How accurately can temperature be measured?

Suppose you have been asked to measure the temperature of an object known to be at around 1000°C. Being a knowledgeable physicist, you know that the most accurate procedure is to obtain the best thermocouple you can and use it carefully, in accordance with the manufacturer’s instructions. The calibration table that comes with the instrument charts emf values to the nearest microvolt, corresponding to about 0.1°C, and with practice you find you can get reproducible readings within this interval. Does this mean that you are measuring temperature to $\text{±0.1°}\mathrm{C}$ accuracy?

This article is only available in PDF format