When Robert Hooke laid down precepts for keeping a meteorological observation journal in the 1660s, few people in Britain or elsewhere had access to instruments for weather measurement. At that time, Hooke and others were first working out how to make reliable thermometers and give them uniform graduated scales. In the 1700s, thermometers and barometers became more widely available, and by the latter part of that century precision instruments of high quality could be obtained by those with the means to pay for them.
Yet in the new United States as Jefferson began his weather diary and encouraged others to do so, it could take some doing to get the equipment needed for weather journaling. When Jefferson advised James Madison to start a weather journal in 1784, Madison responded that he could not begin right away because he lacked “both a Thermometer and Barometer.” Writing in January of that year, Madison’s kinsman of the same name reported from the College of William and Mary in Williamsburg that when the British swept through three years earlier, they “robbed me of my Thermometer and Barometer.” Unable to get replacements yet, he had borrowed a thermometer and “I wish we had [a] Barometer but there is no Possibility of getting one here at present.”
The same limitation arose when Jefferson urged his son-in-law Thomas Mann Randolph, Jr., to keep a meteorological journal in Virginia in 1790. “I am sorry it is not in my power to begin immediately the course of observations you proposed in your last letter,” Randolph replied. “I have not a thermometer even, at present, but shall provide myself directly with one, and as soon as possible with a Barometer.” He felt the need to apologize again almost a year later: “I am sorry not to be able to give you a good account of the Diary you desired me to keep,” he wrote in April 1791. “I could only find one Thermometer in Richmond for Sale which was short, badly graduated and at an exorbitant price.” Randolph finally resorted to using “the old Spirit of Wine Thermometer in the Study” at Monticello. The best place to obtain scientific instruments in the United States was Philadelphia, where fine equipment was both imported and made. There in 1797, Jefferson paid Joseph Gatty $6 each for two thermometers, one for himself and the other for Randolph. Jefferson was fortunate to have the means and opportunity to purchase instruments in Philadelphia and, in the 1780s, in London and Paris.
The concept of measuring relative heat was not new in the late 18th century, but the scales quantifying the measurement were. The simultaneous development of air thermometers throughout Europe began in the early 17th century. Responding to the expansion of air when heated, these thermometers were usually not graduated into degrees and were little more than indicators of temperature change. As scientists began developing glass tube thermometers that employed spirits or mercury they applied scales to measure the change, but there was little unanimity about the reference points used to graduate heat into degrees. Most used the phase change of water to mark one or both ends of the scale, but the congealing of oils, the temperature of caves or cellars, body heat, and the freezing of spirits were also employed. By the time Jefferson began keeping his weather record, three scales were in common use, each based in some way on the phase change of water. Daniel Fahrenheit’s scale used the freezing point of a brine mixture at 32° as the low end and the 96° temperature of “a living man in good health” at the high end. The boiling point of water was not part of Fahrenheit’s scale, although he did determine it to be 212°, as he knew that it changed with atmospheric pressure. Anders Celsius and Carl Linnaeus simultaneously developed a centigrade scale that set melting ice at 0° and boiling water at 100°, and Rene-Antoine Ferchault de Reaumur’s scale put melting ice at 0° and boiling water at 80°.
In 1778, Jefferson informed Giovanni Fabbroni that “Fahrenheit’s thermometer is the only one in use with us.” Yet while in France in 1788, he also began employing the Reaumur scale, as it was the preferred system in that country. In December of 1788, he wrote to William Short that “the thermometer has been 10°. of Reaumur above freezing. This is 8°. of Farenheit above zero” and he acquired thermometers graduated in both scales. However, except for one brief period in November and December 1816, Jefferson only recorded Fahrenheit readings in his weather records.
The placement of thermometers for the correct measurement of air temperature was a matter of discussion in England. In continental Europe, a general consensus had been reached by the mid-17th century that meteorological thermometers should be placed on the north wall of a building, protected from exposure to the sun’s rays or the reflection of heat from other structures. It was not until the 1750s that English meteorologists consistently began to recommend hanging thermometers outside. In 1776, Sir Henry Cavendish reported that the Royal Society’s thermometer, which was hung out a window on the north wall “that it may be more exposed to the air,” was in the best position “the house afforded.” Jefferson left few clues to indicate where he hung his thermometers. The Reverend James Madison, when reporting on his methodology for collecting temperature data requested by Jefferson, stated that his thermometer “stands out all Night, or lies in an open window.” It was a borrowed instrument and he wished to prevent it from “meeting with some accident.” Jefferson’s instructions to Madison have not been found, but this report indicates that he wished the thermometer to be placed out-of-doors. In 1788, Jefferson requested a thermometer “intended to be hung on the outside of a glass window” from William Jones of London, and in 1790 he lamented to Thomas Mann Randolph that “there will not be one position to be had for the thermometer free from the influence of the sun both morning and evening” at the house he had rented in New York. At Monticello Jefferson evidently kept thermometers in the North Portico. In 1799, when the portico was “newly built” and still had an earthen floor, Jefferson “removed the 2 thermometres to approved positions” for fear that the earth had “artificially heated” the space (see Note on “Rejected” Temperatures, ca. June 1799). He may have returned an instrument to the Northeast Portico after the soil was no longer able to affect it. From 1802 to 1803, while living at the President’s House, Jefferson recorded temperatures in his bedroom in addition to his regular readings, another indication that his primary thermometer was placed outside the house (see Bedroom Temperatures, 1802–1804).
