When treating of the mechanical properties of air, we explained the manner in which the expansion occasioned by heat produces the draught up the chimney; and in the article Smoker we considered the circumstances which tend to check, to promote, and to direct this current, so as to free us from the smoke and vitiated air which accompany the consumption of the fuel. Under Heat, the communication and radiation have been fully explained. By the former, the caloric is given from one object to another, and communicated from particle to particle of the same object by contact; by the latter, it emanates in straight lines, through the air, from substance to substance, the emanations being absorbed by some bodies, and reflected by others.

The power which objects possess of receiving and communicating heat by the former of these modes, is termed their conducting power; by the latter, their radiating, their absorbing, and reflecting power. Those which receive it quickly by the former, also give it off quickly in the same way; and those which radiate powerfully also absorb powerfully; of course they reflect little: accordingly the radiating and reflecting powers are opposed to each other. Metals are good conductors; stones, bricks, and porous bodies of a similar nature, are bad conductors. The radiating, reflecting, and absorbing powers depend not so much on the nature of the substance, as on the surface. Highly polished surfaces radiate and absorb little, while dark and rough ones radiate and absorb powerfully; of course, resplendent surfaces are good reflectors.

When fuel is consumed with the view of heating apartments, the communication of heat depends entirely on the manner in which it is consumed. In an open fire-place or common grate, the heat thrown into the apartment is chiefly that emitted by radiation and reflection; the former coming directly from the fuel, and from the parts of the fireplace warmed by it; the latter being the radiated heat reflected by the polished parts of the grate. The rays thus thrown off, being absorbed by the objects in the room, then communicate warmth to the atmosphere which it contains. But part of the warmth must also be received by communication; for the walls adjoining the fire-place being heated by contact, will likewise communicate caloric to the air brought in contact with them; and as it is expanded it will ascend, and allow other particles to flow in, also to receive their supply. This however is small in comparison to that given to the apartment by radiation and reflection.

Much of the heat thus thrown into the room is carried off, indeed may be said to be lost; for, owing to the manner in which the fuel is consumed, a large supply of air is necessary for the combustion, which, rushing in upon the fuel, is itself, after acting on it, carried up the vent. Hence, to warm apartments in this way, a great deal of fuel is required. The quantity must depend very much on the form of the fire-place, and on the materials of which it is constructed.

The principle on which stoves operate is different. A stove, however modified in form, is merely a fire-place enclosed on all sides; the air necessary for the combustion entering from below, and carried off, as in a common grate, by a vent. Now, in this way of consuming fuel, the radiation is trifling in comparison to the communication by contact. Of course much must depend on the kind of stove employed; but it is allowed, that by far the greater part of the heat which the atmosphere of the room receives is that given directly by communication; for the air in contact with the sides of the stove is heated, is expanded, and carried upwards; and thus, by the constant flow of cool air on the stove, the whole of that in the room is warmed. No doubt, the stove being itself warm, must give off heat, not only by contact, but also by radiation; and the proportion thus emitted must depend on the temperature and surface, as already explained. But though heat is thus distributed, yet it must be borne in mind, that it is chiefly by contact that the atmosphere of the room is warmed; and as the air is thus easily heated, much less fuel is required than when it is burned in an open fire-place. The necessary supply of air is therefore not so great; and it is this which principally causes the difference in our feelings when in a room warm- ed by a stove, and by a grate; for in the latter instance, not only have we the exhilarating influence excited by the blaze of the fuel, but there is a more rapid renewal of air through the apartment, and accordingly the atmosphere has not the close unpleasant feeling which it has when warmed by a stove.

The warming by stoves must therefore be conducted on principles different from those adopted in the employment of open fire-places. The general principle is, 1st, to employ the fuel in the most effectual manner for heating the external part of the stove, which is immediately efficient in warming the contiguous air, chiefly by contact; and, 2nd, to keep within the room the air thus warmed, at least as much as is consistent with wholesomeness and cleanliness. It would occupy a volume to describe the immense variety of stoves which ingenuity has constructed. We shall content ourselves with giving a specimen of the two chief classes into which they may be distinguished.

