THERMOMETER

, an instrument for measuring the temperature of the air, &c, as to heat and cold.

The Thermometer and thermoscope are usually accounted the same thing. But Wolfius makes a difference; and he also shews that what we call Thermometers, are really no more than thermoscopes.

The invention of the Thermometer is attributed to several persons by different authors, viz, to Sanctorio, Galileo, father Paul, and to Drebbel. Thus, the invention is ascribed to Cornelius Drebbel of Alcmar, about the beginning of the 17th century, by his countrymen Boerhaave (Chem. 1, pa. 152, 156), and Musschenbroeck (Introd. ad Phil. Nat. vol. 2, pa. 625).— Fulgenzio, in his Life of father Paul, gives him the honour of the first discovery.—Vincenzio Viviani (Vit. de l'Galil. pa. 67; also Oper. di Galil. pref. pa. 47) speaks of Galileo as the inventor of Thermometers.— But Sanctorino (Com. in Galen. Art. Med. pa. 736, 842, Com. in Avicen. Can. Fen. 1, pa. 22, 78, 219) expressly assumes to himself this invention: and Borelli (De Mot. Animal. 2, prop. 175) and Malpighi (Oper. Posth. pa. 30) ascribe it to him without reserve. Upon which Dr. Martine remarks, that these Florentine academicians are not to be suspected of partiality in favour of one of the Patavinian school.

But whoever was the first inventor of this instrument, it was at first very rude and imperfect; and as the various degrees of heat were indicated by the different contraction or expansion of air, it was afterwards found to be an uncertain and sometimes a deceiving measure of heat, because the bulk of the air was affected, not only by the difference of heat, but also by the variable weight of the atmosphere.

There are various kinds of Thermometers, the construction, defects, theory, &c, of which, are as follow.

The Air Thermometer.—This instrument depends on the rarefaction of the air. It consists of a glass tube BE (fig. 1, pl. 34) connected at one end with a large glass ball A, and at the other end immersed in an open vessel, or terminating in a bait DE, with a narrow orifice at D; which vessel, or ball, contains any coloured liquor that will not easily freeze. Aquafortis tinged of a fine blue colour with solution of vitriol or copper, or spirit of wine tinged with cochineal, will answer this purpose. But the ball A must be first moderately warmed, so that a part of the air contained in it may be expelled through the orifice D; and then the liquor pressed by the weight of the atmosphere, will enter the ball DE, and rise, for example, to the middle of the tube at C, at a mean temperature of the weather; and in this state the liquor by its weight, and the air included in the ball and tube ABC, by its elasticity, will counterbalance the weight of the atmosphere. As the surrounding air becomes warmer, the air in the ball and the upper part of the tube, expanding by heat, will drive the liquor into the lower ball, and consequently its surface will descend; on the contrary, as the ambient air becomes colder, that in the ball is condensed, and the liquor, pressed by the weight of the atmosphere, will ascend: so that the liquor in the tube will ascend or descend more or less, according to the state of the air contiguous to the instrument. To the tube is affixed a scale of the same length, divided upwards and downwards, from the middle C, into 100 equal parts, by means of which may be observed the ascent and descent of the liquor in the tube, and consequently the variations also in the temperature of the atmosphere.

A similar Thermometer may be constructed by putting a small quantity of mercury, not exceeding the bulk of a pea, into the tube BC (fig. 4, pl. 33), bent into wreaths, that taking up the less height, it may be the more manageable, and less liable to harm; divide this tube into any number of equal parts to serve for a scale. Here the approaches of the mercury towards the ball A will shew the increase of the degree of heat. The reason of which is the same as in the former.

The defect of both these instruments consists in this, that they are liable to be acted on by a double cause: for, not only a decrease of heat, but also an increase of weight of the atmosphere, will make the liquor rise in the one, and the mercury in the other; and, on the contrary, either an increase of heat, or decrease of the weight of the atmosphere, will cause them to descend.

For these, and other reasons, Thermometers of this kind have been long disused. However, M. Amontons, in 1702, with a view of perfecting the aërial Thermometer, contrived his Universal Thermometer. Finding that the changes produced by heat and cold in the bulk of the air were subject to invincible irregularities, he substituted for these the variations produced by heat in the elastic force of this fluid. This Thermometer consisted of a long tube of glass (fig. 3, pl. 34) open at one end, and recurved at the other end, which terminated in a ball. A certain quantity of air was compressed into this ball by the weight of a column of mercury, and also by the weight of the atmosphere. The effect of heat on this included air was to make it sustain a greater or less weight; and this effect was measured by the variation of the column of mercury in the tube, corrected by that of the barometer, with respect to the changes of the weight of the external air. This instrument, though much more perfect than the former, is nevertheless subject to very considerable defects and | inconveniences. Its length of 4 feet renders it unfit for a variety of experiments, and its construction is difficult and complex: it is extremely inconvenient for carriage, as a very small inclination of the tube would suffer the included air to escape: and the friction of the mercury in the tube, and the compressibility of the air, contribute to render the indications of this instrument extremely uncertain. Besides, the dilatation of the air is not so regularly proportional to its heat, nor is its dilatation by a given heat nearly so uniform as he supposed. This depends much on its moisture; for dry air does not expand near so much by a given heat, as air stored with watery particles. For these, and other reasons, enumerated by De Luc (Recherches sur les Mod. de l'Atmo. tom. 1, pa. 278 &c), this instrument was imitated by very few, and never came into general use.

