Civil Architecture.

As a rule, it may be asserted that the southern province of Italy is singularly deficient in examples of civil or domestic architecture. Great monastic establishments existed there during the Middle Ages which must have possessed buildings befitting their magnificence; but these have either perished and been rebuilt, or have been so restored that their original forms can hardly be recognised. There are, indeed, cloisters at Amalfi and Sorrento; much more remarkable, however, for the beauty of their situation than for their architecture, which is extremely rude and clumsy. There are no chapter-houses: no halls or conventual buildings of any sort. In this respect, the province forms a remarkable contrast with Spain in the same age; though it must be confessed that the North of Italy is also very deficient in conventual buildings of the Middle Ages, the most magnificent and beautiful belonging more to the Renaissance than to the Mediæval period.

At Ravello there is the Casa Ruffolo, a picturesque palace of the 13th century, still nearly entire: a strange mixture of Gothic and Saracenic taste, but so exceptional, that it would not be fair to quote it as a type of any style. It seems to owe its peculiarities more to the taste of some individual patron or architect rather than to any national taste or form of design.

490. Plan of Castel del Monte. (From Schultz.) Scale 100 ft. to 1 in.

There are, however, several Hohenstauffen castles of tolerable preservation, more or less typical of the domestic arts of the day in which they were erected. One of the best preserved of these is that of Castel del Monte, erected by Frederick II., 1240-44. It is an octagon in plan, with octagonal turrets at each angle. It measures 167 ft. across its extreme breadth, and surrounds a courtyard 57 ft. in diameter. Both storeys are vaulted, and all the details throughout are good and pleasing. The whole is an admixture of Italian taste, superimposed on a German design; but it will be observed how little removed the architectural details of the entrance are, even at that early age, from the style of the Renaissance. This is, indeed, the great characteristic of the architectural objects in Southern Italy. Though they adopted Christian forms, they never abandoned the classical feeling in details; and it is this which mainly renders them worthy of study. Whether considered in regard to dimensions, outline, or constructive peculiarities, their churches will not bear a moment’s comparison with those of the North; but in elegance of detail they often surpass purely Gothic buildings to such a degree as to become to some extent as worthy of study as their more ambitious rivals.

491. Part Section, part Elevation, of Castel del Monte. (From Schultz.) Scale 50 ft. to 1 in.

CHAPTER VI.
POINTED ITALIAN GOTHIC.

CONTENTS.

Fresco paintings—Churches at Vercelli, Asti, Verona, and Lucca—Cathedral at Siena—Sta. Maria, Florence—Church at Chiaravalle—St. Petronio, Bologna—Cathedral at Milan—Certosa, near Pavia—Duomo at Ferrara.

CHRONOLOGY.

DATES.
Bologna independentA.D. 1112
Countess Matilda at Florence1115
Obizzo d’Este at Ferrara1184
Enrico Dandolo takes Constantinople1203
War between Genoa and Venice1205
Azzo d’Este at Ferrara1208
Martino della Scala at Verona1259
Martino delle Torre at Milan1260
Visconti Lord of Milan1277
Taddeo de Pepoli at Bologna1334
Conspiracy of Marino Faliero1355
Gian Galeazzo Visconti, Duke of Milan1395
Verona ceded to Venice1409
Cosmo de’ Medici1434

Before the commencement of the 13th century, the Italians had acquired such mastery over the details of their round-arched style, and had worked it into such originality and completeness, that it is surprising that they should so easily have abandoned it for that form of Pointed Gothic which they afterwards adopted. It is true the Italians never rose to the conception of such buildings as the great Rhenish cathedrals, like those of Spires and Worms, or the old churches at Cologne; nor did they perhaps even rival the quasi-classical grace and elegance of the Provençal churches; but at Verona, Modena, and indeed throughout the North of Italy, they had elaborated a complete round-arched style, all the details of which were not only appropriate and elegant, but seemed capable of indefinite development in the direction in which they were proceeding. They had also before their eyes the Romanesque style of Pisa and Lucca with all its elegance, and the example of Rome, where the architects steadily refused to acknowledge the pointed arch during the whole of the Mediæval period. Yet in the beginning of the 13th century—say 1220, when the cathedrals of Amiens, Salisbury, and Toledo were designed—Italy too was smitten with admiration for the pointed arch, and set to work to adapt it to her tastes and uses.

It would be difficult to account for this, were we not aware how deeply the feelings that gave rise to the Ghibelline faction were rooted in the Italian soil. In all the cities, except Rome, the cause of the Ghibellines was throughout the Middle Ages identified with that of freedom and local independence, in opposition to that of the Guelfs, which symbolised the supremacy of the Pope and the clerical party. Knowing how strenuously this was resisted, we naturally expect to find it expressed in the architecture of the country. Two, indeed, of the great churches of Italy, Assisi (1228) and Milan (1385), were erected by Germans in the German style of the day; but these are exceptional. The form which the pointed-arched style took on its introduction, was that of adaptation to the Italian style, in a manner which the Italians thought more consonant with beauty and convenience than that adopted north of the Alps. In this they were certainly mistaken. The elegance of the details employed by a refined and cultivated people, and based on classical traditions, goes far to redeem, in most instances, the defects of their designs; but they never grasped the true principles of Gothic art, and the fatal facility of the pointed arch led them more astray after mechanical clevernesses than even the Germans. Still, it is an original style, and, however imperfect, is well worthy of study.

Before proceeding to describe the style more in detail, it may be well to point out one of the principal causes which led to the more marked features of difference between the Gothic architecture of Italy and that of Germany and France. This was the distaste of the Italians for the employment of painted glass, or at least their want of appreciation of its beauties when combined with architecture.

It will be explained in a future chapter how all-important painted glass was to the elaboration of the Gothic style. But for its introduction, the architecture of France would bear no resemblance to what it was, and is. In Italy, indeed, the people loved polychromy, but always of the opaque class. They delighted to cover the walls of their churches with frescoes and mosaics, to enrich their floors with the most gorgeous pavements, and to scatter golden stars over the blue ground of their vaults; but rarely, if ever, did they fill, or design to fill, their windows with painted glass. Perhaps the glare of an Italian sun may have tended to render its brilliancy intolerable; but more probably the absence of stained glass is owing to its incompatibility with fresco-painting, the effect of which would be entirely destroyed by the superior brightness of the transparent material. The Italians were not prepared to relinquish the old and favourite mode of decoration in which they so excelled. This adherence to the ancient method of ornamentation enabled them, in the 15th and 16th centuries, to surpass all the world in the art of painting, but it was fatal to the proper appreciation of the pointed style, and to its successful introduction into the land.

The first effect of this tendency was that the windows in Italian churches were small, and generally devoid of tracery, with all its beautiful accompaniments. The walls, too, being consequently solid, were sufficient, by their own weight, to abut the thrust of the arches: so that neither projecting or flying buttresses nor pinnacles were needed. The buildings were thus deprived externally of all the aspiring vertical lines so characteristic of true Gothic. The architects, to relieve the monotony arising from the want of these features, were forced to recur to the horizontal cornices of the classical times, and to cover their walls with a series of panelling which, however beautiful in itself, is mere ornament—both unmeaning and inconsistent.

Internally, too, having no clerestory to make room for, and no constructive necessities to meet, they jumped to the conclusion that the best design is that which covers the greatest space with the least expenditure of materials, and the least encumbrance of the floor. With builders this is a golden rule, but with architects it is about the worst that can possibly be adopted. The Germans were not free from this fault, but the Italians carried it still further. If on four or five piers they could support the vault of a whole nave, they never dreamed of introducing more. A French architect, though superior in constructive skill, would probably have introduced eight or ten in the same space. An Italian aimed at carrying the vaults of the side-aisles to the same height as that of the nave, if he could. A Northern architect knew how to keep the two in their due proportion, whereby he obtained greater height and greater width in the same bulk, and an appearance of height and width greater still, by the contrast between the parts, at the same time that he gave his building a character of strength and stability perhaps even more valuable than that of size.

In the same manner the Northern architects, while they grouped their shafts together, kept them so distinct as to allow every one to bear its proportional part of the load, and perform its allotted task. The Italians never comprehended this principle, but merely stuck pilasters back to back, in imitation of the true architects, producing an unmeaning and ugly pier. The same incongruities occur in every part and every detail. It is a style copied without understanding, and executed without feeling. The elegance of the sculptured foliage and other details sometimes goes far to redeem these faults; for the Italians, though bad architects, were always beautiful carvers, and, as a Southern people, were free from the vulgarities sometimes apparent farther north, and never fell into the wild barbarisms which too often disfigure even the best buildings on this side of the Alps. Besides, when painting is joined to sculpture in churches, the architecture may come to occupy a subordinate position, and thus escape the censure it deserves. Unfortunately there are only two examples of any importance in this style that retain all their painted decorations—St. Francis at Assisi, and the Certosa near Pavia. From this circumstance they are perhaps the most admired in Italy. In others the spaces left for colour are still plain and blank. We see the work of the architect unaided by the painting which was intended to set it off, and we cannot but condemn it as displaying at once bad taste and ignorance of the true Gothic feeling.

One of the earliest, or perhaps the very first Italian edifice into which the pointed arch was introduced, is the fine church of St. Andrea at Vercelli, commenced in the year 1219 by the Cardinal Guala Bicchieri, and finished in three years. This prelate, having been long legate in England, brought back with him an English architect called, it is said, Brigwithe, and entrusted him with the erection of this church in his native place.

492. Plan of the Church at Vercelli. (From Osten’s ‘Baukunst in Lombardei.’) Scale 100 ft. to 1 in.

In plan, it is certainly very like an English church, terminating squarely towards the east, and with side chapels to the transepts, arranged very much as we find them at Buildwas, Kirkstall, and other churches of this class and size, only that here they are polygonal, which was hardly ever the case in England. But with the plan all influences of the English architect seem to have ceased, and the structure is in purely Italian style. Externally the pointed arch nowhere appears, all the doors and windows being circular-headed; while internally it is confined to the pier-arches of the nave and the vaulting of the roof. The façade is flanked at its angles by two tall, slender, square towers; and the intersection of the nave and transept is covered by one of those elegant octagonal domes which the Italians knew so well how to use, and which is in fact the only original feature in their designs. The external form of this church is interesting, as displaying the germs of much that two centuries afterwards was so greatly expanded by a German architect in the design of Milan cathedral.

A few years later, in 1229, a church was commenced at Asti, the tower of which was finished in 1266. This allowed time for a more complete development of the pointed style, which here prevails not only internally, but externally. Tall pointed windows appear in the flanks, and even the doorways assume that form, in their canopies, if not in their openings. The porch (Woodcut No. [493]) is a later addition, and a characteristic specimen of the style during the 14th century. This church is also one of the earliest examples in which those elegant terra-cotta cornices of small intersecting arches seem to have been brought to perfection.

493. Church at Asti. (From Chapuy, ‘Moyen-Âge Monumental.’)

The most remarkable church of this age is that of St. Francis at Assisi, commenced in 1228, and finished, in all essentials at least, in 1253. It is said to have been built by a German named Jacob, or Jacopo. Certainly no French or English architect would have designed a double church of this class, though, on the other hand, no Italian could have drawn details so purely Northern as those of the upper church. In the lower church there are hardly any mouldings to mark the style, but its character is certainly rather German than Italian. This church depends for its magnificence and character much more on painting than on architecture. In the first place it is small, the upper church being only 225 ft. long, by 36 in. width; and though the lower one has side-aisles which extend the width to 100 ft., yet the upper church is only 60 ft. in height, and the lower about 30, so that it is far too small for much architectural magnificence. None of its details are equal to those of contemporary churches on this side of the Alps. The whole church is covered with fresco paintings in great variety and of the most beautiful character, which justly render it one of the most celebrated and admired of all Italy. On this side of the Alps without its frescoes, it would hardly attract any attention. It is invaluable as an example of the extent to which the polychromatic decoration may be profitably carried, and of the true mode of doing it; and also as an illustration of the extent to which the Italians allowed a foreign style and mode of ornamentation to be introduced into their country.

494. Plan of Sta. Anastasia, Verona. Scale 100 ft. to 1 in.

495. One Bay of Sta. Anastasia, Verona.

One of the purest and most perfect types of an Italian Gothic church is that of Sta. Anastasia at Verona, commenced apparently in 1260. It is not large, being only 285 ft. in length externally; but its arrangements are very complete, and very perfect if looked at from an Italian point of view. The square of the vault of the nave is the modulus, instead of that of the aisles, as in true Gothic churches: owing to which the pier-arches are further apart than a true artist would have placed them; there are also no buttresses externally, but only pilasters. The consequence of this is, that the arches have to be tied in with iron rods at the springing, which internally adds very much to the appearance of weakness, caused in the first instance by the wide spacing and general tenuity. These bad effects are aggravated by the absence of a string-course at the springing of the vault; and by the substitution of a circular hole for the triforium, and a hexafoiled opening of very insignificant dimensions for the glorious clerestory windows of Northern churches. Altogether, though we cannot help being pleased with the spaciousness and general elegance of design, it is impossible not to feel how very inferior it is to that of churches on this side the Alps.

496. One Bay, externally and internally, of the Church of San Martino, Lucca.

The church of San Martino at Lucca, built about a century after Sta. Anastasia (middle of 14th century), presents a strikingly happy compromise between the two styles. The pier-arches are still too wide—23 ft. in the clear; but the defect is remedied to some extent by the employment of circular instead of pointed arches, and the triforium is all that can be desired; the clerestory, however, is as insignificant as it must be where the sun is so brilliant and painted glass inadmissible. It would be easy to point out other defects; but, taking it altogether, there are few more elegant churches than this, and hardly one in Italy that so perfectly meets all the exigencies for which it was designed.

