Unlike Celsius, the Fahrenheit scale wasn't based on pure water — and that explains why its numbers seem so arbitrary.
Daniel Gabriel Fahrenheit was born in 1686 in the city of Danzig (now Gdańsk, Poland), then part of the Polish-Lithuanian Commonwealth, into a family of prosperous merchants. His childhood was marked by a double tragedy: at age 15, both of his parents died within days of each other, apparently after eating poisonous mushrooms. Orphaned as a teenager, Fahrenheit was sent to work as a merchant's apprentice in Amsterdam, but his real interest always lay in natural science and precision instruments, the field where he would go on to build his entire career.
Fahrenheit settled professionally in the Netherlands, then one of Europe's most advanced centers for scientific instrument manufacturing, and became a renowned maker of thermometers and barometers. He was the one who decisively popularized the use of mercury as a thermometric liquid, replacing the alcohol used until then. This change was not trivial: mercury expands and contracts much more linearly and predictably with temperature than alcohol, and has a much higher boiling point, allowing far more stable, precise, and reproducible measurements across different instruments — a real problem that plagued scientists of the era, since thermometers from different makers frequently gave inconsistent readings.
The scale bearing his name, formally presented in 1724 in a paper published by the Royal Society of London, does not use pure water as its main reference — unlike the approach Anders Celsius would adopt almost two decades later. Fahrenheit set the zero of his scale using a mixture of crushed ice, water, and ammonium chloride (sal ammoniac), the coldest, most stable temperature he could reliably reproduce in his early 18th-century laboratory using the materials available at the time. This type of freezing mixture was known to chemists of the era as a way to reach temperatures below the freezing point of pure water.
The value 96°F, in turn, was originally associated by Fahrenheit with human body temperature (measured in the mouth or under the arm), though modern measurements show that figure was slightly off — average human body temperature is now recognized as closer to 98.6°F. Historians of science speculate that Fahrenheit may have chosen 96, instead of a round number like 100, because 96 is repeatedly divisible by 2 (48, 24, 12, 6, 3), making it easier to precisely graduate the scale manually into subdivisions using only a compass and ruler, without complex decimal calculations.
These unintuitive reference points explain why water freezes at exactly 32°F and boils at 212°F — numbers that seem completely random until you understand the original logic behind them: the distance between these two points (180 degrees) wasn't deliberately planned, but rather an indirect consequence of how the zero and body-temperature reference point were originally defined.
Although today it's associated almost exclusively with the United States, the Fahrenheit scale was once widely used across much of Europe and the English-speaking world during the 18th and 19th centuries, including the United Kingdom, which only officially migrated to Celsius in the mid-20th century, as part of a broader modernization of measurement standards. The United States' resistance to abandoning Fahrenheit — even after official attempts at metric conversion in the 1970s — is today more a matter of cultural inertia and transition cost than any real technical advantage of the scale.
Today, the Fahrenheit scale is officially used as the standard in only a handful of territories — mainly the United States, plus a few smaller territories like the Cayman Islands and the Bahamas — while the rest of the world adopted the simpler two-fixed-point logic of the Celsius scale. Even so, Fahrenheit survives robustly in American technical vocabulary, appearing in weather forecasts, cooking recipes, home thermostats, and even industrial oven specifications — a reminder of how measurement conventions, once culturally rooted in a generation, tend to survive well beyond their original technical usefulness.