Of the Florentine Thermometer.—The academists del Cimento, about the middle of the 17th century, considering the inconveniencies of the air Thermometers above described, attempted another, that should measure heat and cold by the rarefaction and condensation of spirit of wine; though much less than those of air, and consequently the alterations in the degree of heat likely to be much less sensible.

The spirit of wine coloured, was included in a very fine and cylindrical glass tube (fig. 2, pl. 34), exhausted of its air, having a hollow ball at one end A, and hermetically sealed at the other end D. The ball and tube are filled with rectified spirit of wine to a convenient height, as to C, when the weather is of a mean temperature, which may be done by inverting the tube into a vessel of stagnant coloured spirit, under a receiver of the air-pump, or in any other way. When the thermometer is properly filled, the end D is heated red hot by a lamp, and then hermetically sealed, leaving the included air of about 1/3 of its natural density, to prevent the air which is in the spirit from dividing it in its expansion. To the tube is applied a scale, divided from the middle, into 100 equal parts, upwards and downwards.

Now spirit of wine rarefying and condensing very considerably; as the heat of the ambient atmosphere increases, the spirit will dilate, and so ascend in the tube; and as the heat decreases, the spirit will descend; and the degree or quantity of the motion will be shewn by the attached scale.

These Thermometers could not be subject to any inconvenience by an evaporation of the liquor, or a variable gravity of the incumbent atmosphere. Instruments of this kind were first introduced into England by Mr. Boyle, and they soon came into general use among philosophers in other countries. They are however subject to considerable inconveniences, from the weight of the liquor itself, and from the elasticity of the air above it in the tube, both which prevent the freedom of its ascent; besides, the rarefactions are not exactly proportional to the surrounding heat. Moreover spirit of wine is incapable of bearing very great heat or very great cold: it boils sooner than any other liquor; and therefore the degrees of heat of boiling fluids cannot be determined by this Thermometer. And though it retains its fluidity in pretty severe cold, yet it seems not to condense very regularly in them: and at Torneao, near the polar circle, the winter cold was so severe, as Maupertuis informs us, that the spirits were frozen in all their Thermometers. So that the degrees of heat and cold, which spirit of wine is capable of indicating, is much too limited to be of very great or general use.

Another great defect of these, and other Thermometers, is, that their degrees cannot be compared with each other. It is true they mark the variations of heat and cold; but each marks for itself, and after its own manner; because they do not proceed from any point of temperature that is common to all of them.

From these and various other imperfections in these Thermometers, it happens, that the comparisons of them become so precarious and defective: and yet the most curious and interesting use of them, is what ought to arise from such comparison. It is by this we should know the heat or cold of another season, of another year, another climate, &c; and what is the greatest degree of heat or cold that men and other animals can subsist in.

Reaumur contrived a new Thermometer, in which the inconveniences of the former are proposed to be remedied. He took a large ball and tube, the content or dimensions of which are known in every part; he graduated the tube, so that the space from one division to another might contain a 1000th part of the liquor, which liquor would contain 1000 parts when it stood at the freezing point: then putting the ball of his Thermometer and part of the tube into boiling water, he observed whether it rose 80 divisions: if it exceeded these, he changed his liquor, and by adding water lowered it, till upon trial it should just rise 80 divisions; or if the liquor, being too low, fell short of 80 divisions, he raised it by adding rectified spirit to it. The liquor thus prepared suited his purpose, and served for making a Thermometer of any size, whose scale would agree with his standard. Such liquor, or spirits, being about the strength of common brandy, may easily be had any where, or made of a proper degree of density by raising or lowering it.

