On this page
Every time you convert 212 °F into 100 °C, you are using numbers chosen by one instrument maker in the early 1700s. Daniel Fahrenheit's real achievement was not a scale with odd numbers. It was thermometers that agreed with each other. This post covers who he was, how he fixed his points, what he found about freezing water, and how to convert his readings without the common mistakes.
Who was Daniel Fahrenheit?
The Museo Galileo biography records that he was born in Danzig (now Gdańsk), lived from 1686 to 1736, and spent most of his life in Holland. He was a scientific-instrument maker and merchant with a special skill in blowing glass. In 1724 he was elected a Fellow of the Royal Society of London, when he proposed his system for making thermometers. He also invented a type of hydrometer (an instrument for measuring the density of liquids).
He did not write much. His reputation rests on a handful of short Latin papers in the Royal Society's Philosophical Transactions for 1724, and on the instruments themselves.
How did Fahrenheit fix the points on his scale?
His paper on freezing water in a vacuum begins by describing his thermometers. He made two kinds, one filled with spirit of wine (alcohol) and one with quicksilver (mercury). Their lengths varied, but, he wrote, all of them agreed in every degree of the scale and moved between fixed limits. That agreement was the point.
The weather thermometers ran from 0 to 96, set by three fixed points, as you can read in the archive.org copy of the 1724 paper:
0: a mixture of ice, water and sal ammoniac (ammonium chloride) or sea salt. He noted this worked better in winter than in summer.
32: ice and water without the salt, which he called the beginning of freezing.
96: the thermometer held in the mouth or under the armpit of a healthy person until it took up the body's heat.
For fevers he made thermometers that ran to 128 or 132. For boiling liquids he made a scale from 0 to 600, since mercury itself begins to boil at about that point. He used cylinders rather than round bulbs, so the larger surface would respond faster to changes of heat.
Why did he choose mercury?
In a second 1724 paper on boiling points, he explains that about ten years earlier he had read that Guillaume Amontons had found water boils at a fixed degree of heat. He wanted to see it for himself. Remembering that the height of mercury in a barometer changes slightly with temperature, he reasoned that mercury could make a good thermometer, and built one.
He then measured several liquids. On his scale, spirit of wine boiled at 176, rain water at 212, lye at 240, spirit of nitre at 242 and oil of vitriol at 546. He added a careful warning: apart from spirit of wine and water, these numbers might change, especially with larger quantities boiled for longer. He also noted that 48 sat in the middle of his scale, between the cold of the ice and salt mixture and the heat of a healthy person's blood.
What did he find about freezing water?
On 2 March 1721 he left a small sealed glass globe, emptied of air and about half filled with rain water, out in the frost. His thermometer showed 15 degrees, well below 32. The next morning the water was still liquid. When he broke the tip of the tube and let air in, the water suddenly filled with thin plates of ice, and a thermometer placed in the mixture read 32. Today we call liquid water below its freezing point supercooled water.
Worked example: converting Fahrenheit's numbers
The modern relation is °C = (°F − 32) × 5/9. Here 32 is the offset and 5/9 is the ratio of the step sizes, because 180 Fahrenheit degrees span the same range as 100 Celsius degrees.
Rain water boiling, 212 °F: (212 − 32) × 5/9 = 180 × 5/9 = 100 °C.
His healthy body point, 96 °F: (96 − 32) × 5/9 = 64 × 5/9 ≈ 35.6 °C.
Spirit of wine boiling, 176 °F: (176 − 32) × 5/9 = 144 × 5/9 = 80 °C.
Where the scales meet: set F = C. Then C = (C − 32) × 5/9, so 9C = 5C − 160, and C = −40. So −40 °F and −40 °C are the same temperature.
The same method works for any made-up scale in an exam. If a thermometer reads 10 in melting ice and 130 in boiling water, a reading of 70 means (70 − 10)/(130 − 10) × 100 = 60/120 × 100 = 50 °C.
Mistakes to avoid:
Multiplying before subtracting. Take away 32 first, then multiply by 5/9.
Using the offset for a change. A rise of 9 °F is a rise of 5 °C. The 32 applies only to readings, not to differences.
Flipping the ratio. A Celsius degree is bigger than a Fahrenheit degree, so going from °F to °C you multiply by 5/9, not 9/5.
For the scale that replaced his in most of the world, read about Anders Celsius and his scale. Temperature differences matter most in heat problems, which our guide to calorimetry and heat transfer works through. If you get stuck on a conversion with an unusual scale, the Fermi Problem Library has a step-by-step solver that explains the reasoning rather than printing an answer.
What can a student take from how he worked?
Make your results repeatable. He chose fixed points that anyone with ice, salt and water could reproduce.
Borrow clues from other instruments. A small effect in barometers suggested a new thermometer.
State your uncertainty. He warned readers which boiling points might shift and why.
Notice when nature surprises you. Liquid water at 15 degrees was a puzzle, and he tested it by changing one thing: letting the air back in.
Try it now: convert 98.6 °F, 50 °F and −4 °F into Celsius, then check that each answer makes sense.
