Celsius proposes a temperature scale

Eighteenth-century thermometers used different liquids and scales, making records from separate places difficult to compare. Anders Celsius took part in weather observation and tested instruments. He sought fixed points that could be established repeatedly, reducing differences caused by manufacture. His paper of 1742 examined two stable points based on the freezing and boiling of water and investigated how environmental conditions affected those reference temperatures.

References: [1]

Celsius recognized that boiling depends on air pressure, so calibration required specified conditions. He divided the interval between the two points into one hundred parts. His original direction was reversed from familiar practice: boiling stood at zero and freezing at one hundred. Later users inverted it. Shared reference points made thermometers easier to check against one another and supplied a numerical record of heating and cooling in experiments.

References: [1]

Continuing observations in Uppsala produced a temperature series that allowed seasons and years to be compared. Medicine, industry, and laboratories also adopted more standardized measurements. Material changes, the states of water, and human body temperature entered regular instrumental records. Modern scales developed connections with thermodynamics and more precise standards. Calibration retained an essential working requirement: a number obtained from an instrument should have a meaning reproducible in another place and time.

References: [1]