The abbé Nollet made many excellent Thermometers upon Reaumur's principle. Dr. Martine however expresses his apprehensions that Thermometers of this kind cannot admit of such accuracy as might be wished. The balls or bulbs, being large, as 3 or 4 inches in diameter, are neither heated nor cooled soon enough to shew the variations of heat. Small bulbs and small tubes, he says, are much more convenient, and may be constructed with sufficient accuracy. Though it must be allowed that Reaumur, by his excellent scale, and by depriving the spirit of its air, and expelling the air by means of heat from the ball and tube of his Thermometer, has brought it to as much perfection as may be; yet it is liable to some of the inconveniences of spirit Thermometers, and is much inferior to mercurial ones. These two kinds do not agree together in indicating the same degrees of intense cold; for when the mercury has stood at 22° below 0, the spirit indicated only 18°, and when the mercury stood at 28° or 37° below 0, the spirit rested at 25° or 29°. See the description of Reaumur's Thermometer at large in Mem. de l'Acad. dés Scienc. an. 1730, pa. 645, Hist. pa. 15. Ib. an. 1731, pa. 354, Hist. pa. 7. |

Mercurial Thermometer.—It is a most important circumstance in the construction of Thermometers, to procure a fluid that measures equal variations of heat by corresponding equal variations in its own bulk: and the fluid which possesses this essential requisite in the most perfect degree, is mercury: the variations in its bulk approaching nearer to a proportion with the corresponding variations of its heat, than any other fluid. Besides, it is the most easy to purge of its air; and is also the most proper for measuring very considerable variations of heat and cold, as it will bear more cold before freezing, and more heat before boiling, than any other fluid. Mercury is also more sensible than any other fluid, air excepted, or conforms more speedily to the several variations of heat. Moreover, as mercury is an homogeneous fluid, it will in every Thermometer exhibit the same dilatation or condensation by the same variations of heat.

Dr. Halley, though apprized only of some of the remarkable properties of mercury above recited, seems to have been the first who suggested the application of this fluid to the construction of Thermometers. Philos. Trans. Abr. vol. 2, pa. 34.

Boerhaave (Chem. 1, pa. 720) says, these mercurial Thermometers were first contrived by Olaus Roemer; but the claims of Fahrenheit of Amsterdam, who gave an account of his invention to the Royal Society in 1724, (Philos. Trans. num. 381, or Abr. vol. 7, pa. 49) have been generally allowed. And though Prius and others, in England, Holland, France, and other countries, have made this instrument as well as Fahrenheit, most of the mercurial Thermometers are graduated according to his scale, and are called Fahrenheit's Thermometers.

The cone or cylinder, which these Thermometers are often made with, instead of the ball, is made of glass of a moderate thickness, lest, when the exhausted tube is hermetically sealed, its internal capacity should be diminished by the weight of the ambient atmosphere. When the mercury is thoroughly purged of its air and moisture by boiling, the Thermometer is filled with a sufficient quantity of it; and before the tube is hermetically sealed, the air is wholly expelled from it by heating the mercury, so that it may be rarefied and ascend to the top of the tube. To the side of the tube is annexed a scale (fig. 3, pl. 34), which Fahrenheit divided into 600 parts, beginning with that of the severe cold which he had observed in Iceland in 1709, or that produced by surrounding the bulb of the Thermometer with a mixture of snow or beaten ice and sal ammoniac or sea salt. This he apprehended to be the greatest degree of cold, and accordingly he marked this, as the beginning of his scale, with 0; the point at which mercury begins to boil, he conceived to shew the greatest degree of heat, and this he made the limit of his scale. The distance between these two points he divided into 600 equal parts or degrees; and by trials he found at the freezing point, when water just begins to freeze, or snow or ice just begins to thaw, that the mercury stood at 32 of these divisions, therefore called the degree of the freezing point; and when the tube was immersed in boiling water, the mercury rose to 212, which therefore is the boiling point, and is just 180 degrees above the former or freezing point. But the present method of making the scale of these Thermometers, which is the sort in most common use, is first to immerge the bulb of the Thermometer in ice or snow just beginning to thaw, and mark the place where the mercury stands with a 32; then immerge it in boiling water, and again mark the place where the mercury stands in the tube, which mark with the num. 212, exceeding the former by 180; dividing therefore the intermediate space into 180 equal parts, will give the scale of the Thermometer, and which may afterwards be continued upwards and downwards at pleasure.

Other Thermometers of a similar construction have been accommodated to common use, having but a portion of the above scale. They have been made of a small size and portable form, and adapted with appendages to particular purposes; and the tube with its annexed scale has often been enclosed in another thicker glass tube, also hermetically sealed, to preserve the Thermometer from injury. And all these are called Fahrenheit's Thermometers.

