Early Danish thermometers.
The thermometers of Ole Rømer.
Erling Poulsen
In Polish

1681
Rømer returned from Paris to become a professor of astronomy. In Paris he had discovered that light had a limited velocity, and he had designed planetariums and an advanced eclipsarium.
1683
One of his first assignments was to carry out a reform of the system of weights and measurements, all units became founded on one unit, the Rhineland foot.
1685
He became the director of the Roundtower Observatory.
1687
He travelled to England, Holland and France to meet other scientists and to learn more.
1688
When he returned he started making the Observatory up-to-date. There was placed an altazhimuth and an equatorial refractor on the top, both with a revolving dome.
1689
He moved to one of the old and very stable canon-houses (made of bricks).
1690
He constructed his transitinstrument in his new home. From erecting that instrument he learned many things, which later helped him to construct the meridian-circle (1704).
1692
When he used his new instrument he had serious problems with the refraction of light and the temperature. When he used the transitinstrument he had to read both the time on the pendulum clock and the latitude of the star on the graduated arc. But the length of both depended on the temperature, in the summer the clock was too slow and the distance between graduations was too large. To compensate for variations of temperature he made a measurement of the expansion of different materials on December 12th 1692, and described everything in his work papers1), and from his papers (Royal Library, Copenhagen) we learn: He had a thermometer (how it was divided, he doesn't tell, he probably purchased it during his travels). When we compare his measurements with modern values, he made a pretty precise experiment, in which he says that the difference of temperature in all measurements is 24°. When this is compared to modern values we find that 1°? = 1,9°C. In the following years, he uses this measurement to compensate for variations of temperature to his readings from the instrument.
1702
He began producing thermometers, and in his work papers he describes how. First he took a glasstube (18") and examined the internal diameter with a drop of Mercury (the volume of which he knew from its mass). If the diameter was the same all the way through, it was OK, and if it was conical he could use it too. Otherwise, he threw it away. Then he cast a hollow glass-sphere at the end of the tube; to find the diameter of this sphere he gave a relation between the diameter of the tube, the diameter of the sphere and the length the thermometerliquid would expand when heated 10°Rø (spiritus vini, colored with saffron), he also gave some examples with numbers2). Then he filled the thermometer, sealed it, put it in ice-water and made a mark on the tube, put it in boiling water and made another mark on the tube (from his papers we learn that he was only certain about the stability of the boiling point from 1703). Then he divided the volume between the marks in 7 equal volumes and placed one of them under the freezing-point, here is 0°Rø, at the freezing point is 7½°Rø and at the boiling point is 60°Rø3). This gives an average of 1°Rø = 1,9°C, so he may already have used this scale with two fixpoints in 1692. From the relation between the diameters and the length of 10°Rø we can find the average coefficient of expansion of the liquid (66,7*10-5 °C-1), and from that the strength of the "spiritus vini" (39%vol.). Now we can reconstruct his thermometer scale. We find that 0°Rø = -22,5°C. A zero-temperature in agreement with his own temperature readings from the winter of 17094) when compared to other descriptions from this very cold winter.
Drawing of a thermometer from his working papers.
1708
Reconstruction of a Rømer thermometerAll of his thermometer constructions would only have theoretical interest if it had not been for a visit by Daniel Fahrenheit (then 22 years old). Rømer was then 64 years old and a scientist of international reputation (Fahrenheit uses the word voortrefflijken) and received Fahrenheit in his home. Fahrenheit must have heard about his constructions and came to learn5). Rømer showed him the construction of an unusual thermometer, he does not describe this kind of thermometer in his work papers; the thermometer has two fixpoints, the freezing point and the human body temperature (at 22½°Rø, Fahrenheit uses the word blutwarm). The volume between 7½°Rø and 22½°Rø is divided into two, and one part is placed under the freezing-point. Our reconstruction of the thermometer scale gives 22½°Rø = 35,9°C in good agreement with the human body temperature in the mouth. Rømer has a description6) of an experiment concerning the expansion of water and air, in which he uses a thermometerlike glass instrument. If we imagine this instrument filled with his "spiritus vini", it will have the dimensions of one of these thermometers, the instrument could have been an open and empty thermometer. Maybe with the original thermometers Rømer discovered that 22½°Rø was the human body temperature and used this fact to construct short thermometers unable to show high temperatures. It is also easier to fill a thermometer if the temperature of the maximum point is much lower than the boiling-point of the liquid.
Later
Fahrenheit used this scale until 17177), with the only difference that he divided every °Rø in four °F, so the two fixpoints, the freezing point was 30°F and the human body temperature was 90°F. He then changed the scale (because it was difficult to divide in thirtieth)7) to FP = 32°F and HBT = 96°F. He discovered that the temperature of youngsters were higher than that of elderly people, so HBT was not so fixed as he had believed. He changed this fixpoint to the temperature of boiling water, BP = 205°F to 212°F, depending on air pressure, so the new fixpoint gave a scale similar to the old one8). Later, he began making thermometers filled with Mercury and after experiments he preferred this liquid; furthermore, it was difficult to obtain alcohol with the same strength9) and consequently the same expansion, every time he wanted to make thermometers.
1739
This year The University Library got the work papers of Rømer from his widow, as the professor of astronomy and his successor, Peder Horrebow, read the papers, and he wrote some remarks in them. In the papers concerning thermometers he proposes another thermometer scale but founded on Rømer10). Because very low temperatures had been measured in Iceland and Greenland he proposed that the difference between freezing and boiling should be divided into four parts and one part placed under the freezingpoint, the five parts should then be subdivided into twenty degrees with zero as the lowest point, then you would always read positive numbers. This is the principle behind the Kelvin scale, the way to divide from Rømer and Fahrenheit, the boiling point of water from Celsius and eighty degrees from freezing to boiling from Réaumur.
Horrebows scale was never published.
References:
(Adv.) Ole Rømers Adversaria, red. Thyra Eibe and Kirstine Meyer, Bianco Lunos Bogtrykkeri, Copenhagen 1910.
(Fl.) Fahrenheit's letters to Leibniz and Boerhaave, red. Pieter van der Star, Leiden 1983, he writes about his visit in a letter to Boerhaave dated April 17th 1729.

1) p. 119, Adv.
2) p. 202-203, Adv.
3) p. 210, Adv.
4) p. 214, Adv.
5) p. 171, Fl.
6) p. 11, Adv.
7) p. 171, Fl.
8) p. 163, Fl.
9) p. 161, Fl.
10) p. 210-211, Adv.

Analysis