thermometry
/ ther-MOM-uh-tree /
You cannot see temperature directly, so you read it off something that changes with it — mercury rising in a thin tube, a metal coil expanding, the colour of a glowing iron. Thermometry is the whole science of doing this reliably: choosing what to watch, and turning its change into a trustworthy temperature number.
More precisely, thermometry is the measurement of temperature. It relies on a thermometric property — a quantity that varies smoothly and reproducibly with temperature, such as the volume of a liquid, the electrical resistance of a wire, or the voltage across a junction of two metals. The measured property is mapped onto a temperature scale that is anchored to agreed fixed points, like the triple point of water.
Why it matters: temperature governs reaction rates, equilibria, phase changes, and nearly every quantity in physical chemistry, so almost any precise experiment depends on measuring it well. A caveat: a thermometer reports its own temperature, which equals the sample's only once they have come into thermal equilibrium — and inserting a thermometer can itself disturb what you are trying to measure.
In a calorimetry experiment, a chemist needs the temperature rise to a hundredth of a degree, so a simple liquid-in-glass thermometer will not do. They use a platinum resistance thermometer, whose electrical resistance changes predictably with temperature, calibrated against the freezing and boiling points of water beforehand.
Temperature is always read indirectly, through something that changes when it does.
A thermometric property only has to vary reliably with temperature; it need not vary in a straight line. Mercury thermometers happen to be nearly linear, but resistance thermometers and thermocouples follow curved relations, which is why each must be calibrated rather than simply read off.