372

INFLUENCE OF SALTS ON FREEZING-POINT.

[175

bility at low temperatures. In the case of salts which furnish anhydrous crystals, so far as can be judged from the limited number of salts submitted to experiment, the depression of temperature is directly proportional to the quantity of salt present in the liquid. For example, successive additions of 1 per cent. of each of the following salts produce a successive reduction of the freezing-point to the following extent :

It would appear that salts which crystallize with water cause a depression in proportion to the amount of hydrated salt dissolved. Calcic chloride occasions a depression of the freezingpoint of o°408 (0°227 C.) for each addition of 1 per cent. of the salt CaCl,, 6 H2O; baric chloride a depression of o° 345 (0°192 C.) for each addition of 1 per cent. of the compound BaCl2, 2 H2O. Sodic chloride crystallizes below 14° F. (-10° C.) with 2 H2O; but these crystals are rapidly dehydrated as soon as the temperature rises above that point: it is remarkable that, for all temperatures above -10° C., the depression of the freezing-point is proportional to the quantity of anhydrous salt in the solution, but below -10° it is proportionate to the addition of the hydrated salt NaCl, 2 H2O, the depression being equal to o°615 F. (0°·342 C.) for every addition of 1 per cent. of this hydrate to the solu

tion.

In the process of freezing as it usually occurs in nature, the act of solidification goes on, not continuously, but in successive layers, and in the intervals between these layers is a stratum of ice slightly more fusible than the mass either above or below. This is beautifully seen by placing a block of homogeneous transparent ice, such as that from the Wenham Lake, in the sun's rays after concentrating them by a large convex lens. Immediately that this is done, the interior of the mass becomes filled with little flower-shaped figures, each flower having six petals, evidently composed of water, while in the centre is a spot which shines with metallic brilliancy, and which looks like an air-bubble, but is

176.]

REGELATION OF ICE.

373

really a space filled only with aqueous vapour, and produced in consequence of the circumstance that water occupies a smaller volume than the ice which furnished it. These little flowers occur in horizontal planes parallel to the surface of congelation.* -(Tyndall, Phil. Trans. 1858, 211.) Faraday has suggested what seems to be a very probable explanation of these successive planes of freezing-viz., the separation of saline particles from each layer of water as it is frozen (71), so that the salts accumulate below the stratum first frozen, and form a very dilute saline solution, the freezing-point of which is a fraction of a degree below that of pure water; this thin stratum when frozen furnishes a layer of ice more fusible than the rest; a fresh layer freezes beneath, gradually excluding its saline particles, which again accumulate below, forming a fresh more fusible layer, and so on successively.

(176) Regelation of Ice.-It was remarked some years ago by Faraday, that when two pieces of ice at o° C., with moistened surfaces, are placed in contact, they freeze together, and manifest the phenomenon thence designated as the regelation of ice; whereas, if the surfaces be dry, they do not cohere. It is owing to this circumstance that during a thaw the particles of snow cohere firmly into a solid lump, whilst during a frost there is difficulty in forming the dry particles into a compact mass. This regelation of ice will occur when the surfaces of the blocks are in contact, even though the external air may be at a temperature of 25° or 30° C., or even when the ice is immersed in water at this temperature. Certain solids, as flannel, hair, or cotton, will freeze to ice even in a warm atmosphere, though others, such as saline substances, gold leaf, and the metals, will not thus freeze to it. Tyndall has followed up these observations, and made some interesting experiments and deductions from them. He took a sphere of transparent ice, and placed it in a warm room under a small hydraulic press between two pieces of boxwood hollowed out so as to form a flattened lenticular cavity. The ice broke, but, on continuing the pressure, it froze again, and in less than a minute was converted into a flattened, transparent, lenticular mass. This mass was in turn placed in a shallow cylindrical cavity of boxwood and again submitted to pressure; again it was crushed and became reduced to the form of a flat transparent cake; and this again was placed in a hemi

In certain exceptional cases this parallelism is disturbed. Probably this is due to the breaking-up of the original floe, and consolidation of its fragments irregularly, by subsequent regelation.

374

REGELATION OF ICE.

[176. spherical cavity in the wood and subjected to the pressure of a hemispherical plug which fitted the cavity; a third time it was crushed, and after a few seconds it froze again into a transparent cup of ice. Tyndall considers that upon the theory that heat is the result of vibratory motion, the liquefaction of ice, when perfectly homogeneous, must necessarily take place more easily upon the surface than within the mass; and conversely, the freezing of a thin layer of water between two masses of ice should occur more readily than upon the surface of a single mass, and hence he attempts to account for regelation. The explanation appears, however, to be insufficient; since, if true for ice, it should hold good for all substances solidifying after fusion, when two portions of the solid are brought into contact beneath the still liquid mass; and it offers no explanation of the freezing of ice to flannel, which apparently is due to the same cause as the freezing of ice to ice. It has been supposed that the masses of ice are colder within than at the surface, and hence that regelation is the result of the absorption of heat by the internal portions. Tyndall has, however, proved conclusively that this hypothesis is at variance with facts, and is indeed impossible from the conducting nature of ice itself. The ingenious theory of James Thomson, that regelation is due to the lowering of the freezing-point by the mutual pressure of two masses of ice, and that the absorption of heat due to this liquefaction freezes the contiguous layer of water, is also quite inadequate

