Principles. pital and base, is 16 modules, and the height of the entablature four; the latter of which being divided into eight parts, two of them are for the architrave, three for the frieze, and three for the cornice.

In most of the antiques, the Doric column is executed without a base. Vitruvius likewise makes it without one; the base, according to him, having been first employed in the Ionic order, in imitation of the sandal of a woman's foot. Scamozzi blames this practice and most of the modern architects are of his opinion.

In the profile of the theatre of Marcellus, the frieze is enriched with bulks and rosettes; the architrave consists only of one fascia and a fillet; the drops are conical; the metope is enriched with a bull's skull, adorned with a garland of beads, in imitation of those on the temple of Jupiter Tonans at the foot of the Capitol. In some antique fragments, and in a great many modern buildings, the metopes are alternately adorned with ox-skulls and paterae. But they may be filled with any other ornaments, according to the destination of the building.

The Ionic Order

Pl.XXXVIII. Is of a more slender make than the Doric or Tuscan; its appearance is simple, yet graceful and majestic; its ornaments are few; so that it has been compared to a sedate matron, in decent, rather than magnificent, attire.

Among the ancients, the form of the Ionic profile appears to have been more positively determined than that of any other order; for, in all the antiques at Rome (the temple of Concord excepted), it is exactly the same.

The modern artists have likewise been unanimous in their opinions; all of them, excepting Palladio and his imitators, having employed the dentil, cornice, and the other parts of the profile, nearly as they are found in the Colliseum, the temple of Fortune, and the theatre of Marcellus.

The height of the Ionic column is 18 modules, and that of the entablature 4½, or one quarter of the height of the column, as in the other orders, which is a trifle less than in any of the antique Ionics. In all the antiques, the base is Attic; and the shaft of the column may either be plain, or fluted with 24 flutings, or 20 only, as in the temple of Fortune. The plan of the flutings may be a trifle more than a semicircle, as in the forum of Nerva, because they then appear more distinct. The fillets, or intervals between them, must not be broader than one third of the breadth of a fluting, nor narrower than one fourth. The ornaments of the capital must correspond with the flutings of the shaft; and there must be an ove above the middle of each fluting. The volutes ought to be traced according to Mr Goldman's method, which is as follows:

Plate XXIX. fig. 9. Draw the cathetus F C, whose length must be 15 minutes, or one fourth of a module; and, from the point C, describe the eye of the volute A E B D, of which the diameter is to be 6½ minutes; divide it into four equal sectors by the diameters A B, D E. Bisect the radius C A, C B, in 1 and 4; and on the line 1, 4, construct a square 1, 2, 3, 4. From the centre C, to the angles 2, 3, draw the diagonals C 2, C 3, and divide the side of the square 1, 4, into 6 equal parts, at 5, 9, C, 12, 8.

Then through the points 5, 9, 12, 8, draw the lines 5, 6, 9, 10, 12, 11, 8, 7, parallel to the diameter E D, which will cut the diagonals in 6, 7, 10, 11; and the points 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, will be the centres of the volute. From the first centre 1, with the distance 1 F, describe the quadrant F G; from the second centre 2, with the distance 2 G, describe the quadrant G H; and, continuing the same operation from all the 12 centres, the contour of the volute will be completed.

Fig. 10. The centres for describing the fillet are found in this manner. Construct a triangle, of which the side A F is equal to the part of the cathetus contained between A F and the side F V, equal to C t; place the distance F S from F towards A, equal to F S the breadth of the fillet, and through the point S draw the line S T, which will be to C t in the same proportion as A S is to A F; place this line on the diameter of the eye A B; divide it into three equal parts; and, through the points of division, draw lines parallel to the diameter E D, which will cut the diagonals C 2, C 3, and you will have twelve new centres, from whence the interior contour of the fillet may be described, in the same manner as the exterior one was from the first centres.

Of the Corinthian Order.

The proportions of this order are extremely delicate. It is divided into a great variety of members, and enriched with a profusion of ornaments. Scamozzi calls it the virginal order; and indeed it has all the delicacy in its make, and all the delicacy in its dress, peculiar to young girls.

The most perfect model of the Corinthian order is generally allowed to be in the three columns in the Campo Vaccino at Rome, the remains, as it is thought, of the temple of Jupiter Stator.

The Corinthian column should be 20 modules high, and the entablature 5; which proportions are a medium between those of the Pantheon and the three columns. The base of the column may be either Attic or Corinthian: They are both beautiful. If the entablature be enriched, the shaft may be fluted. The flutings may be filled, to one third of their height, with cabling, as in the inside of the Pantheon; which will strengthen the lower part of the column, and make it less liable to injury.

In most of the antiques at Rome, the capital of this order is enriched with olive-leaves; the acanthus being seldom employed but in the Composite. De Cordemoy, however, prefers the acanthus.

The divisions of the entablature bear the same proportions to each other, as the Tuscan, Ionic, and Composite orders.

The Composite

Is, strictly speaking, only a species of the Corinthian; and therefore retains, in a great measure, the same character.

It does not appear that the ancients affected any particular form of entablature to this order. Sometimes the cornice is entirely plain, as in the temple of Bacchus; at others, as in the arch of Septimius Severus, it is enriched with dentils differing very little from the Ionic; and in the arch of Titus, there are both dentils and Principles, and modilions; the whole form of the profile being the same with the Corinthian, as executed in the antiques at Rome.

The modern architects have varied more in this than in any other order, each following the bent of his own fancy.

The height of the Composite column, and parts of the entablature, is the same with that of the Corinthian. The foot of the leaves of the capital ought not to project beyond the upper part of the shaft. The different bunches of leaves should be strongly marked; the sprigs which arise between the upper ones should be kept flat upon the vase; and the ornaments of the volutes must not project beyond the fillets that inclose them.

**Chap. II. Of Pilasters.**

These differ from columns only in their plan; which is a square, as that of columns is round. Their bases, capitals, and entablatures, have the same parts, with the same heights and projections, as those of columns; they are also distinguished in the same manner, by the names of Tuscan, Doric, Ionic, Corinthian, and Composite.

The column is undoubtedly more perfect than the pilaster. However, they may be employed with great propriety on many occasions. Some authors declaim against pilasters, because, according to them, they do not admit of diminution. But this is a mistake; there are many instances, in the remains of antiquity, of their being diminished. Scamozzi always gave his pilasters the same diminution as his columns; Palladio and Inigo Jones have likewise diminished them in many of their buildings.

Pilasters are employed in churches, galleries, halls, and other interior decorations, to save room; for, as they seldom project beyond the solid wall above one quarter of their diameter, they do not occupy near so much space as columns. They are likewise used in exterior decorations; sometimes alone, instead of columns, on account of their being less expensive; and sometimes they accompany columns, being placed behind them to support the architraves, where they enter the building, as in the Pantheon at Rome; or, in the same line with them, to fortify the angles, as in the portico of Septimius.

When pilasters are used alone, they should project one quarter of their diameter beyond the walls. When placed behind columns, especially if they be very near them, they need not project above one eighth of their diameter. But, when placed on a line with columns, their projection must be regulated by that of the columns; and consequently, it can never be less than a semidiameter, even when the columns are engaged as much as possible.

The shafts of pilasters are frequently adorned with flutings, in the same manner as those of columns; the plan of which may be a trifle more than a semicircle: their number must be seven on each face, which makes them nearly of the same size with those of columns. The intervals, or fillets, must either be one third or one fourth of the fluting in breadth.

The capitals of pilasters are profiled nearly in the same manner as those of columns.

**Chap. III. Of Attics.**

These very properly follow the pilasters; being nothing more than square pillars with their cornices. They had their origin in Athens, where it was for many ages a rule in building to conceal the roof. For this purpose, nothing served so well as a kind of low or little order ranged in a continued line, singly, or with the interruption of balusters; which rising above the rest of the work and before the roof, hid it perfectly, and placed something agreeable in view. The place of attics, therefore, is at the uppermost extremity of a building, to which they serve as a crown, or very properly make a finishing for the other orders when they have been used in the structure. They must never stand under anything except such ornaments as are placed at the very top. These Attics should never exceed in height one third of the height of the order on which they are placed, nor be less than one quarter of it. The base, dye, and cornice, of which they are composed, may bear the same proportions to each other as those of pediments do; and the base and cornice may be composed of the same mouldings as those of pediments. Sometimes the Attic is continued throughout; at others, it projects, and forms a pilaster over each column of the order. The breadth of this pilaster is seldom made narrower than the upper diameter of the column below it, and never broader. Its projection may be equal to one quarter of its breadth.

**Chap. IV. Of Persians, Caryatides, and Termi.**

Besides columns and pilasters, it is sometimes customary to employ representations of the human figure, to support entablatures in buildings. The male figures are called Persians; and the female, Carian, or Caryatides.

The Persians are so called from a victory gained over the Persians by Pausanias, who having brought Persian spoils and trophies to the Athenians, they fixed upon Persian figures for those which should support entablatures, and thus kept in mind that there were once Persian slaves in Athens. To represent these conquered people in the lowest state possible, they loaded them with the heaviest entablatures, viz. that of the Doric order. In process of time, however, other figures besides those of Persians were introduced, and other entablatures put over them; but the name was still retained.

The proper Caryatides are women dressed in long robes, after the Asiatic manner; and the origin of the device was as follows.—The Carian had been long at war with the Athenians; but being at length totally vanquished, their wives were led away captives; and, to perpetuate the memory of this event, trophies were erected, in which figures of women dressed in the Caryatic manner, were used to support entablatures like the Persians; and though other female figures were afterwards used in the same manner, the name of Caryatides was always retained.

The ancients made frequent use of Persians and Caryatides, and delighted in diversifying them a thousand ways. The modern artists have followed their example; and there is a great variety of compositions of this kind to be met with in different parts of Europe.

Indecent attitudes, distorted features, and all monstrous Part I. ARCHITECTURE.

Principles. Strous productions, ought to be avoided, of which there are many examples in Gothic buildings. On the contrary, the attitudes should be simple and graceful, the countenance always pleasing, though varied and strongly marked agreeable to the nature of the object represented.

The Caryatides, or female figures, should never much exceed the human size. But the Persians, or male figures, may be of any size; and the larger the better, as they will strike the beholder with the greater awe and astonishment. Persians may be used with propriety in arcades, galleries of armour, &c. under the figures of captives, heroic virtues, &c. Their entablature ought to be Doric, and bear the same proportion to them as to columns of the same height. The entablature for Caryatides ought to be either Ionic or Corinthian, according as the character of the figures is more or less delicate.

Terminals are sometimes employed, instead of Persians or Caryatides, to support the entablatures of monuments, chimney-pieces, and such like compositions. These figures owe their origin to the stones used by the ancients to mark the limits of particular possessions. Numa Pompilius, to render these inviolable, consecrated the terminals into a deity, and instituted festivals and sacrifices to his honour. In a short time, what was formerly only large upright stones, were represented in human shape; and afterwards introduced as ornaments to temples and other buildings. The terminals are now principally used as ornaments for gardens and fields.

CHAP. V. Of Pedestals.

Most writers consider the Pedestal as a necessary part of the order, without which it is not complete. It is indeed a matter of little importance whether it be considered in that light, or as a distinct composition: we shall therefore treat of a pedestal as a distinct body, having no more connection with the order than an attic, a basement, or any other part with which it may on some occasions be associated.

A pedestal consists of three principal parts; the base, the dye, and the cornice. The dye is always nearly of the same figure; being constantly either a cube or a parallelipedon: but the base and cornice are varied and adorned with more or fewer mouldings, according to the simplicity or richness of the composition in which the pedestal is employed. Hence pedestals are, like columns, distinguished by the names of Tuscan, Doric, Ionic, Corinthian, and Composite.

