At the top, the triangles of the meridian chain light up one by one, from Dunkirk to Barcelona. An orange wave of the krypton line runs through the arches of the letter m. Below lies the standard bar with lines engraved at its ends. A red pulse of light travels along it from zero to the 1 m mark and flashes. Above the right-hand line sits a microscope: the lines drift slowly through its cross-hairs. In the N window interference fringes run past.
Quantity data sheet
length rev. 2026-10Metre m
SI base unit · quantity: length L
Seven years on a meridian arc, and the bar came out 0.2 mm short
In June 1792 two astronomers, Delambre and Méchain, left Paris. They had been charged with measuring the arc of the meridian from Dunkirk to Barcelona. From this arc a new unit of length was to be derived: one ten-millionth of the distance from the pole to the equator. The work took seven years.
In 1799 a platinum bar was made from their calculation — the Metre of the Archives. It later turned out to be about 0.2 mm shorter than intended. It was not corrected: rulers had already been checked against it. Since then the definition has changed three times. Each time the length stayed the same; only the way of reproducing it changed.
Today the metre is defined through the speed of light. It is the path light travels in a vacuum in 1/299 792 458 of a second. The number c is fixed exactly, so the metre can be reproduced as precisely as the second.
The symbol is a lower-case Latin m, upright and without a full stop. Russian texts write «м». A capital M does not mean the metre but the prefix mega: Mm is a megametre, a thousand kilometres.
m is the metre, not mass m and not the minute min. mm is the millimetre, not “metre by metre”: the square metre is written m². The nautical mile is written M or NM, and it is confused with the metre more often than you might think.
01 · Definition
meridian · bar · wave · lightThe metre is the length of the path travelled by light in a vacuum during 1/299 792 458 of a second. That is how the definition has read since 1983. Since 2019 the same idea is put through a number: the speed of light c is exactly 299 792 458 m/s.
Such a definition needs no stored standard. Any laboratory can reproduce the metre itself if it has an accurate clock and a light source of known frequency. In practice a frequency-stabilised laser is used and wavelengths are counted.
Before that the metre was defined three times, each time differently. First through the size of the Earth. Then through a platinum bar in a vault beneath Paris. Then through the wavelength of light from a krypton lamp. At each step the new definition was fitted to the old length, and precision rose roughly a hundredfold.
What the metre was measured with, and how precisely
Four tabs — four definitions in order. Steel blue shows the instrument and the standard, orange the light of the krypton lamp and the fringes, red the laser pulse and anything beyond tolerance. On the right of each tab a scale shows how far the metre obtained differs from today’s. The sliders change the conditions of the experiment; the action button runs the measurement again.
Time is measured more precisely than length
By the 1970s atomic clocks had become thousands of times more accurate than any length standard. By then the speed of light had been measured so well that the uncertainty came almost entirely from the krypton metre. So it was decided to work the other way round: declare c an exact number and derive the metre from the second. Since 1983 the speed of light is no longer measured. Measuring c became a check of clocks and interferometers.
02 · Conversion
SI · imperial · historicalThe metre in other units of length
Enter a length and choose a unit. The table converts it into multiples of the metre, into imperial measures or into the old measures of Europe. Since 1959 the yard and the inch have been defined exactly through the metre, so almost all numbers in the imperial tab are exact.
The international prototype of 1889 showed the metre at 0 °C, in melting ice. Today the dimensions of parts and gauge blocks are stated at 20 °C, as the ISO 1 standard requires. A steel part 1 m long is about 0.1 mm longer at 30 °C than at 20 °C. That is more than the tolerance of many fits.
| Unit | Value | How it is defined |
|---|---|---|
| {k} | {v} | {note} |
{text}
04 · Measuring instruments
angle · line · fringe · pulseFour instruments — four metres
Borda’s repeating circle
With this circle Delambre and Méchain measured the angles of the triangles and the altitude of stars above the horizon. An angle was taken many times in a row without returning the circle to zero, and the sum was then divided by the number of readings. In this way the error of a single reading shrank. From the altitude of stars they found the latitude of the ends of the arc.
Comparator: bar against bar
Two microscopes are fixed above a table about a metre apart. The standard and the copy are brought beneath them in turn. In each microscope three lines on the end of the bar are visible. The operator aligns the middle line with the cross-hair and takes a reading on the micrometer screw. The difference between the readings for the two bars is the difference in their lengths, usually a fraction of a micrometre.
Interferometer: counting fringes
Light is split into two beams and brought together again on a screen. If one mirror is moved by half a wavelength, a bright fringe on the screen gives way to the next. The number of fringes that pass, multiplied by λ/2, gives the mirror’s shift. In 1892 Michelson expressed the metre this way in wavelengths of the red line of cadmium.
The metre as time of flight
A telescope sends a laser pulse to a corner-cube reflector left on the Moon by an expedition. The light returns after about 2.56 s. The distance equals half this time multiplied by c. Of millions of photons only a handful come back, but after averaging the distance to the Moon is known to within millimetres.
05 · Writing rules
space · case · prefixesNumber, space, m
A non-breaking space goes between the number and the symbol. The symbol is written upright, without a full stop and without endings. A prefix is written together with m, and only one. The unit is given with each number when there are several quantities.
The prefix kilo is written with a lower-case k: a capital K means the kelvin. “Micron” and “mkm” went out of use in 1967; the correct form is µm. Two prefixes in a row are forbidden, so instead of “mµm” one writes nm. In “3 × 4 m” it is unclear whether the metre applies to both numbers. The abbreviations mps and sec come from everyday English and are not part of the SI.
06 · Neighbouring units
through the second and through cThe metre rests on the second: without a clock it cannot be reproduced. The light year is defined through the same speed of light, and the ångström is simply a tenth of a nanometre, kept for atoms and crystals.
The metre is derived from it and from the number c.
ÅThe size of an atom, convenient for crystallographers.
lyAlso a path of light, only over a year.
07 · Historical section
archive · four definitionsTwo hundred years of one and the same length
{title}
{text}
Each time roughly a hundred times more precise
The length of a bar is how many decimal places of the metre could be reproduced. The scale is logarithmic: each division is one decimal place. For 1799 the discrepancy between the Metre of the Archives and the plan is shown, for the others the uncertainty of reproduction.
In two hundred years precision grew ten-million-fold. Yet the length itself did not change: the metre of 1983 equals the metre of 1889 within the precision of the bar.
Two lengths that could have become the metre
In 1791 the Academy of Sciences had two candidates. The first was the length of a pendulum that swings from one side to the other in exactly one second. Such a pendulum is shorter than today’s metre: about 0.994 m. It was rejected. Gravity varies with latitude, and the pendulum comes out with a different length in Paris and at the equator. Besides, the length would depend on the second, and the second was then a fraction of the day.
The second candidate was one ten-millionth of a quarter of the meridian. That one was chosen, and the arc was to be measured. But the real quarter meridian is about 10 001 966 m. So the intended metre would have been 0.197 mm longer than the Metre of the Archives.
The Metre of the Archives was an end standard: its length was set by its ends. The ends wear with every touch. That is why the 1889 standard was made a line standard: its length is set by lines on the neutral plane of the bar. The length was carried over from the Metre of the Archives, not from the meridian. So the 0.197 mm stayed in the metre for ever.
Catalogue of quantities
metre passport open
Index · all quantities on one pageThe lab at the entrance to the indexA drop of acid changes the colour of the juice, salt colours the flame, water freezes at a tap. Below them are all Simetrium sheets, by discipline and alphabetically.Open the index →
Units of length and everything derived from it
{note}
