Two classes of instruments are employed for the measurement of heat, namely, thermometers and pyrometers. Thermometers are only employed to measure comparatively low temperatures, and we shall confine our remarks entirely to this class in the present article. Modern physics having demonstrated that heat is merely a mode of motion in matter, the principles upon which its measurement depends are, perhaps, more difficult to understand than the false theory prevalent before the establishment of this doctrine. So long as heat was considered a substance, even thoagh an imponderable one, it was not difiicult to understand how its absorption into a body might definitely enlarge that body, as wood is enlarged by the absorption of water. Why a body becomes enlarged by the increased motion of its particles is more difficult of comprehension. If we, however, drop the consideration of the why in this case, and confine ourselves to the law or manner in which this expansion takes place, we may arrive at definite and practical results. It is, nevertheless, proper to state that the ultimatum which science has reached in regard to the reason for this expansion is, that hedt is in some way opposed to cohesion. At present it is entirely futile to seek to go further than this. The fact, however, that such expansion takes place in both solids and liquids, and that it is, within certain limits, sufficiently uniform in certa,in substances to become a means of measuring the temperatures to which these substances are exposed, is the basis of thermometric measurement. But another point must be distinctly borne in mind ; thermometers only measure sensible heat. Thus one pound of steam at 313 Fah. contains heat enough to raise five and one half pounds of water to the same temperature, a fact easily demonstrated by experiment. It follows that the absolute or total number of heat units contained in any substance, must be determined by some other means than the thermometer, and that a degree on the thermometer cannot be considered a unit of heat. What then is a unit of heat It has been agreed to consider the amount of heat necessary to raise one pound of water from 33 Fah. to 33 Fah. as a unit of heat, and although doubtless there are some small sources of error in the metliod, it is sufficiently accurate to regard the amount of heat necessary to raise one pound of water one degree, anywhere between 33 Fah. and 313 Fah. as a constant quantity. It is also a correct inference that any particular substance in a uniform state, as regards the cohesive power of its particles, must exhibit the same temperature so long as it maintains that condition, since heat is the opposite force to cohesion. The more heat the less cohesion, and vice versa. Water, when passing from the liquid to the solid state, maintains such a uniformity of condition ; its temperature may, therefore, be regarded as constant. It also maintains the same uniformity of condition while passing from the liquid state into steam at the boiling point. The freezing and boiling points of water may therefore be considered as the two prominent landmarks of temperature from which the amount of expansion of some uniformly, or nearly uniformly expanding substance, as mercury, immersed in water in the two conditions named, being noted on a scale, divisions may be arbitrarily made each way on the same scale, which will indicate temperatures above or below these points. The Centigrade scale makes the hight of a mercury column immersed in freezing water, zero, and divides the distance between this point and the hight of the same column immersed in boiling water, into one hundred degrees, while the Fahrenheit Scale makes the first named hight 33 degrees above zero, and divides the space between this hight and the hight at which the mercury stands in boiling water, into one hundred and eighty divisions, or degrees. How it is possible to determine the amount of heat in any body from thermometric indication next claims our attention. The following law has been established. The total amottnt of heat in any body is the sum of its latent heat and its sensible heat. The latent heat is determined by the known capacity of the body under examination, at given temperatures to absorb heat, or, in other words, to render it latent. This term, latent heat, is not a good one, though we are still obliged to use it for want of a better. We use it only to distinguish the heat which, acting within a mass of matter, and expending its energy in antagonism to cohesive attraction, cannot be recognized by sensation, Mke the free or sensible heat. The latent, or specific heat of various bodies has been made the subject of careful study, and tables of reference have been constructed to afford a ready means of computation ; but the specific heat of all bodies is changed by any cause which lessens or increases the distance between the particles which make up their mass. The compression of steam lessens its specific heat while it increases its temperature, and vine versa. The specific heat of steam, then, is only constant at a constant pressure. It will now be seen that the total amount of heat contained in any body can be determined by the assistance of a thermometer, only when its specific heat for all temperatures has been predetermined. This has been done for many substances, including water and steam, to which the application of heat measurement is of the highest importance, as it is only by such measurement that questions of economy in steam generators can be settled. The amount of water evaporated from a constant temperature per pound of combustible consumed, under a constant pressure, being the only reliable test of the economy of a steam boiler. When the evaporation takes place at 313 the required uniformity of pressure and consequently of temperature is easily maintained, which would not be the case if an engine were driven by the steam generated, or if an attempt were made to produce the steam at a constant higher temperature. The temperature of the feed water may be easily maintained at a constant point, either at 313, or at a lower temperature, and the amount of this water which a pound of fuel will convert into steam at 313 is an exact index of the power of the boiler to transmit heat through its shell into the contained water.
This article was originally published with the title "The Measurement of Heat with the Thermometer" in Scientific American 21, 19, 297 (November 1869)