Of the many thermometers purchased by Jefferson, only one has been identified. Currently displayed at Monticello, it was made by John and William Jones of London and is most likely one of two Jefferson requested from William Jones in December 1788 with “Fahrenheit’s graduation on the right of the one side of the tube and Reaumur’s on the other.” He asked that it be “not less than 18. inches long, and graduated from boiling water down to the congelation of spirits of wine.” Jefferson paid £1 11s. 6d. for the thermometer. Described on the invoice as an “Inch Thermometer of the very best construction and finishing,” it was delivered to him in Paris in February 1789. It is a mercury thermometer with a brass scale in a wooden frame, 19 inches long by 2.25 inches wide (48.3 x 5.7 cm). The Fahrenheit scale, which is on the left, not the right, is graduated from -40° to 212° and the Reaumur scale on the right is graduated from -32° to 80°. Labels on the scale are “Freezing” at 32° F. (1° Reaumur), “Temperate” at 55° F., “Sumr Heat” at 72° F., “Blood Heat” at 98° F., “Fever Heat” at 112° F., “Spirits Boil” at 175° F., and “Water Boils” at 212° F. (80° Reaumur). At the top of the scale is “Jones & Son” and “Holborn London.” The other thermometer purchased from Jones at that time, described on the invoice as an “Inch Ditto on an Ivory scale,” was less expensive, costing 16 shillings.
Henry Cavendish, “An Account of the Meteorological Instruments Used at the Royal Society’s House,” Philosophical Transactions, 33 (1776), 375-6.
Hasok Chang, Inventing Temperature: Measurement and Scientific Progress (London, 2004), 9-56.
W. E. Knowles Middleton, The History of the Thermometer and Its Use in Meteorology (Baltimore, 1966).
Susan R. Stein, The Worlds of Thomas Jefferson at Monticello (New York, 1993), 363.
The use of the barometer in meteorology dates back to the 17th century. Evangelista Torricelli is credited with its invention when he inverted a mercury filled tube in a dish and observed the fluid level change with variations in the weather. (Incidentally this was also the first instance of a consistently maintained vacuum.) Rene Descartes applied a scale to the tube and Florin Perier demonstrated pressure changes at different elevations. Improvements to the instrument included refinements to make it portable and adjustments to magnify the scale and appear to make it more reactive to changes in atmospheric pressure. The familiar wheel barometer that links changes in pressure to the movement of a needle around a circle was such an instrument, developed by Robert Hooke around 1664. Each division of the circle equaled one hundredth of an inch, thus magnifying minute changes in pressure into significant movement of the needle. All barometers of this period required some correction for scale, capillarity, and temperature. Barometers were marked in inches, but in the 18th century there was no standard inch. Instruments made in England varied slightly from those made in France and Italy. Capillarity, or the tendency of a liquid to move in response to surface tension, results in a bowed surface at the top of a column of mercury, requiring some calculations to determine the exact height of the column. Barometers were also highly susceptible to temperature changes. Each instrument maker issued correction tables, “a necessary companion,” for determining the influence of temperature upon their particular instrument.
Although Hooke, Robert Boyle, and others were using barometers for weather observation by the 1660s, they did not have a clear idea of what those instruments were really measuring or what the readings meant with regard to the weather. What was air pressure, and was it affected by temperature, wind, or water vapor? These questions remained open for debate into the last decades of the 18th century. There was as yet no science of weather prediction drawing on precise readings. Why then did Jefferson make so many entries of barometric pressure readings in his weather journals? The answer is that barometric pressure, like air temperature, provides a datum point for a place and time, and Jefferson was interested in building sets of data for synchronous comparison of different locations. Like other students of what was called natural philosophy, he particularly wanted to explore the relationship of air pressure and elevation. To that end, in September 1776 he took readings at Monticello, the tobacco landing on the Rivanna River, the spring above the house, and the top of Mount Alto to calculate the elevation of each place. He also used barometrical readings to measure the height of the Peaks of Otter in the Blue Ridge Mountains. His goal was to learn as much as possible about relationships between physical geography and climate. For Jefferson, the barometer was a tool in that inquiry, not an implement for understanding the dynamics of weather.