In 1733, M. De l'Isle of Petersburgh constructed a mercurial Thermometer (see fig. 3, pl. 34), on the principles of Reaumur's spirit Thermometer. In his Thermometer, the whole bulk of quicksilver, when immerged in boiling water, is conceived to be divided into 100,000 parts; and from this one fixed point the various degrees of heat, either above or below it, are marked in these parts on the tube or scale, by the various expansion or contraction of the quicksilver in all imaginable varieties of heat.—Dr. Martine apprehends it would have been better if De l'Isle had made the integer 100,000 parts, or fixed point, at freezing water, and from thence computed the dilatations or condensations of the quicksilver in those parts; as all the common observations of the weather, &c, would have been expressed by numbers increasing as the heat increased, instead of decreasing, or counting the contrary way. However, in practice it will not be very easy to determine exactly all the divisions from the alteration of the bulk of the contained fluid. And besides, as glass itself is dilated by heat, though in a less proportion than quicksilver, it is only the excess of the dilatation of the contained fluid above that of the glass that is observed; and therefore if different kinds of glass be differently affected by a given degree of heat, this will make a seeming difference in the dilatations of the quicksilver in the Thermometers constructed in the Newtonian method, either by Reaumur's rules or De l'Isle's. Accordingly it has been found, that the quicksilver in De l'Isle's Thermometers, has stood at different degrees of the scale when immerged in thawing snow: having stood in some at 154°, while in others it has been at 156 or even 158°.

Metallic Thermometer.—This is a name given to a machine composed of two metals, which, whilst it indicates the variations of heat, serves to correct the errors hence resulting in the going of pendulum clocks and watches. Instruments of this kind have been contrived by Graham, Le Roy, Ellicot, Harrison, and other eminent artificers. See the Philos. Traus. vol. 44, pa. 689, and vol. 45, pa. 129, and vol. 51, pa. 823, where the particular descriptions &c may be seen.

M. De Luc has likewise described two Thermometers | of metal, which he uses for correcting the effects of heat upon a barometer, and an hygrometer of his construction connected with them. See Philos. Trans. vol. 68, p. 437.

Oil Thermometers.—To this class belongs Newton's Thermometer, constructed in 1701, with linseed oil, instead of spirit of wine. This fluid has the advantage of being sufficiently homogeneous, and capable of a considerable rarefaction, not less than 15 times greater than that of spirit of wine. It has not been observed to freeze even in very great colds; and it sustains a great heat, about 4 times that of water, before it boils. With these advantages it was made use of by Sir I. Newton, who discovered by it the comparative degree of heat for boiling water, melting wax, boiling spirit of wine, and melting tin; beyond which it does not appear that this Thermometer was applied. The method he used for adjusting the scale of this oil Thermometer, was as follows: supposing the bulb, when immerged in thawing snow, to contain 10,000 parts, he found the oil expanded by the heat of the human body so as to take up a 39th more space, or 10256 such parts; and by the heat of water boiling strongly 10725; and by the heat of melting tin 11516. So that, reckoning the freezing point as a common limit between heat and cold, he began his scale there, marking it 0, and the heat of the human body he made 12°; and consequently, the degrees of heat being proportional to the degrees of rarefaction, or 256 : 725 :: 12 : 34, this number 34 will express the heat of boiling water; and, by the same rule, 72 that of melting tin. Philos. Trans. number 270, or Abridg. vol. 4, par. 2, p. 3.

There is an insuperable inconvenience attending all Thermometers made with oil, or any other viscid fluid, viz, that such liquor adheres too much to the sides of the tube, and so inevitably disturbs the regularity and uniformity of the Thermometer.

Of the fixed points of Thermometers.—Various methods have been proposed by different authors, for finding a fixed point or degree of heat, from which to reckon the other degrees, and adjust the scale; so that different observations and instruments might be compared together. Mr. Boyle was very sensible of the inconveniences arising from the want of a universal scale and mode of graduation; and he proposed either the freezing of the essential oil of aniseeds, or of distilled water, as a term to begin the numbers at, and from thence to graduate them according to the proportional dilatations or contractions of the included spirits.

Dr. Halley (Philos. Trans. Abr. vol. 2, p. 36) seems 10 have been fully apprized of the bad effects of the indefinite method of constructing Thermometers, and wished to have them adjusted to some determined points. What he seems to prefer, for this purpose, is the degree of temperature found in subterranean places, where the heat in summer or cold in winter appears to have no influence. But this degree of temperature, Dr. Martine shews, is a term for the universal construction of Thermometers, both inconvenient and precarious, as it cannot be easily ascertained, and as the difference of soils and depths may occasion a considerable variation. Another term of heat, which he thought might be of use in a general graduation of Thermometers, is that of boiling spirit of wine that has been highly rectified.