497. Plan of the Cathedral at Siena. (From the ‘Églises principales d’Europe.’) Scale 100 ft. to 1 in.

The cathedrals of Siena and Orvieto (the former commenced in 1243, the latter in 1290) are perhaps, taken altogether, the most successful specimens of Italian pointed Gothic. They are those at least in which the system is carried to the greatest extent without either foreign aid or the application of distinctly foreign details. These two buildings, moreover, both retain their façades as completed by their first architects, while the three great churches of this style—the cathedrals of Florence, Bologna, and Milan—were in this respect left unfinished, with many others of the smaller churches of Italy. The church at Siena illustrates forcibly the tendency of the Italian architects to adhere to the domical forms of the old Etruscans, which the Romans amplified to such an extent, and the Byzantines made peculiarly their own. I cannot but repeat my regret that the Italians alone, of all the Western Mediæval builders, showed any predilection for this form of roof. On this side of the Alps it could have been made the most beautiful of architectural forms. In Italy there is no instance of more than moderate success—nothing, indeed, to encourage imitation. Even the example now before us is no exception to these remarks, though one of the boldest efforts of Italian architects. In plan it ought to have been an octagon, but that apparently would have made it too large for their skill to execute, so they met the difficulty by adopting a hexagon, which, though producing a certain variety of perspective, fits awkwardly with the lines of columns, and twists the vaults to an unpleasant extent. Still, a dome of moderate height, and 58 ft. in diameter, covering the centre of the church, and with sufficient space around to give it dignity, is a noble and pleasing feature, the merit of which it is impossible to deny. Combined with the rich colouring and gorgeous furniture of the church, it makes up a whole of great beauty. The circular pier-arches, however, and the black and white stripes by which the exterior is marked, detract considerably from the effect of the whole—at least in the eyes of strangers, though the Italians still consider it a beauty. The façade of this cathedral is represented in Woodcut No. [498]. It consists of three great portals, the arches of which are equal in size, though the centre doorway is larger than those at the sides. Above is the invariable circular window of the Italian architects, and the whole is crowned by steep triangular gables. Beneath the cathedral, or rather under the choir, is the ancient baptistery, now the church of St. John the Baptist; its front is in a much purer style of Gothic than the cathedral.[[310]]

498. Façade of the Cathedral at Siena.

The carved architectural ornaments of the façade are rich and elaborate in the extreme, though figured sculpture is used to a much less extent than in Northern portals of the same age. It is also observable that the strong horizontal lines do not harmonise with the aspiring character of pointed architecture.

The cathedral of Orvieto is smaller and simpler, and less rich in its decorations, than that at Siena, with the exception of its façade, which is adorned with sculpture and painting. Indeed the three-gabled front may be considered the typical one for churches of this class. The façades intended to have been applied to the churches at Florence, Bologna, Milan, and elsewhere, were no doubt very similar to that represented in Woodcut No. [498]. As a frontispiece, if elaborately sculptured and painted, it is not without considerable appropriateness and even beauty; but, as an architectural object, it is infinitely inferior to the double-towered façades of the Northern cathedrals, or even to those with only one great tower in the centre. It has besides the defect of not expressing what is behind it; the central gable being always higher than the roof, and the two others merely ornamental appendages. Indeed, like the Italian Gothic buildings generally, it depended on painting, sculpture, and carving for its effect, far more than on architectural design properly so called.

Among the greatest and most complete examples of Italian Gothic is the church of Sta. Maria dei Fiori, the cathedral of Florence, one of the largest and finest churches produced in the Middle Ages—as far as mere grandeur of conception goes, perhaps the very best, though considerably marred in execution from defects of style, which are too apparent in every part.

499. Plan of Cathedral at Florence. (From Isabelle, ‘Édifices Circulaires.’) Scale 100 ft. to 1 in.

The building of the church was commenced in the year 1294 or 1298 (it is not quite clear which), from the designs and under the superintendence of Arnolfo di Lapo, for unfortunately in this style we know the names of all the architects, and all the churches show traces of the caprice and of the misdirected efforts of individuals, instead of the combined national movement which produced such splendid results in France and England. It is not known how far Arnolfo had carried the building when he died, in 1310, but probably up to the springing of the vaults. After this the works proceeded more leisurely, but the nave and smaller domes of the choir were no doubt completed as we now find them in the first twenty years of the 14th century. The great octagon remained unfinished, and, if covered in at all, it was only by a wooden roof of domical outline externally, which seems to be that represented in the fresco in the convent of San Marco, till Brunelleschi commenced the present dome in 1420, and completed it in all essential parts before his death, which happened in 1444. The building may therefore be considered as essentially contemporary with the cathedral of Cologne, which it very nearly equals in size (its area being 84,802 ft., while that of Cologne is estimated at 91,000), and, as far as mere conception of plan goes, there can be little doubt but that the Florentine cathedral far surpasses its German rival. Nothing indeed can be finer than its general ground-plan. A vast nave leads to an enormous dome, extending into the triapsal arrangement so common in the early churches of Cologne, and which was repeated in the last and greatest effort of the Middle Ages, or rather the first of the new school—the great church of St. Peter at Rome. In the Florentine church all these parts are better subordinated and proportioned than in any other example, and the mode in which the effect increases and the whole expands as we approach from the entrance to the sanctum is unrivalled. All this, alas! is utterly thrown away in the execution. Like all inexperienced architects, Arnolfo seems to have thought that largeness of parts would add to the greatness of the whole, and thus used only four great arches in the whole length of his nave, giving the central aisle a width of 55 ft. clear. The whole width is within 10 ft. of that of Cologne, and the height about the same; and yet, in appearance, the height is about half, and the breadth less than half, owing to the better proportion of the parts and to the superior appropriateness in the details on the part of the German cathedral. At Florence the details are positively ugly. The windows of the side-aisles are small and misplaced, those of the clerestory mere circular holes. The proportion of the aisles one to another is bad, the vaults ill-formed, and altogether a colder and less effective design was not produced in the Middle Ages. The triapsal choir is not so objectionable as the nave, but there are large plain spaces that now look cold and flat; the windows are too few and small, and there is a gloom about the whole which is very unsatisfactory. It is nearly certain that the original intention was to paint the walls, and not to colour the windows, so that these defects are hardly chargeable to the original design, and would not be apparent now were it not that in a moment of mistaken enthusiasm the Florentines were seized with a desire to imitate the true style of Gothic art, and rival Northern cathedrals in the glory of their painted glass. This, in a church whose windows were designed only of such dimensions as were sufficient to admit the requisite quantity of white light, was fatal. Notwithstanding the beauty of the glass itself, which seems to have been executed at Lubeck, 1434, from Italian designs, it is so completely out of place that it only produces irritation instead of admiration, and has certainly utterly destroyed the effect and meaning of the interior it was intended to adorn.

500. Section of Dome and part of Nave of the Cathedral at Florence. Scale 100 ft. to 1 in.

501. Part of the Flank of Cathedral at Florence.

502. Dome at Chiaravalle, near Milan. (From a drawing by Ed. Falkener, Esq.)

Externally the façade was never finished,[[311]] and we can only fancy what was intended from the analogy of Siena and Orvieto. The flanks of the nave are without buttresses or pinnacles, and, with only a few insignificant windows, would be painfully flat except for a veneer of coloured marbles disposed in panels over the whole surface. For an interior or a pavement such a mode of decoration is admissible; but it is so unconstructive, so evidently a mere decoration, that it gives a weakness to the whole, and most unsatisfactory appearance to so large a building. This is much less apparent at the east end, where the outline is so broken, and the main lines of the construction so plainly marked, that the mere filling in is comparatively unimportant. This is the most meritorious part of the church, and, so far as it was carried up according to the original design, is extremely beautiful. Even the plainness and flatness of the nave serve as a foil to set off the varying outline of the choir. Above the line of the cornice of the side-aisles there is nothing that can be said to belong to the original design except the first division of the drum of the dome, which follows the lines of the clerestory. It has long been a question what Arnolfo originally intended, and especially how he meant to cover the great octagonal space in the centre. All knowledge of his intentions seems to have been lost within a century after his death: at least, in the accounts of the proceedings of the commission which resulted in the adoption of Brunelleschi’s design for the dome, no reference is made to any original design as then existing, and no one appears to have known how Arnolfo intended to finish his work. Judging from the structure as far as he carried it, and with the knowledge we now possess of the Italian architecture of that age, we can easily conjecture what his design for its completion may have been. Internally, it probably consisted of a dome something like the present, but flatter, springing from the cornice, 40 ft. lower than the present one, and pierced with large openings on each of its eight faces.

503. Section of Eastern portion of Church at Chiaravalle. (From Gruner’s ‘Terra Cotta Architecture in Italy.’) Scale 50 ft. to 1 in.

Externally, two courses were open to him. The first and most obvious was to hide the dome entirely under a wooden roof, as is done in St. George’s, Thessalonica (Woodcut No. [305]), or in the baptistery in front of the cathedral, and is done in half the baptisteries in Italy—as at Parma, for instance (Woodcut No. 514). Had he done this, the span of the dome might have been very much larger, without involving any constructive difficulties, and the three towers over the choir and transepts might have sufficed to relieve its external appearance sufficiently for architectural effect. On the whole, however, I am rather inclined to believe that something more ambitious than this was originally proposed, and that the design was more like that of Chiaravalle near Milan, built in 1221, and one of the most complete and perfect of this class of dome now existing in Italy. Its external appearance may be judged of from Woodcut [502], and its constructive details from the section, Woodcut No. [503].

If the basement is sufficiently solid—and that at Florence is more than sufficient for any superstructure of the sort—it is evident the architect can dispose of such masses of masonry, that he can counteract any thrust or tendency to spread that can exist in any dome of this sort; and instead of being only 136 ft. across, 150 or 160 might easily have been attempted. Instead of 375 ft., which is the height of the present dome from the floor to the top of the cross externally, it might even with the present diameter have been carried up to at least 500 ft., or as high as the church was long,—70 to 100 ft. above the height of St. Peter’s at Rome.

Had this been done, the three smaller semi-domes must have been intended to be crowned with miniature octagonal spires of the same class with the great dome, and between these the vast substructures show that it was intended to carry up four great spires, probably to a height of 400 ft.

Had all this been done (and something very like it seems certainly to have been intended), neither Cologne Cathedral, nor any church in Europe, ancient or modern, would have been comparable to this great and glorious apse. As it is, the plain, heavy, simple outlined dome of Brunelleschi acts like an extinguisher, crushing all the lower part of the composition, and both internally and externally destroying all harmony between the parts. It has deprived us of the only chance that ever existed of witnessing the effect of a great Gothic dome; not indeed such a dome as might with the same dimensions have been executed on this side of the Alps, but still in the spirit, and with much of the poetry, which gives such value to the conceptions of the builders in those days.

But for this change of plan, the ambition of the Florentines might have been in some measure satisfied, whose instructions to the architect were, that their cathedral “should surpass everything that human industry or human power had conceived of great and beautiful.”

About a century later (1390), the Bolognese determined on the erection of a monster cathedral, which, in so far as size went, would have been more than double that at Florence. According to the plans that have come down to us, it was to have been about 800 ft. long and 525 wide across the transepts; at the intersection was to have been a dome 130 ft. in diameter, or only 6 ft. less than that at Florence; and the width of both nave and transepts was to have been 183 ft.: so that the whole would have covered about 212,000 ft., or nearly the same area as St. Peter’s at Rome, and three times that of any French cathedral! Of this vast design, only about one-third (Woodcut No. [504]), 74,000 sq. ft., was ever carried out; but that fragment is quite sufficient to enable us to judge of the merits or defects of this style in its state of greatest perfection. The only other building in the same style on a sufficient scale to admit of comparison with this is the nave of the cathedral at Florence just described, but that is nearly as may be only half of its dimensions, or 36,000 ft. as compared with 72,000. The chapels, too, at Bologna add practically a fifth aisle, giving great variety and richness to the perspective. The varied heights and proportions of the central and side aisles are singularly pleasing, and there being six arches at Bologna instead of only four as at Florence, and twelve side chapels where none exist in the other example, go far to redeem the lean mechanical look which is the great defect of this style. The great advantage San Petronio has over the Florentine church is in the size and number of its windows, and these not being filled with stained glass the whole church has a bright and pleasing effect that contrasts most favourably with the gloom of its great rival. Notwithstanding this, the nave of San Petronio cannot be considered as a successful work of art. In the first place it is too mechanically perfect. The area of the points of support as compared with the voids is, as far as can be made out from such plans as exist, about one-twelfth, which would be a merit in a railway station, but something more is wanted in a monumental building. In the next there is a singular deficiency of either constructive or constructed ornament. On this side of the Alps an architect with vaulting shafts, string-courses, galleries, and fifty other expedients, would have relieved the bareness of the walls. At Bologna it probably was intended they should be painted, and this never having been executed may account for most of its apparent defects.

504. Plan of the part executed of St. Petronio, Bologna. (from Wiebeking.) Scale 100 ft. to 1 in.

In Gothic architecture one of two systems seems indispensable: either painted glass with strongly-marked carved mouldings over the whole of the interior, or white glass with flat surfaces suitable for opaque paintings. Few cathedrals are complete in both respects at the present day, but in their imperfect state the Northern system has an immense advantage over the Southern. The architecture of our cathedrals is complete and beautiful even in ruins. An Italian church without its coloured decoration is only a framed canvas without harmony or meaning. Were San Petronio as complete in its coloured decoration as the Certosa at Pavia or Monreale at Palermo, it might stand a fair competition with the best interiors on this side of the Alps. As it is, it is only a splendid example of ornamental but unornamented construction, and, as was attempted to be explained in the Introduction, both elements are wanted for success in architectural design.

