In this mini-series of blog posts we have followed the history of the English investigators of and writers on the phenomenon of declination or variation of the magnetic compass. Magnetic variation is the fact that although a magnetic needle aligns north/south the north that it points to, the magnetic north pole, is not the same as the geographical north pole. More than this, the amount that the magnetic north pole varies from the geographical north pole varies from location to location.
Initially navigators held theories or beliefs that these variations where somehow geographically regular and that they could be used to solve the major problem of determining longitude at sea. However, the more that European mariners ventured out onto the earth’s oceans and the more magnetic variations they determined, the more it became obvious that these followed no discernible pattern or regularity and thus could not be used to determine longitude. It seemed that this hope was finally laid to rest with the publication of William Gilbert’s De magnete in 1600, when Gilbert having surveyed all the available evidence categorically rejected the concept of regularity in magnetic variation. However, Gilbert continued believing that variation was constant for a given location and endorsed the suggestion of the Dutch polymath, Simon Stevin (1548–1620) put forward in his De Havenvinding (1599), also published in English as The Haven-Finding Art in the same year and included as an appendix in later editions Edward Wright’s Certaine Errors in Navigation, to provide tables of the correctly measured variation to compare with measured observations as an aid to navigation.
Despite Gilbert’s rejection of any form of regularity in magnetic variation, in the first decade of the seventeenth century, first the French author Guillaume de Nautonier, sieur de Castel-Franc au haut Languedoc, Géographe du roi Henri IV (1560– 1620) in his Mecometrie de leymant cest a dire La maniere de mesurer les longitudes par le moyen de l’eymant (The mecometry of the loadstone or the way to determining the longitude by means of the loadstone…), published in 1603 or 1604, and secondly the English author Anthony Linton, chaplain to Charles, Lord Howard of Effingham, who served as High Admiral from 1585–1618, of his Newes of the Complement of the Art of Navigation. And of the Mightie Empire of Cataia Together with the Straits of Anian, published in 1608, relaunched the idea of determining longitude with magnetic variation.
The idea was finally dealt what should have been a death blow[1] with the discovery in 1631 that the magnetic north displayed by the compass varied from the geographic north pole by differing amounts from location to location but that it also varied over time for a given location. How this was discovered and more importantly how this discovery was actually confirmed is the subject of this episode.
The discovery is attributed to Henry Gellibrand (1597–1637), who published the news in his A discourse mathematical of the variation of the magneticall needle together with its admirable diminution lately discovered (1635), but a total of five people contributed to this discovery over a period of sixty two years.
The story starts with the Elizabethan mathematicus, Thomas Digges (c. 1546–1595), who was the foster son of John Dee (1527–1608 or 1609), who we have met several times in this series, his father, Leonard (c. 1515–c. 1559), another mathematicus, having died when he was just thirteen years old. In 1571, Thomas Digges published a figure of 11° 15’ E in 1571, which was confirmed by William Borough (1536–1599), who made nine measurements in 1580, which he published in his A Discourse on the Variation of the Cumpas, or magneticall needle: Wherein is mathematically shewed, the manner of the obseruation, effects, and application thereof in 1581. This figure was accepted by William Gilbert (1544?–1603) in his work and Edward Wright (1561–1615) in his Certaine Errors.
In 1619, the mathematicus Edmund Gunter (1581–1626) (who I have dealt with in detail here) was appointed Gresham Professor of Mathematics, on the recommendation of Henry Briggs (1561–1630), and in 1622 he redetermined the magnetic variation of London and found it to be 6° 15′ E, significantly different to Borough’s value. Disturbed by this substantial difference to Borough’s value Gunter inquired as to the exact place where in Limehouse, Borough had made his observations. Then, with ‘a Quadrant of three feet Semidiameter, and two Needles, the one above 6 inches and the other 10 inches long’, had proceeded to the spot. Towards the night of 13 June 1622, he made observations of the sun’s azimuth from several parts of the ground and he stated, had obtained the following results:
The mean of theses observations 5° 55′ E is less by 20′ than Gunter’s previous observations and is almost half that of Borough’s of 1580.[2] Gunter, maybe not unnaturally, assumed that Borough’s observations were inaccurate, after all the history of magnetic variation up till that time had been littered with false or inaccurate determinations.
