An unusual Barothermometer

An unusual instrument, named “Barothermometer”, belonging to the collection of Villa Vigoni, on the Como Lake, Italy, is analysed. The instrument is composed of a U-shaped thermoscope and a spirit thermometer. Both instruments have their tubes placed side-by-side on the same frame and use the same scale. It measures the air temperature and the tendency of the barometric pressure to predict weather changes. A study is made to clarify the operating principle and produce the instructions for use. The development of the ideas to which the instrument has been inspired follows a tree starting with the Galileo’s thermoscope, the Amontons air thermometer and finally the Adie sympiesometer. The measuring method is typical of the first half of the 19th century, but the construction details in aluminium, the fonts used in the printed instructions and the colour fading suggest that the instrument was built around 1930-40.  


Introduction
Villa Vigoni, on the Como Lake, Italy, is seat of workshops and cultural activities promoting the Italian-German relations within a European context, and preserves the original furnishing and decorations, left by the founder Heinrich Mylius (1769 -1854) and later augmented by the Vigoni Family [Meda-Riquier et al., 2019]. The furniture is enriched by a library, paintings, statues and various pieces of art, and includes some technical devices, e.g. a small telescope, a watch aneroid barometer and an unusual instrument with name tag reporting the inscription «barotermometro» (i.e. barothermometer, in Italian) (Figure 1).
The name may suggest a similarity with other instruments, as follows.
(i) The traditional baro-thermometer is a combination of a barometer (e.g. J-shaped siphon barometer by Robert Hooke, or an aneroid capsule) and a metallic thermometer, or a Bourdon tube, mounted on the same frame ( Figure   2a), often associated with hair hygrometer and clock. They could be produced separately and miniaturized watchsize ( Figure 2b) to be carried in the pocket [Negretti and Zambra, 1864].
(ii) The air-baro-thermometer proposed by Schreiber [1875] consists of a balance barometer, an air thermometer to record the temperature of the external air, and an air thermometer to record the temperature of the recording instrument. The thermometers are composed of a tube, closed at the top and open below, filled with air, dipping into mercury, and balanced by a weight on a string passing over a pulley (Figure 2c).
(iii) The thermo-barometer, also called thermometrical barometer or hypsometer was invented by Wollaston [1817] and later improved by Regnault [1847]. It is a device to determine height measurements without a barometer.
It consists of a thermometer with a narrow tube, through which the smallest temperature changes can be observed with resolution of a thousandth of a degree (Figure 2d). This thermometer is equipped with a pot of water heated to boiling with a spirit lamp. The thermometer is surrounded by the hot steam, and the boiling point of the water at the unknown altitude will be read with high accuracy. The physical principle is that liquids boil when their saturation vapour pressure equals the pressure exerted by the atmosphere on the free liquid surface, and the temperature at which this occurs is called boiling point. If the atmospheric pressure is lowered, e.g. climbing to higher altitudes above the sea level, lower saturation vapor pressures will be needed, and the boiling point will occur at lower temperatures. The boiling point at the higher elevation will be used as an input to the Clausius-Clapeyron equation to calculate the unknown altitude [Camuffo, 2019]. The thermo-barometer is essentially a thermometer, used to calculate a pressure and, therefore, a site elevation, and this explains the name [Negretti and Zambra, 1864].

Dario Camuffo
2 Figure 1. The Barothermometer (digitally restored) indicating 18.5° C and stormy weather. Letters to recognize the parts of which the instrument is composed (see text).
The instrument in Villa Vigoni is none of the above-mentioned instruments. It is composed of a thermometer but does not include a traditional barometer, although it can detect pressure trends.
The aim of this paper is to recognize the measuring principle and the operation protocol of this particular instrument.