* These observations have been ingeniously applied by Tyndall to account for the motion of glaciers. These frozen rivers of ice, in descending from the mountain sides, constantly have to force their way through contracted gorges in the rock, and gradually flow onwards, melting away at their base, whilst fresh portions of ice are forced downwards from the upper regions of the mountain by the weight of the superincumbent ice. It was ascertained by Prof. J. D. Forbes, in a series of beautiful observations, that during the descent of the glacier through its channel, the central portions of the mass move more quickly than the portions on its sides: and he likened the flow to the descent of a viscous liquid, and propounded what has been known as the viscous theory of glacier motion. Viscosity, however, is not a property which is exhibited by ice; and Tyndall has shown that all the phenomena of glacier motion are accurately accounted for by this process of crushing, and subsequent regelation into solid transparent ice. Graham has since suggested that ice may exist in two conditions-the crystalline, which is brittle, and the vitreous or colloid, in which it possesses a certain viscosity. (Phil. Trans. 1861, 222.) This view, however, as yet remains unsupported by direct experiments. Ice just at the freezing-point is, however, less hard than when it is reduced to a lower temperature; and the experiments of Person (Ann. Chim. Phys. 1850 [3], xxx. 73) show a continued evolution of latent heat by ice as it is cooled a few degrees below o° C., which is probably connected with a molecular change subsequent to the first freezing.

178.]

ABSORPTION OF HEAT DURING EVAPORATION.

375

to account for the effect, even if pressure were a necessary element in effecting regelation, which Faraday and others have shown it At present therefore the phenomenon needs further

is not. elucidation.

(177) Evolution of Heat during Solidification.-When liquids return to the solid form, their latent heat is again given out. Water, if undisturbed, may be cooled down in a narrow tube even 10° C. below the freezing-point without congealing; but the least agitation causes a portion to solidify suddenly, and the latent heat emitted at the moment by the portion which freezes raises the temperature of the whole mass to o° C. According to Dufour, this cooling of water below its freezing-point is easily effected by suspending the water in the midst of a liquid of the same density as itself, such as a mixture of chloroform and oil of almonds in suitable proportions, and exposing them to the cold of a freezing-mixture: contact with a fragment of ice causes the instant solidification of the water, though agitation, or stirring with a metallic rod, does not always do so. In like manner, sulphur, or phosphorus, if suspended in solution of zincic chloride, remains liquid many degrees below its point of solidification until touched with a fragment of its own substance. Acetic or sulphuric acid, as well as many other substances, admits, like water, of being cooled down several degrees below its point of solidification; but if agitated, or if touched with a portion of its own substance in the solid form, it immediately solidifies with evolution of heat. When sulphur vapour is condensed on glass, small drops are formed which often remain liquid for days.

A similar evolution of heat occurs when a supersaturated solution of sodic sulphate (73) is made to crystallize suddenly by agitation, the mass becoming sensibly warm to the hand. The solidification of metallic bodies is attended with a like evolution of heat.

(178) Disappearance of Heat during the Formation of Vapour.-In the change from the liquid to the gaseous state, the disappearance of heat is found to occur to an extent still greater than in the liquefaction of a solid. A vessel containing water, such as the boiler of a common still, if placed over a source of heat which is tolerably uniform in temperature, receives in equal times nearly equal accessions of heat; the water at first rises steadily in temperature, but at length it boils, and the thermometer becomes stationary: no matter how much the heat be urged, provided that the steam be allowed to escape freely, the temperature of the boiling liquid cannot be raised beyond a

376

HEATING BY STEAM.

[178. certain point. If the vapour be made to pass through the worm of the still, which is cooled by immersion in water, the steam will transfer part of its heat to the water in the condenser, which rises rapidly in temperature, whilst the vapour returns to the liquid form; but the quantity of water that is raised in the worm-tub to nearly 100° C. is very much greater than the quantity that is condensed into the form of liquid in the receiver of the still.

The large amount of latent heat contained in steam, renders it possible to use steam as a convenient and economical mode of warming buildings and apparatus which do not require to be raised to a temperature beyond that of boiling water. In practice it is found convenient, in warming a building which is used for domestic purposes, to allow one square foot of radiating surface in the steam-pipe for every 200 cubic feet of space to be heated. This estimate, however, is liable to modification, because the greater the extent of radiating and conducting surface exposed by the windows in proportion to the cubic contents of the apartment may be, the more rapid is the loss of heat.

The maintenance of a steady temperature which cannot rise above 100° C., is often required in the laboratory in the prosecution of various inquiries, espe cially in such as relate to organic chemistry, and for this purpose a small steambath, such as is represented at 1, fig. 135, is extremely useful; it may also be

employed to assist in effecting the filtration of hot liquids, where it is important to maintain their high temperature. In drying organic substances, a kind of double oven, or hot closet, made of copper, as exhibited at 2, is a convenient mode of applying heat; the interval between the internal and external plates of copper is filled with water which is heated by the gas flame below; if a higher temperature than this be required, the interval may be filled with oil; the temperature in the latter case may be regulated by a thermometer, introduced at a; at b is a tube for the escape of vapour; this tube communicates with the drying chamber.