Some authors are adverse to pedestals, and compare a column raised on a pedestal to a man mounted on stilts; imagining that they were introduced merely from necessity, and for want of columns of a sufficient length. It is indeed true, that the ancients often made use of artifices to lengthen their columns; as appears by some that are in the Baptistery of Constantine at Rome; the shafts of which, being too short for the building, were lengthened and joined to their bases by an undulated sweep, adorned with acanthus leaves. Nevertheless, there are many occasions where pedestals are evidently necessary; and some in which the order, were it not so raised, would lose much of its beautiful appearance. Thus, in the inside of churches, if the columns that support the vault were placed immediately on the ground, the seats would hide their bases and a good part of their shafts; and, in the theatres of the ancients, if the columns of the scene had been placed immediately on the stage, the actors would have hid a part of them from the audience. In interior decorations, a pedestal diminishes the parts of the order, which otherwise might perhaps appear too clumsy, and hath the advantage of placing the column in a more favourable view, by raising its base nearer the level of the spectator's eye. In a second order of arcades, there is no avoiding pedestals; as without them it is impossible to give the arches any tolerable proportion.

With regard to the proportion that pedestals ought to bear to that of the columns they support, it is by no means fixed. Both the ancients and moderns vary greatly on this head. Vignola's proportions are generally reckoned the best. He makes his pedestals in all the orders of the same height, viz. one third of the column; and as their breadth of course increases or diminishes in the same degree as the diameters of their respective columns do, the character of the order is always preserved, which, according to any other method, is impossible.

As to the divisions of the pedestal; if the whole height be divided into nine parts, one of them may be given to the height of the cornice, two to the base, and the six remaining to the dye. The breadth of the dye is always made equal to that of the plinth of the column. The projection of the cornice may be made equal to its height; and the base being divided into three parts, two of them will be for the height of the plinth, and one for the mouldings, whose projection must be less than that of the cornice. These measures are common to all pedestals. See Plate XXXII.

CHAP. VI. Of Intercolumniations.

Columns are either engaged, or insulated; and, when insulated, are either very near the wall, or at a considerable distance from it. Engaged columns, or such as are near the walls of a building, are not limited in their intercolumniations, as these depend on the breadths of the arches, windows, niches, or other decorations placed between the columns. But columns that are entirely detached, and perform alone the office of supporting the entablature, as in peristyles, porches, and galleries, must be near each other, for the sake both of real and apparent solidity.

The intercolumniations among the ancients were various. Those used in the Ionic and Corinthian orders were the pycnolyle, of which the interval was equal to one diameter and a half of the column; the fyllyle, whose interval was equal to two diameters; the euclyle, to two and a quarter; the diafyle to three, and the arwofyle to four. In the Doric order, they used other intercolumniations, regulating them by the triglyphs, one of which was always placed directly over the middle of each column; so that they were either fyllyle, monotriglyph, of one diameter and a half; diafyle, of two diameters and three quarters; or arwofyle, of four diameters; and the Tuscan intervals were very wide, some of them being above seven diameters, which was very practicable, as the architraves were of wood.

Among these different intercolumniations, the pycnolyle and fyllyle are too narrow; for although the ancients ancients made frequent use of them, that ought rather to be ascribed to necessity than choice. For, as the architraves were composed of single stones, extending from the middle of one column to the middle of another, it would have been difficult, especially in large buildings, to find blocks of sufficient length for diastyle intervals. With regard to the arcuflute and Tuscan intercolumniations, they are by much too wide, and can only be used in rustic buildings, where the architraves are of wood; neither is the diastyle sufficiently foiled in large compositions. The euflute is a medium between the narrow and broad intervals; and, being at the same time both spacious and foild, hath been preferred to any of the rest by the ancients as well as the moderns.

Vignola observed nearly the same proportion in all his intercolumniations; which practice, though condemned by several writers, is certainly preferable to any other; as it preserves the character of each order, and maintains in all of them an equal degree of real solidity. Setting aside therefore the pycnoflute and euflute dispositions on account of their want of space, and the arcuflute for its deficiency in point of strength, it may be established, that the diastyle and euflute intercolumniations (the latter of which, on most occasions, ought to have the preference) may be employed in all the orders without distinction, excepting the Doric; in which the most perfect interval is ditriglyph; neither the monostiglyp, nor the arcuflute, being to be suffered but in cases of necessity.

Sometimes, on account of the windows, doors, niches, and other decorations, which correspond with the intercolumniations of the euflute, or gallery, it is not possible to make the intervals so narrow as euflute, or even as diastyle: wherefore the moderns, authorized by some few examples of the ancients, where grouped columns are employed, have invented a manner of disposing them, called by Perrault arcuflute, which admits of a larger interval, without any detriment to the apparent solidity of the building. This kind of disposition is composed of two euflute intercolumniations; the column that separates them being approached towards one of those at the extremities, sufficient room only being left between them for the projection of the capitals; so that the great space is three diameters and a half wide, and the little one half a diameter.

In peristyles, galleries, or porticos, all the intercolumniations must be equal; but in a logio, or porch, the middle interval may be broader than the others, by a triglyph or modillion, or three or four dentils; unless the columns at the angles be coupled, or grouped with pilasters; in which case, all the intervals should be of the same dimensions.

When buildings are very small, as is frequently the case in temples and other inventions used for ornamenting gardens, the intercolumniations may be broader, in proportion to the diameter of the columns, than usual; because, when they are nearer each other than three feet, there is hardly room for a bulky person to pass between them.

**CHAP. VII. Of Arches.**

Arches are not so magnificent as colonnades; but they are more solid and less expensive. They are proper for triumphal entrances, gates of cities, of palaces, of gardens, and of parks, and in general for all openings that require an extraordinary breadth.

There are various manners of adorning arches. Sometimes their piers are rutilated; sometimes they are adorned with pilasters, termini, or caryatides; and sometimes they are made sufficiently broad to admit niches or windows. The circular part of the arch is either surrounded with rustic key-stones, or with an archivolt enriched with mouldings; which, in the middle, is sometimes interrupted by a console, a mull, serving at the same time as a key to the arch, and as a support to the architrave of the order. The archivolt is sometimes supported by an impost, at the head of the pier; and at others by columns placed on each side of it, with a regular entablature, or architrave and cornice. There are likewise instances of arcades without piers, the arches being turned on single columns, as in the temple of Faunus at Rome, &c. This practice, however, ought to be seldom imitated, as it is neither solid nor handsome.

When arches are large, the key-stone should never be omitted, but cut in the form of a console, and carried close under the soffit of the architrave, which, on account of its extraordinary length, requires a support in the middle. The imposts of arches should never be omitted; at least, if they be, a platform ought to supply their place. If columns are employed without pedestals in arcades, they should always be raised on a plinth. In all arches, the circular part ought not to spring immediately from the impost, but take its rise at such a distance above it as is necessary in order to have the whole curve seen at the proper point of view.

The void or aperture of arches should never be higher, nor much lower, than double their breadth; the breadth of the pier should exceed two thirds, nor be less than one third, of the breadth of the arch; and the angular pier ought to be broader than the others, by one half, one third, or one fourth; the impost should not be more than one seventh, nor less than one ninth of the aperture; and the archivolt must not be more than one eighth, nor less than one tenth of it. The breadth of the console must, at the bottom, be equal to that of the archivolt; and its sides must be drawn from the centre of the arch: the length of it must not be less than one and a half of its smallest breadth, nor more than double. The thickness of the pier depends on the breadth of the portico; for it must be strong enough to resist the pressure of its vault. But with regard to the beauty of the building, it should not be less than one quarter of the breadth of the arch, nor more than one third. These are the general dimensions of arches.

**CHAP. VIII. Of Orders above Orders.**

When, in a building, two or more orders are employed, one above another, the laws of solidity require the strongest should be placed lowest. Hence the Tuscan must support the Doric, the Doric the Ionic, the Ionic the Composite or Corinthian, and the Composite the Corinthian.

This rule, however, is not always strictly adhered to. Most authors place the Composite above the Corinthian. There are likewise examples where the same order Part I.

Principles.

order is repeated, as in the theatre of Statilius Taurus, and the Coliseum; and others, where an intermediate order is omitted, and the Ionic placed on the Tuscan, or the Corinthian on the Doric. But none of these practices ought to be imitated.

In placing columns above one another, the axis of all the columns ought to correspond, or be in the same perpendicular line, at least in front.

With regard to the proportions of columns placed above each other, Scamozzi's rule, That the lower diameter of the superior column should constantly be equal to the upper diameter of the inferior one, is universally esteemed the best, and gives all the columns the appearance of one long tapering tree, cut into several pieces. According to this rule, the Doric column will be to the Tuscan, as $13\frac{1}{2}$ to $14$; the Ionic to the Doric, as $15$ to $16$; the Composite or Corinthian to the Ionic, as $16\frac{1}{2}$ to $18$; and the Corinthian to the Composite, as $16\frac{1}{2}$ to $20$.

In Britain there are few examples of more than two stories of columns in the same aspect: and though in Italy, and other parts of Europe, we frequently meet with three, and sometimes more; yet it is a practice by no means to be imitated; for there is no possibility of avoiding many striking inconsistencies, or of preserving the character of each order in its intercolumnial decorations.

Chap. IX. Of Basements.

Instead of employing several orders one above the other in a composition, the ground-floor is sometimes made in the form of a basement, on which the order that decorates the principal story is placed. The proportion of these basements is not fixed, but depends on the nature of the rooms on the ground-floor. In the palace of the Porti in Vicenza, the height of the basement is equal to that of the order. In some buildings, its height exceeds two thirds of that of the order; and, in others, only half the height of the order. It is not, however, advisable to make the basement higher than the order it supports; neither should it be lower than one half of the order.

The usual method of decorating basements is with rustics of different kinds. The best, where neatness and finishing is aimed at, are such as have a smooth surface. Their height, including the joint, should never be less, nor much more, than half a module of the order placed on the basement. Their figure may be from a square to a lefovaltera; and their joints may be either square or chamfered. The square ones should not be broader than one eighth of the height of the rustic, nor narrower than one tenth; and their depth must be equal to their breadth; those that are chamfered must form a rectangle; and the breadth of the whole joint may be from one fourth to one third of the height of the flat surface of the rustic.

Chap. X. Of Pediments.

Pediments, among the Romans, were used only as coverings to their sacred buildings, till Cæsar obtained leave to cover his house with a pointed roof, after the manner of temples. In the remains of antiquity we meet with two kinds of pediments, the triangular and the circular. The former of these are promiscuously applied to cover small or large bodies: But the latter, being of a heavier figure, are never used but as coverings to doors, niches, windows, or gates.

As a pediment represents the roof, it should never be employed but as a finishing to the whole composition.

The ancients introduced but few pediments into their buildings, usually contenting themselves with a single one to adorn the middle or principal part. But some of the moderns, and particularly the Italians, have been so immoderately fond of them, that their buildings frequently consist of almost nothing else.

The girder being a necessary part in the construction of a roof, it is an impropriety to intermit the horizontal entablature of a pediment, by which it is represented, to make room for a niche, an arch, or a window.

In regular architecture, no other form of pediments can be admitted, besides the triangular and circular. Both of them are beautiful; and when a considerable number of pediments are introduced, as when a range of windows are adorned with them, these two figures may be used alternately, as in the niches of the Pantheon, and in those of the temple of Diana at Nîmes.

The proportion of pediments depends upon their size; for the same proportions will not do in all cases.

When the base of the pediment is short, its height must be increased; and when the pediment is long, the height must be diminished. The best proportion for the height is from one fifth to one fourth of the base, according to the extent of the pediment, and the character of the body it covers. The materials of the roof must also be attended to; for if it be covered with tiles, it will be necessary to raise it more than one quarter of the base, as was the custom of the ancients in their Tuscan temples.

The tympan is always on a line with the front of the frieze; and, when large, admits of various ornaments.

Chap. XI. Of Balustrades.

Balustrades are sometimes of real use in buildings; and at other times they are only ornamental. Such as are intended for use, as when they are employed in stair-cases, before windows, or to inclose terraces, &c. must always be nearly of the same height; never exceeding three feet and a half, nor ever less than three. But those that are principally designed for ornament, as when they finish a building, should be proportioned to the architecture they accompany: and their height ought never to exceed four fifths of the height of the entablature on which they are placed; nor should it ever be less than two thirds thereof, without counting the zoccolo, or plinth, the height of which must be sufficient to leave the whole balustrade exposed to view.