The air of a room may be equally warmed by applying it either to the surface of a small stove made very hot, or to the surface of a much larger stove more moderately heated. The first kind is chiefly used in Holland, Flanders, and the milder climates of Germany and Poland. The last are used in the frozen climates of Russia and Sweden. The first are generally made of cast-iron, the last partly of iron and partly of brick-work.

Fig. 1 represents a small German stove, fully sufficient for warming a room of 24 feet by 18. The base is about three feet broad and fourteen inches deep, that is, from back to front, and six or seven feet high. The decoration is in the fashion of that country; but the operative structure of it will admit of any style of ornament. A is the fire-place, and the wood or charred coal is laid on the bottom, which has no bars. Bars would admit the air too freely among the fuel, and would both consume it too fast and raise too great a heat. That no heat may be uselessly expended, the sole of the fire-place and the whole bottom of the stove is raised an inch or two above the floor of the room, and the air is therefore warmed by it in succession, and rises upwards. For the same reason, the back of the stove is not in contact with the wall of the room, or of the niche in which it is placed. The fire-place is shut up by a door which fits closely to its case, and has a small wicket at the bottom, whose aperture is regulated by a sliding plate, so as to admit no more air than what suffices for slowly consuming the fuel. The flame and heated air rise to the top of the fire-place, three or four inches above the arch or mantle-piece, and get out laterally by two narrow passages B, B, immediately below the top-plate of the base. The current bends downward on each side, passes at C, C, under the partition-plates which divide the two side-chambers, and then rises upwards through the pater division of each, and passes through narrow slits D, D, in the top-plate, and from thence along the two hollow piers, E, E. The two lateral currents unite at the top of the arch, and go through the single passage F into the larger hollow behind the escutcheon G. From this place it either goes straight upwards into the vent in the wall by a pipe on the top of the stove, or it goes into the wall behind by a pipe inserted in the back of the stove. The propriety of this construction is very obvious. The current of hot air is applied to the exterior of the stove everywhere except in the two side-chambers of the base, where the partition-plates form one side of the canal. Even this might be avoided by making each of these side-chambers a detached hollow pillar. But this would greatly increase the trouble of construction and joining together, and is by no means necessary. The arch H has a graceful appearance, and affords a very warm situation for anything that requires it, such as a drink in a sick person's bed-chamber, &c. Persons of a certain class use this place for keeping a dish warm; nay, the lower part of the arch is frequently occupied by an enclosed chamber, where the heat rises high enough even for dressing victuals, as will be easily imagined when we reflect that the sole of it is the roof of the fire-place.

The stove now described is supplied with fuel and with air by the front door opening into the room. That there may be space for fuel, this middle part projects a few inches before the two side-chambers. These last, with the whole upper part of the stove, are not more than ten inches deep. The passages therefore from the fire-place are towards the back of it; so that if we have a mind to see the fire, which is always cheerful, the door may be thrown open, and there is no danger of the smoke coming out after the current has once warmed the upper part of the stove. When the stove is of such dimensions that the base is about two feet and a half or three feet high, the fire-place may be furnished with a small grate in the British style. If the door is so hung that it can not only be thrown back, but lifted off its hinges, we have a stove-grate of the completest kind, fully adequate, in our mild climate, to warm a handsome apartment, even with an open fire; and when we hang on the door, and shut up the fire-place, a stove of the dimensions already given is almost too much for a large drawing-room.

A very simple form of stove is that represented in the annexed cut. It consists of a square box of iron, A B C D, resting on feet, and having a projecting hearth-plate at E. FF is an inner box projecting into the outer one. G is the chimney. The fuel is burned at A; and the flame passes to the chimney around the inner box, which may be used as an oven for cooking. At H there is a large door for the introduction of the fuel, in which there is also a smaller door. Both of these are generally kept shut when the furnace is in use; but when a greater heat is required, the smaller door is opened.