Gunter died in 1626 and, once again, on the recommendation of Henry Briggs, Henry Gellibrand was appointed as his successor. Gellibrand born in the parish of St Botolph, Aldersgate, London the eldest son of Henry Gellibrand (d. 1615), fellow of All Souls College, Oxford. He matriculated at Trinity College, Oxford 1616 graduating BA in 1619, and MA 1623. He took holy orders and was appointed to a curacy at Chiddingstone, Kent. However, inspired by a lecture by Henry Savile’s lectures he devoted his life to mathematics. Settling in Oxford he befriended Henry Briggs and when Briggs died in 1630 Gellibrand took over editing his Trigonometria Britannica, which he published in 1633.
Gellibrand devoted much of his time to trying to find a solution to the longitude problem, publishing An appendix concerning longitude in The Strange and Dangerous Voyage of Captaine Thomas James (1633).
As already noted Gellibrand remeasured the magnetic variation at exactly the same place as Gunter in 1631 and found it to now be 4° 05′ E, even less that Gunter’s value. There existed a problem with these values, Borough had been long dead when Gunter and Gellibrand made their measurements and Gunter dead when Gellibrand made his so it would seem that nobody could confirm that the measurements made by the others had been made correctly. However, there were two qualified witnesses to both the measurements of Gunter and Gellibrand, the first of these was the instrument maker, John Marr (fl. 1614–1647), who played a central role in the whole story.
John Marr is one of those minor figures about whom we know very little but on this occasion played a significant role in the history of science and technology. Marr or Mair was a Scotsman who served both James I and Charles I as a compass-maker, dialist and dial maker. According to the astrologer William Lilly (1602–1681), Marr was an excellent mathematician and geometrician. Lilly also claimed that Marr was familiar with both Henry Briggs and John Napier (1550–1617) and had been present at the first meeting of the two compilers of logarithmic tables in Edinburgh in 1614. In 1631, Marr and Elias Allen (c. 1588–1653) had supplied magnetic needles of special quality to Captain Thomas James (1593–1635) for his Arctic voyage. It was in the account of this voyage that Gellibrand published his work on longitude, as mentioned above.
John Marr worked together with Edmund Gunter and had taken part in his measurement of the magnetic variation in Limehouse in 1622. Also present on that occasion was John Wells (fl. 1606–35), who held the post of storekeeper for life at the Navel Storehouses at Deptford. He was also a mathematical practitioner who had an excellent reputation as a dial maker. Like Marr he belonged to the circle around Henry Briggs, Edmund Gunter , and Henry Gellibrand. Wells wrote one book, Sciographia or the Art of Shadows (1634), on the construction of dials. Originally only a manuscript written for a friend, Edmund Gunter read it and urged him to publish it. After Gunter deaths it was still unpublished and now Henry Gellibrand renewed the urging, writing a preface for the published book.[3]
In 1622, when Gunter and Marr found the substantial difference between their determination of variation and that of Borough forty years earlier, they decided to cross the river and repeat their measurements in John Wells’ garden. This meant that both Marr and Wells could accurately report on Gunter’s measurement procedure and his results.
In fact, it was Marr, who originally repeated Gunter’s measurements, this time in Whitehall, with one of Gunter’s needles from the earlier trials in 1631 and obtained the new value of 4° 05′ E. He took this disturbing result to Gellibrand and a team including Gellibrand, Marr and Wells repeated exactly as possible Gunter’s measurements in Limehouse and also across the riven in Wells’ garden confirming that the magnetic variation had indeed changed once again.
As noted above, Gellibrand then published the results of these trials in 1635 and is generally given credit alone for the discovery but without the testimony of Marr and Wells, Gellibrand’s discovery might well not have been accepted as it was.
Of course, none of those involved could actually explain why the variation varied over time or even why the variation existed in the first place. As we will see in the next post there would be various explanations offered over the years without any of them actually being correct. As we now know the variation in variation over time is caused by the fact that the pole is not static but wanders.
True polar wander represents the shift in the geographical poles relative to Earth’s surface, after accounting for the motion of the tectonic plates. This motion is caused by the rearrangement of the mantle and the crust in order to align the maximal inertia with the current rotation axis. (Wikipedia)
[1] We will see that hopes dies eternal in the next post in this series.
[2] David Waters, The Art of Navigation in England in Elizabethan and Early Stuart Times, Yale University Press, New Haven, 1958
[3] The brief accounts of John Marr and John Wells are taken from E. G. R. Taylor, The Mathematical Practitioners of Tudor & Stuart England, CUP), 1970 Marr p. 203-204. Wells p. 199
The Renaissance Mathematicus 