The best proportion for balustrades is to divide the whole given height into thirteen equal parts; eight of these for the height of the balustrade, three for the base, and two for the cornice or rail; or into fourteen, (if it be required to make the baluster less), giving eight parts to the balustrade, four to the base, and two to the rail. One of these parts may be called a module; and being divided into nine minutes, may serve to determine the dimensions of the particular members.

In ballustrades, the distance between two balusters should not exceed half the diameter of the baluster measured in its thickest part, nor be less than one third of it.

The breadth of the pedestals, when they are placed on columns or pilasters, is regulated by them; the dye never being made broader than the top of the shaft, nor much narrower; and when there are neither columns nor pilasters on the front, the dye should not be much lower than a square, and seldom higher. On stairs, or any other inclined planes, the same proportions are to be observed as on horizontal ones.

**Chap. XII. Of Gates, Doors, and Piers.**

There are two kinds of entrances, viz. doors and gates. The former serve only for the passage of persons on foot; but the latter likewise admit horsemen and carriages. Doors are used as entrances to churches and other public buildings, to common dwelling houses, and apartments: And gates serve for inlets to cities, fortresses, parks, gardens, palaces, &c. The apertures of gates being always wide, they are generally made in the form of an arch, that figure being the strongest. But doors, which are generally of small dimensions, are commonly parallelograms, and closed horizontally.

The general proportion for the apertures, both of gates and doors, whether arched or square, is, that the height be about double the breadth.

The most common, and indeed almost the only ornaments for gates are the piers by which they are supported, and which were originally no more than bare posts into which the hinges of the gate were driven. Though this, however, is the only proper use of piers, it must be concealed as much as possible, and they must seem as if placed there only for ornament. As they are to be fixed to the wall before the house, so they must also be proportioned to it; and as they are to be seen in the same view with the front of the house, their correspondence with it is equally necessary. They are to be placed on a plinth, and something must be allowed by way of ornament and finishing at the top. All the luxuriance of fancy may be employed in the decoration of piers: but it will be proper to observe this general rule, that the pier being an inferior building, it must never be richer than the front of the house. If, for instance, the front of the house is ornamented with columns of the Doric order, the Ionic must not be used in the piers; and it will be found better to omit columns altogether, than to make use of the Tuscan order for piers in any case. If the Ionic or Corinthian orders are employed in the front of the house, the Doric or Ionic may be used with propriety in the piers. One piece of ornament is almost universal in piers, namely, a niche with its seat, made as if for the convenience of weary travellers. On this account, it will be proper to raise the columns on pedestals, because the continued moulding from their cap will be a good ornament under the niche. The base of the columns ought always to be the attic.

Inside-doors, however small the building may be, should never be narrower than two feet nine inches; nor should they ever, in private houses, exceed three feet six inches in breadth, which is more than sufficient to admit the bulkiest person. Their height should at least be six feet three or four inches; otherwise a tall person cannot pass without stooping. In churches, palaces, &c. where there is a constant ingress and egress of people, the apertures must be larger. The smallest breadth that can be given to a gate is 8½ or 9 feet, which is but just sufficient for the passage of a coach.

Plate XXXIII. fig. 1. Is a rustic door, composed by the celebrated Vignola, in which the aperture occupies two thirds of the whole height, and one half of the whole breadth; the figure of it being a double square. The rustics may be either smooth or hatched; their joints must form a rectangle, and the breadth of each joint may be one third, or two sevenths, of the vertical surface of a rustic. The joints of the claveaux, or key-stones, must be drawn to the summit of an equilateral triangle, whose base is the top of the aperture. The architrave surrounding the aperture may be composed either of a large ogee and fillet, or of a platband and fillet. Its whole breadth must be one tenth of the breadth of the aperture; the remaining part of each pier being for the rustics. The entablature is Tuscan: the cornice is to be one fifteenth of the whole height of the door; and what remains below it being divided into 21 equal parts, the two uppermost of them will be for the frieze and architrave, and the remaining 19 for the rustics and plinth at the foot of the door.

Fig. 2. Is a disposition of Michael Angelo's. The windows of the Capitol at Rome are of this kind; and Sir Christopher Wren hath executed doors of the same kind under the semicircular porches in the flanks of St Paul's. The figure of the aperture may be a double square; the architrave one fifth of the breadth of the aperture; and the whole entablature one quarter of its height. The front of the pilasters or columns, on each side, must be on a line with the fascia of the architrave; and their breadth must be a semidiameter.

Fig. 3. Is likewise a design of Vignola's. It is of the Corinthian order, and executed in the Cancellaria at Rome. The height is equal to double its breadth; and the whole ornament at the top is equal to one third of the height of the aperture. The architrave is in breadth one fifth of the breadth of the aperture; and the pilasters that support the consoles are half as broad as the architrave. The whole is well imagined, but rather heavy; and it will be best to reduce the architrave to one sixth of the aperture, diminishing the entablature proportionally.

Fig. 4. Is a design of Serlio's. The aperture may be either twice as high as broad, or a trifle less. The diameter of the columns may be equal to one quarter of the breadth of the aperture; and their height may be from eight diameters to eight and a half. The entablature must be somewhat less than one quarter of the height of the columns; and the height of the pediment may be one quarter of its base.

Fig. 5. Is a door in the salon of the Farnese at Rome, designed by Vignola. The aperture forms a double square. The entablature is equal to three elevenths of its height, the architrave being one of these elevenths; and the whole ornaments on the sides, con- Part I. ARCHITECTURE.

Principles. The first consideration with regard to windows is their size, which varies according to the climate, the destination of the building, &c. In Britain, the windows of the smallest private houses are commonly from 3 to 3½ feet broad; and being generally twice their breadth in height, or somewhat more, in the principal apartments, they generally rise to within a foot or two of the ceilings of the rooms, which are frequently no higher than 10 feet, and at most 12 or 13. But, in more considerable houses, the apartments are from 15 to 20 feet high, and sometimes more; and in these the windows are from 4 to 5 and 5½ feet broad, and high in proportion. These dimensions are sufficient for dwelling-houses of any size in this country; when they are larger, they admit too much of the cold air in winter. But churches, and other buildings of that kind, may have larger windows, proportioned to the size of the structures.

The proportions of the apertures of windows depend upon their situation. Their breadth in all the stories must be the same; but the different heights of the apartments make it necessary to vary the height of the windows likewise. In the principal floor, it may be from 2½ of the breadth to 2½, according as the rooms have more or less elevation. In the ground-floor, where the apartments are lower, the apertures of the windows seldom exceed a double square; and, when they are in a rustic basement, they are frequently made much lower. The height of the windows of the second floor may be from 1½ of their breadth to 1½; and Attics and Mezzanines may be either a perfect square, or somewhat lower.

The windows of the principal floor are generally how ornately enriched. The simplest method of adorning them mentioned, is, with an architrave surrounding the aperture, and crowned with a frieze and cornice. The windows of the ground-floor are sometimes left entirely plain, without any ornament; and at others they are surrounded with rustics, or a regular architrave with a frieze and cornice. Those of the second floor have generally an architrave carried entirely round the aperture; and the same is the method of adorning Attic and Mezzanine windows; but the two last have seldom either frieze or cornice; whereas the second-floor windows are often crowned with both.

The breasts of all the windows on the same floor should be on the same level, and raised above the floor from two feet nine inches to three feet six inches at the very most. When the walls are thick, the breasts should be reduced under the apertures, for the convenience of looking out. In France, the windows are frequently carried quite down to the floor. When the building is surrounded with gardens, or other beautiful objects, this method renders the rooms exceeding pleasant.

The interval between the apertures of windows depends in a great measure on their enrichments. The breadth of the aperture is the least distance that can be between them; and twice that breadth should be the largest in dwelling-houses; otherwise the rooms will not be sufficiently lighted. The windows in all the stories of the same aspect must be placed exactly above one another.

Plate XXXIV. fig. 1. Is a design of P. Lefcot, abbot of Clagny, executed in the old Louvre at Paris. The apertures may be a double square, or a trifle more; the architrave from one sixth to one seventh of the breadth of the aperture; the pilaster is equal to that breadth, when the architrave is narrow; or less, by one quarter, or one fifth, when it is broad. The whole entablature should not exceed one quarter of the height of the aperture, nor be much lower. The consoles may be equal in length to half the breadth of the aperture at most, and to one third of it at least.

Fig. 2. Is a design of Palladio's, executed at the Chiericato in Vicenza: its proportions are not much different. different from the following. The plat-band that supports the window is equal to the breadth of the architrave.

Fig. 3. Is likewise a design of Palladio's, executed by him in many of his buildings. The aperture is a double square. The breadth of the architrave is one sixth of the breadth of the aperture; and the frieze and cornice together are double the height of the architrave. The breadth of the consoles is two thirds of the breadth of the architrave.

Fig. 4. Is a design of Ludovico da Cigoli; and executed in the ground-floor of the Rannunzini palace at Florence.

Fig. 5. Is a design of Inigo Jones, executed at the Banqueting-house. The aperture may be a double square; the architrave may be one sixth of its breadth; the whole entablature one quarter of its height; and the breadth of the consoles two thirds of the breadth of the architrave.

Fig. 6. Is a design of M. Angelo Buonorati, executed at the Farnese.

**CHAP. XIV. Of Niches and Statues.**

It hath been customary, in all ages, to enrich different parts of buildings with representations of the human body. Thus the ancients adorned their temples, baths, theatres, &c. with statues of their deities, heroes, and legislators. The moderns still preserve the same custom, placing in their churches, palaces, &c. statues of illustrious persons, and even groups composed of various figures, representing occurrences collected from history, fables, &c. Sometimes these statues or groups are detached, raised on pedestals, and placed contiguous to the walls of a building, or in the middle of a room, court, or public square. But they are most frequently placed in cavities made in the walls, called niches. Of these there are two sorts: the one formed like an arch in its elevation, and semicircular or semi-elliptical in its plan; the other is a parallelogram both in its plan and elevation.

The proportion of both these niches depends on the characters of the statues, or the general form of the groups placed in them. The lowest are at least a double square in height; and the highest never exceed 2½ of their breadth.

With regard to the manner of decorating them, when they are alone in a composition, they are generally inclosed in a panel, formed and proportioned like the aperture of a window, and adorned in the same manner. In this case, the niche is carried quite down to the bottom; but on the sides and at the top, a small space is left between the niche and the architrave of the panel. And when niches are intermixed with windows, they may be adorned in the same manner with the windows, provided the ornaments be of the same figure and dimensions with those of the windows.

The size of the statues depends on the dimensions of the niches. They should neither be so large as to have the appearance of being rammed into the niches, as in Santa Maria Maggiore at Rome; nor so narrow as to seem lost in them, as in the Pantheon. The distance between the outline of the statue and side of the niche should never be less than one third of a head, nor more than one half, whether the niche be square or arched; and when it is square, the distance from the top of the head to the ceiling of the niche should not be greater than the distance on the sides. Statues are generally raised on a plinth, the height of which may be from one third to one half of a head; and sometimes, where the niches are large, the statues may be raised on small pedestals.

The character of the statue should always correspond with the character of the architecture with which it is surrounded. Thus, if the order be Doric, Hercules, Jupiter, Mars, Esculapius, and all male statues representing beings of a robust and grave nature, may be introduced; if Ionic, then Apollo, Bacchus, &c.; and if Corinthian, Venus, Flora, and others of a delicate nature, should be employed.

**CHAP. XV. Of Chimney-pieces.**

Among the ancients, there are very few examples of chimney-pieces to be met with. Neither the Italians nor French have excelled in compositions of this kind. Britain, by being possessed of many able sculptors at different times, has surpassed all other nations, both in taste of design, and workmanship.