The first trace that occurs of the method of actually applying fixed points or terms to the Thermometer, and of graduating it, so that the unequal divisions of it might correspond to equal degrees of heat, is the project of Renaldini, professor at Padua, in 1694: it is thus described in the Acta Erud. Lips. “Take a slender tube, about 4 palms long, with a ball fastened to the same; pour into it spirit of wine, enough just to fill the ball, when surrounded with ice, and not a drop over: in this state seal the orifice of the tube hermetically, and provide 12 vessels, each capable of containing a pound of water, and somewhat more; and into the first pour 11 ounces of cold water, into the second 10 ounces, into the third 9, &c; this done, immerge the Thermometer in the first vessel, and pour into it one ounce of hot water, observing how high the spirit rises in the tube, and noting the point with unity; then remove the Thermometer into the second vessel, into which are to be poured 2 ounces of hot water, and note the place the spirit rises to with 2: by thus proceeding till the whole pound of water is spent, the instrument will be found divided into 12 parts, denoting so many terms or degrees of heat; so that at 2 the heat is double to that at 1, at 3 triple, &c.”

But this method, though plausible, Wolsius shews, is deceitful, and built on false suppositions; for it takes for granted, that we have one degree of heat, by adding one ounce of hot water to 11 of cold; two degrees by adding 2 ounces to 10, &c: it supposes also, that a single degree of heat acts on the spirit of wine, in the ball, with a single force; a double with a double force, &c: lastly it supposes, that if the effect be produced in the Thermometer by the heat of the ambient air, which is here produced by the hot water, the air has the same degree of heat with the water.

Soon after this project of Renaldini, viz, in 1701, Newton constructed his oil Thermometer, and placed the base or lowest fixed point of his scale at the temperature of thawing snow, and 12 at that of the human body, &c, as above explained.—De Luc observes, that the 2d term of this scale should have been at a greater distance from the first, and that the heat of boiling water would have answered the purpose better than that of the human body.

In 1702, Amontons contrived his universal Thermometer, the scale of which was graduated in the foling manner. He chose for the first term, the weight that counterbalanced the air included in his Thermometer, when it was heated by boiling water: and in this state he so adjusted the quantity of mercury contained in it, till the sum of its height in the tube, and of its height in the barometer at the moment of observation, was equal to 73 inches. Fixing this number at the point to which the mercury in the tube rose by plunging it in boiling water, it is evident that if the barometer at this time was at 28 inches, the height of the column of mercury in the Thermometer, above the level of that in the ball, was 45 inches; but if the height of the barometer was less by a certain quantity, the column of the Thermometer ought to be greater by the same quantity, and reciprocally. He formed his scale on the supposition, that the weight of the atmosphere was always equal to that of a column of mercury of 28 inches, and he divided it into inches | from the point 73 downward, marking the divisions with 72, 71, 70, &c, and subdividing the inches into lines. But as the weight of the atmosphere is variable, the barometer must be observed at the same time with the Thermometer, that the number indicated by this last instrument may be properly corrected, by adding or subtracting the quantity which the mercury is below or above 28 inches in the barometer. In this scale then, the freezing point is at 51 1/2 inches, corresponding to 32 degrees of Fahrenheit, and the heat of boiling water at 73 inches, answering to 212 of Fahrenheit's; and thus they may be easily compared together.

The fixed points of Fahrenheit's Thermometer, as has been already observed, are the congelation produced by sal ammoniac and the heat of boiling water. The interval between these points is divided into 212 equal parts; the former of these points being marked 0, and the other 212.

Reaumur in his Thermometer, the construction of which he published in 1730, begins his scale at an artificial congelation of water in warm weather, which, as he uses large bulbs for his glasses, gives the freezing point much higher than it should be, and at boiling water he marks 80 degrees, which point Dr. Martine thinks is more vague and uncertain than his freezing point. In order to determine the correspondence of his scale with that of Fahrenheit, it is to be considered that his boiling water heat, is really only the boiling heat of weakened spirit of wine, coinciding nearly, as Dr. Martine apprehends, with Fahrenheit's 180 degrees. And as his 10 1/4 degrees is the constant heat of the cave of the observatory at Paris, or Fahrenheit's 53°, he thence finds his freezing point, instead of answering just to 32°, to be somewhat above 34°.

De l'Isle's Thermometer, an account of which he presented to the Petersburgh Academy in 1733, has only one fixed point, which is the heat of boiling water, and, contrary to the common order, the several degrees are marked from this point downward, according to the condensations of the contained quicksilver, and consequently by numbers increasing as the heat decreases. The freezing point of De l'Isle's scale, Dr. Martine makes near to his 150°, corresponding to Fahrenheit's 32, by means of which they may be compared; but Ducrest says, that this point ought to be marked at least at 154°.