505. Section of San Petronio, Bologna. (From Wiebeking.) Scale 50 ft. to 1 in.

The exterior of the church is in too unfinished a state to enable us to judge of what its effect might have been if completed, but many of its details, especially of the façade, are of very great beauty, in many respects superior to what is to be found on this side of the Alps. Its central dome, however, never could have been a feature worthy of so vast a church. In diameter it is equal, or nearly so, to that of Florence, but the points of support are so small, and so far apart, that it must have been mainly if not wholly of wood. No such towering structure as Arnolfo’s vast substructures show that he intended, could have stood on the slim supports of the Bolognese church.[[312]]

506. Plan of the Cathedral of Milan. (From ‘Chiesi Principali d’Europa.’) Scale 100 ft. to 1 in.

The cathedral of Milan—at once the most remarkable and one of the largest and richest of all the churches erected in the Middle Ages—was commenced in the year 1385, by order of Gian Galeazzo, first Duke of Milan, and consecrated in 1418, at which date all the essential parts seem to have been completed, though the central spire was not finished till about the year 1440, by Brunelleschi.

The design is said to have been furnished by a German architect, Heinrich Arlez von Gemunden, or as the Italians call him, “da Gamondia,”—a statement which is corroborated by the fact that the details and many of the forms are essentially Northern; but it is equally certain that he was not allowed to control the whole, for all the great features of the church are as thoroughly Italian as the details are German: it is therefore by no means improbable that Marco da Campione, as the Italians assert, or some other native artist, was joined with him or placed over him.

In size it is, except Seville, the largest of all Mediæval cathedrals, covering 107,782 ft. In material it is the richest, being built wholly of white marble, which is scarcely the case with any other church, large or small; and in decoration it is the most gorgeous—the whole of the exterior is covered with tracery, and the amount of carving and statuary lavished on its pinnacles and spires is unrivalled in any other building of Europe. It is also built wholly (with the exception of the façade) according to one design. Yet, with all these advantages, the appearance of this wonderful building is not satisfactory to any one who is familiar with the great edifices on this side of the Alps. Cologne is certainly more beautiful; Rheims, Chartres, Amiens, and Bourges leave a far more satisfactory impression on the mind; and even the much smaller church of St. Ouen will convey far more pleasure to the true artist than this gorgeous temple.

The cause of all this it is easy to understand, since all or nearly all its defects arise from the introduction of Italian features into a Gothic building; or rather, perhaps, it should be said, from a German architect being allowed to ornament an Italian cathedral. Taking the contemporary cathedral of St. Petronio at Bologna as our standard of comparison, it will be seen that the sections (Woodcuts Nos. [505], [507]) are almost identical both in dimensions and in form, except that at Milan the external range is a real aisle instead of a series of side chapels; but, at the same time, it will be perceived that the German system prevailed in doubling the number of the piers between the nave and side-aisles. So far, therefore, the German architect saved the church. The two small clerestories, however, still remain; and although the design avoids the mullionless little circles of Bologna, there is only space for small openings, which more resemble the windows of an attic than of a clerestory. The greater quantity of light being thus introduced by the tall windows of the outer aisle, the appearance is that of a building lighted from below, which is fatal to architectural effect.

The model still preserved on the spot shows that the German architect designed great portals at each end of the transepts. This, however, was overruled in favour of two small polygonal apses. Instead of the great octagonal dome which an Italian would have placed upon the intersection of the whole width of the nave and transepts, German influence has confined it to the central aisle, which is perhaps more to be regretted than any other mistake in the building. The choir is neither a French chevet nor a German or Italian apse, but a compromise between the two, a French circlet of columns enclosed in a German polygonal termination. This part of the building, with its simple forms and three glorious windows, is perhaps an improvement on either of the models of which it is compounded.

507. Section of the Cathedral of Milan.[[313]] (From Wiebeking.) Scale 50 ft. to 1 in.

This is the nearest approach to the French chevet arrangement to be found in all Italy. It is extremely rare in that country to find an aisle running round the choir, and opening into it, or with the circlet of apsidal chapels which is so universal in France. The Italian church is not, in fact, derived from a combination of a circular Eastern church with a Western rectangular nave, but is a direct copy from the old Roman basilica.

508. View of the Interior of Milan Cathedral. (From Rosengarten.)

The details of the interior of Milan cathedral are almost wholly German (Woodcut No. [508]). The great capitals of the pillars, with their niches and statues, are the only compromise between the ordinary German form and the great deep ugly capitals—fragments, in fact, of classical entablatures—which disfigure the cathedrals of Florence and Bologna, and so many other Italian churches. Had the ornamentation of these been carried up to the springing of the vault, they would have been unexceptionable; as it is, with all their richness, their effect is unmeaning.

Externally, the appearance is in outline not unlike that of Sta. Maria dei Fiori; the apse is rich, varied, and picturesque, and the central dome (excepting the details) similar, though on a smaller scale, to what I believe to have been the original design of the Florentine church. The nave is nearly as flat as at Florence, the clerestory not being visible; but the forest of pinnacles and flying buttresses and the richness of the ornamentation go far to hide that defect. The façade was left unfinished, as was so often the case with the great churches of Italy. Pellegrini was afterwards employed to finish it, and a model of his design is still preserved. It is fortunate that his plan was not carried out. The façade was finished, as we now see it, from the designs of Amati, by order of Napoleon. It is commonplace, as might be expected from its age, but inoffensive. The doorways are part of Pellegrini’s design, and the Mediæval forms being placed over those of the cinque-cento, produce a strangely incongruous effect. For the west front several original designs are still preserved. One of these, with two small square towers at the angles, as at Vercelli and elsewhere, was no doubt the Italian design. The German one (Woodcut No. [509]) is preserved by Bassi:[[314]] had this been executed, the façade would have been about one-third (viz. 100 ft.) wider than that of Cologne. Had the height of the towers been in the same proportion, they would have been the tallest in the world. In that case the effect here, as at Cologne, would have been to shorten and overpower the rest of the building to a painful extent. A design midway between the two, with spires rising to the same height as the central one, or about 360 ft., would perhaps have the happiest effect. At any rate, the want of some such features is greatly felt in the building as it stands.

509. Design for Façade of Milan Cathedral. (From Bassi.)

The Certosa, near Pavia, was commenced about the same date (1396) as the cathedral at Milan. It is seldom that we find two buildings in the Middle Ages so close to one another in date and locality, and yet so dissimilar. There is no instance of such an occurrence on this side of the Alps, till modern times; and it shows that in those days the Italians were nearly as devoid of any distinct principles of architecture as we have since become.

510. View of the Certosa, near Pavia. (From a Photograph.)

The great difference between Pavia and Milan is that the former shows no trace of foreign influence. It is as purely Italian as St. Petronio, and by no means so complete or consistent in design. Nothing, in fact, can be more painful than the disproportion of the parts, the bad drawing of the details, the malformation of the vaults, and the meanness of the windows; though all these defects are completely hidden by the most gorgeous colouring, and by furniture of such richness as to be almost unrivalled. So attractive are these two features to the majority of spectators, and so easily understood, that nine visitors out of ten are delighted with the Certosa, and entirely forget its miserable architecture in the richness and brilliancy of its decorations.

Externally the architecture is better than in the interior. From its proximity to Pavia, it retains its beautiful old galleries under the roof. Its circular apses, with their galleries, give to this church, for the age to which it belongs, a peculiar character, harmonising well with the circular-headed form, which nearly all the windows and openings present. Even in the interior there are far more circular than pointed arches.

The most beautiful and wonderful part of the building is the façade. This was begun in 1473, and is one of the best specimens in Italy of the Renaissance style. It would hardly, therefore, be appropriate to mention it here, were it not that the dome over the intersection of the nave and transepts is of the same age and style, but reproduces so exactly (except in details) what we fancy the Mediæval Italian Gothic dome to have been, that it may be considered as a feature of the earlier ages. Referring to Woodcut No. [502], it will be seen how like it is to that of Chiaravalle in outline. It is less tall, however, and, if translated into the details of the great church at Florence, would fit perfectly on the basement there prepared for such a feature.

Like many other churches in Northern Italy, the principal parts of the Certosa are built in brick, and the ornamental details executed in terra-cotta. Some of the latter, especially in the cloisters, are as beautiful as any executed in stone in any part of Italy during the Middle Ages; and their perfect preservation shows how suitable is the material for such purposes. It may not be appropriate for large details or monumental purposes, but for the minor parts and smaller details, when used as the Italians in the Middle Ages used it, terra-cotta is as legitimate as any material anywhere used for building purposes; and in situations like the alluvial plains of the Po, where stone is with difficulty obtainable, its employment was not only judicious but most fortunate in its results.

It would be a tedious and unprofitable task to attempt to particularise all the churches which were erected in this style in Italy, as hardly one of them possesses a single title to admiration beyond the very vulgar one of size. To this Santa Croce, at Florence, adds its association with the great men who lie buried beneath it, and Sta. Maria Novella can plead the circumstance—exceptional in that city—of possessing a façade;[[315]] but neither of these has anything to redeem its innate ugliness in the eyes of an architect.

There are two great churches of this period at Venice, the San Giovanni e Paolo (1246-1420) and the Frari (1250); they are large and richly ornamented fabrics, but are both entirely destitute of architectural merit.

511. Duomo at Ferrara. (From Hope’s ‘Architecture.’) Scale 50 ft. to 1 in.

A much more beautiful building is the cathedral at Como, the details of which are so elegant and so unobtrusively used as in great measure to make up for the bad arrangement and awkward form of the whole. In design it is, however, inferior to that of the Duomo at Ferrara (Woodcut No. [511]). The latter does not display the richness of the façades of Siena or Orvieto, nor the elegance of that last named; but among the few Italian façades which exist, it stands pre-eminent for sober propriety of design and the good proportions of all its parts. The repose caused by the solidity of the lower portions, and the gradual increase of ornament and lightness as we ascend, all combine to render it harmonious and pleasing. It is true it wants the aspiring character and bold relief of Northern façades; but these do not belong to the style, and it must suffice if we meet in this style with a moderate amount of variety, undisturbed by any very prominent instances of bad taste.

The true type of an Italian façade is well illustrated in the view of St. Francesco at Brescia (Woodcut No. [512]), which may be considered the germ of all that followed. Whether the church had three aisles or five, the true Italian façade in the age of pointed architecture was always a modification or extension of this idea, though introduced with more or less Gothic feeling according to the circumstances of its erection.

At Florence there is a house or warehouse, converted into a church,—Or (horreum) San Michele, which has attracted a good deal of attention, but more on account of its curious ornaments than for beauty of design—which latter it does not, and indeed can hardly be expected to, possess. The little chapel of Sta. Maria della Spina at Pisa owes its celebrity to the richness of its niches and canopies, and to the sculpture which they contain. In this the Italians were always at home, and probably always surpassed the Northern nations. It was far otherwise with architecture, properly so called. This, in the age of the pointed style, was in Italy so cold and unmeaning, that we do not wonder at the readiness with which the Italians returned to the classical models. They are to be forgiven in this, but we cannot so easily forgive our forefathers, who abandoned a style far more beautiful than that of Italy to copy one which they had themselves infinitely surpassed; and this only because the Italians, unable either to comSprehend or imitate the true principles of pointed art, were forced to abandon its practice. Unfortunately for us, they had in this respect in that age sufficient influence to set the fashion to all Europe.

512. View of St. Francesco, Brescia. (From Street’s ‘Brick and Marble in the Middle Ages.’)

Of late work in Dalmatia the most remarkable is the Cathedral of Sebenico (described in Mr. Jackson’s work), built entirely in stone and marble, and without any brick or timber in its construction. It is a cruciform building, covered over by a waggon-vault of stone, visible both inside and outside. It was commenced from the design of Messer Ambrosia, a Venetian architect, in 1435, to whom may be attributed the nave and aisles up to the string-course above nave arches. The work was continued after 1441 by another architect, Messer Giorgio, also from Venice, who died in 1475, leaving the building still incomplete. The style of the work is late Venetian Gothic, influenced in its later portions by the Renaissance revival. The cloisters of the Badia at Curzola, and of the Dominican and Franciscan convents at Ragusa, are also beautiful specimens of late Italian Gothic.

END OF VOL. I.

LONDON: PRINTED BY WILLIAM CLOWES AND SONS LIMITED,

STAMFORD STREET AND CHARING CROSS.

[1]. The first volume was published in 1865; the second in 1867.

[2]. ‘Mémoire sur les Fouilles exécutés au Madras’en,’ Constantine, 1873.

[3]. ‘Monumentos Arquitectonicos de España.’ Folio. Madrid, 1860, et seqq.

[4]. Parcerisa, ‘Recuerdos y Bellezas de España.’ Folio. Madrid.

[5]. ‘Gothic Architecture in Spain,’ by G. E. Street. Murray. 1865.

[6]. ‘Denkmäler der Kunst des Mittelalters in Unter Italien,’ by H. W. Schulz. Dresden, 1860. Quarto. Atlas, folio.

[7]. ‘Syrie Centrale,’ by Count M. De Vogüé. Paris.

