The story of Invar [Basic Metrology]

Material properties are affected by temperature. This can be a major obstacle when, for instance, measuring precise lengths with a metal ruler. If the ruler was calibrated by one laboratory at a reference temperature— say 15°C—but then used at a different temperature in another laboratory, then the calibration certificate might include the value of the ruler's coefficient of thermal expansion, so that a correction could be made for the thermal expansion between the lower and higher temperatures. But what if the second “laboratory” was not a laboratory at all? What if it was an open field and the length to be measured was a 10 km geodesic baseline needed for surveying? This was a serious problem from the late 19th century to the late 20th century.