The size of the chimney must be regulated by the proportions of the room where it is placed. In the smallest apartments, the breadth of the aperture should never be less than three feet, or three feet six inches. In rooms from 20 to 24 feet square, or of equal superficial dimensions, it may be from 4 to 4½ feet broad; in those of 24 to 27, from 4½ to 5; and, in such as exceed these dimensions, the aperture may even be extended to 5½ or 6 feet.

The chimney should always be situated so as to be immediately seen by those who enter the room. The middle of the partition wall is the most proper place in halls, salons, and other rooms of passage; but in drawing-rooms, dressing-rooms, and the like, the middle of the back-wall is the best situation. In bed-rooms, the chimney is always in the middle of one of the partition-walls; and in closets, and other very small places, to save room, it is put in a corner. Where-ever two chimneys are used in the same room, they should be placed either directly facing each other, if in different walls, or at equal distances from the centre of the wall in which they both are.

The proportion of the apertures of chimney-pieces of a moderate size is generally a perfect square; in small ones, it is a trifle higher; and in large ones, a trifle lower. Their ornaments consist in architraves, friezes, cornices, columns, pilasters, termini, caryatides, consoles, and all kinds of ornaments of sculpture, representing animals and vegetables, &c. likewise vases, chalices, trophies of arms, &c. In designing them, regard must be had to the nature of the place where they are to be employed. Such as are intended for halls, salons, guard-rooms, galleries, and other large places, must be composed of large parts, few in number, of distinct and simple forms, and having a bold relief; but chimney-pieces for drawing-rooms, dressing-rooms, &c. may be of a more delicate and complicated nature.

Chimney-pieces are composed of wood, stone, or marble; the last of which ought to be preferred, as figures or profiles are best represented in a pure white.

Plate XXXV. fig. 1, 2, 3, and 4, are different designs. Chap. XVI. Of the Proportions of Rooms.

The proportions of rooms depend in a great measure on their use, and actual dimensions: but, with regard to beauty, all figures, from a square to a selaquilateral, may be employed for the plan.

The height of rooms depends on their figure. Flat ciled ones may be lower than those that are coved. If their plan be a square, their height should not exceed five sixths of the side, nor be less than four fifths; and when it is oblong, their height may be equal to their breadth. But coved rooms, if square, must be as high as broad; and when oblong, they may have their height equal to their breadth, more one fifth, one quarter, or even one third of the difference between the length and breadth: and galleries should at least be in height one and one third of their breadth, and at most one and a half, or one and three-fifths.

The coldness of the British climate is a strong objection to high rooms; so that it is not uncommon to see the most magnificent apartments not above 15, 16, or at most 18 feet high; though the extent of the rooms would require a much more considerable elevation. But, where beauty is aimed at, this practice ought not to be imitated.

When rooms are adorned with an entire order, the entablature should never exceed one sixth of the whole height in flat-ciled rooms, and one fifth of the upright part in coved ones; and when there are neither columns nor pilasters, but only an entablature, its height should not be above one seventh of these heights. If the rooms be finished with a simple cornice, it should never exceed one fourteenth, nor ever be less than one fifteenth part of the above-mentioned height.

Chap. XVII. Of Ceilings.

Ceilings are either flat, or coved, in different manners. The simplest of the flat kind are those adorned with large compartments, surrounded with one or several mouldings, either let into the ceiling, or projecting beyond its surface: and when the mouldings that form the compartments are enriched, and some of the compartments adorned with well-executed ornaments, such ceilings have a good effect, and are very proper for common dwelling-houses, and all low apartments. Their ornaments and mouldings do not require a bold relief; but, being near the eye, they must be finished with taste and neatness. For higher rooms, a flat ceiling which has the appearance of being composed of various joints framed into each other, and forming compartments of various geometrical figures, should be employed. The sides of the joints forming the compartments are generally adorned with mouldings, and represent either a simple architrave, or an architrave-cornice, according to the size of the compartments and the height of the room.

Coved ceilings are more expensive; but they are likewise more beautiful. They are used promiscuously in large and small rooms, and occupy from one fifth to one third of the height of the room. If the room be low in proportion to its breadth, the cove must likewise be low; and when it is high, the cove must be so likewise: by which means the excess of the height will be rendered less perceptible. But, where the architect is at liberty to proportion the height of the room to its superficial dimensions, the most eligible proportion for the cove is one fourth of the whole height. In parallelogram-figured rooms, the middle of the ceiling is generally formed into a large flat panel. This panel, with the border that surrounds it, may occupy from one half to three fifths of the breadth of the room. The figure of the cove is commonly either a quadrant of a circle or of an ellipse, taking its rise a little above the cornice, and finishing at the border round the great panel in the centre. The border projects somewhat beyond the coves on the outside; and, on the side towards the panel, it is generally made of sufficient depth to admit the ornaments of an architrave, or architrave and cornice.

In Britain, circular rooms are not much in use; but they are very beautiful. Their height must be the same with that of square rooms; their ceilings may be flat; but they are handsomer when coved, or of a concave form.

Arches doubleaux, or soffits of arches, when narrow, are ornamented with guilloches, or frets; but, when broad, they may be adorned in a different manner.

When the profiles of the room are gilt, the ceilings ought likewise to be gilt. The usual method is to gild all the ornaments, and to leave the grounds white, pearl colour, light blue, or of any other tint proper to set off the gilding to advantage. Painted ceilings, so common in France and Italy, are but little used in Britain.

Chap. XVIII. Of Stairs and Stair-cases.

There are many kinds of stair-cases: for, in some, the steps are made straight; in others, winding; in others, mixed of both. Of straight stairs, some fly directly forward, others are square, others triangular. Others are called French flights, or winding stairs (which in general are called spirals, or cockle-stairs); of which some are square, some circular or round, and some elliptical or oval; and these again are various, some winding about a solid, others about an open newel. Stairs mixed of straight and winding steps are also of various kinds; some are called dog-legged; some there are that wind about a solid newel; and others that fly about a square open newel.

Great care ought to be taken in placing of the stair-case in any building; and therefore stair-cases ought where to be described and accounted for justly when the plan of a building is made. For want of this, sometimes unpardonable errors have been committed; such as having a little blind stair-case to a large house, or, on the other hand, a large spacious stair-case to a little one.

Palladio says, in placing stair-cases, the utmost care ought to be taken; it being difficult to find a place convenient for them, that will not at the same time prejudice the rest of the building. But commonly the stairs are placed in the angle, wing, or middle of the front.

To every stair-case are required three openings. First, the door leading thereto. Second, Secondly, the window, or windows that give light to it;

And, thirdly, the landing.

First, the door leading to the stair-case should be so placed, that most of the building may be seen before you come at the stairs, and in such a manner that it may be easy for any person to find out.

Secondly, for the windows; if there be but one, it must be placed in the middle of the stair-case, that thereby the whole may be enlightened.

Thirdly, the landing of stairs should be large and spacious, for the convenient entering into rooms: in a word, stair-cases should be spacious, light, and easy in ascent. The height of large steps must never be less than six inches, nor more than seven inches and a half.

The breadth of steps should never be less than ten inches, nor more than eighteen inches; and the length of them not less than three feet, nor more than twelve.

Plate XXXVI. fig. 1. A stair-case of two flights.—A shews the manner of drawing the ramp, which is to rise equal to the height of the first step of the next flight, and as much as its kneeling; as is shewn by the ramp intersecting the rail of the second flight.

Fig. 2. Shews the straight rail intersecting a circular cap.

Fig. 3. Section of two different hand-rails.

Fig. 4. Shews the manner of dovetailing the rifer into the step.

Plate XXXVII. fig. 1. Represents a stair-case, with flights, and its landing-rail.

Fig. 2. Shews the solid part of the step out of which the scroll is formed; where a represents the overfall of the step; b, The thickness of the bracket, with its mitring to the rifer; and, c, The string-board.

Fig. 4. Shews the scale for drawing the scroll of fig. 3.—To perform which, take the distance from 1 to the centre, in fig. 3, and set it from 1 to the centre in fig. 4; divide that extent into three parts, then set four such parts on the upper side of the scale, and draw the line from 4 to 1; set one foot of your compasses at 4, and strike the circular line; let that be divided into twelve equal parts, and then draw lines from 4 through those divisions to the upright line.

The scale being thus made, draw the scroll of fig. 3, by it in the following manner.

Set one foot of your compasses in 1, and describe a stroke at c; take the same distance, and with one foot in 2, cross the stroke at c; then from c, turn the part from 1 to 2, and proceed in the same manner; for if the distance were taken in the scale from 1 to the centre, it would strike the circle too flat; and if taken from 2, it would strike the circle too quick.

When this is well understood, there will be little difficulty in drawing the scroll below fig. 2.; which throws itself out farther in proportion than that in fig. 3.; for this will always be the case when the upper line of the scale, which consists of four divisions in fig. 4., is made but with three divisions or less; whence it appears, that the upper line of the scale may be drawn at what length you please, according as you would bring in or keep out the scroll.

Plate XXXVIII. Shews the manner of squaring twist-rails.

Fig. 2. Exhibits the pitch-board, to shew what part of the step the twisted part of the rail contains; the three dotted lines drawn from the rail to the pitch-board represent the width of the rail, which is to be kept level. The dotted lines a and b shew how much half the width of the rail turns up from its first beginning to 3.

Fig. 3. Shews the same pitch-board, with the manner of the rail's turning up. If the sides of the twisted part of the rail be shaped by the rail-mould, so that they direct down to its ground-plan, that is, the upper side of the rail being first struck by the mould, then apply the mould to the under side, as much back as the level of the pitch-board shews, by being struck on the side of the rail, and then fig. 3., being applied to the outside of the rail, from its first twisting part to 3, will show how much wood is to be taken off.

Fig. 5. Exhibits the square of the rail, with the raking line of the pitch-board drawn through the middle on the upper side; then draw the depth of the side the rail parallel to this, and the dotted lines from the diagonal of the rail; these lines shew what quantity of wood will be wanting on the upper and lower sides of the rail. Set your compasses at c, and draw the circular stroke from the raking part of the pitch-board to b; take the distance a b and transfer it from a to b, in fig. 7. The several distances thus found may be set at any number of places, ranging with the straight part of the rail; and it then forms the width of the mould for the twisting part of the rail.

Fig. 7. Shews the sweep of the rail. The rail cannot be fixed less than one fourth part from the nosing or front of the step.

The remaining part of the pitch-board may be divided into any number of parts, as here into four; from these divisions draw lines across the pitch-board to the raking line; then take the distances from the ground-line of the pitch-board to the plan of the rail, and set them perpendicular from the raking line of the pitch board; and these divisions, when the rail is in its proper position, lie directly over the divisions on the ground plan.

In this figure l, m, and n, rise as much above o as the dotted line in fig. 5., does above the width of the rail; and they sink as much below o as the other dotted line in fig. 5., falls below the width of the rail; the same thicknecks must be glued upon o, though the greatest part will come off in squaring. The reason of placing the letters l, m, and n, where they are, is, that they might not obstruct the small divisions of the rail-mould.

Fig. 4. Shews how to find the rail when it takes more than one step. The remaining part of the pitch-board is divided into four parts, as before in fig. 7 and it takes in two such parts of the next step. Draw lines from these divisions to the diagonal of the pitch-board as in fig. 7; then take the distance a b, and set it from c to d, and so proceed with the other divisions.

Another way to find the outside of the rail-mould is, to draw all the divisions across the plan of the rail; then take the distance from the ground-line of the pitch-board to 4, transfer it from the diagonal of the pitch-board to 4 on the rail; and so proceed with the other distances. Now, when the rail is put in its proper situation, c will be perpendicular to b, and all the divisions, as 1, 2, 3, 4, &c., in the rail, will be perpendicular to 1, 2, 3, 4, &c., in the ground-plan.

Fig. 6. Shews the plan of a rail of five steps. Part II. ARCHITECTURE.

To find the rail.—Set five divisions, as from e to b, which is the height of the five steps; draw the diagonal b to the plan of the rail; then take the distance e f, and transfer it to g h, and proceed in the same manner with the other seven distances.