[8]. ‘Byzantine Architecture,’ by Chev. Texier. London, 1864.

[9]. ‘Mission to the Court of Ava in 1855,’ by Colonel Yule. 4to. London, 1858.

[10]. ‘Travels in Siam and Cambodia,’ by Henri Mouhot. London: John Murray. 1864.

[11]. The number of illustrations in the chapters of the Handbook comprised in this first volume of the History was 441. They now stand at 536 (1874); and in the second volume the ratio of increase will probably be even greater.

[12]. It may be suggested that the glory of a French clerestory filled with stained glass made up for all these defects, and it may be true that it did so; but in that case the architecture was sacrificed to the sister art of painting, and is not the less bad in itself because it enabled that art to display its charms with so much brilliancy.

[13]. The numbers in the table must be taken only as approximative, except 2, 4, 6, and 7, which are borrowed from Gwilt’s ‘Public Buildings of London.’

[14]. The Isis-headed or Typhonian capitals cannot be quoted as an exception to this rule: they are affixes, and never appear to be doing the work of the pillar.

[15]. See woodcuts further on.

[16]. Max Müller, who is the facile princeps of the linguistic school in this country—in an inaugural lecture which he delivered when, it was understood, he was appointed to a chair in the Strasburg University—gave up all that has hitherto been contended for by his followers. He admitted that language, though an invaluable aid, did not suffice for the purposes of the investigation, and that the results obtained by its means were not always to be depended upon.

[17]. The term “Persistent Varieties” has recently been introduced, instead of “race,” in ethnological nomenclature, and, if scientific accuracy is aimed at, is no doubt an improvement. It is an advantage to have a term which does not even in appearance prejudge any of the questions between the monogenists and polygenists, and leaves undecided all the questions how the variations of mankind arose. But it sounds pedantic; and “race” may be understood as meaning the same thing.

[18]. The whole of this subject has been carefully gone into by the Author in a work entitled ‘Rude Stone Monuments’ published in 1872, to which the reader is referred.

[19]. All round the shores of the Mediterranean are found the traces of an art which has hitherto been a stumbling-block to antiquarians. Egyptian cartouches and ornaments in Assyria, which are not Egyptian; sarcophagi at Tyre, of Egyptian form, but with Phœnician inscriptions, and made for Tyrian kings; Greek ornaments in Syria, which are not Greek; Roman frescoes or ornaments, and architectural details at Carthage, and all over Northern Africa, which however are not Roman. In short, a copying art something like our own, imitating everything, understanding nothing. I am indebted to my friend Mr. Franks for the suggestion that all this art may be Phœnician, in other words, Semitic, and I believe he is right.

[20]. Had there been no Pelasgi in Greece, there probably would have been no Architecture of the Grecian period.

[21]. The derivation of the two words Heathen and Pagan seems to indicate the relative importance of these two terms very much in the degree it is here wished to express. Heathen is generally understood to be derived from ἔθνος, a nation or people; and Pagan from Pagus, Pagani, a village, or villagers. Both are used here not as terms of reproach, but as indicative of their being non-Christian, which is what it is wished to express, and was the original intention of the term.

[22]. ‘Rude Stone Monuments,’ 1 vol. 8vo. Murray, 1872.

[23]. The above scheme of Egyptian Chronology was published by me in the ‘True Principles of Beauty in Art,’ in 1849; and the data on which it was based were detailed in the Appendix to that work. As there seems to be nothing in the subsequent researches or discoveries which at all invalidates the reasoning on which the table was founded, it is here reproduced in an abridged form as originally set forth.

[24]. Syncellus, Chron. p. 98, ed. Dindorff, Bonn, 1829.

[25]. ‘Josephus contra Apion,’ i. 14, 16 and 26.

[26]. Vyse, ‘Operations on the Pyramids at Gizeh in 1837,’ vol. i. p. 297, et seq.

[27]. At Wady Meghara, in the Sinaitic peninsula, a king of the 4th dynasty is represented as slaying an Asiatic enemy. It is the only sign of strife which has yet been discovered belonging to this ancient kingdom. Lepsius, Abt. ii. pl. 39.

[28]. By a singular coincidence, China has been suffering from a Hyksos domination of Tartar conquerors, precisely as Egypt did after the period of the Pyramid builders, and, strange to say, for about the same period—five centuries. Had the Taepings been successful, we should have witnessed in China the exact counterpart of what took place in Egypt when the 1st native kings of the 18th dynasty expelled the hated race.

[29]. Col. H. Vyse, ‘Operations carried on the Pyramids of Gizeh in 1837.’ Lond. 1840-43.

[30]. This will be best understood by looking at the section (Woodcut [7]), in which it will be seen that the so-called coping or casing-stones were not simply triangular blocks, filling up the angles formed by the receding steps, and which might have been easily displaced, but stones from 7 to 10 feet in depth, which could not have been supported unless the work had been commenced at the bottom. On the other hand, it is difficult to understand how the casing-stones for the upper portion could have been raised up the sloping portion completed. It is probable, therefore, that the casing was commenced at the angles and was carried up in vertical planes, thus leaving a causeway of steps in the middle of each face, which diminished in width as the work proceeded; this causeway, a few feet wide only, on each face being then encased from the top downwards after the apex blocks had been laid.—Ed.

[31]. ‘The Pyramids and Temples of Gizeh by W. M. Flinders Petrie. Lond. 1883.

[32]. On the north side the paving is carried under the lowest course.

[33]. Except the spires of Cologne Cathedral.

[34]. They are situated in latitude 30° N.

[35]. ‘Pyramids and Temples of Gizeh,’ p. 199.

[36]. Mr. Petrie says, p. 117: “All the chambers of this pyramid are entirely hewn in the rock.”

[37]. ‘Medum,’ by M. Flinders Petrie. D. Nutt, London, 1892.

[38]. Diodorus, i. 51.

[39]. M. Mariette’s discoveries in these tombs were only in progress at the time of his death: but his manuscript notes and drawings of the hieroglyphics and figures have since been published in facsimile under the title of ‘Les Mastabas de l’Ancienne Empire’ Paris 1889. They are, however, incomplete; some of the plates referred to could not be found, and M. Maspero, who edited the work, has unfortunately given no preface of his own, which might have rendered them more intelligible. At present no sufficient data exist to enable others to realise and verify the extraordinary revelation it presents to us. It is 2000 years older, and infinitely more varied and vivid, than the Assyrian pictures which recently excited so much interest.

[40]. The false door is a niche in the side of the mastaba, the back of which is carved in imitation of a wooden door.

[41]. Lucian, ‘De Syria Dea,’ ed. Reetzin, tom. iii. p. 451, alludes to the fact of the old temples of the Egyptians having no images.

[42]. The roof slabs are gone, but the lower portions of the slits are still uninjured.

[43]. The plan and particulars relating to this temple are taken from Mr. W. M. Petrie’s work before referred to.

[44]. The tablet discovered at Gizeh, in which Khufu, the builder of the Great Pyramid, is recorded to have made some repairs to the Sphinx, is stated by Mr. Petrie to be a forgery of the 20th dynasty, and his reasons are given in section 118 of his work.

[45]. Lepsius, ‘Denkmaler,’ Abt. ii. pls. 115, 116.

[46]. Syncellus, p. 69; Euseb. Chron. p. 98.

[47]. ‘Hawara, Biahmun, and Arsinoe’ by W. M. Flinders Petrie, 1889.

[48]. ‘Kahun, Garob, and Hawara,’ by W. M. Flinders Petrie, 1890.

[49]. ‘Illahun, Kahun, and Gurob,’ by W. M. Flinders Petrie, 1891.

[50]. Ibid.

[51]. The researches of Mr. Petrie at Kahun have shown that originally this form of column was in wood, which would account for the base on which, in Egyptian work, it is always placed.

[52]. In a tomb of the 4th dynasty found at Sakkara is a wall decoration in which the lotus column is used in a frieze, examples of it being carved in low relief to separate the figures in a procession (see plate 10, ‘Voyage dans la Haute Égypte,’ by F. A. F. Mariette. Cairo, 1878). The polygonal or Proto-Doric column has also been found as a hieroglyph in an inscription of the 4th dynasty. This carries back the date of the two columns to a period some twelve centuries prior to the example at Beni-Hasan.

[53]. ‘Revue Archæologique,’ vol. iii., 1861, p. 97, and v., 1862, p. 297.

[54]. 518 years: ‘Josephus contra Apion.,’ I. 26.

[55]. Layard, ‘Nineveh and Babylon,’ 281.

[56]. Tacitus, Ann. II. 60.

[57]. ‘Revue Archéologique,’ vol. x. 1864, p. 170, and vol. xiii. 1866, p. 73.

[58]. Now in Sir John Soane’s Museum, in Lincoln’s-Inn-Fields.

[59]. ‘Egyptian Archæology,’ by G. Maspero, translated from the French by Amelia B. Edwards. London, 1887.

[60]. The information regarding these temples is principally derived from Hoskins’s ‘Travels in Ethiopia,’ which is the best and most accurate work yet published on the subject.

[61]. Herodotus. iii. 24. Diodorus, ii. 15.

[62]. Woodcuts 982 and 1091 in the first edition of this History.

[63]. Published in the ‘Rheinischer Museum’ vol. viii. p. 252, et seq.

[64]. ‘Josephus contra Apion,’ i. 14.

[65]. If the Greeks traded to Naucratis as early as the 1st Olympiad.

[66]. When the ‘Handbook of Architecture’ was published in 1855, there existed no data from which these affinities could be traced. It is to the explorations of Sir Henry Rawlinson and Messrs. Taylor and Loftus that we owe what we now know on the subject; but even that is only an instalment.

[67]. The chronology here given is based on the various papers communicated by Sir Henry Rawlinson to the ‘Journal of the Royal Asiatic Society,’ vol. x. et seq., and to the ‘Athenæum’ journal. The whole has been abstracted and condensed in his brother’s ‘Five Great Monarchies of the Ancient world;’ from which work the tables here given are taken in an abridged form.

[68]. Loftus, ‘Chaldæa and Babylonia,’ p. 167.

[69]. Journal R. A. S., vol. xv. p. 260, et seq.

[70]. Journal R. A. S., vol. xviii. p. i, et seq., Sir H. Rawlinson’s paper, from which all the information here given regarding the Birs is obtained.

[71]. Flandin and Coste, ‘Voyage en Perse,’ vol. iv. pl. 221.

[72]. I have ventured to restore the roof of the cella with a sikra (ziggur or ziggurah, according to Rawlinson’s ‘Five Ancient Monarchies,’ vol. I, p. 395, et passim), from finding similar roofs at Susa, Bagdad, Keffeli, &c. They are certainly indigenous, and borrowed from some older type, whether exactly what is represented here is not clear, it must be confessed. It is offered as a suggestion, the reason for which will be given when we come to speak of Buddhist or Saracenic architecture.

[73]. Rich gives its dimensions: On the north, 600 feet; south, 657; east, 546; and west, 408. But it is so ruinous that only an average guess can be made at its original dimensions. [Mr. George Smith, in the ‘Athenæum’ of February 1876, wrote a letter giving an account of a tablet of the Temple of Belus at Babylon he had deciphered, which constitutes the only description found giving the dimensions thereof. The bottom stage was 300 feet square and 110 feet high, the second, with raking sides, 260 feet square and 60 feet high, the third 200 feet square and 20 feet high, the fourth, fifth, and sixth each 20 feet high and 170, 140, and 110 feet respectively. The top stage, which was the sanctuary, was 80 × 70 feet and 50 feet high, the whole height being thus 300 feet, the same as the width of the base. Mr. W. R. Lethaby, in his work on ‘Architecture, Mysticism, and Myth,’ gives as a frontispiece a restoration according to these dimensions, the appearance of which is more impressive and probably approaches more closely to the actual proportions of a ziggurat than any previously published, excepting that at Khorsabad, with which in general proportion it coincides.—Ed.]

[74]. Strabo, xvi. p. 738.

[75]. There is a slight discrepancy in the measures owing to the absence of fractions in the calculation.

[76]. Loftus, ‘Chaldæa and Babylonia,’ p. 188.

[77]. This chapter and that next following may be regarded as, in all essential respects an abridgment or condensation of the information contained in a work published by the author in 1851, entitled, ‘The Palaces of Nineveh and Persepolis Restored,’ the only real difference being that the more perfect decipherment of the inscriptions since that work was published has caused some of the palaces and buildings to be ascribed to different kings and dynasties from those to whom they were then assigned, and proved their dates to be more modern than was suspected, for the oldest at least. The order of their succession, however, remains the same, and so consequently do all the architectural inferences drawn from it. Those readers who may desire further information on the subject are referred to the work alluded to.

[78]. Published in 1862, in the ‘Athenæum’ journal, No. 1812.

[79]. This plan, with all the particulars here mentioned, are taken from Layard’s work, which is the only authority on the subject, so that it is not necessary to refer to him on every point. The plan is reduced to the usual scale of 100 ft. to 1 inch, for easy comparison with the dimensions of all the other edifices quoted throughout this work.

[80]. The whole of the information regarding Khorsabad is taken from M. Botta’s great work on the subject, and its continuation, ‘Ninive et l’Assyrie,’ by M. Victor Place.

[81]. These particulars are all borrowed from M. Place’s great work, ‘Ninive et l’Assyrie,’ folio. Paris, 1865.

