The effect is the same as that produced when messages are passed on and on: very soon they have changed so that they have no longer much in common with what was originally said. This may happen to all things which are passed on from one stage to the next. In metrology, this would be disastrous. It has, for example, been agreed on the international level to determine the mass by the exactly defined unit of mass, the kilogram. The international prototype kilogram (the only SI base unit still realized by a simple solid body) is stored under a glass bell in the Bureau International des Poids et Mesures 
BIPM near Paris. Copies of it are maintained at the national metrology institutes, and these copies in turn are copied, passed on, copied again, and so on. In this way, each mass standard is ultimately traced back to the Paris prototype kilogram – unless errors occurred sometime during copying and have been passed on – the effect referred to above.
Paths ensuring the traceability of measurements
To prevent this effect, one meticulously checks whenever a copy is made whether the two standards are in fact identical. Random checks carried out later at regular intervals show whether the two bodies have changed. An entire 
hierarchy of standards which differ only minimally is established in this way. And it is for this reason that whoever wants to have a measuring instrument calibrated (and whose requirements for the uncertainty of measurement are not extremely high) may use for this purpose also a standard from the lower categories. If the measuring instrument determines the standard's data with sufficient accuracy, traceability to the national standard – and ultimately to the international standard as well – has been proved.
Depending on the legal situation there are two paths which both lead to the BIPM: the path of 
calibration and – for all measuring instruments subject to the Verification Act – that of verification.
© Physikalisch-Technische Bundesanstalt, last update: 2011-11-17, Volker Großmann
Printview, 
PDF