To find the width of the rail-mould.—Draw the lines across the plan of the rail, as at k; set that distance from the diagonal to i; and so proceed with the rest, as was shewn in fig. 4.

Having formed the sides of the rail perpendicular to its ground-plan, and having squared the lower end of the rail, then take a thin lath, and bend it with the rail, as is represented by m fig. 1.

This is the readiest method of squaring a solid rail; but if the rail be bent in the thicknesses, the noting of the steps must be drawn upon a cylinder, or some other solid body of a sufficient width to contain the width of the rail or string-board.

r Represents the depth of the rail, touching the nose of each step. Take a sufficient number of thicknesses of this width, to make the thickness of your rail; glue them all together upon your cylinder or templet, con-

Plate XXIX. Fig. 1. Shews the form of a trussed roof, with three ring-ports, that may carry seventy feet, or upwards.

Fig. 2. Exhibits an M roof, capable of carrying as great an extent as the former. Indeed both these designs are capable of carrying almost any extent.

Fig. 3. Represents two different sorts of trusses.

Fig. 4. Shews the manner of piecing timber. Sometimes the joint may be extended as far as a, with another bolt through it. To the right is shewn a different sort of joint.

Fig. 5. Shews the manner of trussing a girder. If the trusses are full long, with the pieces b and c you may make them as light as you please.

Fig. 6. Represents the manner of trussing partitions.

Part II. PRACTICE OF ARCHITECTURE.

Having thus described and given rules for the most generally received proportions of the different parts of buildings, both of the useful and ornamental kind, we must next give an account of the method of erecting different kinds of edifices; and here the judgment of the architect must necessarily be very much employed, as no fixed rules have been laid down by which he can be directed in all cases. As a necessary preliminary, however, to the construction, we must first consider,

Chap. I. The Situations of Houses.

Though it must be, in many cases, impossible to choose such a situation as might be agreeable either to the architect or the proprietor, yet, where a choice can be made, there are certainly a great many circumstances that will determine one situation to be preferable to another. These circumstances depend entirely on the person who is to inhabit the house. A farmer, for instance, ought to dwell in the most central part of his farm; an independent gentleman must regard the healthiness, the neighbours with whom he can converse, the prospect from his house, and also the aspect of the ground near it. To answer these purposes of health and pleasure, an open elevated situation is the best, as the air is there pure, and the prospect extensive; but too elevated a situation is disagreeable, as being both difficult of access, and exposed to cold and bleak winds. To build in bottoms between hills is both unhealthful and unpleasant, the house being in a manner buried, and the ground near it generally marshy from the rain-water which runs down from the hills, which renders the air unwholesome. As a garden also is a very necessary article to a country habitation, the soil is by no means a matter of indifference; and therefore it may be concluded, that an elevated situation on a gravelly loam, near some running water, is the best situation for a country house.

Chap. II. Of the Construction of Edifices in general.

The proper situation of a house, or any other building, being chosen, according to its intended nature, the next thing to be considered is to lay the foundation in a proper manner. The only security of a house, or any other building whatever, is in having a good foundation, and no error is so dangerous as that which is committed here; as the shrinking of the foundation but the breadth of a straw may cause a rent of five or six inches wide in the superstructure. To guard against errors of this kind, the qualities of the ground for a considerable depth must be carefully observed.

The best foundation is that which consists of gravel or stone; but, in order to know whether the inferior strata are sufficient for the support of the building, it will be advisable to sink wells at some little distance. By attending to what is thrown up in digging these, the architect will be acquainted with what lies under the stony or gravelly bed which on the surface promises so much security, and will know what measures to take.

But though a stony or gravelly bottom is undoubtedly the most sure and firm, where all is found beneath, there is no kind of ground which may prove more fallacious, or occasion such terrible accidents. The reason of this is, that such kind of ground often contains absolute vacuities; nor is rock itself, though a foundation upon a rock is strong even to a proverb, free from dangers of the same kind. Caverns are very frequent in rocky places; and should an heavy building be erected over one of these, it might suddenly fall down altogether. To guard against accidents of this kind, Palladio advises the throwing down great weights forcibly on the ground, and observing whether it sounds hollow, or shakes; and the beating of a drum upon it, by the sound of which an accustomed ear will know whether the earth is hollow or not.

Where the foundation is gravel, it will be proper to examine examine the thickness of the stratum, and the qualities of those that lie under it, as they have appeared in digging. If the bed of gravel is thick, and the under strata of a sound and firm kind, there needs no affluence; if otherwise, we must have recourse to various methods in order to supply the defect.

The other matters which may occur for a foundation are clay, sand, common earth, or rotten boggy ground. Clay will often both raise and sink a foundation; yet it has a solidity which, with proper management, is very useful. The marshy, rotten, or boggy ground is of all others the worst; yet even upon this great buildings may be raised with perfect safety, provided proper care be taken. In case of boggy earths, or unfirm land, piling is one of the most common methods of securing a foundation; and, notwithstanding the natural disadvantage of the earth, piles, when properly executed, are one of the firmest and most secure foundations.

In foundations near the edge of waters, we should always be careful to found to the very bottom, as many terrible accidents have happened from the ground being undermined by rivers. The same method is to be followed when the ground on which we build has been dug or wrought before. It ought never to be trifled in the condition in which it is left; but we must dig through it into the solid and unmoved ground, and some way into that, according to the weight and big-

Defect in St. Peter's at Rome.

Before the architect, however, begins to lay the foundation of the building, it will be proper to construct such drains as may be necessary for carrying off the rain, or other refuse water that would otherwise be collected and lodge about the house. In making of drains for carrying off this water, it will be necessary to make large allowances for the different quantities that may be collected at different times. It must also be considered, that water of this kind is always loaded with a vast quantity of sediment, which by its continual falling to the bottom will be very apt to choke up the drain, especially at those places where there happen to be angles or corners in its course. The only method of preventing this is by means of certain cavities disposed at proper distances from one another. Into these the sediment will be collected, and they are for that reason called sumps. With regard to these, the only directions necessary are, that they be placed at proper distances, be sufficiently large, and placed so as to be easily cleaned. It is a good rule to make a sump at each place where the water enters the drain; as by this means a considerable quantity of sediment will be prevented from entering the channel at all. Others are to be made at proper distances, especially where there are any angles. They must be made sufficiently large; the bigger, in moderation, the better; and they must also be covered in such a manner as to be easily got at in order to be cleaned. But, as putrid water is exceedingly noxious, it will be necessary to carry up a brick funnel over every sump, in order to prevent the collection of the putrid effluvia, which would otherwise occasion the death of the person who cleaned it.

All drains ought to be arched over at top, and may proportions be most conveniently built of brick. According to their different sizes, the following proportions of height and thickness may be observed. If the drain is 18 inches wide, the height of the walls may be one foot, and their thickness nine inches; the bottom may be paved with brick laid flatwise, and the arch turned four inches. If the drain is 22 inches wide, the side walls are then to be one foot three inches in height, and the rest constructed as before. If it is 14 inches wide, the height of the walls may be 9 inches, and the sweep of the arch four. A drain of a yard wide should have the same height, and the arch turned over it ought to be 9 inches thick. Upon the same principles and proportions may other drains of any size be constructed.

The sewers and drains being constructed in a manner proportioned to the size of the intended building, the architect may next proceed to lay the foundation of the walls. Here the first care must be, that the floor of the foundation be perfectly smooth and level. The Italians begin with laying over it an even covering of strong oak plank; and upon that they lay, with the most exact care, the first course of the materials. Whether we take this method, or begin upon the naked floor, all must be laid with the most exact truth by rule and line. When the board plat is laid, a course of stone is the best first bed, and this is to be laid without mortar; for lime would make the wood decay, which otherwise, in a tolerably good soil, will last for ages. After this, all the courses should follow with the same perfect evenness and regularity. If the materials are brick, they should be laid on with an equal, and not too great, quantity of mortar; if stone, they ought to be placed regularly, and in the same situation in which they lay in the quarry; for many stones, which will bear any weight flatwise, and in their natural position, are of such a grain, that they will split otherwise. The joinings of the under course must be covered by the fold of the next course all the way up; and the utmost care must be taken that there be no vacancy left in the wall, for the weight will most certainly crush it in. The least mortar there is in a foundation, the better. Its use is to cement the bricks and stones together; and the fewer they are, the less will be required for that purpose. Where mortar is used to fill up cavities, it becomes part of the wall; and not being of equal strength with the solid materials, it takes from the firmness of the building. For the same reason, nothing can be more absurd than to fill up a foundation with loose stones or bricks thrown in at random; and where this is done, the ruin of the building is inevitable. Where the foundation of a principal wall is laid upon piles, it will be necessary also to pile the foundations of the partitions, though not so strongly.

The thickness of foundation-walls in general ought to be double that of the walls which they are to support. The looser the ground, the thicker the foundation wall ought to be; and it will require the same addition also in proportion of what is to be raised upon it. The plane of the ground must be perfectly level, that the weight may press equally every where; for when it inclines more to one side than another, the wall will split. The foundations must diminish as they rise, but the perpendicular is to be exactly kept in the upper and lower parts of the wall; and this caution ought to be observed all the way up with the same strictness. In some ground, the foundation may be arched; which will save materials and labour, at the same time that the superstructure has an equal security. This practice is peculiarly serviceable where the foundation is piled.

As the foundation-walls are to diminish in thickness, so are those which are built upon them. This is necessary in order to save expense, but is not absolutely so to strengthen the wall; for this would be no less strong though it was continued all the way to the top of an equal thickness, provided the perpendicular was exactly kept. In this the ancients were very expert; for we see, in the remains of their works, walls thus carried up to an exorbitant height. It is to be observed, however, that, besides perfect truth in their perpendiculums, they never grudged iron work, which contributed greatly to the strength of their buildings. The thickness and diminution of walls is in a great measure arbitrary. In common houses built of brick, the general diminution from the bottom to the top is one half the thickness at the bottom; the beginning is two bricks, then a brick and an half, and lastly one brick, thickness. In larger edifices, the walls must be made proportionally thicker; but the diminution is preferred much in the same manner. Where stones are used, regard must be had to their nature, and the propriety of their figures for holding one another. Where the wall is to be composed of two materials, as stone and brick, the heaviest ought always to be placed undermost.

There is one farther particular regarding the strength of a plain wall, and that is, the fortifying its angles. This is best done with good stone on each side, which gives not only a great deal of strength, but a great deal of beauty. Pilasters properly applied are a great strengthening to walls. Their best distance is about every 20 feet, and they should rise five or six inches from the naked of the wall. A much lighter wall of brick with this affluence, is stronger than a heavier and massier one built plain. In brick walls of every kind, it is also a great addition to their strength to lay some chief courses of a larger and harder matter; for these serve like fines to keep all the rest firmly together, and are of great use where a wall happens to sink more on one side than another. As the openings in a wall are all weakenings, and as the corners require to be the strongest parts, there should never be a window very near a corner. Properly, there should always be the breadth of the opening firm to the corner. In the most perfect way of forming, the diminution of walls, the middle of the thinnest part being directly over the middle of the thickest, the whole is of a pyramidal form; but where one side of the wall must be perpendicular and plain, it ought to be the inner, for the sake of the floors and cross walls. The diminished side, in this case, may be covered with a fascia or cornice, which will at once be a strength and ornament.

Along with the construction of walls, that of the chimneys must also be considered; for errors in the construction of these will render the most elegant building extremely disagreeable. The common causes of smoke- ing are either that the wind is too much let in above at the mouth of the shaft, or the smoke is stifled below; and sometimes a higher building, or a great elevation of the ground behind, is the source of the mischief; or, lastly, the room in which the chimney is may be so small or close, that there is not a sufficient current of air to drive up the smoke. Almost all that can be done while the walls are constructing to prevent smoke is, to make the chimney vent narrower at bottom than at top; yet this must not be carried to an extreme; because the smoke will then linger in the upper part, and all the force of the draught will not be able to send it up.

—As for the methods of curing smoky chimneys in houses already built, see the article Chimney.