[82]. Space will not admit of my entering into all the reasons for this restoration here. If any one wishes for further information on the subject, I must refer him to my ‘Palaces of Nineveh and Persepolis Restored,’ published in 1851. Nothing has occurred during the twenty-three years that have elapsed since that work was published that has at all shaken my views of the correctness of the data on which these restorations were based. On the contrary, every subsequent research has served only more and more to convince me of their general correctness, and I cannot now suggest any improvement even in details. [It should be noted that the author’s theory as to the covering over of the Assyrian halls with a flat roof carried on columns has never been accepted by foreign archæologists, and no trace has ever been found of the foundations which would be required to carry such columns. M. Place, who conducted the excavations at Khorsabad, and Messrs. Perrot & Chipiez, who, among others, have devoted much time and research to the subject, are of opinion that the halls were vaulted. It would be difficult now to determine the possibility of building vaults of thirty feet span in crude or unburnt brick, because we have no means of testing the resistance to crushing which such bricks might afford. The brick voussoirs found by M. Place in the arches of the town gates had been prepared in special moulds, and so completely dried that liquid clay had been used to cement them together. In some of the large halls, far away from the walls, and in some cases in the centre of the rooms, huge blocks of hard clay were found with their lower surface curved and covered with a layer of stucco; these masses were sometimes many metres long, one to two metres wide, nearly a metre thick. According to M. Place they formed part of a barrel vault covering the halls, and their size would account for the immense thickness of the walls constructed to carry them and resist their thrust, as well as for the peculiar shape of the halls; that is, their length as compared with their breadth. The sculptured slabs would seem to have been carved to be seen by a high side-light, which suggests openings of some kind, just above the springing of the vault, and above the flat roof of the smaller halls round.—Ed.]

[83]. These gateways are extremely interesting to the Biblical student, inasmuch as they are the only examples which enable us to understand the gateways of the Temple at Jerusalem as described by Ezekiel. Their dimensions are nearly the same, but the arrangement of the side chambers and of gates generally are almost identical. These gates had been built 100 years at least before Ezekiel wrote.

[84]. Layard’s excavations here furnish us with what has not been found or has been overlooked elsewhere, e.g., a ramp or winding staircase leading to the upper storey (‘Nineveh and Babylon,’ 461). As explained above, I believe the tops of the walls, which are equal to the floor space below, formed such a storey. This ramp at Koyunjik would just suffice to lead to them, and goes far to prove the theory. If it was similarly situated at Khorsabad it would be in the part fallen away.

[85]. [This assumption is speculative, no trace of such dwarf columns having been found; to raise a solid wall thirteen feet thick to carry a gallery seems unlikely.—Ed.]

[86]. This façade, as I read it, is identical with the one I erected at the Crystal Palace as a representation of an Assyrian façade, long before this slab was exhumed.

[87]. See Rawlinson, ‘Ancient Monarchies,’ vol. i. p. 398.

[88]. It is called tomb by Strabo, lib. xvi., and Diodorus, xvii. 112, 3; temple, Herodotus, i. 181, Arrian, vii. 17, 2, Pliny, vi. 26.

[89]. Texier shows columns on the fourth side.

[90]. Mr. Weld Blundell in 1892 found a column with fluted base and Doric capital, but it did not apparently belong to the palace.

[91]. [It follows from what has already been pointed out in a note respecting the roofs of the Assyrian palaces; if, as is contended by French archæologists, the great halls were vaulted, Mr. Fergusson’s theory respecting the origin of the Persian columns partly falls to the ground; in that case it would seem more probable that the Persians owed their columnar architecture to prototypes of wooden posts, covered with metal plates, such as are described as existing in the Median palaces of Ecbatana, where Cyrus, the first Persian monarch, passed so many years of his life.—Ed.]

[92]. The woodcuts in this chapter, except the restorations, are taken from Flandin and Coste’s ‘Voyage en Perse,’ except where the contrary is mentioned.

[93]. It is curious that neither Ker Porter, nor Texier, nor Flandin and Coste, though measuring this building on the spot, could make out its plan. Yet nothing can well be more certain, once it is pointed out.

[94]. ‘Palaces of Nineveh and Persepolis Restored,’ p. 126.

[The prayer platform or talar represented on the tomb of Darius is extremely unlike any constructional feature such as an upper storey, and may have been placed there only to give dignity and importance to the figure of the king: the hall of the Palace of Darius could easily have been lighted by clerestory windows over the roofs of the smaller chambers on each side.—Ed.]

[95]. It is very strange that this similarity, like the plan of the square halls, should hitherto have escaped observation. Had any one looked at the matter as a whole we should have been spared some restorations which are too absurd even to merit exposure.

[The restorations referred to are those in which the columns of the Great Hall and of the porticoes are shown as isolated features standing on the platforms. The authors of these designs would appear to have been misled by Messrs. Flandin and Coste’s plan, in which the drains are shown as if they ran under the line of the wall proposed by Mr. Fergusson, the enclosing wall of the Great Hall. Mr. Weld Blundell’s researches (1891), however, have shown that the main drain really lies under the hall, and between the enclosure wall and the first row of columns, and that the vertical rain-water shafts which were built into the wall communicated direct with this main drain. These shafts, cut in stone, in some cases rise above the level of the platform, which show that they were not intended to carry off the surface water from the platform. Mr. Weld Blundell discovered also the traces of the foundation of walls at the angles where shown by Mr. Fergusson. It would seem that in course of time the platforms have become coated with so hard and uniform a covering as to suggest its being the natural surface; when once broken through, however, the evidences of foundations of various walls are abundant.—Ed.]

[96]. M. Dieulafoy’s work on the Acropolis of Susa has just (1893) appeared, but, so far as the palace is concerned, his discoveries do not add much to our knowledge. He appears to have arrived at the conclusion that the great hall (which in plan resembles that of the palace of Xerxes—Woodcut [94]) was not enclosed on the south side, but was left open to the court in the same way as the great reception halls of the later Parthian and Sassanian kings at Al Hadhr, Firouzabad, and Ctesiphon.

[97]. It is now generally considered that these two buildings were tombs; the projecting bosses, as shown on woodcut, are in reality sinkings, and were probably decorative only.—Ed.

[98]. M. Dieulafoy claims to have traced the plan of a temple at Susa which consisted of a sanctuary the roof of which was supported by four columns, with a portico-in-antis in front, and a large open court, measuring about 50 ft. by 40 ft., in the middle of which was placed the fire-altar. The whole building was enclosed with a corridor or passage, with entrances so arranged that no one could see inside the temple from without.—Ed.

[99]. Mr. Flinders Petrie’s latest excavations at Medum have resulted in the discovery of small brick arches over a passage in the sepulchral pit of Rahotep of the 4th dynasty.

[100]. Wilkinson’s ‘Egypt and Thebes,’ pp. 81 and 126.

[101]. ‘Manners and Customs of the Egyptians,’ vol. iii. p. 263.

[102]. 1 Kings vii. 1-12. Josephus, B. J. viii. 5.

[103]. Josephus, Ant. viii. 5. § 2.

[104]. The details of this restoration are given in the ‘Dictionary of the Bible,’ sub voce ‘Temple,’ and repeated in my work entitled ‘The Holy Sepulchre and the Temple at Jerusalem.’ Murray, 1865.

[105]. ‘Speaker’s Commentary on the Bible,’ vol. ii. p. 520; note on verse 15, chap. vii. 1 Kings.

[106]. For a restoration of this screen see ‘Tree and Serpent Worship,’ Appendix i., p. 270.

[107]. Since the article on the Temple in Smith’s ‘Dictionary of the Bible’ was written, from which most of the woodcuts in this chapter are taken, I have had occasion to go over the subject more than once, and from recent explorations and recently discovered analogies have, I believe, been able to settle, within very narrow limits of doubt, all the outstanding questions with reference to this celebrated building. I have in consequence written and published a monograph of the Temple, but have deemed it more expedient to leave the illustrations here as they are.

[108]. 2 Chronicles xx. 5.

[109]. Hecateus of Abdera, in ‘Müller’s Fragments,’ ii. 394.

[110]. Josephus, Ant. xi. 4, § 2.

[111]. Josephus, B. J. v. 5, § 4.

[112]. Dawkins and Wood, ‘The Ruins of Palmyra,’ Lond. 1753.

[113]. Texier, ‘Arménie et la Perse,’ vol. i. pl. 62 and 68.

[114]. Texier, ‘Asie Mineure,’ pl. 10 to 21.

[115]. Herodotus, i. 93.

[116]. Lydischen Königsgräber, I. F. M. Olfers, Berlin, 1859.

[117]. “Toward the centre of the monument two large stones were found leaning at an angle the one against the other, and forming a sort of tent, like in Woodcut [124], under which was presently discovered a small statue of Minerva seated on a chariot with four horses, and an urn of metal filled with ashes, charcoal, and burnt bones. This urn, which is now in the possession of the Comte de Choiseul, is enriched in sculpture with a vine branch, from which is suspended bunches of grapes done with exquisite art.”—‘Description of the Plain of Troy,’ translated by Dalzel, Edin. 1791, p. 149.

If this is so, this is no doubt the vessel mentioned, ‘Iliad,’ xvi. 221, xxiii. 92; ‘Od.,’ xxiv. 71, and elsewhere. But where is it now? and why has not the fact of its existence been more insisted upon?

[118]. One of the most interesting facts brought to light in Dr. Schliemann’s excavations is that between the age of the “Ilium Vetus” of Homer, rich in metals and in arts, and the “Ilium Novum” of Strabo, a people ignorant of use of the metals, and using only bone and stone implements, inhabited the mound at Hissarlik which covered both these cities. This discovery is sufficient to upset the once fashionable Danish theory of the three ages—Stone, Bronze, and Iron—but, unfortunately, adds nothing to our knowledge of architecture. These people, whoever they were, built nothing, and must consequently be content to remain in the “longa nocte” of those who neglect the Master Art.

[119]. Fergusson’s ‘History of Indian and Eastern Architecture.’ John Murray, London 1876, page 108 et seq.

[120]. This tomb is considered by M. Renan (Mission de Phœnicie, Paris 1864) to be of Phœnician origin, who remarks generally on their work: “Phœnician tombs are generally excavated in the solid rock; their architecture is the carved rock without columns; they obtained all they could out of the solid rock, leaving it as they found it, with more or less attempt to make it graceful; the fact that it was worked before being transported suggests that as it left the quarry so it remained, no sound of hammer or saw being heard during its erection.” There is another tomb at Marathos also attributed to the Phœnicians, which is partly cut out of the rock and partially built in large blocks of masonry.

[121]. In reality the monument stands exactly over the centre of the rock-cut sepulchre. The section-line must, therefore, be understood to be carried back about 10 feet from the face of the monument.

[122]. Josephus, Ant. xvi. 7, § 1.

[123]. Beule’s excavations have proved that the outer gate of the Acropolis was in front, not at the side, as here shown. ‘Acropole d’Athènes.’ Paris, vol. i. pl. i. and ii.

[124]. For details of this see Bötticher, ‘Baumkultus der Hellenen.’ Berlin, 1856.

[125]. Pausanias, ix. 38.

[126]. It appears that on the back of the stones laid in horizontal courses were others of great size piled on the top.

[127]. The same scroll exists at New Grange in Ireland, in the Island of Gozo near Malta, and generally wherever chambered tumuli are found.

[128]. A cast of these is to be found in the South Kensington Museum.

[129]. These antæ (parastades) or responds were destined in the first case to protect the angles of the wall, and in the second case to support the beams carried by them and the columns between, the sun-dried brick wall being not to be relied on; in the later Greek temples the walls were built in stone and marble, and the parastades became therefore no longer constructional necessities, being retained only as decorative features, of which so many others are found in the style.

[130]. Pausanias, vi. 19.

[131]. The dimensions are 94 feet by 45, covering consequently only 4230 feet.

[132]. This refers only to the columns and antæ; the lower portion of the walls, 3 feet 6 inches high, were in stone; above this clay bricks were employed in building the walls, and it was to the disintegration of these that we owe the preservation of the Hermes of Praxiteles, which was found embedded in a thick layer of clay. At first it was thought that this clay had been washed down from the neighbouring slopes of the hill of Kronos.

[133]. M. J. Thacher Clarke, who directed the American expedition in 1881, is now occupied with a monograph on the subject, and a report by him was published in 1882. Boston and London. J. Trübner.

[134]. A proto-Ionic capital of early date was found in 1882 on the summit of Mount Chigri, in the Troad, by Mr. J. Thacher Clarke, and is described in the American Journal of Archæology, Baltimore. 1886. Another example ascribed to Phœnician artists was found at Trapeza in Cyprus, and is now in the Louvre; both are of the same type as that which is represented in the ivory carvings from the north-western palace of Nimroud, now in the British Museum, so that the Asiatic origin of the order is thus confirmed.

[135]. Pausanias, viii. 45.

[136]. Bohn.

[137]. [The earliest example in stone at Benihasan is of less diameter than the columns at Kalabscheh, so that it is difficult to draw this distinction; we have already shown also (p. 115 [note]) that wooden shafts of the twelfth dynasty have been found at Kahun, and this and the existence of the base proves their wooden origin. If therefore the Greek Doric column was derived originally from Egypt, as Mr. Fergusson believed, then its earlier wooden parentage must be accepted. Further evidence on this subject however has been afforded by the discoveries at Olympia, and the references in consequence made to Greek authors; all these show without doubt that the columns of the temple of Hera were originally in wood, and were gradually replaced by stone. The theory that the pillars in Egypt or early Greece were built in brickwork or rubble masonry is not borne out by the discoveries at Tiryns, for the walls of the palace there, in rubble and clay mortar, were of such weak construction that posts of timber were required to carry the epistyle or beam, either isolated as columns or built up against the wall as antæ.