The effect of stoves as now described may be greatly increased, as is frequently done, by having the mouth communicating with or joined to an opening of the same dimensions, formed in the wall; and the door is in this case on the other side of the wall, in an antechamber or lobby. In some places the apartments are disposed round a spacious lobby, in which the doors of all the stoves are situated, and through which the fuel and air necessary for the combustion are supplied. But this method, though it warms the apartment, is very unfavourable to health and cheerfulness; for the same air confined and repeatedly breathed, and adulterated with the volatile emanations of the room, loses the refreshing quality which is so desirable, and even so necessary for health.

Something of this kind, it has been already mentioned, is unavoidable in all rooms warmed by stoves; and the hotter the surface of the stove, the more and more unpleasant does the air in it become; because, in addition to the slight renewal of air consequent on their use, when the surface becomes very hot, the impurities constantly floating in the atmosphere are decomposed, and emit offensive effluvia. The stove already described is almost always made of metal, and this objection applies particularly to it; hence the necessity of being attentive when cleaning it outside, to avoid the contact of greasy or oleaginous matter, which is so easily decomposed by heat, and gives off offensive and deleterious vapours. The objections stated to the stoves as now described, have given rise to the use of those constructed of brick-work, or other materials of a similar nature. These are much used in Flanders, Holland, and Germany, where they are made of the most elegant forms, and finely decorated; but it is evident that they cannot be so effectual, nor equally warm a room with the same expense of fuel; earthen ware being inferior to metal in its conducting power. In addition to this, they are liable to a very great objection, the difficulty of preventing their cracking when heated; for different parts of the stove being of different heats, they expand unequally; and no cement can be expected to withstand this, especially when we recollect, that the heat which causes the baked earthen ware to expand, causes a contraction of the clay or cement with which the parts of the stove are joined together. Accordingly stoves of this kind do not stand long without cracking, even when strengthened by iron hoops and cramps judiciously disposed within them; nor does hooping them externally prevent it. When a crack is thus occasioned, it is not only unsightly, but may be dangerous, from its allowing the vitiated air to escape into the apartment.

For these and other reasons, we can scarcely hope to make stoves of brick-work or pottery which shall bear the necessary heat without cracking; and their use must therefore be confined to cases where very moderate heat is sufficient. We need not describe their construction. It is evident that it should be more simple than that of iron stoves; and we imagine, that in the very few cases in which they are likely to be employed in this country, a single fire-place, and an arch over it, divided by a partition or two of thin tile to lengthen the flue, will be quite enough. If the stove is made in whole or in part of potter's ware, a base for the fire-place, with an urn, column, obelisk, or pyramid above it, for increasing the surface, will also be sufficient. The failure commonly happens at the joinings, where the different pieces of a different heat, and perhaps of a different baking, are apt to expand unequally, and by their working on each other, one of them must give way. Instead of making the joints close and using any cement, the upper piece should therefore stand in a groove formed in the undermost, having a little powdered chalk or clay sprinkled over it, which will effectually prevent the passage of any air; and room being thus given for the unequal expansion, the joint remains entire. This may be considered as a general direction for all furnace-work, where it is in vain to attempt to hinder the mutual working of the parts. When fitted up with these precautions, the brick or pottery stoves are incomparably more sweet and pleasant than the iron ones.

But in the intense colds of Russia and Sweden, or even for very large rooms in this kingdom, stoves of these small dimensions are not sufficiently powerful; and we must follow the practice of those countries where they are made of great size, and very moderately heated. It is needless to describe their external form, which may be varied at pleasure. We shall only enlarge a little on the peculiarities connected with the general principle of their construction.