After the walls are finished, the roof is the next consideration; but concerning it very little can be said; only that its weight must be proportioned to the strength of the walls. It must also be so contrived as to press equally upon the building; and the inner walls must bear their share of the load as well as the outer ones. A roof ought neither to be too natty, nor too light; as being necessary for keeping the walls together by its prelude, which it is incapable of doing while too light; and if too heavy, it is in danger of throwing them down. Of these two extremes, however, the last is to be accounted the worst.

With regard to the floors, they are most commonly made of wood; in which case, it will be necessary that it should be well seasoned by being kept a considerable time before it is used. The floors of the same story should be all perfectly on a level; not even a threshold rising above the rest; and if in any part there is a room or closet whose floor is not perfectly level, it ought not to be left so, but raised to an equality with the rest; what is wanting of the true floor being supplied by a false one.

In mean houses, the floors may be made of clay, ox blood, and a moderate portion of sharp sand. These three ingredients, beaten thoroughly together and well spread, make a firm good floor, and of a beautiful colour. In elegant houses, the floors of this kind are made of plaster of Paris, beaten and fitted, and mixed with other ingredients. This may be coloured to any hue by the addition of proper substances; and, when well worked and laid, makes a very beautiful floor. Besides these, halls, and some other ground-rooms, are paved or floored with marble or stone; and this either plain or dotted, or of a variety of colours: but the universal practice of carpeting has in a great measure set aside the following any ornamental workmanship upon floors. In country buildings, also, floors are frequently made of bricks and tiles. These, according to their shapes, may be laid in a variety of figures; and they are also capable of some variation in colour, according to the nature of the earth from which they were made. They may be laid at any time; but for those of earth or plaster, they are best made in the beginning of summer, for the sake of their drying.

CHAP. III. Of the Distribution of the Apartments of Houses, with other conveniences.

As houses are built only for the sake of their inhabitants, the distribution of the apartments must of necessity be directed by the way of life in which the inhabitants... habitants are engaged. In the country, this is commonly farming; and here, besides the house for the family, there is also necessary a barn for the reception of the produce of the ground, a stable for cattle, a cart-house for keeping the utensils under cover, and sheds for other uses.—To accomplish these purposes, let a piece of ground be taken of five times the extent of the front of the house, and inclosed in the least expensive manner. Back in the centre of this let the house be placed, and in the front of the ground the barn and the stable, with the adjoining sheds. These are to be set, one on each side, to the extreme measure of the inclosed ground; they will thus fill up a part of the entrance, and will leave all about the house some inclosed ground by way of yard. From the barn to the stable may be extended a fence with a gate in the middle, and this gate ought to front the door of the house.

This much being settled, the plan of the house and out-buildings may be made as follows. The door may open into a plain brick passage, at the end of which may be carried up a small stair-case. On one side of the passage may be a common kitchen; and on the other side a better or larger room, which will serve the family by way of parlour. Beyond this may stand on one side the pantry, and on the other the dairy-room, the last being twice the size of the former. They are placed on the same side with the parlour, on account of the heat of the kitchen, which renders it improper to be near them. On the kitchen side, a brew-house may very conveniently be placed. More rooms may be added on the ground-floor as occasion requires; and the upper story is to be divided into bed-chambers for the family, with garrets over them for the servants.—A house of this kind is represented 2d Plate XXXIX. fig. 1.; and (fig. 2.) one of a somewhat better kind, where a private gentleman who has a small family may find conveniency.

3d Plate XXXIX. fig. 1. represents a gentleman's country-seat, built on a more elegant plan. Here the front may extend 65 feet in length, the depth in the centre being 40 feet, and in each of the wings 45. The offices may be disposed in wings; the kitchen in the one, and the stables in the other; both of which, however, may correspond in their front with the rest of the building, which they ought also to do with one another. These wings may have a projection of 13 feet from the dwelling-house, to which they ought to be connected, not by straight lines, but by curves, as represented fig. 2.

The best proportion of these offices to a house extending 65 feet in front, is 35 feet. If they are smaller, the house will look gigantic; if larger, they will lessen its aspect. To a front of 35 feet, a depth of 48 is a very good proportion. There ought also to be a covered communication between the dwelling-house and offices, which must not appear only to be a plain blank wall, but must be ornamented with gates, as in the figure. The arch by which the offices are joined to the dwelling-house must be proportioned to the extent of the buildings; and there cannot be a better proportion than five feet within the angles of the buildings. By this means the wings, which have only a projection of 13 feet, will appear to have one of 18, and the light will be agreeably broken.

With regard to the internal distribution of a house of this kind, the under story may be conveniently divided into three rooms. The hall, which is in the centre, will occupy the whole of the projecting part, having a room on each side. The length of the hall must be 24 feet, and its breadth 12; the rooms on each side of it must be 16 feet long, and 11 wide. Of these two front rooms, that on the right hand may be conveniently made a waiting-room for persons of better rank, and that on the left hand a dressing-room for the master of the house. Behind the hall may run a passage of four feet and an half, leading to the apartments in the hinder part of the house, and the stair-case. These may be disposed as follows. Directly behind the hall and this passage the space may be occupied by a saloon, whose length is 24 feet, and its breadth 17. On the left hand of the passage, behind the hall, is to be placed the grand stair-case; and as it will not fill the whole depth, a pleasant common parlour may terminate on that side of the house. On the other side, the passage is to lead to the door of the great dining parlour, which may occupy the whole space.

A plan of a house of the same kind, but somewhat different in the distribution, is represented fig. 2. The 3d Plate front here extends 68 feet, and the wings project 28 feet; their depth is 48, and their breadth 36. The hall may be 26 feet long, and 17 broad. On the left hand of the hall may be a waiting-room 16 feet long, and 10 broad; behind which may be a handsome dining-room. The passage into this waiting-room should be at the lower end of the hall; and it must have another opening into the room behind it. On the right hand of the hall is the place of the great stair-case, for which a breadth of 16 feet three inches is to be allowed. In the centre of the building, behind the hall, may be a drawing-room 26 feet long, and 16 broad; and behind the stair-case will be room for a common parlour of 16 feet square. The passage of communication between the house and wings may be formed into colonnades in a cheap manner behind: a flight of steps, raised with a sweep, occupying the centre of each, and leading up to a door, and the covering being no more than a shed supported by the plainest and cheapest columns.

The two wings now remain to be disposed of. That on the right hand may contain the kitchen and offices belonging to it, and the other the stables. The front of the right-hand wing may be occupied by a kitchen entirely, which will then be 30 feet long, and 16½ wide; or it may be made smaller, by setting off a small room to the right. Twenty-two feet by 16 will then be a good signal. The other room will then have the same depth of 16 feet, and the width to the front may be 7½. Beyond the kitchen may stand the stair-case, for which 7½ feet will be a proper allowance; and to the right of this may be a scullery 12 feet 10 inches deep from the back front by 7 in breadth. To the left of the stairs may be a servants hall 16 feet square, and behind that a larder 12 feet 10 by 14 feet 6. In the centre of the other wing may be a double coach-house: for which there should be allowed the whole breadth of the wing, with 10 feet 6 inches in the clear; and on each side of this may be the stables. The external decorations of the front and wings will be better understood from the figure than they can be by any description.

4th Plate XXXIX. shews the plan and elevation of the house of Francis Charteris, Esq.; at Newmills. The proportions... portions of the rooms are marked in the plan; and the front, being decorated with columns of the Ionic order, will sufficiently show in what manner any of the five orders may be induced with propriety and elegance.

**Chap. IV. Of Aquatic Buildings.**

1. Of Bridges.

These are constructed either of wood or stone; of which the last are evidently the strongest and most durable, and therefore in all cases to be preferred where the expense of erecting them can be borne. The proper situation for them is easily known, and requires no explanation; the only thing to be observed is, to make them cross the stream at right angles, for the sake of the boats that pass through the arches, with the current of the river; and to prevent the continual striking of the stream against the piers, which in a long course may endanger their being damaged and destroyed in the end.

Bridges built for a communication of high roads, ought to be so strong and substantial as to be proof against all accidents that may happen, to have a free entrance for carriages, afford an easy passage to the waters, and be properly adapted for navigation, if the river admits of it. Therefore the bridge ought to be at least as long as the river is wide in the time of its greatest flood: because the sloping of the water above may cause too great a fall, which would prove dangerous to the vessels, and occasion the under graveling the foundation of the piers and abutments; or, by reducing the passage of the water too much in time of a great flood, it might break through the banks of the river, and overflow the adjacent country, which would cause very great damages; or, if this should not happen, the water might rise above the arches, and endanger the bridge to be overfet, as has happened in many places.

When the length of the bridge is equal to the breadth of the river, which is commonly the case, the current is lessened by the space taken up by the piers: for which reason, this thickness should be no more than is necessary to support the arches; and it depends, as well as that of the abutments, on the width of the arches, their thickness, and the height of the piers.

The form of the arch is commonly semicircular; but when they are of any great width, they are made elliptical, because they would otherwise become too high. This has been done at the Pont Royal, at Paris, where the middle arch is 75 feet, and its height would have been 37.5 feet, instead of which it is only 24 by being made elliptical.

Another advantage of much more importance arises from the oval figure, which is, that the quantity of masonry of the arches is reduced in the same proportion as the radius of the arch is to its height. That is, if the radius is 36 feet, and the height of the arch 24, or three fourths of the radius, the quantity of masonry of the arches is likewise reduced to three-fourths; which must lessen the expense of the bridge considerably. Notwithstanding these advantages, however, the latest experiments have determined segments of circles to be preferable to curves of any other kind; and of these the semicircle is undoubtedly the best, as pressing most perpendicularly on the piers.

When the height of the piers is about six feet, and the arches are circular, experience has shown, says Mr Belidor, that it is sufficient to make the thickness of the piers the fifth part of the width of the arch, and two feet more; that is, the thickness of the piers of an arch of 36 feet, ought to be 8 feet; those of an arch of 48 feet, to be 10.

When the arches are of a great width, the thickness of the piers may be reduced to the sixth part of that of the piers' width; but the depression of the two feet is not done at once; that is, in an arch of above 48 feet, 3 inches are taken off for every 6 feet of increase of the width of the arch. For instance, the thickness of the piers supporting an arch of 72 feet wide, should be 14 feet, according to the preceding rule; but by taking off 3 inches for every 6 feet, above an arch of 48 wide, the thickness of the piers is reduced to 13 feet; consequently, by following the same rule, the thickness of the piers supporting an arch of 16 fathoms wide, will be 16 feet; all the others above that width are the sixth part of the width.

After this, Mr Belidor gives a rule for finding the thickness of the piers which support elliptic arches, and makes them stronger than the former: the abutments he makes one sixth part more than the piers of the largest arch. But it is plain, that these rules are insufficient, being merely guess-work, determined from some works that have been executed.

The thickness of the arch-stones is not to be determined by theory, nor do those authors who have written on the subject agree amongst themselves. Mr Gautier, an experienced engineer, in his works, makes the length of the arch-stones, of an arch 24 feet wide, two feet; of an arch 45, 60, 75, 90 wide, to be 3, 4, 5, 6, feet long respectively, when they are hard and durable, and something longer when they are of a soft nature: on the contrary, Mr Belidor says, they ought to be always one twenty-fourth part of the width of the arch, whether the stone be hard or soft; because, if they are soft, they weigh not so much.

But that the length of the arch-stones should be but a foot in an arch of 24 feet wide, 2, 3, 4, in arches of 48, 72, 96, feet, seems incredible; because the great weight of the arches would crush them to pieces, by the pressure against one another; and therefore Mr Gautier's rule appears preferable: as he made the length of the arch-stones to increase in a slower proportion, from 10 to 45 feet wide, than in those above that width, we imagine that the latter will be sufficient for all widths, whether they are great or little: therefore we shall suppose the length of the arch-stones of 30 feet in width to be two feet, and to increase one foot in 15, that is, 3 feet in an arch of 45 feet, 4, 5, 6, in an arch of 60, 75, and 90 feet; and so the rest in the same proportion. Rectangular piers are seldom used but in bridges over small rivers. In all others, they project the bridge by a triangular prism, which presents an edge to the stream, in order to divide the water more easily, and to prevent the ice from sheltering there, as well as vessels from running foul against them: that edge is terminated by the adjacent surfaces at right angles to each other at Westminster-bridge, and make an acute angle at the Pont Royal of about 60 degrees; but of late the French terminate this angle by two cylindrical surfaces, whose bases are arcs of 60 degrees, in all their new bridges.

