TABLE VI.-Linear Dilatation from the Freezing to the Boiling Point of Water.

The expansion in volume may be obtained, without sensible error, by trebling the number which expresses its increase in length, where, as in the above instances, the fraction of its length is small. By the same change of temperature, the following liquids augment their volumes in the annexed proportions:

But every aeriform substance, provided it be not in contact with a liquid, expands in the same proportion; one thousand parts of air becoming 1373, by being heated from the freezing to the boiling point of water.

These expansions take place gradually, and when the heat is withdrawn the bodies return to their original bulks, by corresponding regular contractions.

§ 106. Now the perfection of science depends upon accurate measurement and instrumental precision there can be no advance of a single step without these essential means. The unassisted senses are by no means sufficient to determine

:

quantities with anything approaching to the necessary degree of exactness. The hand, or the application of muscular force, may indeed inform us that one body differs very greatly from another in gravity; but without the balance to determine weights which would be wholly inappreciable in this way, the science of chymistry could not have existed. So our sensations may inform us that one body is hotter or colder than another; but without some means of measuring differences of temperature which would wholly fail to affect our senses, our knowledge of heat would have been small indeed.

It is by observing that expansion or enlargement of volume is always produced by the same causes, which affect us with the sensation of heat, that we come to regard expansion as the indication of heat, and as this is an effect which can be ascertained with the utmost precision, we adopt its measure as that of the cause which produces it.

If a certain quantity of air, or of a liquid, or solid, undergo an augmentation of volume which we can determine when exposed to a certain source of heat, and, when we expose it to another source of heat, that expansion is doubled, there is reason to infer that the intensity of the second source is double that of the first. And this is the principle of that useful instrument the thermometer.

107. The honour of the first invention is generally ascribed to Sanctorio, an Italian physician, about the year 1590; but the same contrivance probably suggested itself in an independent manner to Cornelius Drebel, about the year 1610. One thing, however, is certain, viz., that it dates from about the beginning of the seventeenth century.

The thermometer of Sanctorio consisted of a hollow glass globe, attached to a long stem, open at the opposite extremity; a portion of air was expelled from it by the expansive force of heat, and the end of the tube was then immersed in a colour

461676 A

ed liquid.(21) As the included air cooled and returned to its former volume, the liquid rose in the tube, from the superior elasticity of the external air pressing upon the external surface of the liquid. A scale of equal parts was applied to the stem, by which the expansion of the included air from heat, or its contraction from cold, could be measured, by the movement of the column of liquid. It was liable, however, to the objection of being acted upon, not only by the expansion of the included air, but by the barometric changes of the exterior atmosphere.

§ 108. Air has not only the advantage of being extremely regular in its expansion, but also of indicating very minute changes of temperature, by the great alterations of volume which it undergoes from being heated and cooled; and a modification of the air-thermometer is now very often used in delicate researches. It was invented in the year 1676, but was chiefly brought into notice by the admirable experiments of the late Sir John Leslie. It consists of two thin equal glass balls, united together by a tube bent twice at right angles, the balls being situ

B

ated at the top of the perpendicular arms, and cut off from any communication with the atmosphere. Both balls contain air, but the greater part of the tube is filled with a colour

(21) Air thermometer of Sanctorio. B is the stem of the instrument, the upper end of which terminates in a capacious ball, while the lower dips below the surface of the liquid in the vessel c. When a portion of air has been expelled from the ball by heat, and it is afterward cooled, the superior elasticity of the outward air raises a column of liquid in the stem.

ed liquid.(22. This instrument cannot be affected by any change of temperature acting alike upon both the balls, for the pressure on the opposite surfaces of the included liquid will in such cases be always equal; but it will instantly indicate the slightest difference of temperature between the two balls; for the elasticity of the air on one side being greater than on the other, the liquid will be forced towards the side where the temperature is the lowest, and the difference may be measured upon a scale. The instrument has hence been named the differential thermometer.

109. The first great improvement in the thermometer for ordinary purposes was made by the members of the Italian Academy del Cimento, in 1660, who substituted the expansions of a liquid as the measure of heat; enclosing it in a glass ball and tube, which was afterward cut off from all communication with the variable atmosphere by softening the extremity of the glass at the flame of a lamp, and hermetically sealing it: the instrument thus at

once became more accurate and more manageable. Spirits of wine was the liquid first employed, and quicksilver was afterward used by Halley and Sir Isaac Newton: both liquids are now, at times, employed for different purposes.

§ 110. Still the thermometer wanted much of per

fection; for different instruments thus constructed could not afford comparable results, although experiments made with the same_instrument were comparable with one another. For the last great improvement we are indebted to Newton. It had been observed by another eminent philosoper, Hook, that the temperature of melting ice was always fixed and permanent; and that the temperature of boiling water was equally invariable, provided the pressure of the atmosphere did not change. The sagacity of Newton pointed out the application of these observations to the completion of the instrument. If we immerse the mercurial thermometer in melting snow or ice, the liquid will gradually contract, and sink in the stem to a certain point, and then stop; and, however long we may allow it to remain in the ice, it will sink no lower. The experiment, repeated at any time or in any place, will afford the same results; the liquid will always sink to the same part, and no farther. By marking this we obtain one fixed point, which must be the same in every thermometer which is subjected to the trial. If we now remove the instrument into a vessel where it may be surrounded with boiling water, while the barometer indicates an unvarying pressure of thirty inches, the liquid in the glass will expand till the mercurial column reaches a certain height, where it will again become perfectly stationary, and afford another invariable point of comparison. The distance between these two points measures the amount of expansion of the whole quantity of the included liquid; or, rather, is the measure of the difference of the expansion of the liquid and the glass; for both expand. Upon the supposition that the expansion of the two is equable, and that the bore of the glass tube, in which the liquid moves, is equal throughout, the distance between the two points may be divided into any number of equal parts, and the equal amounts of