Ducrest, in his spirit Thermometer, constructed in 1740, made use of two fixed points; the first, or 0, indicated the temperature of the earth, and was marked on his scale in the cave of the Paris Observatory; and the other was the heat of boiling water, which that spirit in his Thermometer was made to endure, by leaving the upper part of the tube full of air. He divided the interval between these points into 100 equal parts; calling the divisions upward, degrees of heat, and those below 0, degrees of cold.—It is said that he has since regulated his Thermometer by the degree of cold indicated by melting ice, which he found to be 10 2/5.

The Florentine Thermometers were of two sorts. In one sort the freezing point, determined by the degree at which the spirit stood in the ordinary cold of ice or snow (probably in a thawing state) and coinciding with 32° of Fahrenheit, fell at 20°; and in the other sort at 13 1/2. And the natural heat of the viscera of cows and deer, &c, raised the spirit in the latter, or less sort, to about 40°, coinciding with their summer heat, and nearly with 102° in Fahrenheit's; and in their other or long Thermometer, the spirit, when exposed to the great midsummer heat in their country, rose to the point at which they marked 80°.

In the Thermometer of the Paris Observatory, made of spirit of wine by De la Hire, the spirit always stands at 48° in the cave of the observatory, corresponding to 53 degrees in Fahrenheit's; and his 28° corresponded with 51 inches 6 lines in Amontons' Thermometer, and consequently with the freezing point, or 32° of Fahrenheit's.

In Poleni's Thermometer, made after the manner of Amontons', but with less mercury, 47 inches corresponded, according to Dr. Martine, with 51 in that of Amontons, and 53 with 59 1/2.

In the standard Thermometer of the Royal Society of London, according to which Thermometers were for a long time constructed in England, Dr. Martine found that 34 1/2 degrees answered to 64° in Fahrenheit, and 0 to 89.

In the Thermometers graduated for adjusting the degrees of heat proper for exotic plants, &c, in stoves and greenhouses, the middle temperature of the air is marked at 0, and the degrees of heat and cold are numbered both above and below. Many of these are made on no regular and fixed principles. But in that formerly much used, called Fowler's regulator, the spirit fell, in melting snow, to about 34° under 0; and Dr. Martine found that his 16° above 0, nearly coincided with 64° of Fahrenheit.

Dr. Hales (Statical Essays, vol. 1, p. 58), in his Thermometer, made with spirit of wine, and used in experiments on vegetation, began his scale with the lowest degree of freezing, or 32° of Fahrenheit, and carried it up to 100°, which he marked where the spirit stood when the ball was heated in hot water, upon which some wax floating first began to coagulate, and this point Dr. Martine found to correspond with 142° of Fahrenheit. But by experience it is found that Hales's 100 falls considerably above our 142.

In the Edinburgh Thermometer, made with spirit of wine, and used in the meteorological observations published in the Medical Essays, the scale is divided into inches and tenths. In melting snow the spirit stood at 8 2/10, and the heat of the human skin raised it to 22 2/10. Dr. Martine found that the heat of the person who graduated it, was 97 of Fahrenheit.

As it is often of use to compare different Thermometers, in order to judge of the result of former observations, I have annexed from Dr. Martine's Essays, the table by which he compared 15 different thermometers. See Plate 34, fig. 3.

There is a Thermometer which has often been used in London, called the Thermometer of Lyons, because | M. Cristin brought it there into use, which is made of mercury: the freezing point is marked 0, and the interval from that point to the heat of boiling water is divided into 100 equal degrees.

From the abstract of the history of the construction of Thermometers it appears, that freezing and boiling water have furnished the distinguishing points that have been marked upon almost all Thermometers. The inferior fixed point is that of freezing, which some have determined by the freezing of water, and others by the melting of ice, plunging the ball of the Thermometer into the water and ice, while melting, which is the best way. The superior fixed point of almost all Thermometers, is the heat of boiling water. But this point cannot be considered as fixed and certain, unless the heat be produced by the same degree of boiling, and under the same weight of the atmosphere; for it is found that the higher the barometer, or the heavier the atmosphere, the greater is the heat when the water boils. It is now agreed therefore that the operation of plunging the ball of the Thermometer in the boiling water, or suspending it in the steam of the same in an inclosed vessel, be performed when the water boils violently, and when the barometer stands at 30 English inches, in a temperature of 55° of the atmosphere, marking the height of the Thermometer then for the degree of 212 of Fahrenheit; the point of melting ice being 32 of the same; thus having 180 degrees between those two fixed points, so determined. This was Mr. Bird's method, who it is apprehended first attended to the state of the barometer, in the making of Thermometers. But these instruments may be made equally true under any pressure of the atmosphere, by making a proper allowance for the difference in the height of the barometer from 30 inches. M. De Luc, in his Recherches sur les Mod. de l'Atmosphere, from a series of experiments, has given an equation for the allowance on account of this difference, in Paris measure, which has been verified by Sir George Schuckburgh, Philos. Trans. 1775 and 1778; also Dr. Horsley, Dr. Maskelyne, and Sir George Shuckburgh have adapted the equation and rules, to English measures, and have reduced the allowances into tables for the use of the artist. Dr. Horsley's rule, deduced from De Luc's, is this: , where h denotes the height of a Thermometer plunged in boiling water, above the point of melting ice, in degrees of Bird's Fahrenheit, and z the height of the barometer in 10ths of an inch. From this rule he has computed the following table, for finding the heights, to which a good Bird's Fahrenheit will rise, when plunged in boiling water, in all states of the barometer, from 27 to 31 English inches; which will serve, among other uses, to direct instrument makers in making a true allowance for the effect of the variation of the barometer, if they should be obliged to finish a Thermometer at a time when the barometer is above or below 30 inches; though it is best to fix the boiling point when the barometer is at that height.

