go with Eddington, because both assistants at his observatory had been killed in the war. All three telescopes were mounted horizontally under shelters, with a moving mirror (a coelostat) in front of each to feed in the view of the sky. Everyone knew in advance that these mirrors could distort the images if they warmed up in the sun.

What the plates showed

Each telescope produced its own result, corrected to the edge of the Sun. The eclipse report, written by Dyson, Eddington and Davidson and published in the Philosophical Transactions of the Royal Society in 1920, gave the following (errors are "probable errors", the standard way of stating uncertainty at the time; modern standard deviations are about 1.48 times larger):

The Sobral 4-inch plates were the best. All of them had sharp star images. The Sobral astrograph plates were a problem. The report said the images were "diffused and apparently out of focus," and, worse, that this change was temporary: the telescope was back in focus for the evening check plates and the July comparison plates. In other words, something had changed in the instrument during the eclipse itself, most likely the sun heating the mirror, so the eclipse plates and the comparison plates were not truly comparable. The report described the separate measures from these plates as "discordant by an amount much beyond the limits of accidental error". The Greenwich team under Dyson decided not to use them in the final result.

To give a sense of scale, Dyson told a Royal Astronomical Society meeting in December 1919 that the quantities being measured on the plates were "of the order of 1/100th of a millimetre," and that astronomers familiar with stellar photographs would agree that such quantities "are readily measured, if only good photographs are secured." In a 1921 article in Nature he added that the displacements, though small, were ten times as large as those measured in routine determinations of stellar parallax (the method used to find the distances of nearby stars).

The announcement

The results were presented to a joint meeting of the Royal Society and the Royal Astronomical Society in London on 6 November 1919, chaired by the President of the Royal Society, the physicist J. J. Thomson. Dyson presented the observations. The conclusion was that the deflection matched Einstein's prediction and did not match the half value or zero. The press reported the meeting the next day, and on 7 November 1919 Einstein became a household name around the world. Jeans wrote that the expeditions "went and returned, bringing back news which changed, and that irrevocably, the astronomer's conception of the nature of gravitation and the ordinary man's conception of the nature of the universe in which he lives."

The later controversy, and what modern re-analysis found

At the time, the expert astronomical community accepted the result. Later, though, a story grew up that the 1919 data had been massaged. In 1980 the philosophers of science John Earman and Clark Glymour published a study arguing that the decision to discard the Sobral astrograph data was not justified, that those data were as good as the Príncipe data, and that, with them included, the result did not clearly favour Einstein. Their argument was taken up and widely repeated, including in the sociologists Harry Collins and Trevor Pinch's popular book The Golem. The accusation usually fell on Eddington, who was portrayed as a relativity enthusiast who saw what he wanted to see.

Two lines of later work have pushed back strongly. The historian Daniel Kennefick showed that the two expeditions' plates were reduced and analysed separately, and that the Sobral analyses, including the decision to exclude the astrograph results, were made by the Greenwich team under Dyson, not by Eddington. Kennefick argued that there were good grounds for thinking Dyson made the scientifically correct call, given his well-known caution about errors that could not be quantified.

In 2020 the Cambridge astronomers Gerard Gilmore and Gudrun Tausch-Pebody re-ran the 1919 calculations from the data published in the eclipse report, in a paper in the Royal Society's history journal Notes and Records. They found that the original analysis was mathematically sound and that, in modern statistical terms, the Sobral 4-inch result was highly inconsistent with the half value and with zero, while being consistent with Einstein. They concluded that Earman and Glymour's case rested on conceptual and mathematical errors, including treating the defective astrograph data as if they had a reliable uncertainty when the report had pointedly declined to give one. They also noted that a 1979 re-measurement of surviving Sobral plates by the astronomer Harvey, published in The Observatory, confirmed that the star positions had been reliably measured. (The Príncipe plates have been lost.)

Later eclipse measurements and, from the 1960s and 1970s, far more precise radio measurements of light bending by the Sun have confirmed Einstein's value. Whatever one thinks of the arguments about 1919, the prediction Dyson set out to test has held up.

The six pips: putting Greenwich time on the radio

Why time came from an observatory

Before the twentieth century, the most accurate clock available was the sky itself. An observatory could fix the time by timing the moment a star crossed the meridian, the north-south line overhead, and then keep the time with precision pendulum clocks between observations. Distributing that time was another matter. Each town had once kept its own local time, with noon arriving four minutes later for every degree of longitude westward. The railways pushed for a single standard, and in 1880 Greenwich time became the legal time across Great Britain. The 1884 International Meridian Conference in Washington adopted Greenwich as the world's prime meridian. Greenwich had been sending time signals by telegraph since the nineteenth century, and companies could subscribe through the General Post Office to receive them.

How the pips began

When the British Broadcasting Company began radio broadcasts in 1922, it marked the hour first with Big Ben's chimes played on tubular bells and then with an hourly signal from the Standard Time Company. According to Royal Museums Greenwich, the idea of something better came from Frank Hope-Jones, who had counted down the last five seconds before 10 pm in a 1923 broadcast. The BBC's head, John Reith, decided the public deserved the best available time signal, one coming directly from the Royal Observatory. Dyson agreed and designed the system.