Mr. Fergusson’s theory that a pillar, originally copied from the wooden post, is slenderer at first, and gradually departs from the wood form as the style advances, is borne out by the evidence of the Egypt lotus column; this, as found in the rock-cut tombs of Benihasan, is of very small diameter, and quite unequal to carry the weight of any stone superstructure; whereas afterwards in the temples at Thebes it assumes a proportion nearer that of the earliest Greek Doric example at Corinth.—Ed.]

[138]. These facts have all been fully elucidated by Mr. Penrose in his beautiful work containing the results of his researches on the Parthenon and other temples of Greece, published by the Dilettanti Society.

[139]. For measurements we depend on Penrose, ‘Principles of Athenian Architecture,’ &c., fol.; and Cockerell, ‘The Temples of Egina and Bassæ,’ Lond. 1860. The details of the system were first publicly announced by Watkiss Lloyd, in a paper read to the Institute of British Architects in 1859; afterwards in an appendix to Mr. Cockerell’s work, and in several minor publications.

[140]. The pyramid-building kings of Lower Egypt seem to have had some distinct ideas of a system of definite proportions in architectural building, and to have put it into practice in the pyramid, and possibly elsewhere, but it has not yet been sought for in the other buildings of that age.

At times I cannot help suspecting more affinity to have existed between the inhabitants of Lower Egypt and those of Greece than is at first sight apparent.

[141]. It was called Zoophorus (life or figure bearer).

[142]. [The reasons which induced the late Mr. Fergusson to suggest an “opaion,” or clerestory, were fully set forth in the ‘True Principles of Beauty in Art,’ in 1849. A paper on the same subject was communicated by him to the Royal Institute of British Architects in 1861, and published in their “Transactions” for that year. Since his death, however, Mr. Penrose’s discovery that the Temple of Jupiter Olympius at Athens was really octastyle has thrown a new light on the question of hypæthral temples; and, as Dr. Dorpfield remarks in his essay on the “Hypæthral Temple” (communicated to the R. I. B. A. on Dec. 19): “The words of Vitruvius have now received quite another interpretation, through the excavation of the Olympieion at Athens, to that which they have had up to the present. The most important proof of the hypæthral lighting of the temples of antiquity has now turned into a proof against the same;” and he concludes his arguments by stating: “After it has been shown by the excavations that the Olympieion at Athens is the sole example of a great hypæthral temple mentioned by Vitruvius, we can answer this much-vexed question of the lighting of the temples of antiquity in this way—that a few great dipteral hypæthral temples existed, but that the Greek and Roman temples had as a rule no light from above, and were only lighted from the door.”—Ed.]

[143]. See Woodcuts Nos. [22], [24], [27].

[144]. Vitruvius, lib. i. ch. 1.

[145]. Boeckh, Corpus Inscript. Græc. No. 109.

[146]. Attica, xxvi.

[147]. Historia, viii, 41.

[148]. Among the many attempts made to restore the interior of this temple, the last and most elaborate is that by the late E. Beulé, ‘Acropole d’Athènes,’ 1854, vol. ii. pl. ii.; but it is also one of the worst. Indeed it is quite painful to see how the author twists his authorities to meet a preconceived theory. Without going into it, there is one objection which seems fatal to the whole.

Like most antiquaries when in difficulties for lighting Greek temples, he takes off the roof and makes the Temple of Pandrosus an open courtyard, in which he plants the olive. This is so opposed to the whole spirit of Greek art as to be inadmissible on general grounds, but in this instance it introduces the further absurdity that the Greeks opened three windows in the west wall of the temple to light this courtyard which was already open to the sky! The mode of lighting a temple by vertical windows is so exceptional that it would not have been introduced here had any other means existed of lighting the interior, and consequently the combination shown by M. Beulé seems simply impossible.

[149]. “Universo Templo longitudo est ccccxxv. pedum, latitudo ccxx. Columnæ centum viginti septem a singulis regibus factæ, lx. pedum altitudine: ex iis xxxvi. cælatæ, una a Scopa.”—H. N. xxxvi. 14.

[150]. [Mr. Wood places two in the pronaos and two in the posticum, thus reducing the depth of the opisthodomus; beyond the pronaos he places a vestibule and omits the staircases as shown on plan 159. In 1883, Mr. Fergusson returned to the subject again, and published in the Transactions of the Institute (session 1882-83) a revised plan, to which we refer our readers.—Ed.]

[151]. The finial ornament is triangular in plan, and there are three scrolls on the roof with mortices in them, showing that something must have stood on them to support the projecting angles. Dolphins and various other objects have been suggested. My own conviction is that they were winged genii, most probably in bronze, and gilt like the neckings of the capitals.

[152]. [Dr. Dorpfield is of opinion that in the Greek theatres of the best period there was no proscenium, or raised stage, and that the actors played their parts in the orchestra on the same level as the chorus. Professor Middleton also points out that in the earliest Greek theatres built in the 5th and 4th centuries B.C. the orchestra was a complete circle, the space being gradually diminished by the bringing forward of the stage.—Ed.]

[153]. It will not be necessary to enter here into all the details of this restoration. They will be found in a separate work published by me on the subject, to which the reader is referred. [The student should also refer to the restoration suggested by M. Pullan in the work published by him and Sir Charles Newton (‘Discoveries at Halicarnassus, 1862’). In the arrangement and design of the podium it accords better with other examples of Greek tombs than Mr. Fergusson’s. The three columns as shown at the angle of Mr. Fergusson’s peristyle would be quite repugnant to any student of Greek architecture.—Ed.]

[154]. Hist. Nat. xxxvi. 5.

[155]. The figures given in the text are all Greek feet: the difference between them and English feet, being only 11⁄4 per cent., is hardly perceptible in these dimensions, without descending to minute fractions, and disturbing the comparison with Pliny’s text.

[156]. The circumstance of Asoka, the Buddhist king of India B.C. 250, having formed an alliance with Megas of Cyrene for the succour of his co-religionists in the dominions of the latter, points to such a conclusion even if nothing else did.—‘Journal Asiatic Society of Bengal,’ vii. p. 261; J. R. A. S. xii. p. 223 et seq.

[157]. Beechy’s ‘Journey to Cyrene,’ p. 444; see also Smith and Porcher, pl. 37.

[158]. Vitruvius, iv. 7.

[159]. Dionysius, iv. 61.

[160]. For more detail, see ‘The True Principles of Beauty in Art,’ p. 446 et. seq.

[161]. The Etruscan and Roman origin of the circular temple is now known to be erroneous, as remains of large circular temples have been discovered at Epidaurus and Olympia.

[162]. Even in more modern times I know of no building showing a trace of these forms except the tomb of Theodoric at Ravenna. This, however, is Etruscan both in form and detail, as will be seen farther on.

[163]. Plin. ‘Hist.’ xxxvi. 13.

[164]. A diagram is given in ‘The True Principles of Beauty in Art’ p. 459, which shows at least that there is no difficulty in designing a monument in perfect accordance with the text. Whether the latter is to be depended upon or not is another matter.

[165]. These dimensions, with all those that follow, unless otherwise specified, are taken from Taylor and Cresy’s ‘Architectural Antiquities of Rome,’ London, 1821. They seem more to be depended upon than any others I am acquainted with.

[166]. These two temples, like almost all the others of Rome, have recently been renamed by the Roman or rather German antiquaries. The Jupiter Tonans is now the Temple of Saturn, and the Jupiter Stator is decreed to have been the Temple of Castor and Pollux. The names by which they are currently known has been adhered to, as the architecture is of more importance here than the archæology.

[167]. Laborde, ‘Monumens de la France,’ vol. i. pls. xxix. xxx. p. 68.

[168]. IMP. CÆS. M. AVRELIVS ANTONINVS PIVS FELIX AVG. TRIB. POTEST V. COS. PROCOS. PANTHEVM VETVSTATE CORRVPTVM CVM OMNI CVLTV RESTITVERVNT. Isabelle, ‘Édifices Circulaires,’ p. 37, pl. xii.

[169]. When the first edition of this work was written I believed the rotunda to have been added to the portico by Severus; and if this were so it would get over many of the difficulties arising from its size and the character of its brickwork. My personal examination, however, has forced me very unwillingly to give up this hypothesis. It certainly is, however, very astonishing that such a vault should have been attempted at so early an age.

[There seems to be some probability that Mr. Fergusson’s first belief was correct, and that the Rotunda was built by Hadrian, bricks with the stamp of his period having been found in the casing and in the bond courses in the solid concrete both of the drum and in the dome. The discovery is due to M. Chedanne, one of the “Grand Prix” students in the Villa Medici, who had selected the subject for his “Envoi de Rome,” and was allowed to superintend certain repairs and restorations which were required in the Pantheon. It would seem that the portico erected by Agrippa preceded a temple with cella of the ordinary form, the pavement of which has been found nearly seven feet below the floor of the present church. From this it follows that when the Rotunda was erected in the first half of the second century, the portico, which is undoubtedly of Agrippa’s time, must have been taken down and rebuilt on to it, and this explains Mr. Fergusson’s reasons for insisting that the portico was built on to the Rotunda. The theory as to the Pantheon forming part of Agrippa’s bath is thus disposed of. Independently of that, however, Prof. Middleton has pointed out that the discoveries made in 1882, by the removal of the block of houses at the back, showed that there was no connection whatever between the two buildings. Traces exist of the original marble lining, and of cornices which were continued round the dome, showing that originally the complete circuit was exposed to view. “Moreover,” Prof. Middleton states, “if further proof were wanting to contradict the theory that the Pantheon was over the Calidarum or Laconicum of the bath, this is supplied by the fact that there is no trace of any hypocaust under the floor, but merely an ancient drain to carry away the rain-water that fell through the opening in the dome. The Pantheon, too, is on the north side of the Thermæ—a very improbable position for the Laconicum, or hot room, which was usually placed on the sunny side of the buildings.”—Ed.]

[170]. The bronze plates which were removed by Pope Urban VIII. in 1626 to make cannon, and also for the great Baldachino in St. Peter’s, were taken from the portico; the coffers of the interior of the dome were decorated, according to Prof. Middleton, with mouldings in stucco painted and gilt.

[171]. This building is commonly called a temple, though it is not known to what deity it was dedicated. My own impression is that it was a tomb, or at least a funereal monument of some sort.

[172]. Owing to a misreading of Vitruvius’s statement respecting the temple it had always been classed as decastyle. See Mr. Penrose’s researches published in the ‘Transactions of the Royal Institute of British Architects,’ vol. iv. New Series. 1888.

[173]. See ‘The True Principles of Beauty in Art,’ where the reasons for this arrangement will be found stated at length. [See note on page 272.—Ed.]

[174]. Canina, in his restoration, shows a flat roof with coffers, so there is probably no exact authority for its form, though it seems to be generally agreed that the centre was not hypæthral.

[175]. This basilica is generally represented as having an apse at either end; but there is no authority whatever for this, and general analogy would lead us rather to infer that it was not the case. Prof. Middleton, however, is of opinion that an apse existed at both ends, and shows the same in his restoration of the plan of Trajan’s form.—‘The Remains of Ancient Rome,’ by J. H. Middleton, Fig. 52, vol. ii.

[176]. One of the pillars of this basilica remained in situ till the year 1614, when it was removed by Carlo Maderno, by order of Paul V., and re-erected in the piazza of St. M. Maggiore, where it now stands as a monumental column, supporting a statue of the Virgin. The column, with its base and capital, is as nearly as may be 60 ft. in height; the whole monument, as it now stands, 140 ft.

[177]. As it was sunk slightly below the pavement of the peristyle, and drains leading from it were traced by Mr. Ashpitel, it was probably hypæthral.

[178]. The theatres of Curio and Scaurus were in timber, except the proscenium of the latter, which was partly decorated with marble and mosaics. The Theatre of Pompey, B.C. 54, was in stone, and parts of it still exist (Prof. Middleton). The Theatre of Marcellus was begun by Julius Cæsar, but not completed till 13 B.C., when it was opened by Augustus. It was subsequently restored after a fire by Vespasian, but the purity and simplicity of the architecture, and the refinement of the details, in comparison with those of the Colosseum, 70-80 A.D., are in favour of the earlier date assigned to it. Prof. Middleton quotes another theatre, that of Cornelius Balbus (13 B.C.), built to the north-west of the Theatre of Marcellus.

[179]. According to Prof. Middleton the Amphitheatre of Sutrium is of Roman origin, and but little earlier than the Colosseum at Rome. “There is really no evidence,” he says (p. 76), “that amphitheatres were built by the Etruscans; and there can be little doubt that they were purely Roman inventions.”

[180]. At the Crystal Palace it has always been found necessary to allow 6 sq. ft. to each person.

[181]. Considerable difference of opinion seems to exist as to the extent of the velaria which sheltered the arena; this was supported by masts fixed outside the upper part of the walls, resting on brackets, 14 ft. below the cornice, which was cut away to allow the mast to fit close against the wall. M. Gérôme suggests, in his well-known picture of the Roman gladiators, that the velaria extended over a portion of the arena only. Prof. Middleton states, “The awning did not, as has been sometimes supposed, cover the whole amphitheatre, a thing which would have been practically impossible, owing to the enormous strain of so long a bearing, far beyond what any ropes could bear. It simply sloped down over the spectators in the cavea, leaving the whole central arena uncovered.” In case of rain, however, this might have been inconvenient, and it would not have protected the spectators from the sun, supposing that the performances lasted the whole day. Besides, there is no reason why the masts should have been carried so high above the wall, as shown in the restoration in Prof. Middleton’s book, p. 70. Mr. Alma Tadema is of opinion that the velarium extended over the whole arena, and was suspended on a principle similar to that of a suspension bridge, the ridge, or highest portion lying between the foci of the ellipse. This accounts in a much more satisfactory way for the height of the masts, and would afford facilities for the draining off of the rain on to the top of the gallery round.