When the banks of the rivers are pretty high, the slope of the bridge is made quite level above, and all the arches of bridge on an equal width; but where they are low, or for the sake of navigation a large arch is made in the middle of the stream, then the bridge is made higher in the middle than at the ends; in this case, the slope must be made easy and gradual on both sides, so as to form above one continued curve line, otherwise it appears disagreeable to the eye. Mr Belidor will have the descent of that slope to be one twenty-fourth part of the length; but this is undoubtedly too much, as one fiftieth part of the length is quite sufficient for the descent.

The width commonly allowed to small bridges is 30 feet, &c., but in large ones near great towns, these 30 feet are allowed clear for horses and carriages, besides a banquet at each side for foot passengers of 6 to 9 feet each, raised about a foot above the common road; the parapet-walls on each side are about 18 inches thick, and 4 feet high; they generally project the bridge with a cornice underneath; sometimes balustrades of stone or iron are placed upon the parapet, as at Westminster; but this is only practised where a bridge of a great length is made near the capital of a country.

The ends of bridges open from the middle of the two large arches with two wings, making an angle of 45 degrees with the rest, in order to make their entrance more free and easy; these wings are supported by the same arches of the bridge next to them being continued in the manner of an arch, of which one pier is much longer than the other.

How the work is to be carried on.

As the laying the foundation of the piers is the most difficult part of the whole work, it is necessary we should begin with an easy case, that is, when the depth of the water does not exceed 6 or 8 feet; and then proceed to those which may happen in a greater depth of water.

One of the abutments with the adjacent piers is inclosed by a dyke called batardeau by the French, of a sufficient width for the work, and room for the workmen. This batardeau is made by driving a double row of piles, whose distance is equal to the depth of water, and the piles in each row are 3 feet from each other; they are fastened together on the outside by bonds of 6 by 4 inches: this being done, frames of about 9 feet wide are placed on the inside to receive the boards which are to form the inclosure: the two uprights of these frames are two boards of an inch and half thick, sharpened below to be driven into the ground, and fastened together by double bonds, one below, and the other above, each separated by the thickness of the uprights; these bonds serve to slide the boards between: after these frames have been driven into the ground as hard... hard as can be, then the boards themselves are likewise driven in till they reach the firm ground underneath.

Between every two piles tie-beams are fastened to the bonds of the piles, to fasten the inside wall to the outside one; these tie-beams are let into the bonds and bolted to the adjacent piles; this being done, the bottom is cleared from the loose sand and gravel, by a machine like those used by ballast-heavers; and then well prepared clay is rammed into this coffer very tight and firm, to prevent the water from oozing through.

Sometimes these inclosures are made with piles only driven close to each other; at others, the piles are notched or dove-tailed one into the other; but the most usual method is to drive piles with grooves in them, 5 or 6 feet distant from each other, and boards are let down between them.

This being done, pumps and other engines are used to draw the water out of the inclosure, so as to be quite dry; then the foundation is dug, and the stones are laid with the usual precautions, observing to keep some of the engines always standing, in order to draw out the water that may ooze through the batardeau.

The foundation being cleared, and every thing ready to begin the work; a course of stones is laid, the outside all round with the largest stretchers and headers that can be had, and the inside filled with ashlers well jointed, the whole laid in terracotta mortar; the facings are cramped together, and set in lead; and some cramps are also used to fasten the facings with the inside. The same manner is to be observed throughout all the courses to the height of low-water mark; after which the facings alone are laid in terracotta mortar, and the inside with the best of the common fort. When the foundation is carried to the height of low-water mark, or to the height where the arches begin, then the shaft or middle wall is to be carried up nearly to the height of the arches, and there left standing till all the piers are finished, in order that the masonry may be sufficiently dry and settled before the arches are begun.

As the piers end generally with an arch at each end, it is customary to lay the foundation in the same manner; which is not so well as to continue the base rectangular quite to the ends of the piers, and as high as low-water mark; both because the foundation becomes then so much broader, and also because the water will not be able to get under it; for when the current sets against a flat surface, it drives the sand and mud against it, so as to cover it entirely; whereas if a sharp edge be presented to the stream, it carries everything away, and exposes the foundation to the continual action of the water, which in course of time must destroy it.

After the intervals between the arches are filled up with stones laid in a regular manner without mortar, and the gravel is laid over them; two drains or gutters are to be made lengthwise over the bridge, one on each side next to the foot-path, about 6 feet wide and a foot deep; which being filled with small pebble stones, serve to carry off the rain-water that falls on the bridge, and to prevent its filtering through the joints of the arches, as often happens.

How to build in water with Coffers.

The former method of laying the foundation by means of batardeaus is very expensive, and often meets with great difficulties; for when the depth of water is 8 feet or more, it is scarcely possible to make the batardeaus so tight as to prevent the water from oozing through them; and in that case the number of engines required, as well as the hands to work them, become very expensive; and if part of the batardeau should break by some extraordinary wind or tide, the workmen would be exposed to very great danger.

The next and best method therefore is to build with coffers, when it is practicable, such as were used at Westminster bridge. Here the height of water was 6 feet deep at a medium when lowest, and the tide rose about 10 feet at a medium also; so that the greatest depth of water was about 16 feet. At the place where one of the piers of the middle or great arch was to be, the workmen began to drive piles of about 13 or 14 inches square, and 34 feet long, thod with iron, so as to enter into the gravel with more ease, and hooped above to prevent their splitting in driving them; these piles were driven as deep as could be done, which was 13 or 14 feet below the surface of the bed of the river, and 7 feet distant from each other, parallel to the short ends of the pier, and at about 30 feet distant from them; the number of these piles was 34, and their intent to prevent any vessels or barges from approaching the work, and in order to hinder boats from passing between them, booms were placed so as to rise and fall with the water.

This being done, the ballast-men began to dig the foundation under the water, of about 6 feet deep, and 5 wider all round than the intended coffer was to be, with an easy slope to prevent the ground from falling in; in order to prevent the current from washing the sand into the pit, short grooved piles were driven before the two ends and part of the sides, not above 4 feet higher than low-water mark, and about 15 feet distant from the coffer; between these piles, rows of boards were let into the grooves down to the bed of the river and fixed there.

The bottom of the coffer was made of a strong grate, consisting of two rows of large timbers, the one longitudinal, and the other crosswise, bolted together with wooden trunnels, ten feet wider than the intended foundation. The sides of the coffer were made of fir timbers laid horizontally close one over another, pinned with oaken trunnels, and framed together at the corners, excepting at the two salient angles, where they were secured with proper irons, so that the one half might be loosed from the other if it should be thought necessary; these sides were lined on the inside as well as on the outside with three-inch planks placed vertically; the thickness of these sides was 18 inches at the bottom, reduced to 15 above, and they were 16 feet high; besides, knee timbers were bolted at the angles, in order to secure them in the strongest manner. The sides were fastened to the bottom by 28 pieces of timber on the outside, and 18 within, called straps, about 8 inches broad, and 3 or 4 inches thick, reaching and lapping over the ends of the sides; the lower part of these straps had one side cut dove-tail fashion, in order to fit the mortises made near the edge of the bottom to receive them, and were kept in their places by iron wedges; which being drawn out when the sides were to be taken away, gave liberty to clear the straps from the mortises. Before the coffer was launched, the foundation was examined, in order to know whether it was level; for which purpose several gauges were made, each of which consisted of a stone of about 15 inches square, and 3 feet long. The foundation being levelled and the coffer fixed directly over the place with cables fastened to the adjacent piles, the masons laid the first course of the stones for the foundation within it; which being finished, a sluice made in the tide was opened near the time of low-water; on which the coffer sunk to the bottom; and if it did not set level, the sluice was shut, and the water pumped out, so as to make it float till such time as the foundation was levelled; then the masons cramp the stones of the first course, and laid a second; which being likewise cramped, a third course was laid; then the sluice being opened again, proper care was taken that the coffer should settle in its due place. The stone-work being thus raised to within two feet of the common low-water mark, about two hours before low-water the sluice was shut, and the water pumped out so far as that the masons could lay the next course of stone, which they continued to do till the water was risen so high as to make it unsafe to proceed any farther; then they left off the work, and opened the sluice to let in the water. Thus they continued to work night and day at low-water, till they had carried their work some feet higher than the low-water mark; after this, the sides of the coffer were loosed from the bottom, which made them float; and then were carried adrift to be fixed to another bottom, in order to serve for the next pier.

It must be observed, that the coffer being no higher than 16 feet, which is equal to the greatest depth of water, and the foundation being 6 feet under the bed of the river; the coffer was therefore 6 feet under water when the tide was in; but being loaded with three courses of stones, and well secured with ropes fastened to the piles, it could not move from its place. By raising it no higher, much labour and expense were saved; yet it answered the intent full as well as if it had been high enough to reach above the highest flood.

The pier being thus carried on above low-water mark, the masons finished the rest of it during the intervals of the tides in the usual way; and after all the piers and abutments were finished in a like manner, the arches were begun and completed as mentioned before; the whole bridge was built in about seven years; without any accidents happening either in the work or to the workmen, which is seldom the case in works of this nature.

It may be observed, that all the piers were built with solid Portland stone, some of which weighed four tons. The arch-stones were likewise of the same sort; but the rest of the masonry was finished with Kentish rag-stones; and the paths for foot passengers were paved with parquet, which is the hardest stone to be had in England, excepting Plymouth marble.

This method of building bridges is certainly the easiest and cheapest that can be thought of, but cannot be used in many cases: when the foundation is so bad as not to be depended upon without being piled, or the depth of water is very great, with a strong current and no tide, it cannot then be practised. For, if piles are to be used, it will be next to impossible to cut them off in the same level five or six feet below the bed of the river, notwithstanding that saws have been invented for that purpose: because if they are cut off separately, it will be a hard matter to do it so nicely that the one shall not exceed the other in height; and if this is not done, the grating or bottom of the coffer will not be equally supported, whereby the foundation becomes precarious; neither can they be cut off all together; for piles are to be driven as far as the bottom of the coffer extends, which at Westminster bridge was 27 feet; the saw must have three feet play, which makes the total length of the saw 30 feet; now if either the water is deeper than it is there, or the arches are wider, the saw must still be longer; so that this method is impracticable in any such case.

In a great depth of water that has a strong current, and no tide, the coffers must reach above the water, which makes them very expensive, and unwieldy to manage, as well as very difficult to be secured in their places, and kept steady; so that there is no probability of using them in such a case.

In some cases, when there is a great depth of water, and the bed of the river is tolerably level, or where it tho'd. can be made so by any contrivance, a very strong frame of timber about four times as large as the bale of the piers may be let down with stones upon it round the edges to make it sink: after fixing it level, piles must be driven about it to keep it in its place; and then the foundation may be laid in coffers as before, which are to be kept steady by means of ropes tied to the piles.

This method has frequently been used in Russia; and though the bed of the river is not very solid, yet such a grate, when once well settled with the weight of the pier upon it, will be as firm as if piles had been driven under the foundation; but to prevent the water from gulling under the foundation, and to secure it against all accidents, a row of dove-tail piles must be driven quite round the grating: this precaution being taken, the foundation will be as secure as any that can be made.