Parts of which the barothermometer is composed
The barothermometer (Figure 1) of Villa Vigoni is handcrafted, without a logo or producer name. It is composed of • a wooden frame (WF) • a blue spirit-in-glass thermometer (ST); • a U-shaped glass tube (UT) with an ampulla on the top of the left arm, while the right arm lies on the side of the thermometer tube. The upper end (UE) of the tube is shaped as a reversed bell with a spherical glass stopper on the top. The liquid inside is yellow, probably deriving from a discoloured red liquid.
• an aluminium shield (AS) to protect the ampulla from accidental bumps; • a thermometric scale (TS) from -10 to 50°C located behind the two tubes; • a mobile metal slit (MS) to adjust vertically the UT tube in correspondence of the elevation of the site. A black mark on the slit should be located at the corresponding elevation; 3 An unusual Barothermometer • an aluminium strip with indicated values from 0 to 700. This refers to the elevation of the site (SE), expressed in m above mean sea level; • a circular graduated dial (GD) with a hand-adjustable arrow fixed to a knob. There is an upper clockwise scale from 0 to 10, and a lower counter-clockwise scale also from 0 to 10. Units are not specified, and "+" or "-" signs are missing. To avoid confusion, in this context the upper scale will be indicated with positive values, and the lower with negative. The following headings (in Italian) are reported: -2 to +2: «variable» (traditionally used in Italian barometers for «change»); +3 to +5 «fair»; +6 to +10 «stable» (traditionally used in Italian barometers for «very dry»); on the lower side: -3 to -6 «rain»; -6 to -10 «stormy». Some additional words give a friendly explanation. On the upper side: «red below the blue colour»; on the lower side: «blue above the red colour».
• an instruction table (IT) in aluminium, entitled «Weather Forecast» (Figure 3a). It reports examples of the weather classes deduced from the relative heights of two bars, coloured (faded) red and blue, with the following headings. Blue bar much higher than red: «stable»; blue bar higher than red: «fair -clear»; blue and red bars at the same level: «variable»; blue bar lower than red: «rain or snow»; blue bar much lower than red: «stormy». In the bottom, three notes, i.e. (i) «when the red bar falls below the blue, weather is getting better»; (ii) «when the red bar rises above the blue, weather is getting worse»; (iii) «the blue bar represents [temperature] in centigrade degrees».
• an instruction paper sheet is glued on the back of the frame (Figure 4), with a few notes, as follows. Install the instrument far from heat sources. When the instrument is installed, remove the cork from the upper end of the tube and substitute it with the glass closure with spherical handle. Adjust the metal slit of the tube in correspondence of the elevation of the site. Take advantage of the instruction table and the circular graduated dial.
The used language indicates that the instrument was built in Italy. The colour fading occurred to the liquid in UT and the IT table (in Figure 1 the faded red colour has been digitally restored) indicates that the instrument is aged. The use of aluminium in various parts suggests a building date after 1930. The instruction sheet glued on the back has the title in Birch font that was created in 1879 but was popular in the first half of the 20 th century; the body uses a style that was very popular in Italy around 1930s-40s, i.e. typefaces sans serif inspired to Art Deco, e.g. capital G, M, N, P and R.

Operating principle of the Barothermometer
A premise should be done about the limitations concerning the instrument inspection. In order to guarantee the safety of this vulnerable instrument, the study has been limited to photographic and visual analysis, avoiding any physical contact and making impossible the chemical analysis of the yellow liquid in UT, as well as the efficiency of the system to equalize the atmospheric pressure with the deliberate leakage around the ground glass stopper on the top of the tube (UE). However, this has not been essential to discover the physical principle on which the instrument is based.
From the above description, it is clear that the instrument is composed of a traditional spirit thermometer (ST) and an unknown device (i.e. UT) to measure the atmospheric pressure ( Figure 5). This means that the UT glassware is in connection with the external air, and the part devoted to this aim is the upper end of the tube (UE), with a permanent small opening on the glass stopper, or a temporary one every time an observation is made.
A tube with permanent opening would require the use of non-volatile liquids, because the evaporation would cause a drift to the system. To this aim, the most popular liquids were mercury or oil. Mercury should be excluded for the yellow colour and transparency. Linseed oil is unlikely after the negative experience of the Newton linseed oil thermometer [Newton, 1701] because it sticks on the internal side of the capillary, making difficult or even impossible readings after a certain period of time [Camuffo and della Valle 2017]. Almond oil is the most likely candidate, and was used in the Adie [1819] sympiesometer. It leaves clean the interior of the glass tube, and may be coloured with a red dye (e.g. dragon blood dracena draco resin) but fades over time, even in indoor conditions, for the discolouring effect of light. As opposed, if a volatile liquid is used, e.g. wine-spirit, it is necessary to hinder evaporation with a hermetic glass stopper, and remove it for a very short time before an observation is made, to equalize the pressure. However, the glued sheet does not include this recommendation. In addition, it explains that the instrument had a cork for the transport, that had to be substituted with a glass stopper sent with the instrument.
This implies that this stopper allows some leakage and the liquid is non-volatile, very likely almond oil. The ancestor of the UT device is the Galileo thermoscope (Figure 6a) built in 1593 [Middleton, 1966;Beaurepaire, 1994;1995;Camuffo and Bertolin, 2012;Benicasa et al., 2019;Camuffo, 2019]. The thermoscope consisted of an ampulla with an air pocket and the pressure exerted by the air pocket moves the liquid column in the tube connected with the ampulla. The other end of the tube is open, to allow the free liquid motion, and is exposed to the atmospheric pressure. In the ampulla, the volume (V) of the air pocket has a direct relationship with the air temperature (T) and an opposite one with the atmospheric pressure (P), i.e. V will increase at increasing T or decreasing P, and vice-versa. The thermoscope reading is given by the change ΔH of the height H of the liquid column, which responds to changes in temperature ΔT and pressure ΔP according to the formula

