The advantages of mercury over alcohol and air, as a measure of tem perature, are that its expansions are more nearly proportional to the increments of heat than those which take place in either of the other fluids; it is easily deprived of air, and its power to conduct heat being considerable, the changes of its volume by changes of temperature in the surrounding medium take place more rapidly than those of any other fluid except the gases.
At first the scales for measuring degrees of tempe rature were arbitrary, and consequently no two thermo meters could be compared together. The scale of the Florentine thermometer was determined by marking the place where the top of the spirit column stood in the tube when the latter was immersed in snow, and the place at which it stood at the time of the greatest beats in Florence : the interval between the points was divided into 60 parts. Subsequently, in this country, Mr. Boyle and Sir Isaac Newton formed scales for determining the expansion of the spirit or oil by making the space included in each degree of the tube equal to a certain portion of the whole volume. Thus, supposing the ball of the thermometer and part of the tube to be divided into 10,000 equal 'parts, and to be wholly occupied by the oil when the instrument is plunged in melted ice, Sir Isaac found that by the heat of the human body the oil expanded 256 such parts, and by that of boiling water 725 parts ; then, considering the point at which the top of the column stood in the tube when the latter was placed in the ice, as the zero of the scale, he divided the interval between this point and that at which the top of the column stood when the ball of the thermometer was placed under the arm of a man, into 12 parts. After wards by proportion be found that the distance from the ice-point to that of boiling water was equal to 34 such parts (` Phil. Trans.,' vol. This method, being of difficult execution, was soon abandoned.
The scale which has been in general use in this country since the year 1724, is supposed to have been invented by Fahrenheit. It is quite unknown on what ground he made choice of the fixed points on his scale, or of the number of graduations between them ; but it is thought that one of tho fixed points was that of boiling water, and that the other, which is the zero of the scale, was that at which the top of the column stood when the instrument was exposed to an intense cold in Iceland, in 1709. The extent of the scale between this last point and that of boiling water is divided into 212 parts, and the point of freezing water is at the thirty-second division from the zero point. See the scale on the right of the tube in the
above figure.
M. Reatimur constructed a thermometer in which spirit of wine was employed, and be formed a scale in a manner nearly similar to that which had been put in practice by Sir Isaac Newton. lle computed the volume of the gLlaa ball, and graduated the tube so that the space between two divisions was equal to one-thousandth part of that volume: be then found the zero of the scale by marking the place whore the top of the column stood when the thermometer was placed, in water just freezing : and afterwards, plunging the instrument in boiling water, he observed whether or not the spirit rose exactly eighty divisions. If not, he strengthened or diluted the spirit till it did so ; and the point at which the top of the spirit stood became the point of boiling water. Of this instrument an account was published in the 316moires ' of the Academy of Sciences for 1730, but the construction has been long since abandoned ; for, besides the difficulty of giving a proper degree of strength to the spirit, it is well known that the latter cannot be made to take the temperature of boiling water, so that the determination of the upper point in the scale must be very erroneous, That which is now called Itstaumur'a thermometer is an improvement on the former, by 31. Deltic, who determined the points of freezing and boiling water by experiment, and divided the distance between therr into eighty parts, the zero of the scale being at the former point. Se( the scale on the left of the tube in the above figure.
A third scale, called "Centigrade;' has been much in use among tin philosophers of the Continent within the last eighty years: it wag invented by Celsius, a Swede, and it differs from that of Rdaumur of Deltic, only in the distance between the points of freezing and boiling water being divided into 100 parts. The length of each degree in thin thermometer, as well as in that of Ithumur, is greater than in till scale of Fahrenheit ; and consequently the indications of temperature when the top of the spirit or mercury is between the lines of divider are rather uncertain, from the difficulty of estimating them accuratel; by the eye ; also, the temperatures required to be determined bein1 often below the point of freezing-water, the employment of negatiss signs is of more frequent occurrence with these thermometers that with those of Fahrenheit.