Jefferson recorded purchasing a barometer for £4 10s. on 8 July 1776 from Philadelphia merchant John Sparhawk, but he did not describe the type of instrument he acquired. Barometrical observations appear in Jefferson’s records from 12 April 1778 to 18 December 1799, and all 2,670 readings were taken at Monticello. In 1784, Jefferson asked James Madison to include barometrical readings in his weather diary. In 1790, when describing his method of record keeping to Thomas Mann Randolph, he was less concerned with atmospheric pressure. He indicated that if Randolph should “chuse to take barometrical observations” he could insert the appropriate columns in his weather register.
No barometer owned or used by Jefferson has been identified.
W. E. Knowles Middleton, The History of the Barometer (Baltimore, 1964).
George Adams, A Short Dissertation on the Barometer, Thermometer, and other Meteorological Instruments (London, 1790), 3-17.
Jan Golinski, British Weather and the Climate of Enlightenment (Chicago, 2007), 112-36.
The effects of humidity on materials such as salt or wood and on atmospheric visibility were common folk methods of predicting the weather. Evaporation and condensation had been observed by scientists for centuries, but how humidity came to affect salt or visibility was not generally understood until the 17th century, when René Descartes postulated the presence of water molecules existing as vapor in the air. Robert Hooke designed the first practical hygrometer for measuring humidity around 1664. In the late 18th century, Jean André Deluc and Horace Bénédict de Saussure developed mechanical hygrometers that relied on the tendency of certain materials to change length when absorbing moisture from the air. Saussure used a human hair under tension in an instrument that measured its tautness on a scale of 1, very dry, to 100, saturated. Deluc did the same with a piece of whalebone (baleen), also marking his hygrometers in degrees with 100 being the number attained after soaking the whalebone in water. George Adams, a London instrument maker, in describing the two hygrometers declared Saussure’s “probably the best” but extremely difficult to manufacture, while Deluc’s was “less liable to accidents.” What Deluc and Saussure did not recognize was that their measurements represented relative humidity. They believed that they were measuring the absolute amount of vapor in the air. This discrepancy, in addition to the individuality of their scales, makes the readings produced by their hygrometers difficult to compare with those of modern instruments.
There are hygrometer readings in Jefferson’s records that he did not link to a specific type of hygrometer. The 1,259 readings, taken between 1 November 1806 and 3 March 1809 while Jefferson was at the President’s House in Washington, are all in the single digits and thus markedly different from readings that he recorded as from the Deluc and Saussure instruments. They resemble figures obtained when using an instrument now called a psychrometer. The psychrometer compares the temperature read from a thermometer with its bulb either dipped in water or covered with wet material to the temperature measured by a dry thermometer. Scientists had begun experimenting with the use of wet and dry bulb thermometers to measure humidity in the late 18th century. Dr. James Hutton of Scotland is credited with overcoming the “want of an accurate hygrometer” by using the temperature read on a wet bulb thermometer as a measure of the dryness of the air, the degree of cold “being proportional to the dryness of the air.” Shortly thereafter, John Leslie invented a hygrometer that combined the wet and dry bulb principle into one instrument. Whether Jefferson was using two separate thermometers or a single instrument is not known, although he seems to have thought of the apparatus as a hygrometer, for he headed the column in his record “Hygr.”
Although there is no extant hygrometer that has been attributed to Jefferson’s ownership, it is known that he possessed a Deluc whalebone hygrometer, purchased for him by Benjamin Vaughan from Nairne & Blunt of London in 1788. Nairne & Blunt also constructed a hygrometer for Jefferson from Vaughan’s design based on Benjamin Franklin’s observations on the effects of moisture on mahogany wood, which prompted Jefferson to make his own design for one using beechwood. In Paris on 1 September 1788 Jefferson paid Ferdinand Grand for acquiring a “de Saussure’s hygrometer.” Jefferson was particularly interested in comparing humidity across climates in his efforts to refute the Comte de Buffon’s assertion that the North American climate with its high humidity was responsible for the “degeneracy” of New World species.
“Letter to Mr. Nairne, of London, from Dr. Franklin, Proposing a Slowly Sensible Hygrometer for Certain Purposes,” Transactions of the American Philosophical Society, 2 (1786), 51-6.
W. E. Knowles Middleton, Invention of the Meteorological Instruments (Baltimore, 1969), 81-132.