The first broadcast went out live at 9.30 pm on Tuesday 5 February 1924, in a programme lasting fifteen minutes, with Dyson as a guest who counted "1, 2, 3, 4, 5, 6". The signal consists of six short tones at one-second intervals, with the sixth marking the exact moment of the hour (or half hour). The magazine Punch, Tenn reports, quipped that this enabled people to miss their morning trains by a smaller margin.

The clock behind the first pips was "Dent 2016", an astronomical regulator originally made for the 1874 Transit of Venus expeditions. In 1923 it became the primary standard for the signal, sending electrical impulses down a telephone line to the BBC, where they were turned into audible tones. A second regulator, Dent 2012, was upgraded as a backup a year later. At first newspapers called the signal the "ticks" or the "dots". The name "six pips" first appeared in print in 1929.

Better clocks: the Shortt free-pendulum clock

Dyson also brought in a new kind of clock that transformed Greenwich timekeeping. Invented by W. H. Shortt and known as the Shortt-Synchronome clock, it used two pendulums. The "master" pendulum swung in a sealed, low-pressure, temperature-controlled case and did no mechanical work at all, because almost anything a pendulum has to do disturbs its timing. A second, "slave" pendulum in an ordinary clock did all the work of counting and sending out signals, and was kept in step by the master through a small electrical link. Tenn records that Dyson liked to point out that the slave occasionally got to give the master a kick. Sources date the introduction of Shortt clocks at Greenwich to between 1924 and 1928. These clocks were accurate enough to begin revealing small irregularities in the Earth's own rotation, the subject of Dyson's 1929 paper.

From Britain to the world

Dyson pushed the time service well beyond Britain. Royal Museums Greenwich notes that from 1927 the time signal was broadcast worldwide, initially mainly for shipping, and Dyson arranged regular wireless transmission of Greenwich Mean Time from the Post Office's powerful radio station at Rugby. For ships at sea, an accurate time signal is a navigation tool: comparing local solar time with Greenwich time gives a ship's longitude, the very problem the Royal Observatory had been founded to solve.

For his work on time, Dyson served for several years as President of the British Horological Institute, the professional body for clockmakers, and received its gold medal in 1928.

What happened to the pips later

The signal outlived Dyson by decades. On 1 January 1972 it switched from Greenwich Mean Time to Coordinated Universal Time, the atomic-clock-based standard. When the Royal Greenwich Observatory moved to Cambridge in 1990 its time department was disbanded, and the last six pips transmitted from the Observatory went out on 5 February 1990, exactly 66 years after the first. Since then the BBC has generated the pips itself, from atomic time standards. The format Dyson designed is still heard on BBC radio.

International astronomy after the war

The First World War split the scientific world. The International Astronomical Union, founded in 1919, at first excluded scientists from the former Central Powers. Dyson was an internationalist who often remarked that science knows no frontiers. Tenn notes that only four years after the war he nominated a German astronomer for the Bruce Medal. He attended every General Assembly of the IAU from 1922 to 1935 and served as its President from 1928 to 1932.

His standing was recognised with a string of honours: the Royal Society's Royal Medal in 1921, the Bruce Medal of the Astronomical Society of the Pacific in 1922, the Gold Medal of the Royal Astronomical Society in 1925, and appointment as a Knight Commander of the Order of the British Empire (KBE) in the 1926 Birthday Honours. In November 1925 a portrait of him by Ernest Moore was shown in an exhibition called "Men of Yorkshire".

Retirement and death at sea

Dyson retired on 28 February 1933, at 65. He was succeeded by Harold Spencer Jones. Retirement was not idle. He spent much of his remaining time advising on the move of the Radcliffe Observatory from Oxford to Pretoria in South Africa, where the clearer southern skies offered better observing. He co-wrote a book, Eclipses of the Sun and Moon, published by Oxford in 1937, and visited his eight children and many grandchildren.

His wife Caroline died in 1937. In 1939 Dyson travelled to Australia. On the voyage home aboard the ship Ascanius he died at sea on 25 May 1939, and he was buried at sea the same day. Tenn notes that his predecessor Christie had also died and been buried at sea.

Family

Frank and Caroline Dyson had eight children, two sons and six daughters. The Wikipedia list of their names, which is not cited to a source there, gives them as Stella, Evelyn, Sylvia, Margaret, Palemon, Watson, Elsie and Ruth. In 1951 Margaret Wilson published a full-length biography, Ninth Astronomer Royal: The Life of Frank Watson Dyson (Cambridge), which remains the main book about him.

Not related to Freeman Dyson

People often assume that Frank Dyson was related to the physicist and writer Freeman Dyson (1923–2020). As far as is known, they were not. Both families' fathers came from West Yorkshire, where the surname is common. Freeman Dyson did, however, credit Frank with sparking his early interest in astronomy, and as a boy in 1931 Freeman wrote a story, "Sir Phillip Robert's Erolunar Collision", that featured a thinly disguised version of the Astronomer Royal.

Legacy

Dyson's legacy runs along several lines.

