like mercury, oil, and other liquids, its density went on augmenting to its freezing point, the cold air would continne to rob the mass of its heat till the whole sank to 32°, and it would suddenly set into a solid rock of ice, and every living animal within it would perish. In these climates a lake so frozen could never again be liquefied; for the process of thawing necessarily beginning above, the heated and light water would lie upon the surface, and effectually prevent the convection of heat to the lower strata.

We are naturally struck with this wonderful proof of design in a superintending Providence: for although proofs of the most perfect contrivances abound in every stone which we tread beneath our feet, and in every breath of air which we draw, we here see that the Almighty in his working is not rigidly bound by the laws which He has framed for the order of the material universe, but that He can maintain that order and effect his beneficial purposes by exceptions to those laws, when it seems fit to his perfect wisdom.

This is the course in accumulations of fresh water: for the waters of the "great deep," another protection has been provided. Saline matter in solution in water, it has been ascer tained, lowers both the point of freezing, and the point of maximum density. The ocean, on that account, and because of its great depth, which renders it an almost inexhaustible store of heat, resists freezing still more effectually than the deepest natural reservoirs of fresh water, and is scarcely known to freeze, except in latitudes where the most intense cold prevails. Even then, it is the watery particles alone which congeal to the exclusion of the saline, which, increasing the density of the lower strata, arrest their circulation, and thus preserve them from the superficial cold.

HEAT OF COMPOSITION.

§ 170. Heat and Temperature we have hitherto used as nearly synonymous terms, and all the effects of the subtle force, to which we have been directing our attention, have been accompanied by its free development, and have been measured by our sensations, and by the thermometer and pyrometer. We have now to trace it, entering, as it were, into the composition of bodies, losing its character of temperature, and becoming latent to our instruments and our feelings.

Equal volumes of the same liquid, at different temperatures, afford, upon mixture, the mean temperature of the two. A pint

of water at 50°, being mixed with a pint at 100°, a thermometer immersed in the mixture will indicate a temperature of 75°. This result has already, indeed, been adduced in confirmation of the accuracy of the instrument (§ 143). If, however, a measure of quicksilver at 100° be agitated with an equal measure of water at 40°, the resulting temperature of the two will not be 70°, or the mean, but 10° lower, or 60°; so that the quicksilver will lose 40°, whereas the water will only gain 20°: yet the water must contain the whole heat which the quicksilver has lost. Hence, it appears that water has a greater capacity for heat than quicksilver: it requires a larger quantity of heat to raise it to a given temperature. The confirmation of this view may be obtained by the converse of the experiment; for if a measure of water at 100° be agitated with an equal measure of quicksilver at 40°, the resulting temperature will be 80°: the water will fall 20° in temperature, but in this fall will give out sufficient heat to raise the quicksilver 40°.

The same comparison may be made by weight, and will lead to the same conclusion. Thus, if a pound of quicksilver at 40° be agitated with a pound of water at 156°, the resulting temperature will be 152°.3: the water will lose 3°.7 of temperature, but enough heat will be evolved to raise the metal 112°3. Now, the proportion of 3°.7: 112°.3, is the same as 0.033 1; hence, adopting water as the standard of comparison, we call the specific heat of quicksilver 0.033, designating by the term specific heat the heat peculiar to the species of matter compared with the standard.

Again: If a pound of water at 100°, and the same weight of oil at 50°, be mixed together, the resulting temperature will not be the mean, 75°, but 83°; the water, therefore, will lose 16%, while the oil will gain 33°, or reversing the temperatures, the mean will be 66°, so that the oil will give out 33°, and the water will rise only 16%. Hence, the heat which will raise the temperature of oil 2°, will raise an equal weight of water only 10; and the specific heat of oil will therefore be 0.5.

§ 171. This different capacity of different bodies for heat must have a considerable influence upon their rates of heating or cooling: those which have the highest specific heat increasing or diminishing their temperatures most slowly under equal circumstances. Thus, if equal weights of water and quicksilver be placed at equal distances before a fire, the metal will be

more rapidly heated than the water; and again will cool down a certain number of degrees more rapidly when exposed in a cold place. Conversely, the specific heats of different bodies may be determined by carefully observing the time in which they cool down a certain number of degrees, and comparing them with water under similar circumstances. This method is susceptible of great accuracy, and may obviously be applied where mixture is impossible.

