ABSTRACT
The measurements used in scientific practice, as in everyday life, are either measurements of the geometrical properties of bodies or are derived from them. Measurement must also be distinguished from counting. The distinction then between measurement proper, such as that of lengths, weights, densities and so on, and quasi-measurements such as that of hardness lies in just this, that in the former case we know enough about the laws of the properties to manipulate bodies so as to add or subtract the property in question. The measurement of temperature, before the thermo-dynamic scale was devised, is an example of a more satisfactory but still not perfect approach to measurement. Thermo-dynamic theory enabled the arbitrary element to be removed and made temperature quite strictly measurable. Supposing a system of direct weight measurement without balances and independent of mechanical theory had been developed, it would ultimately be necessary to correlate weight measurements with geometrical and mechanical knowledge independently obtained.