The stove is intended as a sort of magazine, in which a great quantity of heat may be quickly accumulated, to be afterwards slowly communicated to the apartment. The stove is therefore built extremely massive, and is found to be more powerful when coated with clay as wet as can be made to hang together. We imagine the reason of this to be, that very wet clay, and more particularly stucco, must become exceedingly porous when dry, and therefore a very slow conductor of heat. Instead of sticking on the glazed tiles with no more clay or stucco than is sufficient to attach them, each tile has at its back a sort of box, baked in one piece, about two or three inches deep. It is represented in fig. 3. This is filled with mortar, and then stuck on the brick-work of the stove, which has a great number of iron pins or hooks driven into the joints, which may sink into this clay and keep it firmly attached when dry. This coating, with the massive brick-work, forms a great mass of matter to be heated by the fuel. The lowest chamber, which is the fire-place, is somewhat wider, and considerably thicker than the stories above, which are merely flues. When the fire-place is finished, and about to be arched over, a flat iron bar of small thickness is laid along the top of the side-wall on both sides, a set of finishing bricks being moulded on purpose, with a notch to receive the iron bar. Cross bars are laid over these, one at each end and one or two between, having a bit turned down at the ends, which takes hold of the longitudinal bars, and keeps them from being thrust outwards either by the pressure of the arch or by the swelling in consequence of the heat. In fig. 4, A is the cross section of one of the long bars, and BC is part of one of the cross bars, and CD is the clench which confines the bar A. This precaution is chiefly necessary, because the contraction of the stove upwards obliges the walls of the other stories to bear a little on the arch of the fireplace. The building above is kept together in like manner by other courses of iron bars at every second return of the flue. The top of the stove is finished by a pretty thick covering of brick-work. The last passage for the air has a ring lining its upper extremity, and projecting an inch or two above it. The flat round it is covered with sand. When we would stop this passage, a cover shaped like a basin or cover for dishes at table is wheeled over it. The rim of this, resting on the sand, effectually prevents all air from coming through and getting up the vent. Access is had to this damper by a door which can be shut tight enough to prevent the heated air of the room from wasting itself up the vent. When the room is too warm, it may be very rapidly cooled by opening this door. The warm air rushes up with great rapidity, and is replaced by cool air from without.

In a stove of this kind the fire is kindled early in the morning, after which the stove-door is shut, and the air-aperture below left open for some time to blow it up; but in the course of a short time, to prevent the too rapid consumption of the fuel, the fire-door is opened, by which the draught is checked. In this way the combustion is allowed to go on, and the materials of the stove become warmed, after which the air-passages are shut, so as to prevent any abstraction of heat by the current that would otherwise be occasioned. The stove thus becomes a great mass of heated matter, which is gradually pouring warmth into the apartment during the whole of the day; and as the temperature of the surface never becomes very high, the impurities in the atmosphere are not decomposed, and consequently it is free from those offensive effluvia unavoidable when metal stoves are used. One precaution, however, is necessary in the management of these stoves, which does not apply to the metal ones; we must take care that when the air-apertures are closed, there is no back-draught to carry the products of combustion into the apartment, which might be attended with fatal consequences. These being almost free from smoke, give little or no warning of their presence, and when inhaled for some time, produce giddiness and lassitude, and in some cases a dangerous state of insensibility. Hence the necessity of allowing the fire to be burned down, or nearly so, before closing the air-aper- tures. Should the escape of the noxious gases into the room be suspected, the furnace-door and air-apertures must be opened, and a draught established through the stove, by kindling some wood shavings in it, the door and windows of the room being at the same time thrown open. This attention to back-draught is of course more necessary when charcoal or coke is used instead of coal, the products of combustion being then entirely free from smoke. When coal is used in stoves, the vent-tubes are apt to become choked with soot; but this may in a great measure be prevented by giving for a short time every day a brisk draught, by which the soot will be burned.

To prevent the injurious or disagreeable effects of the first kind of stoves, and at the same time to secure their advantages, numerous modifications have been proposed, not only with the view of preventing the too great heat of the external surface, but also of avoiding unnecessary expenditure of fuel. Perhaps the most important of these is Arnott's stove; the principle of which consists in allowing the fuel to burn very slowly, the admission of air for the combustion being regulated by an adjustment connected with the stove, and influenced by the degree of heat produced. Numerous forms and modifications of this stove are now in use, but in their general structure they are the same. The stove consists of a square or cylindrical box of iron, generally lined inside with a thick layer of fire-clay, and having a grating near the bottom, on which the fuel is burned; or the fuel may be contained in a small fire-box within the stove. Sometimes the fuel is burned within a hollow cylinder of fire-clay, in which case the stove is not lined with it. There is an ash-pit below for the reception of the ashes, and the products of the combustion are, as usual, carried off by a vent.