[182]. Maffei, ‘Verona Illustrata,’ vol. vii. p. 84 et seq.

[183]. See [note] on p. [321].

[184]. These baths have been carefully measured by M. Blouet, who has also published a restoration of them. This is, on the whole, certainly the best account we have of any of these establishments.

[185]. According to Prof. Middleton this magnificent hall appears to have been what Spartianus calls the cella soliaris, the ceiling of which he says was formed of interlaced bars of gilt bronze. When the excavations in this hall were being made, many tons of fragments of iron girders were found. These were (according to Prof. Aitchison) compound girders, formed of two T bars riveted together, and then cased in bronze. A sort of lattice-work ceiling had been formed with these bronze-cased girders, the panels being probably filled in with concrete made of light pumice-stone, worked with fine stucco reliefs, painted and gilt. Prof. Middleton is of opinion that the central part over the swimming-bath was left open for the admission of light. In the upper part of the walls deep sinkings to receive the ends of the great girders which supported the ceiling are clearly visible.

[186]. St. George’s Hall at Liverpool is the most exact copy in modern times of a part of these baths. The Hall itself is a reproduction both in scale and design of the central hall of Caracalla’s baths, but improved in detail and design, having five bays instead of only three. With the two courts at each end, it makes up a suite of apartments very similar to those found in the Roman examples. The whole building, however, is less than one-fourth of the size of the central mass of a Roman bath, and therefore gives but little idea of the magnificence of the whole.

[187]. The left-hand wing of this arch has since been restored by M. Viollet-le-Duc, and the right-hand wing cleared of the square building in front of it.

[188]. These two buildings are described further on (p. [544]) as Christian edifices.

[189]. Professor Middleton states: “This building appears to be a nymphæum, or a part of some baths of about the time of Gallienus (263-268 A.D.).” It was known in the Middle Ages as the “Terme de Gallucio.” The site of the real Temple of Minerva Medica was discovered in 1887 (according to the same authority) between the new Via Macchiavelli and the Via Buonarroti, about 7 ft. below the present ground-level.

[190]. See p. [114], and Woodcut [15].

[191]. M. de Saulcy has recently attempted to prove that these tombs are those of the kings of Judah from David downwards. Their architecture is undoubtedly as late as the Christian era, and the cover of the sarcophagus which is now in the Louvre under the title of that of David is probably of the same date as these tombs, or if anything more modern.

[192]. ‘Voyage dans la Marmarique, la Cyrénaique, &c.’ Didot, Paris, 1827-29.

[193]. Though the dates of all these tombs at Cyrene are so uncertain, there seems little doubt that if any one thoroughly versed in the style were to visit the place, he could fix the age of all of them with approximate correctness. The one difficulty is, that a chronometric scale taken from the buildings at Rome, or even in Syria, will not suffice. Local peculiarities must be taken into account and allowed for, and this requires both time and judgment.

[194]. ‘Le Tombeau de la Chrétienne,’ par A. Berbrugger, Alger. 1867, from which the above particulars are taken.

[195]. It is understood that it too has been explored, but no account of the result has yet reached this country, and such rumours as have reached are too vague to be quoted. Even its dimensions are not known.

[196]. ‘De Situ Orbis,’ I. vi. p. 38. edit. Leyden, 1748.

[197]. For plan of same, see Prof. Middleton’s ‘Ancient Rome,’ 1891.

[198]. By an oversight this difference is not expressed in the woodcut.

[199]. See p. [323].

[200]. These are well epitomised by Gibbon, Book xlvi. vol. v. p. 528.

[201]. Journal of the Royal Geographical Society, ix. pl. 9. p. 476.

[202]. The sixth great Oriental monarchy; or the geography, history and antiquities of Parthia, &c., 1873.

[203]. These inscriptions were all copied by Consul Taylor, and brought home to this country. I never could learn, however, that they were translated. I feel certain they were never published, and cannot find out what has become of them.

[204]. These are expedients for filling up the corners of square lower storeys on which it is intended to place a circular superstructure. They somewhat resemble very large brackets or great coves placed in an angle. Examples of them are shown on page [434] when speaking of Byzantine architecture, and others will be found in the chapter on Mahomedan Architecture in India, in vol. iii.

[205]. These three buildings probably date as near as may be one century from each other, thus—

SerbistanA.D. 350
Firouzabad450
Ctesiphon550
To which we may now add
Mashita620

A bare skeleton, which it will require much time and labour to clothe with flesh and restore to life.

[206]. ‘The Land of Moab,’ by H. B. Tristram, M. A., &c. Murray, 1873. As all the information respecting the palace is contained in that book, pp. 195 to 215, all the illustrations here used are taken from it, it will not be necessary to refer to it again. For further information on the subject the reader is referred to that work.

[207]. Rich, ‘Residence in Koordistan,’ ii. 251 et seq.

[208]. The plan made by Dr. Tristram’s party, which is all we yet have, was only a hurried sketch, and cannot be depended upon for minute details.

[209]. Flandin and Coste, vol. iv. pls. 214, 215.

[210]. Texier and Pullan. ‘Byzantine Architecture.’ 4to. 1864. Pl. iv. p. 40 et seq.

[211]. Ruskin, ‘Stones of Venice,’ vol. ii. pls. 3, 4, and 5.

[212]. ‘L’art Antique de la Perse,’ by Marcel Dieulafoy. Paris.

[213]. In the Museum at Pesth are a number of objects of Egyptian art, said to have been found in this quarter. Is it too much to assume the pre-existence of a Phœnician or Egyptian colony here before the Roman times?

[214]. As a matter of fact, 12th century would be more exact; nearly all the chief problems of pointed arch construction in intersecting vaulting having been worked out before the close of that century.

[215]. [The domical construction of the vaults of the two great cisterns erected by Constantine, the Binbirderek, or thousand-and-one columns, and the Yeri Batan Seraï, both in Constantinople, suggests that there already existed in the East a method of vaulting entirely different from that which obtained in Rome, and which may have been a traditional method handed down even from Assyrian times.—Ed.]

[216]. ‘Syrie Centrale: Architecture civile et religieuse du Ier au VIIme Siècle. Par le Comte Melchior de Vogüé.’

[217]. ‘Byzantine Architecture,’ by Texier and Pullan. Folio, London, 1864.

[218]. De Vogüé, ‘Églises de la Terre Sainte,’ p. 101.

[219]. For a careful analytical description of the church, see Professor Willis, ‘Architectural History of the Holy Sepulchre,’ London, 1849.

[220]. The particulars for these churches are taken from Texier and Pullan’s splendid work on Byzantine architecture published by Day, 1864.

[221]. Another very small church, that of Moudjeleia, though under 50 ft. square, seems to have adopted the same hypæthral arrangement.

[222]. A great deal of very irrelevant matter has been written about these “giant cities of Bashan,” as if their age were a matter of doubt. There is nothing in the Hauran which can by any possibility date before the time of Roman supremacy in the country. The very earliest now existing are probably subsequent to the destruction of Jerusalem by Titus.

[223]. The constructive dimensions of the porch at Chillambaram (p. 353. History of Indian and Eastern Architecture, 1876.) are very similar to those of this church: both have flat stone roofs, but in the Indian, though a much more modern example, there is no arch.

[224]. These are all given in colours in Texier and Pullan’s beautiful work on Byzantine architecture, from which all the particulars regarding this church are taken.

[225]. A wayside retreat or shelter.

[226]. A restoration of the church from Procopius’s description, ‘De Ædificiis,’ lib. i. ch. iv., will be found in Hübsch, ‘Altchristliche Baukunst,’ pls. xxxii. and xxxiii.

[227]. See vol. iii., in chapter on Indian Saracenic Architecture.

[228]. The Renaissance dome which fits best to the church on which it is placed is that of Sta. Maria at Florence; but, strange to say, it is neither the one originally designed for the place, nor probably at all like it. All the others were erected as designed by the architects who built the churches, and none fit so well.

[229]. [The apses on each side of central apse are said to be additions to the original structure. The triple apses in Greek churches are found, according to Dr. Freshfield (‘Archæologia,’ vol. 44), only in churches erected subsequent to Justin II. In St. Simeon Stylites and St. Sergius at Bosra the side apses have been added afterwards.—Ed.]

[230]. Strictly speaking, circular with flattened sides, for the pendentive has a longer radius than half the diagonal of the square.

[231]. The two eastern cupolas have been raised in Arab times, and a cylindrical drum inserted with windows pierced in them to give more light to the interior.

[232]. There are numerous examples of this class of structure in North Syria, but whether they are memorials or tombs is not known. See ‘Reisen Kleinasien und Nord Syria’ by Karl Humann and Otto Puchstein.

[233]. [This rule cannot be made a hard and fast one. Procopius states that in the central dome of the Church of the Apostles, Constantinople, “the circular building standing above the arches is pierced with windows, and the spherical dome which over-arches it seems to be suspended in the air.” In the church of St. Sergius at Constantinople the walls of the octagon, which are pierced with windows, are carried up to the vault, and in the church of Sta. Sophia at Thessalonica the windows are pierced in an upright dome cylindrical internally. In all these cases, however, there is a marked distinction between these examples and those of the lofty cylindrical drums which were employed in the Neo-Byzantine churches. Mr. Fergusson’s rule, therefore, with these exceptions, may be taken as absolute.—Ed.]

[234]. They are found in the Mustaphapacha mosque at Constantinople dating from 430 A.D., but rebuilt in the 13th century.

[235]. [It is now considered that the Church of the Holy Apostles was the original model. This church, rebuilt by Justinian, was pulled down in 1464 A.D. by Mohammad II. to furnish a site for his mosque.—Ed.]

[236]. [This work has lately been undertaken by Messrs. Barnsley and Schultz, who are preparing their drawings for publication, and hope to follow up the task with a survey of the more important churches in Mount Athos.—Ed.]

[237]. ‘Die Kunst in den Athos Kirchen,’ Leipzig, 1890.

[238]. ‘Athos; or, the Mountain of the Monks,’ by Athelstan Riley, M.A., 1887.

[239]. See the photogravure of the interior of the Catholicon at Dochiariu.

[240]. ‘Églises Byzantines en Grèce.’

[241]. ‘Expédition scientifique de la Morée.’

[242]. There would seem however to have been a revival in the 11th century, possibly a reflex of that which was taking place in West Europe. And it was during this period that the churches of St. Luke in Phoeis, the church at Daphné and the churches of St. Nicodemus and St. Theodore in Athens were erected.

[243]. C. Texier, ‘Arménie et la Perse.’ 2 vols. folio. Paris.

[244]. Dubois de Montpereux, ‘Voyage autour du Caucase.’ 6 vols. 8vo. Paris, 1839, 1841.

[245]. Brosset, ‘Voyage Archéologique dans la Georgie et l’Arménie.’ St. Pétersbourg, 1849.

[246]. D. Grimm, ‘Monuments d’Architecture en Georgie et Arménie.’ St. Pétersbourg, 1864.

[247]. Texier gives three dates to this church. In the ‘Byzantine Architecture,’ p. 174, it is said to be of the 7th, and at p. 4, of the 9th century. In the ‘L’Arménie et la Perse,’ at p. 120, the date is given as 1243. My conviction is that the first is correct.

[248]. Flandin and Coste, ‘Voyage en Perse,’ pls. 214, 215.

[249]. Texier and Pullan, ‘Byzantine Architecture,’ pp. lix., lx.

[250]. I am a little doubtful regarding the scales of these two buildings. They are correctly reduced from M. Brosset’s plates. But are these to be depended upon?

[251]. Even if it should be asserted that this is no proof that the inhabitants of these countries were Buddhists in those days, it seems tolerably certain that they were tree-worshippers, which is very nearly the same thing. Procopius tells us that “even in his day these barbarians worshipped forests and groves, and in their barbarous simplicity placed the trees among their gods.” (‘De Bello Gotico,’ Bonn, 1833, ii. 471.)

[252]. The principal part of the information regarding these excavations is to be found in the work of Dubois de Montpereux, passim.

[253]. [See paper by Mr. Wm. Simpson in R. I. B. A. Transactions, vol. vii., 1891.—Ed.]

[254]. All the plans and information regarding the churches at Kief are obtained from a Russian work devoted to the subject, procured for me on the spot by Mr. Vignoles, C.E.

[255]. The first bay, as shown on plan (Woodcut No. [382]), is the narthex; the five domes come beyond it.

[256]. The particulars and illustrations of this church are taken from a paper by Heinrich Keissenberger, in the ‘Jahrbuch der K. K. Commission für Enthaltung der Baudenkmale,’ 1860. A model of it, full size, was exhibited at the Paris Exhibition of 1867.