An unusual Barothermometer
where a 1 and a 2 are proportionality coefficients, characteristic of each individual instrument.

Dario Camuffo
6 Figure 5. Scheme of the assembly constituted by the U-shaped device (UT) and the spirit thermometer (ST).
As the thermoscope is sensitive to both T and P variables, it behaves like a thermometer when P is constant (i.e. = 0), and like a barometer when T is constant (i.e. = 0). In general, however, both T and P are variable and the readings are difficult to interpret. The thermoscope was applied to a variety of uses, but with poor success, and was abandoned after the liquid-in-glass thermometer was invented in Florence in 1641 [Magalotti, 1666].
The thermoscope was improperly called air-thermometer, because the bulb was composed of an air pocket. However, the opposite end of the tube was open and the equilibrium level of the liquid column was determined by the combined effect of the air temperature and the atmospheric pressure. Therefore, it was also called thermo-baro-scope.
At the beginning of the 18 th century, Amontons [1702] found a solution to the problem of the twofold dependence of thermoscopes from T and P. He built a thermoscope ( Figure 6b) filled with mercury, with the same size of his barometer. Having clear in mind equation (1), he thought that the real T could be obtained by summing the heights of two mercury columns read on the thermoscope (H) and the barometer (P), i.e.
The Amontons thermometer was popular in the first decades of the 18 th century. Later, it was abandoned for a number of problems connected with this instrument, e.g. the observer needed to have a thermometer and a barometer, and to sum the two readings; there was not a well defined scale; the thermometer was not linear; it could not be moved for spot measurements; the moisture in the air pocket could condense for the pressure and/or temperature [Camuffo and Jones, 2002;Camuffo et al., 2016] [Camuffo and Jones, 2002].
An opposite approach was the sympiesometer (Figure 6c) invented by Adie [1819], i.e. a marine barometer inspired by the Amontons thermometer, but finalized to obtain P instead of T. Amontons used a thermoscope and corrected for P; Adie used the same methodology, but corrected for T. The sympiesometer is the combination of a thermoscope and a thermometer. This thermoscope had a complex construction to be adapted to the marine use and smooth the adverse effects of yaw, pitch and roll of the ship. Adie used a phial with hydrogen instead of air and almond oil instead of mercury to increase the viscous friction and reduce the free oscillations of the liquid.
A free volume is located on the upper end of the tube to act as a reservoir and keep the liquid in case of violent oscillations or tropical temperatures. This volume has a hole on the top, to allow pressure equalization. The instrument responded to both T and P, and the unwanted T contribution was removed using a traditional mercury thermometer that was fixed to the same frame.
The baro-thermometer patented by Bodeur [1838] (Figure 6d) is similar to the Adie sympiesometer, equally based on the comparison of the readings taken on a thermoscope and a thermometer. Two differences may be found.
Instead of the linear scale, it has a circular graduated dial to calculate the pressure changes and recognize the weather types. Instead of almond oil, mercury was used. In the next two decades, Bunten 1 [Alluard, 1847;Pionnier, 1844;Pouillet, 1856] (Figure 6e), Silbermann and Gaudin [Daguin, 1861] built other very similar instruments, with some minor changes to improve the scale readability.
The Villa Vigoni barothermometer includes a thermoscope and a thermometer, and its aim is the same as the sympiesometer by Adie, Bodeur, Bunten, Silbermann and Gaudin, but simplified. The scale of the thermoscope is neither mobile, not calibrated. It cannot obtain accurate barometric values. It is only aimed to point out general pressure tendencies, i.e. increasing pressure and fine weather or falling pressure and bad weather. Possibly, the Italian manufacturer was inspired by the description reported by Pagliani [1891], with text and drawing close to Daguin [1861], but written in Italian.
An overview of the operating principles and key features of the above instruments are summarized in