Margen
You plant one foot either side of the brass strip at the Royal Observatory in Greenwich, London, you take out your phone, and the screen tells you that you are at 0° 00′ 05.3″ west. You are not on the prime meridian. The zero your phone uses lies 102 metres further east, where there is no strip, no plaque and no queue of tourists. Neither of the two lines is wrong.
The brass strip marks the Airy meridian, named after George Airy, the British Astronomer Royal who in 1850 installed his transit circle there, a fixed telescope that only looks straight up and times the stars as they cross a single point of the sky. That was the instrument enshrined in 1884, when the International Meridian Conference brought delegates from twenty-five countries together in Washington to settle once and for all where the world counted longitude from. Greenwich won with twenty-two votes in favour and one against.
The GPS zero dates from 1984 and is counted from the IERS reference meridian, which is driven into no ground at all: it is worked out from the average position of dozens of observatories scattered across the planet. The satellites do not measure against the real Earth, with its mountains and its hollows, but against a smooth ellipsoid, a ball of arithmetic called WGS84.
What separates the two lines is a plumb line.
Airy levelled his telescope the way everything was levelled in 1850: by hanging a weight from a thread and accepting that this was the vertical. And it was, the vertical of that spot. A plumb line obeys the gravity immediately beneath it, and beneath Greenwich there is rock of uneven density that tugs the thread slightly to one side. Geodesists call it the deflection of the vertical, and at Greenwich it puts some 5.3 seconds of arc between longitude measured by the stars and longitude measured by satellite: an angle that would fit on a pinhead seen from ten metres away and that, projected down to the ground, turns into those hundred-odd metres.
The whole calculation was done in 2015 by the geodesist Stephen Malys with four other researchers, in the Journal of Geodesy. Until then the difference was explained any old how — that GPS had corrected the British maps, that somebody had laid the plaque in the wrong place — and nobody was in a hurry to check, because it got in nobody's way, until tourists began carrying a GPS receiver in their pocket. Malys compared nineteenth-century astronomy with modern coordinates, measured the deflection and put a number on the gap: 102.5 metres.
The observatory has not moved the strip and is not going to move it. It is the exact spot where the telescope that set the world's clock stood, and it is still selling the photograph. Which is more than France did.
Of the twenty-five countries of 1884, two abstained, and one of the two carried on regardless: France kept the Paris meridian for another twenty-seven years for its time, and until 1914 for the longitude on its nautical charts. When it gave up the time, in 1911, it avoided writing the word Greenwich in the official gazette. The law defined French legal time as “Paris mean time retarded by nine minutes and twenty-one seconds”, which is Greenwich, said without naming it.