The numbers in the first column of this table express heights of the quicksilver in the barometer in English inches and decimal parts: the 2d column shews the equation to be applied, according to the sign prefixed, to 212° of Bird's Fahrenheit, to find the true boiling point for every such state of the barometer. The boiling point for all intermediate states of the barometer may be had with sufficient accuracy by taking proportional parts, by means of the 3d column of differences of the equations. See Philos. Trans. vol. 64, art. 30; also Dr. Maskelyne's paper, vol. 64, art. 20.

Sir Geo. Shuckburgh (Philos. Trans. vol. 69, pa. 362) has also given several tables and rules relating to the boiling point, both from his own observations and De Luc's, form whence is extracted the following table, for the use of artists in constructing the Thermometer.

The Royal Society too, fully sensible of the importance of adjusting the fixed points of Thermometers, appointed a committee of seven gentlemen to consider of the best method for this purpose; and their report may be seen in the Philos. Trans. vol. 67, art. 37.

They observe, that although the boiling point be placed so much higher on some of the Thermometers now made, than on others, yet this does not produce any considerable error in the observations of the weather, at least in this climate; for an error of 1 1/2 degree in the position of the boiling point, will make an error only of half a degree in the position of 92°, and of not more | than a quarter of a degree in the point of 62°. It is only in nice experiments, or in trying the heat of hot liquors, that this error in the boiling point can be of much signification.

In adjusting the freezing, as well as the boiling point, the quicksilver in the tube ought to be kept of the same heat as that in the ball. When the freezing point is placed at a considerable distance from the ball, the pounded ice should be piled up very near to it; if it be not so piled, then the observed point, to be very accurate, should be corrected, according to the following table.

The correction in this table is expressed in 1000th parts of the distance between the freezing point and the surface of the ice: ex. gr. if the freezing point stand 6 inches above the surface of the ice, and the heat of the room be 62, then the point of 32 should be placed 6 X .00261, or .01566 of an inch lower down than the observed point.

The committee farther observe, that in trying the heat of liquors, care should be taken that the quicksilver in the tube of the Thermometer be heated to the same degree as that in the ball; or if this cannot be done conveniently, the observed heat should be corrected on that account; for the manner of doing which, and a table calculated for that purpose, see Philos. Trans. vol. 67, art. 37.

It was for some time thought, especially from the experiments at Petersburgh, that quicksilver suffered a cold of several hundred degrees below o before it congealed and became fixed and malleable; but later experiments have shewn that this persuasion was merely owing to a deception in the experiments, and later ones have made it appear that its point of congelation is no lower than — 40°, or rather — 39°, of Fahrenheit's scale. But that it will bear however to be cooled a few degrees below that point, to which it leaps up again on beginning to congeal; and that its rapid descent in a Thermometer, through many hundred degrees, when it has once passed the above-mentioned limit, proceeds merely from its great contraction in the act of freezing. See Philos. Trans. vol. 73, art. *20, 20, 21.

It is absolutely necessary that those who would derive any advantage from these instruments, should agree in using the same liquor, and in determining, according to the same method, the two fundamental points. If they agree in these fixed points, it is of no great importance whether they divide the interval between them into a greater or a less number of equal parts. The scale of Fahrenheit, in which the fundamental interval between 212°, the point of boiling water, and 32° that of melting ice, is divided into 180 parts, should be retained in the northern countries, where Fahrenheit's Thermometer is used: and the scale in which the fundamental interval is divided into 80 parts, will serve for those countries where Reaumur's Thermometer is adopted. But no inconvenience is to be apprehended from varying the scale for particular uses, provided care be taken to signify into what number of parts the fundamental interval is divided, and the point where o is placed.