The principal feature of this stove is the contrivance by which the air is admitted for the combustion. When the stove-door or ash-pit door is open, the combustion is lively, and the fuel is soon consumed; but when these are shut, and they ought to be so made as to shut quite close, then air must be admitted otherwise, and this is done by the air-tube, furnished, as already stated, with a regulator.

The annexed figure represents the stove of its simplest form. ABCD is the outer casing of iron, in which there is the fire-box E, with its grating. Over the fire-box there is a dome, h, with a funnel to carry the products of combustion into the chimney; i is the stove-door, and g is the thermometer regulator, or adjustment by which the air is admitted for combustion. A great variety of these regulators has been described. Perhaps the best of these is represented by the annexed cut. ABC is a glass tube, shut at A, containing air from A to B, and filled with mercury from B to C. On the mercury at C is placed a float, from which there proceeds an upright rod D, kept steady by passing through a support at H. From this upright wire there descends another, FGH, terminated by the plate-valve F. LE is the air-tube of the stove. When the heat within is great, the air in the shut limb of the regulator at A is expanded, and forces up the mercury at C, raising the rods and plate-valve F, and thus bringing it near to, or in contact with, the mouth of the tube, by which more or less air is admitted to the stove, according to the heat within. If the combustion is proceeding too slowly, then the air in the tube A is not much expanded, consequently air is allowed to enter the stove more freely; but when, owing to this, the combustion becomes lively, and the temperature too high, then, by the elevation of the plate-valve, air enters in smaller quantity, and the temperature is moderated. Instead of the flat plate a conical valve is sometimes used, which, passing more or less into the mouth of the tube, allows the passage of less or more air; and sometimes this contrivance is restored to. ed is the wire raised by the movement of the mercury; ab is a plate hung on an axis within the air-tube, and acting like a throttle-valve. When the edge of the plate is turned to the current, air is admitted freely; but when the heat becomes high, then the mercury rises and makes the plate revolve, by which more and more of the area of the tube is shut, and consequently less and less air is admitted. Perhaps the simplest of these regulators is that here represented. abc is a bent tube shut at a, where it contains air, and open at c, where it is cup-shaped. The bent part at b is occupied by mercury. From c there proceeds a bent tube to supply air to the stove. When the heat within is great, the air in a is expanded, and forces the mercury up in c, and thus, bringing it in contact with the mouth, prevents the free admission of air to the stove. Numerous other contrivances of a similar nature have been recommended, all acting in the same way. Instead of these, the admission of air is sometimes regulated by a semicircular slide over the mouth of the air-tube, which may be placed so as to admit to the stove a greater or smaller quantity, as occasion requires.

The principle on which this stove operates, whatever modification of it may be used, is merely the slow combustion of the fuel, by which the stove itself is warmed; and there is thus a reservoir of heat to be communicated to the air in the apartment. As the quantity of fuel consumed is small, there is no necessity for frequent supply. When one of the ordinary dimensions for a room is in use, it will require to be supplied morning and evening, supposing that it is kept constantly burning, and the ashes removed once a day. These, instead of falling into the ash-pit, may be received in a box placed there for the purpose, which is taken out and emptied, so as to avoid dust in the apartment.

The quantity of fuel consumed must of course depend on circumstances. During the severe winter of 1836-37, Dr Arnott kept his library at a temperature varying from 60° to 63° of Fahrenheit, by about six pounds of coal a day; which, supposing the coal to be twenty shillings per ton, is at the daily expense of less than a penny.

Though the Arnott stove answers well the purpose for which it is intended, that is, economy of fuel, for most undoubtedly a room may be kept warm at a very moderate expense, yet it is liable to the objection already stated with regard to the unpleasant feeling consequent on the use of all stoves of the kind, and indeed with it more than others; for, owing to the very slight expenditure of fuel, there is little or no change in the atmosphere. It is generally allowed that a pound of coal on an average requires about 150 cubic feet of air for combustion; but as part of the air passes off without being acted on, 200 may be allowed, and this is a large allowance where the combustion is going on so slowly as in the stove. Now suppose the apartment in which the stove is placed to be fifteen feet long by twelve wide and eleven in height, its cubic contents are 1980 feet; and suppose six pounds of coal per day to be the consumption, each pound requiring about 200, that is in all 1200 feet of air for combustion. This quantity must pass through the stove and be carried off by the vent, so that in the course of twenty-four hours the atmosphere of the apartment is not once completely changed, and consequently renewed, by the direct influence of the fire. Hence it is that the apartment is so easily warmed; but it is this which necessarily renders it so unpleasant.

