The thermometric scale in general use in this country was formed by Fahrenheit, in which the cypher or zero is supposed to be indicated by a mixture of common salt and pounded ice. The freezing point of water is fixed at 32°, and the boiling pbint at 212°, so that the interval between the freezing and boiling of water comprehends 180°. In France, before the Revolution, the thermometer of Reau mur was generally employed. Here the freezing of water is the zero, or the commencement of the scale, and boiling water is assumed as the 80th degree. Since the Revolu tion, the centigrade scale is employed, which had been previously adopted by Celsius, a Swedish philosopher. Here freezing water is marked 0°, and boiling water 100°. Perhaps none of them are without considerable objections. Fahrenheit and Rcaumur's scale are completely arbitrary; while the centigrade has its degrees too large for many purposes, and it has also the defect of beginning too high, so that we have very frequent occasion to employ the tents + and which may lead to confusion. Of the three, that of Fahrenheit is certainly the best ; both because it is, in a great measure, free from this last objection, and because the degrees are not too large. Nor is it of much conse quence that the commencement and termination are so en tirely arbitrary ; for when we are once accustomed to it, the mind refers to it as a term of comparison, without ever considering the reason for fixing upon time numbers that are employed.
The fluids employed in the formation of thermometers are either mercury or alcohol. Where we wish to ascer tain very low temperatures, the latter is preferable, because it cannot he frozen but by the most intense cold, see COLD, whereas mercury becomes solid at — 39°. But for every other purpose, mercury has a decided advan tage. Its expansions from the same additions of heat are mote uniform ; it is also more sensible to slight varia tions of temperature ; and its boiling point is so elevated, that it possesses a range in the upper part of scale, which is applicable to most practical purposes. Except, therefore, for measuring very low temperatures, mercury is now always preferred. See THERMOMETER.
A great deal of attention has been paid to the question, whether mercury, or any other fluid used for thermometri cal instruments, was an exact measure of the additions of heat ? whether the expansion was the same, by adding a certain quantity of heat at the lower as in the higher part of the scale? The experiments of Dcluc on this point al e regarded as the most correct. Ile observed the height to which the mercury rose in the tube, by immersing it in two equal portions of water, at temperatures considerably different from each other; then he mixed the two portions of fluid together, and examined whether the thermometer indicated the mean between the two. The result of De luc's investigation was, that no fluid is an exact measure of temperature ; but that, in all cases, the augmentation of bulk is in a greater ratio than the actual addition of heat. Nercury appeared to be the most uniform ; but even in this fluid, the temperature of a mixture of two portions, one at the freezing, and the other at the boiling point, was 23" less than the mean. The real temperature in this case is but, according to the common graduation, the ther mometer stands at 1244°, (Ree/2erehes, t. i. p 311.) Hence, in order to obtain a precise measure of the increments of heat, we ought to have a thermometer so graduated as to allow for this deviation; and although we are not aware that any instruments have been constructed upon this principle, yet tables have been calculated by which a proper scale might be very easily formed.
The thermometer, however, in its most improved form, is only applicable to a range of temperature below that at which mercury boils; and although this includes most cases that usually fall under our observation, yet we have not un frequently occasion to measure much higher degrees of heat.
Newton attempted to accomplish this point, without any apparatus directly adapted to the purposve, by a species of computation. Ile ascertained the rate at which bodies cool, compared to the heat of the surrounding medium, at tem peratures within the reach of the thermometer, and he then extended the same ratio to the higher degrees of heat. By this very ingenious method he discovered the temperature of red hot iron, and the melting point of many of the me tals, with a considerable degree of precision. Since his time, different instruments, called pyrometers, have been invented, of which the most valuable, and that which has come into the most general use, was invented by Wedge wood. In the course of his experiments on the manufac ture of earthen ware, he perceived that the different kinds of earths, called clays, into the composition of which a large quantity of alurnine enters, had the remarkable property of contracting by exposure to a great heat, and likewise that this contraction was in proportion to the degree of beat em ployed. When the clay has once contracted, its bulk re mains permanently diminished, so that by applying it to a gauge, previously graduated for the purpose, we can as certain to what degree of heat the piece of clay in question has been exposed. The gauge by which the contraction of the clay is ascertained, consists of two straight pieces of brass, fixed on a brass plate in such a manner that they are nearer together at one end than at the other ; and thus the more the clay has been contracted, the farther will it slip between the brass bars. Wcdgewood then proceeded to examir.e the relation between his pyrometer and the common mer curial thermometer, and lie determined that the 0° of his scale was equivalent td the 1077° of Fahrenheit's, were it so far extended. The degrees. the size of which was arbi trary, were each of them equal to 130' of Fahrenheit's ; and the highest temperature which he actually measured was 160' of his pyrometer, corresponding to above 21 800° of Fahrenheit, which was Found to be the temperature of a particular air furnace which he employed in his experi ments : (Phil. Trans. 1782, p. 303, et seq.) This pyrometer has generally been regarded as a very valuable addition to unr philosophical apparatus; but there is one circumstance which unfortunately much impairs its ust —the difficulty of procuring the same kind of clay on which Wcdgewood operated for the test pieces, as the mass which he miginal ly employed is exhausted ; and it does not appear cer tain, whether the imitation of it that has been since prepar ed, is possessed of precisely the same physical in operties. Sec CHEMISTRY, Several other pyrometers have been invented, depending upon different principles, and possessing diffelent degrees of accuracy. Musehenbr0ek, Ferguson, Ellicot, and Smea. ton, have all exercised their ingenuity upon this point; but the only apparatus which we shall fat ther notice is that of Guy ton. In this instrument along rod of platina is fixed in an horizontal groove, formed in a mass of clay that had been hardened by exposure to a strong heat ; one end of the rod extends beyond the groove, and presses against another rod of platina, which acts as a lever ; its longer arm being extended, so as to form an index, which moves upon a graduated arc. As the rod is more or less expanded by the heat, the index is carried along the scale, and serves to Mint out very minute variations in the bulk of the metal : (?11m. C'him. torn. xlvi. p. 276.) We are not certain whether this instrument has ever been employed in England; but it appears to possess many advantages as an accurate measure of high temperatures. See Prito NETER.