The test of general relativity. Dyson did not invent the eclipse test, and he did not travel to either site. But he identified the 1919 eclipse as uniquely suitable, kept the plans alive through the war, provided the Greenwich team and instruments, oversaw the Sobral analysis, and presented the results. Without him there would have been no 1919 measurement. Modern historians, notably Kennefick and Gilmore and Tausch-Pebody, have also restored his role in the decisions about the data, which were for decades wrongly attributed to Eddington alone.

Time for everyone. The six pips put a precise, authoritative time signal into ordinary homes for the first time. The system he designed outlasted the Observatory's own time department and is still broadcast. The move to free-pendulum clocks and worldwide radio time signals was a real step toward the era of precision timekeeping that followed with quartz and atomic clocks.

Star positions and motions. His work on the Astrographic Catalogue and on proper motions provided data that the first generation of galactic astronomers used to work out how the Milky Way's stars move.

Solar physics. His measurements of the chromosphere's spectrum were among the best of their time.

Institutions. He ran Greenwich through a world war, modernised it, and was by all accounts a respected and well-liked director. He also helped rebuild international astronomy after 1918.

His name is preserved in the lunar crater Dyson and the asteroid 1241 Dysona. In the television drama Longitude, broadcast in 2000, he was played by the actor Alec McCowen. His working papers are held with the Royal Greenwich Observatory archives at Cambridge University Library.

Frequently asked questions

Was Frank Dyson the ninth or the eleventh Astronomer Royal?

The ninth. He followed William Christie (the eighth) in 1910 and was followed by Harold Spencer Jones (the tenth) in 1933. At least one reference source gives "eleventh", which is an error; Margaret Wilson's biography is titled Ninth Astronomer Royal.

Did Dyson go on the 1919 eclipse expeditions?

No. He planned and organised the British effort and directed it from Greenwich. The Sobral expedition in Brazil was carried out by Charles Davidson and Andrew Crommelin of Greenwich, and the Príncipe expedition by Arthur Eddington and Edwin Cottingham.

Who decided to throw out some of the 1919 data?

The Sobral astrograph plates were set aside by the Greenwich team under Dyson, not by Eddington, according to the historian Daniel Kennefick. The plates were out of focus during the eclipse but in focus before and after, which pointed to a change in the instrument during totality. A 2020 re-analysis by Gilmore and Tausch-Pebody found the original decision and calculations sound.

What exactly are the "six pips"?

Six short tones broadcast one second apart, with the start of the sixth marking the exact hour. Dyson designed the format at the request of the BBC's John Reith, following a suggestion from Frank Hope-Jones, and the first broadcast was on 5 February 1924. The pips were generated at Greenwich until 5 February 1990 and have been produced by the BBC since then.

Why was time a job for an observatory?

Until atomic clocks arrived, the most reliable way to know the time was to watch the stars cross the meridian and keep that time with precision clocks. The Royal Observatory had been founded in 1675 to help sailors find their longitude, which depends on knowing the time accurately, and it kept Britain's official time for centuries.

What is a Shortt clock?

A precision pendulum clock with two pendulums: a "free" master pendulum in a sealed, low-pressure case that did no work, and a "slave" pendulum that did the counting and signalling and was kept in step by the master. Dyson introduced them at Greenwich in the mid-to-late 1920s, and they were the best mechanical clocks of their era.

How big was the light bending Dyson's teams measured?

About 1.98 seconds of arc from the best Sobral instrument and 1.61 seconds of arc from Príncipe, compared with Einstein's prediction of 1.75. One second of arc is 1/3,600 of a degree. On the photographic plates this meant shifts of around a hundredth of a millimetre.

Was Frank Dyson related to Freeman Dyson?

Not as far as anyone knows. The name is common in West Yorkshire, where both families came from. Freeman Dyson said Frank inspired his boyhood interest in astronomy.

Where is Dyson buried?

At sea. He died aboard the Ascanius on 25 May 1939 on the voyage home from Australia and was buried at sea that day.

Sources

• Frank Watson Dyson – Wikipedia
• Joseph S. Tenn, "Frank W. Dyson: The Seventeenth Bruce Medalist", Mercury, March/April 1993 (Sonoma State University)
• Sir Frank Dyson, 1868-1939 – Royal Museums Greenwich collections
• The BBC pips and the Royal Observatory Greenwich time service – Royal Museums Greenwich
• Gilmore & Tausch-Pebody, "The 1919 eclipse results that verified general relativity and their later detractors: a story re-told", Notes and Records of the Royal Society (2020), arXiv:2010.13744
• Dyson, Eddington & Davidson, "A determination of the deflection of light by the Sun's gravitational field" (1920), Phil. Trans. R. Soc. A 220
• José P. S. Lemos, "Shadow of the Moon and general relativity: Einstein, Dyson, Eddington and the 1919 light deflection" (2019), arXiv:1912.05587
• Dyson, Frank Watson – Encyclopedia.com (Complete Dictionary of Scientific Biography)
• A. S. Eddington, "Frank Watson Dyson, 1868-1939", Obituary Notices of Fellows of the Royal Society (1940)