§ 172. A third method of ascertaining specific heats was devised by MM. Lavoisier and La Place, who contrived an apparatus for the purpose, to which they gave the name of Calorimeter. This instrument was liable, however, to some practical objections, which have limited its use. The principle, upon which it was constructed, will afford another illustration of the nature of the phenomenon (40). A certain weight of water, for instance, was surrounded with ice in a convenient vessel, and in passing from the temperature of 212° to 32°, the quantity melted was found to be a pound; an equal weight of oil in cooling down through the same range of temperature thawed only half a pound: and from this experiment we arrive at the same conclusion, as from mixture and cooling, that the specific heat of water being reckoned as 1°, that of the oil is only 0.5°.

(40) The calorimeter consists of two similar metallic vessels, the one contained within the other, and kept separate by small pieces of wood. The interval between the two is filled with ice, broken small, and packed close. By constantly renewing this ice as it melts by the heat of the atmosphere, the interior vessel will be kept constantly at the temperature of 32°. The water which is formed is removed by a stop-cock placed at the lower part of the interval between the two vessels. Within the interior vessel another still smaller is suspended, formed of iron net, designed to hold the body to be cooled. The interval between this third vessel and the second is also filled with ice and the water which this latter produces in melting, flows out of the lateral stop-cock into a vessel which receives it, that it may be accurately weighed.

The following table exhibits the specific heats of equal weights of various bodies referred to this standard, from the best authorities:

§ 173. It has been shown by the careful experiments of MM. Dulong and Petit, that the specific heat of bodies increases as their temperature rises; so that it requires more heat to raise them a certain number of degrees when at a high than when at a low temperature. The specific heat of iron, for instance, was found as set down in the following table:

TABLE XVI. Specific Heat of Iron.

A similar law is maintained in other bodies, as shown in the following table:

TABLE XVII. Progressive Specific Heat.

§ 174. It is probably from changes effected in the specific heat of bodies that condensation or approximation of their particles is attended with elevation of temperature, and dilatation or expansion with the opposite effect.

When spirits of wine and water are mixed together in equal measures, it may be shown that the bulk of the mixture is less

than that of the two liquids in their separate states (§96); and, in consequence, the temperature rises so as to become sensibly warm to the hand. In the same way oil of vitriol and water contract on mixture, and so much heat becomes free, that some inflammable substances may be kindled, or water boiled, by its application. The sudden compression of air by the piston of a small syringe disengages heat enough to kindle tinder exposed to it: while compressed air suddenly allowed to expand will become so cold as to condense all the vapour, with which it may be mixed, in the form of a cloud. If a delicate thermometer again be suspended in the receiver of an air-pump, it will be found to sink during the process of exhaustion, and the cloud which commonly forms at the same time is owing to the same absorption of the heat of temperature.

§ 175. The increased capacity which air acquires by rarefaction has an important influence in modifying the temperature of the atmosphere. The air becoming rarer as it ascends, absorbs its own free heat, and hence becomes cold in proportion as it recedes from the surface of the earth, from which it chiefly derives its heat. The average depression of temperature has been found to be about 1° of Fahrenheit's scale for each 300 feet of ascent. Sir John Leslie investigated the subject and proposed a formula, the results of which agree admirably with observation and experiment; even the extreme result of the ignition of the tinder to which we have just referred being indicated by it. It may be expressed as follows: Multiply the constant co-efficient 45 by the difference between the density of the air and its reciprocal, and the result will represent the measure of heat upon Fahrenheit's scale due to the change of condition. This result may be plus or minus: it may express the heat liberated in the condensation of air, or the heat absorbed during its opposite rarefaction.

Thus let it be required to estimate the heat liberated from air when its density is tripled :

which is the measure of the heat liberated; and the same quantity will be absorbed either when the air recovers its former density, or when air of the ordinary state is expanded into triple its volume. By this constitution of the atmosphere, heat, so to speak, is economised: for if, instead of thus being absorbed and laid up in store, it had remained free, it would soon have become dissipated and lost. Other most important