The French engineers make use of another method in raising the foundations of masonry under water; which is, to drive a row of piles round the intended place, nearer to, or farther from each other, according as the water is more deep or shallow: these piles, being strongly bound together in several places with horizontal tie-beams, serve to support a row of dove-tail piles driven within them: when this is done, and all well secured according to the nature of the situation and circumstances, they dig the foundation by means of a machine with scoops, invented for that purpose, until they come to a solid bed of gravel or clay; or if the bed of the river is of a soft consistence to a great depth, it is dug only to about 6 feet, and a grate of timber is laid upon it, which is well secured with piles driven into the opposite corners of each square, not minding whether they exceed the upper surface of the grate much or little.

When the foundation is thus prepared, they make a kind of mortar called beton, which consists of twelve parts of pozolano or Dutch terras, six of good sand, nine of unslaked lime the best that can be had, thirteen of stone splinters not exceeding the bigness of an egg, and three parts of tile-dust, or cinders, or else scales of iron out of a forge: this being well worked together... Part II.

Practice. together must be left standing for about 24 hours, or till it becomes so hard as not to be separated without a pick-ax.

This mortar being thus prepared, they throw into the coffer a bed of rubble-stone, not very large, and spread them all over the bottom as nearly level as they can; then they sink a box full of this hard mortar, broken into pieces, till it come within a little of the bottom; the box is so contrived as to be overfed or turned upside down at any depth; which being done, the pieces of mortar soften, and to fill up the vacant spaces between the stones; by these means they sink as much of it as will form a bed of about twelve inches deep all over: then they throw in another bed of stone, and continue alternately to throw one of mortar and one of stone till the work approaches near the surface of the water where it is levelled, and then the rest is finished with stones in the usual manner.

Mr Belidor says, in the second part of his hydraulics, vol. ii. p. 188, that Mr Milet de Montville having filled a coffer containing 27 cubic feet, with masonry made of this mortar, and sunk it into the sea, it was there left standing for two months, and when it was taken out again it was harder than stone itself.

We have hitherto mentioned such situations only where the ground is of a soft nature; but where it is rocky and uneven, all the former methods prove ineffectual; nor indeed has there yet been any one proposed which can be always used upon such occasions, especially in a great depth of water. When the water is not so deep but that the unevenness of the rock can be perceived by the eye, piles strongly fixed with iron may be raised and let fall down, by means of a machine, upon the higher parts, so as to break them off piece by piece, till the foundation is tolerably even, especially when the rock is not very hard; which being done either this or any other way that can be thought of, a coffer is made without any bottom, which is let down and well secured, so as not to move from its place: to make it sink, heavy stones should be fixed on the outside; then strong mortar and stones must be thrown into it; and if the foundation is once brought to a level, large hewn stones may be let down so as to lie flat and even; by these means the work may be carried on quite up to the surface of the water. But when the water is so deep, or the rock so hard as not to be levelled, the foundation must be founded, so as to get nearly the risings and fallings; then the lower part of the coffer must be cut nearly in the same manner, and the rest finished as before. It must however be observed, that we suppose a possibility of sinking a coffer; but where this cannot be done, no method that we know of will answer.

Among the aquatic buildings of the ancients none appears to have been more magnificent than Trajan's bridge. Dion Cassius gives the following account of it: "Trajan built a bridge over the Danube, which in truth one cannot sufficiently admire; for though all the works of Trajan are very magnificent, yet this far exceeds all the others; the piers were 20 in number, of square stone; each of them 150 feet high above the foundation, 60 feet in breadth, and distant from one another 150 feet. Though the expense of this work must have been exceeding great, yet it becomes more extraordinary by the river's being very rapid, and its bottom of a soft nature: where the bridge was built, was the narrowest part of the river thereabout, for in most others it is double or treble this breadth; and although on this account it became so much the deeper and the more rapid, yet no other place was so suitable for this undertaking. The arches were afterwards broken down by Adrian; but the piers are still remaining, which seem as it were to testify that there is nothing which human ingenuity is not able to effect." The whole length then of this bridge was 1590 yards; some authors add, that it was built in one summer, and that Apollodorus of Damascus was the architect, who left behind him a description of this great work.

Where stone bridges cannot be erected on account of the expense, very strong and durable ones may be bridged constructed of wood: in which case, they ought to be so framed, as that all the parts may press upon one another like the arch of a stone bridge; and thus, instead of being weakened by great weights passing over them, they will become the stronger. How this is to be accomplished, will be better understood from § Plate XXXIX. fig. 3, which represents a wooden bridge constructed after this manner, than it can be by any description.

2. Of Harbours.

In these, the first thing to be considered is the situation; which may be some large creek or basin of water, in or near the place where the harbour is intended to be made, or at the entrance of a large river, or near the sea: for a harbour should never be dug entirely out of dry land, unless upon some extraordinary occasions, where it is impossible to do otherwise, and yet a harbour is absolutely necessary. When a proper place is found, before it is fixed upon, it must be considered whether ships can lie there safe in stormy weather, especially when those winds blow which are most dangerous upon that coast; whether there be any hills, rising ground, or high buildings, that will cover it; in these cases, the situation is very proper: but if there be nothing already that will cover the ships, it must be observed whether any covering can be made at a moderate expense, otherwise it would be useless to build a harbour there.

The next thing to be considered is, whether there be a sufficient depth of water for large ships to enter with safety, and lie there without touching the ground; and if not, whether the entrance and inside might not be made deeper at a moderate expense: or, in case a sufficient depth of water is not to be had for large ships, whether the harbour would not be useful for small merchantmen; for such a one is often of great advantage, when situated upon a coast much frequented by small coasting vessels.

The form of the harbour must be determined in such a manner, that the ships which come in when it is stormy weather may lie safe, and so as there may be sufficient room for as many as pass that way: the depths of water where the piers are to be built must be taken at every 10, 15, or 20 feet distance, and marked upon piles driven here and there, in order that the workmen may be directed in laying the foundation.

This being done, it must be considered what kind of materials are to be used, whether stone, brick, or wood. When stones are to be had at a moderate price, they ought ought to be preferred, because the work will be much stronger, more lasting, and need fewer repairs, than if made with any other materials; but when stones are scarce, and the expense becomes greater than what is allowed for building the harbour, the foundation may be made of stone as high as low-water mark, and the rest finished with brick. If this manner of building should still be too expensive, wood must be used; that is, piles are driven as close as is thought necessary, which being fastened together by cross-bars, and covered with strong oaken planks, form a kind of coffer, which is filled with all kinds of stones, chalk, and shingles.

The manner of laying the foundation in different depths of water, and in various soils, requires particular methods to be followed. When the water is very deep, the French throw in a great quantity of stones at random, so as to form a much larger base than would be required upon dry land; they continue to within 3 or 4 feet of the surface of the water, where they lay the stones in a regular manner, till the foundation is raised above the water; they then lay a great weight of stones upon it, and let it stand during the winter to settle; as likewise to see whether it is firm, and resists the force of the waves and winds; after that, they finish the superstructure with large stones in the usual manner.

As this method requires a great quantity of stones, it can be practised only in places where stones are in plenty; and therefore the following one is much preferable. A coffer is made with dove-tail piles about 30 yards long, and as wide as the thickness of the foundation is to be; then the ground is dug and levelled, and the wall is built with the best mortar.

As soon as the mortar is tolerably dry, those piles at the end of the wall are drawn out, the side-rows are continued to about 30 yards farther, and the end inclosed; then the foundation is cleared, and the stones laid as before. But it must be observed, that the end of the foundation finished is left rough, in order that the part next to it may incorporate with it in a proper manner; but if it is not very dry, it will incline that way of itself, and bind with the mortar that is thrown in next to it; this method is continued till the whole pier is entirely finished.

It must likewise be observed, that the piers are not made of one continued solid wall; because in deep water it would be too expensive; for which reason, two walls are built parallel to each other, and the interval between them is filled up with shingle, chalk, and stone. As these walls are in danger of being thrust out or over-set, by the corps in the middle, together with the great weight laid at times on the pier, they are tied or bound together by cross-walls at every 30 or 40 yards distance, by which they support each other in a firm and strong manner.

In a country where there is a great plenty of stones, piles may be driven in as deep as they will go, at about two or three feet distance; and when the foundation is sunk and levelled, large stones may be let down, which will bed themselves; but care must be taken to lay them close, and so as to have no two joints over each other; and when the wall is come within reach, the stones must be cramp together.

Another method practised, is to build in coffers much after the same manner as has been done in building the piers of Westminster-bridge; but as in this case the ends of the coffers are left in the wall, and prevent their joining so well as to be water-tight, the water that penetrates through and enters into the corps may occasion the wall to burst and to tumble down. Another inconvenience arising from this manner of building is, that as there are but few places without worms, which will destroy wood wherever they can find it; by their means the water is let into the pier, and consequently makes the work liable to the same accident as has been mentioned above.

To prevent these inconveniences, the best method is, to take the wood away, and joggle the ends of the walls together with large stones, pouring terras-mortar into the joints; when this is done, the water between the two walls may be pumped out, and the void space filled up with stone and shingle as usual: or if these joggles cannot be made water-tight, some dove-tail piles must be driven at each end as close to the wall as can be done, and a strong sail-cloth put on the outside of them, which, when the water is pumped out, will stick so close to the piles and wall, that no water can come in. This method is commonly used in Russia.

The thickness of a pier depends on two considerations: it ought to be both such as may be able to resist the shock of the waves in stormy weather; and also to be of a sufficient breadth above, that ships may be laden or unladen whenever it is thought necessary. Now, because the specific gravity of sea-water is about one half that of brick, and as 2 to 5 in comparison of stone; and since the pressure of stagnated water against any surface is equal to the weight of a prism of water whose altitude is the length of that surface, and whose base is a right angled isosceles triangle, each of the equal sides being equal to the depth of the water; therefore a pier built with bricks, whose thickness is equal to the depth of the water, will weigh about four times as much as the pressure of the water against it; and one of stone of the same breadth, about 6 times and a quarter as much. Now this is not the force to be considered, since this pressure is the same within as without the pier; but it is that force with which the waves strike against the piers, and that depends on the weight and velocity of the waves, which can hardly be determined; because they vary according to the different depths of water, the distance from the shore, and according to the tides, winds, and other causes. Consequently the proper thickness of the piers cannot be determined by any other means than by experience.

Practitioners suppose, that if the thickness of a pier is equal to the depth of the water, it is sufficient; but for a greater security they allow 2, 3, or 4 feet more. This might probably do, if piers were built with solid stones cramp together; but as this is hardly ever the case, and on the contrary, as the inside is filled up with shingle, chalk, or other loose materials, their rule is not to be depended upon: besides it makes the space above too narrow for loading and unloading the ships, unless in a great depth of water; so that it does not appear that their method can be followed, excepting in a very few cases where the water has but very little motion.

When stone can be had, no other materials should be used, because they being of a larger bulk than brick, will will better resist the waves by their own weight, till such time as the mortar is grown hard; for after this is effected, brick will resist better against the action of sea-water than soft stones.

The wall must be built with terracotta mortar from the bottom to the height of low-water mark, and the rest finished with cinder or tile-dust mortar, which has been found sufficiently good in those places where the wall is wet and dry alternately. The upper part of the pier should be paved with flat hewn stones laid in strong mortar, in order to prevent any water from penetrating into it; iron rings ought also to be fixed here and there at proper distances, to fasten the ships, and prevent them from striking against the pier when agitated by the waves.

Wooden fenders or piles should be driven at the inside close to the wall, and cramp to it with iron, to prevent the ships from touching them, and from being worn by the continual motion. Where the sea breaks against the piers with great violence, breakers should be made at proper distances; that is, two rows of piles are driven nearly at right angles to the piers for the length of about 12 or 15 feet, and at about 8 or 10 feet distant from each other; and then another to join the two former: these piles being covered with planks, and the inside being filled with shingle and rubble-stones, then the top is paved with stones of about a foot in length, set long-wise to prevent the waves from tearing them up. This precaution is absolutely necessary where the water rushes in very strongly.

---

**ARC**

Military Architecture, the same with what is otherwise called fortification. See Fortification.

Naval Architecture, the art of building ships. See Ship-building.

Architallassus, or admiral-shell, a synonyme of a species of voluta. See Voluta.