The stoves now described answer for small apartments. When the place to be heated is large, or when several apartments in the same building are to be warmed, a different kind is used. The heat in these cases is in general not communicated to the apartment or apartments by the direct influence of the stove, but by air heated by its external surface (of course not brought into contact with the fuel), and then conveyed by tubes or otherwise to the places to be warmed, on the principle already illustrated, that when heated it expands and ascends, and consequently rushes along the tubes, the supply being kept up by the constant flow of cold air upon the hot sides of the stove. Stoves for this purpose are made of iron or of brick-work, and sometimes of both. Fig. 9 represents a cast-metal stove of this kind. It may be considered as a double stove; an outer case, and a furnace or inner stove. The fuel is burnt in the inner stove, and the smoke produced during the process of combustion is carried off by a chimney, which passes through the top of the outer stove, and is conveyed to the outside of the building. The outer case includes not only the furnace or inner stove, but also a considerable space, occupied by the air of the atmosphere, which is freely admitted through a number of openings placed around it; and when any current of air is produced, it passes off from the space between the outer case and inner stove, and is conveyed by tubes through the body of the apartment. But we shall first describe the different parts of which the stove is composed, after which we shall be better able to understand its mode of operation.

Fig. 9 exhibits a perspective view of this stove. AB is the body, which is about three feet high, and of a circular form. C is a square pedestal, on which the stove is placed, and which contains the ash-pit. The height of the pedestal is about a foot, and it is nearly insulated by resting on the spherical supports a, also of cast iron. EEE are openings in front of the ash-pit, through which the air enters to support the combustion. These openings can be enlarged or diminished, or opened and shut, at pleasure. FF is the door of the furnace, through which the fuel is introduced. This door is attached to the inner furnace, and is double. It is one foot broad, and eleven inches high. GG is the chimney, which passes from the furnace within, through the outer case, and conveys the smoke out of the building. HH are openings in the outer case, and are eight in number, through which the air enters, and being heated, is greatly rarefied, and passes off through the funnel or pipe III. This pipe communicates only with the outer stove, and being shut at the end K, the air rushes out from the small tubes LL, inserted into the side of the pipe III, and thus mixes with the cold air of the apartment. The diameter of the outer case at the bottom is about two feet, and the diameter of the furnace within is about sixteen inches.

The length of the body of a church in which two stoves of the form and dimensions now described are erected, is about sixty feet, and the breadth is about forty-five feet. The tubes III are conveyed along the lower edge of the gallery, about half the length of the church. The fires are lighted about four or five o'clock on the Sunday morning during the earlier part of the cold season, but as the season advances it is usual to light them earlier. From this time till the congregation assembles for the afternoon service, the furnaces are constantly supplied with fuel. By this management the air in the church is kept comfortably warm during the coldest season of the year.

When only one apartment is to be heated by a stove of this kind, the stove ought to be placed in the apartment, because then it is supplied with air from it, which has been previously heated; whereas, when it is placed in another situation, the supply of heated air thrown into the apartment being derived from without, that within must be forced out to make way for it, and consequently at a great expenditure of fuel. When several apartments are to be warmed, suppose those of a dwelling-house, the stove ought to be situated in some part of the lobby, from which the tubes proceed to the rooms. Of course the size of the stove, and the distribution and size of the tubes, must depend on the supply of heat required.

There is one very great objection to the stove as now described. It has been already stated, that when air is projected against a red-hot surface, the impurities constantly floating in it are burned, and emit offensive effluvia; and this is generally the case with metallic hot-air stoves. Hence the necessity of having them so placed that they shall be supplied with air as free as possible from impurities. Instead of having the whole of the stove made of metal, the inner part is sometimes metal and the outer covering brick-work; in which case the chamber between them is generally larger than in the others, consequently the temperature does not become so high, and the objection urged against the metal stoves is in a great measure removed.