With regard to the choice of tubes, it is best to have them exactly cylindrical through their whole length. The capillary tubes are preferable to others, because they require smaller bulbs, and they are also more sensible, and less brittle. The most convenient size for common experiments has the internal diameter about the 40th or 50th of an inch, about 9 inches long, and made of thin glass, that the rise and fall of the mercury may be better seen.

For the whole process of filling, marking, and graduating, see De Luc's Recherches &c, tom. 1, p. 393, &c.

The following is a table of some observations made with Fahrenheit's Thermometer, the barometer standing at 29 inches, or little higher.

On the general subject of Thermometers also see Martine's Essays, Medical and Philosophical. Desaguliers's Exp. Phil. vol. 2, p. 289. Musschenbroeck's Int. ad Phil. Nat. vol. 2, p. 625, ed. 1762. De Luc's Recherches sur les Modif. &c, tom. 1, part 2, ch. 2. Nollet's Leçons de Physique, tom. 4, p. 375.

Thermometers for particular uses.—In 1757, lord Cavendish presented to the Royal Society an account of a curious construction of Thermometers, of two different forms; one contrived to shew the greatest degree of heat, and the other the greatest cold, that may happen at any time in a person's absence. Philos. Trans. vol. 50, p. 300.

Since the publication of Mr. Canton's discovery of the compressibility of spirits of wine and other fluids, there are two corrections necessary to be made in the result given by lord Cavendish's Thermometer. For in estimating, for instance, the temperature of the sea at any depth, the Thermometer will appear to have been colder than it really was: and besides, the expansion of spirits of wine by any given number of degrees of Fahrenheit's Thermometer, is greater in the higher degrees than in the lower. For the method of making these two corrections by Mr. Cavendish, see Phipps's Voyage to the North Pole, p. 145.

Instruments of this kind, for determining the degree of heat or cold in the absence of the observer, have been invented and described by others. Van Swinden (Diss. sur la Comparaison du Therm. p. 253 &c) describes one, which he says was the first of the kind, made on a plan communicated by Bernoulli to Leibnitz. Mr. Kraft, he also tells us, made one nearly like it. Mr. Six has lately, viz, in 1782, proposed another construction of a Thermometer of the same kind, described in the Philos. Trans. vol. 72, p. 72 &c.

M. De Luc has described the best method of constructing a Thermometer, fit for determining the temperature of the air, in the measuring of heights by the barometer. He has also shewn how to divide the scale of a Thermometer, so as to adapt it for astronomical purposes in the observation of refractions. See Recherches &c, tom. 2, p. 35 and 265.

Mr. Cavallo, in 1781, proposed the construction of a Thermometrical Barometer, which, by means of boiling water, might indicate the various gravity of the atmosphere, or the height of the barometer. This Thermometer, he says, with its apparatus, might be packed up into a small portable box, and serve for determining the heights of mountains &c, with greater facility, than with the common portable barometer. The instrument, in its present state, consists of a cylindrical tin vessel, about 2 inches in diameter, and 5 inches high, in which vessel the water is contained, which may be made to boil by the flame of a large waxcandle. The Thermometer is fastened to the tin vessel in such a manner, as that its bulb may be about an inch above the bottom. The scale of this Thermometer, which is of brass, exhibits on one side of the glass tube a few degrees of Fahrenheit's scale, viz, from 200° to 216°. On the other side of the tube are marked the various barometrical heights, at which the boiling water shews those particular degrees of heat which are set down in Sir Geo. Shuckburgh's table. With this instrument the barometrical height is shewn within one 10th of an inch. The degrees of this Thermometer are rather longer than one 9th of an inch, and therefore may be divided into many parts, especially by a Nonius. But a considerable imperfection arises from the smallness of the tin vessel, which does not admit a sufficient quantity of water; but when the quantity of water shall be sufficiently large, as for instance 10 or 12 ounces, and is kept boiling in a proper vessel, its degree of heat under the same pressure of the atmosphere is very settled; whereas when a Thermometer is kept in a small quantity of boiling water, the mercury in its stem does not stand very steady, sometimes rising or falling so much as half a degree. Mr. Cavallo proposes a farther improvement of this instrument, in the Philos. Trans. vol. 71, p. 524.

The ingenious Mr. Wedgwood, so well known for his various improvements in the different sorts of pottery ware, has contrived to make a Thermometer for measuring the higher degrees of heat, by means of a distinguishing property of argillaceous bodies, viz, the diminution of their bulk by fire. This diminution commences in a low red heat, and proceeds regularly, as the heat increases, till the clay becomes vitrified. The total contraction of some good clays which he has examined in the strongest of his own fires, is considerably more than one-fourth part in every dimension. By measuring the contraction of such substances then, Mr. Wedgwood contrived to measure the most intense heats of ovens, furnaces, &c. For the curious particulars of which, see Philos. Trans. vol. 72, p. 305 &c.