Margen
1791 — The French Academy of Sciences, the institution that gathered the country's scientists, was handed a task both political and material: to give the world a common measure. It defined the metre as one ten-millionth of the quarter meridian, the arc running from the equator to a pole. The definition looked clean; to build the measure, the real size of that arc had to be known with precision.
June 1792 — Jean-Baptiste Delambre set out northward and Pierre Méchain southward. They were to measure by triangulation the stretch of meridian between Dunkirk and Barcelona, with Rodez as their meeting point. A baseline of known length would support a chain of triangles: once their angles were measured, each side could be calculated from the last without stretching a ruler across the whole territory.
Summer 1792 — The astronomers carried Borda repeating circles. The instrument allowed the same angle to be observed several times and the readings accumulated to reduce the effect of small errors. Every station meant climbing a tower, a summit or a belfry with a view of the neighboring points, setting up signals and waiting for adequate conditions.
Late 1792 — The French Revolution turned that work into a suspicious activity. Delambre traveled the roads with instruments, lights and papers issued by institutions whose authority kept changing. Patrols and local committees detained him several times. At each interruption he had to explain that the signals aimed at another tower belonged to a geodetic survey.
Winter 1792 to 1793 — Méchain installed his instruments in the fortress of Montjuic. From there he observed stars to determine Barcelona's latitude and close the southern end of the arc. A plumb line, the hanging weight that establishes the instrument's vertical, served as the reference for orienting the observations. Méchain repeated the readings until he had gathered a series he considered precise enough.
1793 — War between France and Spain closed the border and left Méchain stranded in the south. During the wait he suffered an accident in Barcelona with a hydraulic pump that caused severe chest injuries. He survived, but spent months recovering while his permission to return depended on two countries at war.
Early 1794 — Before leaving Barcelona he made a second determination of the latitude. This time he observed from the Fontana de Oro, the inn where he was staying. The result did not match the one obtained at Montjuic: the difference was about three seconds of arc, roughly equivalent to a hundred metres on the ground. Two points in the same city had produced verticals less concordant than expected.
1794 — The reconstruction by the historian Ken Alder, who examined Méchain's notebooks and published his conclusions in The Measure of All Things in 2002, documented what happened next. Méchain set aside the observations that revealed the discrepancy and prepared a corrected series for the mission. The papers kept the original calculations, but the Barcelona difference was not then laid out beside the submitted result.
The anomaly did not mean he had misplaced kilometres, nor that the whole triangulation was ruined. It affected one delicate, small astronomical determination. But the mission aimed to tie a universal unit to the shape of the Earth, and Méchain treated those seconds as a personal fault he had to correct before delivering his definitive data.
1794 to 1795 — The northern half also suffered an official interruption. The new government temporarily removed Delambre from the mission during a political purge and later reinstated him. On his return he resumed stations, rebuilt signals and went on linking the triangles. The expedition now depended as much on decrees and borders as on instruments.
1796 to 1797 — Méchain returned to France with the Barcelona notebooks, but delayed handing them over. In a private letter he admitted that the measurement tormented him and expressed his wish to repeat it. Alder found in that correspondence and in the withheld pages the joint trail of the confession, the concealment and the rewriting. Delambre, for his part, accepted working with series that included documented uncertainties.
1798 — The triangulation chain was completed between Dunkirk and Barcelona. Seven years had passed since the departure. Belfries, fortresses and summits formed a calculated succession across France and the northeastern peninsula. At the ends of the network, the baselines of Melun and Perpignan, each about twelve kilometres long, had been measured directly with rules laid end to end for weeks.
The calculations brought together the length of the land chain and the astronomical observations at its ends. From that stretch the full quadrant of the meridian was estimated. The operation required correcting for curvature and adopting a value for the flattening of the Earth, whose shape approximates a sphere widened at the equator and compressed at the poles.
1799 — An international commission gathered in Paris examined the results and a platinum bar was made to embody the calculated unit. The Metre of the Archives was deposited on June 22 in the Archives of the Republic. From then on, everyday length could be copied from a stable physical standard without repeating the expedition every time someone needed a reference.
The bar settled the practical use, but Méchain went on holding back the Barcelona problem. The discrepancy between the two latitudes had not been the direct basis of any public correction. He wanted to return to the observation points, measure them again and replace the series he considered flawed with one he could hand over without reservations.
1803 — Méchain went back to Spain to extend the triangulation and revise his work. The official expedition had already produced the platinum standard; this second journey answered the need to redo the southern observations. He moved down the coast seeking new stations and preparing measurements that would connect the arc with territories further south, on the way to the Balearics.
1804 — An epidemic of yellow fever reached the region where he was working. Méchain fell ill and died in Castellón on September 20 without returning to the two Barcelona points to close the discrepancy. His papers passed to those who had to put the mission in order and explain what had actually been measured.
After 1804 — Delambre received his colleague's notebooks and found the set-aside Barcelona observations. He decided to publish them along with the rest of the work. He also argued that the divergence could stem from the physical conditions of each station and from the limits of a vertical determined by plumb line, not necessarily from poor execution by the astronomer.
Local gravity does not point identically everywhere. The distribution of mass beneath the ground and around the observer can slightly deflect the direction a plumb line marks. The geodesy of the late eighteenth century could not yet describe those variations with precision. Montjuic and the Fontana de Oro could yield somewhat different astronomical latitudes even if the readings had been made with care.
1806 to 1810 — Delambre published the Base du système métrique décimal in three volumes. The work gathered the procedures, tables and observations of the expedition, including the awkward Barcelona figures. The complete record allowed others to examine both the construction of the metre and the decisions taken while reducing the data.
2002 — Ken Alder reconstructed the concealment while studying the archives for The Measure of All Things. By collating notebooks, letters and published results, he showed that Méchain had kept the discrepancy to himself and had tried to redo the series he was delivering. The documentation moved the episode from an anecdote about perfectionism to a verifiable sequence of observations, withheld pages and correspondence.
The modern review also separated those three seconds from the error in the standard's length. The estimate of the quadrant depended on the shape adopted for the Earth, and the value used for its flattening was inaccurate. The metre of 1799 came out roughly 0.2 millimetres short because of that misjudged flattening.