The annexed figure represents a vertical section of a stove of this kind. A is the stove of metal, with the door at B and ash-pit at C. DD is the vent, making a turn downwards, and carried into the chimney at E. FGHI is the outer casing of brick-work, completely enclosing the stove, and also the descending part of the vent; K is the opening for the admission of air to the hot chamber. L is the tube for carrying off the warm air, and from this it is conveyed by other tubes to the rooms to be heated. The stove from which this figure is taken is two feet and a quarter in width, six feet in height, and the sides three fourths of an inch in thickness. The brick casing is at the distance of six inches from the metal, and the descending vent within is six inches in diameter. It is used for warming a lecture-room thirty-five feet long, twenty-seven broad, and twenty high, also a large apartment thirty feet long, twenty-seven broad, and eighteen high, besides two smaller rooms and a staircase. The fire is kindled during winter at seven in the morning, and kept burning till four in the afternoon, when it is allowed to go out. The quantity of coal amounts daily, on an average, to rather less than half a hundredweight. The temperature of the air from the tubes varies from $120^\circ$ to $180^\circ$, according to the state of the fire. The temperature of the different apartments is kept at about $60^\circ$. When first erected, the supply of air for the hot chamber was brought from without; but now the air for the fuel and for the hot chamber are both taken from the apartment in which the stove is placed, which is generally at the temperature of $70^\circ$.

When the place to be heated is spacious, or when there are several apartments, it is customary to have the part of the stove immediately exposed to the fire lined with fire-brick, in order to prevent the direct action of the fuel on the metal; for in these instances the stove is much larger than that described, and consequently the fire more powerful. Its size must depend on the size and number of the apartments. In its general construction it resembles the other, consisting of the inner part for the fuel, with its grating, ash-pit, and vent, and of the outer casing, with the aperture for the supply of air to the chamber, and the tube or tubes for its transmission to the apartments.

The great advantage attending the use of stoves of this kind is, that they do not become so warm as to decompose the impurities in the air flowing into the hot chamber, and consequently there are no offensive effluvia generated. As the temperature is not so high as that of metal stoves, more of the warm air is requisite, by which there is a frequent renewal of that in the apartment, and the temperature throughout is more uniform than when a smaller quantity of hotter air is admitted. It has already been mentioned that the temperature of the air from the hot chamber of the stove described is from 120° to 180°. Perhaps this is higher than it ought to be. Many prefer having the stove so constructed that the temperature of the air which it throws into the apartment shall not exceed 70°. Of course, in this case, a much larger quantity of it is necessary. By throwing in a sufficient supply, that of the apartments may be maintained at about 60°. The mouths of the hot air-tubes are made to terminate as near the floor as possible, that the air may rise, and gradually mix with the atmosphere of the apartment.

In erecting stoves for the supply of hot air in this way, there are many circumstances to be considered, to which it would here be useless to allude. Much must depend on local situation, the size and number of the apartments, and the draught through them. These considerations must be left to the skill and ingenuity of the workman, who must be guided by keeping in view the general principles which determine the supply of air to the furnace, the ascensional force of the column of warm air to be conveyed through the tubes, the manner in which the apartments to be warmed are disposed, &c.

In heating buildings by these stoves when they are not in constant use, as is the case in churches, the time for keeping them going must also depend on circumstances. If the object is merely to throw in a supply of warm air, then the stove must be kindled a few hours before the apartment is to be occupied, especially if it is not expected that there shall be much change in its atmosphere; but if the object is to keep the place warm, then the stove ought to be kept in constant use, so as to be constantly throwing in warm air. During night the fire can be damped, and it is again made brisk when required. This is much better than allowing the fire to go out, because in again heating the stove to the requisite temperature, there must be a considerable waste of fuel. After the fire is in good condition, and the stove well heated, the combustion should be allowed to proceed slowly, which is accomplished by the proper regulation of the draught.