[257]. [It has been assumed that the Roman basilicas were taken possession of by the early Christians for their own religious services, but as Mr. G. G. Scott points out in his ‘Essay on the History of English Church Architecture,’ “there is no well-authenticated instance of the conversion of any Pagan basilica into a Christian church, whilst there are abundant examples of Pagan temples converted into Christian sanctuaries” (see Texier and Pullan’s ‘Byzantine Architecture,’ pp. 75, 103). Indeed, it is, as Mr. Scott observes, “on the face of it improbable, if we reflect that the conversion of the government to Christianity had no tendency to render the existing basilicas less necessary for legal business, after the peace of the church, than they had been before that event. Christianity, unfortunately, could not abolish the litigious instincts of our nature, and after fifteen centuries of the gospel the legal profession still flourishes.” The buildings which were rendered useless by the official recognition of the new faith were not the basilicas but the temples, the fact being that the class of building known as a basilica (a term never used by either the writers or architects of Byzantine times), with its wide central nave and aisles with galleries over them lighted by clerestory or side windows, and covered with a timber roof, constituted the simplest and most economical building of large size which could be constructed to hold a vast assembly of worshippers; especially as the only features which can be looked upon as having any architectural pretensions, viz., the columns and their capitals, could be taken wholesale from temples and other Roman buildings. The semicircular apse, which alone in the Roman basilica served as a court of law, became the tribune for the bishop and presbyters.

Mr. Scott is even inclined to assign an earlier and more independent origin for the basilican form. According to his theory the germ of the Christian basilica was a simple oblong aisleless room divided by a cross arch, beyond which lay an altar detached from the wall. This germ was developed by the addition of side aisles, and sometimes an aisle returned across the entrance, and over these upper aisles were next constructed and transepts added, together with the oratories or chapels in various parts of the building. Mr. Butler, in his work on ‘The Ancient Coptic Churches of Egypt,’ accepts this theory, as the churches of Egypt are rich in evidence that favours it. At the same time, the first great basilica erected by Constantine, viz., the Vatican (St. Peter’s), and the Lateran, (St. John Lateran), are of too great importance to warrant the suggestion that their origin should be sought for in the very small though possibly earlier examples in Egypt or the East.—Ed.]

[258]. This probably refers to its foundation, for M. Cattaneo, in his work ‘L’architecture en Italie, 1890,’ judging by its ornamental detail, places the church in the second half of the seventh century.

[259]. ‘Antiquités,’ vol. i. pl. 97.

[260]. Eodem, vol. iv. pl. 67.

[261]. Mr. Alfred J. Butler’s work, already referred to, has thrown considerable light on the subject, though, as he was unable to visit any of the Coptic churches up the Nile, we are still left in doubt as to the age of the convent near Siout and other buildings. From comparison of the plans and descriptions given in Denon, Curzon and Pococke of these buildings, with those in Cairo and Old Cairo, Mr. Butler ascribes them to the fourth century, that which in fact is claimed for them as having been founded by Sta. Helena. On this subject he says, p. 365: “Were there no more of evidence besides to determine the truth of this tradition, the plan of the Haikal (the central of the three chapels in a Coptic church) would decide it beyond question. The persistence with which certain churches are ascribed to Sta. Helena by a people utterly ignorant of history and architecture is in itself remarkable, and it is still more remarkable to find that these churches are always marked by a particular form of Haikal. Indeed, so regular is the coincidence, that a deep apsidal haikal with recesses all round it and columns close against the wall may be almost infallibly dated from the age of Sta. Helena.”

[262]. The older church has been so altered and ruined by the subsequent rebuildings that it is extremely difficult to make out its history. It seems, however, to have been built originally above the site of an old Mithraic temple, which has recently been cleared out, and probably before the time of Gregory the Great. It was apparently rebuilt, or nearly so, by Adrian I., 772, and burnt by Robert Guiscard, 1084. The upper church seems to have been erected by Paschal, 1099-1118. The question is, to what age do the frescoes found on the walls of the older church belong? Some of the heads and single figures may, I fancy, be anterior even to the time of Adrian; but the bulk of the paintings seem certainly to have been added between his age and 1084, and nearer the latter than the former date. If it had not been entirely ruined in 1084 Paschal would not have so completely obliterated it a century afterwards. A considerable quantity of the materials of the old church were used in the new, which tends further to confuse the chronology.

[263]. Gutensohn and Knapp, ‘Die Basiliken des Christlichen Roms.’

[264]. Cicero de Legg., ii. 24; Festus, s. v.; Smith’s ‘Dictionary of Classical Antiquities.’

[265]. The dates here given generally refer to the building now existing or known, and not always to the original foundation.

[Mr. G. G. Scott, in his work before referred to (p. 506), after giving a full quotation from Eusebius of Constantine’s basilica at Jerusalem, in which he points out that the orientation of primitive times is the reverse of that which has become general in later times, continues his enquiry into the evidence afforded by the numerous early basilicas in Rome itself. Of about fifty churches of early date, in forty of them the sanctuary is placed at the western end, and of the remaining ten (one of which is the great church of St. Paolo fuori le Mura), there are only seven which appear to have retained their original form, and which have an eastward sanctuary.

The exact orientation of the sanctuary in each case has been added to the list.—Ed.]

[266]. ‘Il Vaticano discritto da Pistolesi,’ vol. ii. pls. xxiv. xxv.

[267]. The new church which superseded this one is described in the History of the Modern Styles of Architecture, vol. i., page 111, woodcut 45.

[268]. It should be observed that the dosseret is first found in Italy in the Church of St. Stefano Rotondo, built 468-482, and is there of similar design to examples in Thessalonica.

[269]. ‘L’architecture en Italie du vie au xie siècle.’ Venice, 1891.

[270]. ‘Altchristlichen Kirchen nach Baudenkmalen und alteren Beschreibungen,’ von D. Hubsch. Carlsruhe, 1862.

[271]. These piers were built in the 12th century, taking the place of the columns of the original Basilican church of the 9th century, and the arches date from the same period (Cattaneo).

[272]. It is now called S. Martino in Cielo d’Oro, from its having been decided in the twelfth century that the other church in Classe possessed the true body of the saint to which both churches were dedicated.

[273]. A. F. von Quast, ‘Die Altchristlichen Bauwerke von Ravenna.’

[274]. The basilica Pudenziana at Rome has similar arcades externally.

[275]. The twenty-four marble columns are said to have been brought over from Constantinople, but they were probably obtained from Greek quarries.

[276]. [The narthex as shown in Woodcut No. [409] is of much later date than the church, and has been partially rebuilt on two or three occasions. It is now (1892) being taken down, and the removal of the central portion has uncovered the triple window which originally lighted the nave.—Ed.]

[277]. “La basilica di San Marco in Venezia,” by Cattaneo, continued by Boito. Venezia, 1890.

[278]. Probably owing to its having been utilized to receive the relics of St. Mark, which were temporarily hidden there.

[279]. This church, built by Justinian, no longer exists, having been pulled down in 1464 by Mohammed II. to make way for his mosque. From the description of it, however, given by Procopius, the plan was similar to that adopted in St. Mark, being that of a Greek Cross with central and four other domes. Procopius speaks of the church being surrounded within by columns placed both above and below, probably referring to galleries similar to those in St. Sophia of Constantinople. In St. Mark’s the columns exist in one storey only, and the main wall is carried up at the back of the aisles to give increased size inside.

[280]. Originally, according to M. Cattaneo, his was the vestibule to the atrium from the south, but it is now blocked up by an altar.

[281]. [They are shown in the mosaic of the doorway of St. Alipe, executed at the end of the 13th century, as also the filling in of the great west window.—Ed.]

[282]. ‘Dalmatia, the Quarnero and Istria,’ by T. G. Jackson, M.A. Oxford, 1887.

[283]. In support of this statement he points out that twice during Christian times it had been found necessary to raise the floor of the church. The nave floor, which in 1857 was two steps below that of the aisles, was raised in 1881 to the same level; but two feet nine inches below the nave floor before it was raised there existed, according to Prof. Eitelberger, another mosaic pavement, which must have been the floor of the first basilica erected, and which was pulled down by Bishop Euphrasius in 543. This lower pavement extended also under the three chapels of the confessio, which suggests that these are part of the first basilica.

[284]. The same polygonal form is found in the apses of St. Agatha, St. Apollinare in Classe, St. Apollinare in Nuovo, St. Spirito, and St. Vitale, all in Ravenna, and St. Fosca, Torcello.

[285]. The apses of two churches, of the 4th and 6th century respectively, in the island of Paros, are similarly fitted with marble seats: in the 6th century church there are eight rows, so that the apse looks like a small amphitheatre.

[286]. That is on the supposition that the word kirk is derived from the Latin word “circus,” “circular,” as the French term it, “cirque.” My own conviction is that this is certainly the case. The word is only used by the Barbarians as applied to a form of buildings they derived from the Romans. Why the Germans should employ κυρίου οἶκος, when neither the Greeks nor the Latins used that name, is a mystery which those who insist on these very improbable names have as yet failed to explain.

[287]. The Tholos at Epidaurus seems to be an exception to this rule.

[288]. Isabelle, ‘Édifices Circulaires,’ plates 26 and 27.

[289]. M. Cattaneo states that it was built by Pope St. Simplice, 468-482.

[290]. Above the capitals are impost blocks or dosserets, the earliest known examples of that feature in Italy.

[291]. [The Vaults over the outer aisle of St. Stefano Rotondo were built with hollow pots, the remains of which can still be traced in the outer walls of the 2nd aisle.

Prof. Middleton points out also the existence of rings of earthen pots in the vault of the tomb of Sta. Helena (Woodcut No. [227]), and also in the vaults of the Circus of Maxentius, on the Via Appia.—Ed.]

[292]. In this building they now show a sarcophagus of ancient date, said to be that of Galla Placidia, daughter of Theodosius. She, however, was certainly buried at Ravenna; but it may be of her time, and in these ages it is impossible to distinguish between baptisteries and tombs.

[293]. Frederick Von Osten, ‘Bauwerke in der Lombardei.’ Darmstadt, 1852.

[294]. By an oversight of the engraver, the vault of the nave, which ought to be made hexapartite, is drawn as quadripartite. [The nave was so completely restored in the 14th century as to render doubtful the original existence of a vault.—Ed.]

[295]. Étude de l’Architecture Lombarde,’ par F. de Dartein. Paris, 1878.

[296]. These are incorrectly shown on woodcut. The central pier is nearly 4 feet wide and carried a transverse rib of the same size and of two orders.

[297]. Ferrario, ‘Monumenti Sacri e Profani dell’ I. R. Basilica di S. Ambrogio,’ Milan, 1824.

[298]. “Quid dicamus columnarum junceam proceritatem? Moles illas sublimissimas quasi quibusdam erectis hastilibus contineri substantiæ qualitate concavis canalibus excavatas vel magis ipsas æstimes esse transfusas. Ceris judices factum quod metallis durissimis videas expolitum. Marmorum juncturas venas dicas esse genitales, ubi dum falluntur oculi laus probatur crevisse miraculis.” In the above, metallum does not seem to mean metal as we now use the word, but any hard substance dug out of the ground. (Cassiodorus, Variorum, lib. vii. ch. 15.)

[299]. See vol. i. p. 372.

[300]. ‘The Land of Moab,’ by Dr. Tristram (Murray, 1873), pp. 376 et seqq. [The small triangular marble panels referred to in Murano are of a very elementary character in their carving, and have scarcely the importance attached to them by Mr. Fergusson. Besides, the same wall decoration in brickwork is found in the apse of St. Fosca, Torcello (c. 1008), where, however, the triangular recesses are simply covered with stucco and painted; being closer to the eye in Murano, they filled the spaces with incised marble slabs: in other words, it seems more probable that the slabs were made for the triangular panels than the converse, which is suggested by Mr. Fergusson.—Ed.]

[301]. The typical example of this class is the San Giorgio at Venice, though it is not by any means the one most like St. Pietro; many attempts were made before it became so essentially classical as this (see Woodcut No. 39, Vol. I. in the ‘History of Modern Architecture’).

[302]. From the boldness of the construction, M. Cattaneo is induced to place the erection of the building at the end of the 11th or beginning of the 12th century.

[303]. The four square towers of San Lorenzo, Milan, and the circular campanile by the side of the cathedral of Ravenna, are the earliest examples known, the latter dating from the commencement of the 5th century.

[304]. [The tower of St. Satiro at Milan (879 A.D.), is considered by Cattaneo to be the most ancient campanile known in which the wall surface is broken up with flat pilasters or vertical bands in relief, and divided into storeys by horizontal string courses, with ranges of small blind arches below, carried on corbels, and may be regarded as the prototype of the most characteristic Lombard towers.—Ed.]

[305]. ‘History of Medieval Art,’ by Dr. F. M. Reber, translated by J. T. Clarke. New York, 1887.

[306]. ‘Dalmatia, the Quarnero and Istria,’ by T. G. Jackson, A.R.A. Oxford, 1887.

[307]. Schultz, ‘Denkmäler der Kunst der Mittelalters in Unter-Italien.’ Folio, 1860.

[308]. The polygonal form given to the apse externally shows the direct influence of Byzantine art.

[309]. The cornice projects 1 ft. 10 in., and consequently overhangs the base by 13 ft.

[310]. The present cathedral is only a portion, viz. the transept of a much vaster edifice which was never completed; but the beautiful unfinished south front and portions of the gigantic nave and aisles still exist on the western side of the present cathedral, and the drawings of it are preserved in the archives of the Duomo.

[311]. [Since this was written the façade has been completed to harmonize with the rest, but not in accordance with the original design, if we may judge by the painting in Sta. Maria Novella, which shows side gablets similar to those of the cathedral of Siena.—Ed.]

[312]. If we may trust Wiebeking, the first two bays of the nave from the front were vaulted in 1588, but the work was suspended till 1647, and completed only in 1659. Yet no difference can be perceived in the details of the design.

[313]. The plan and section being taken from two different writers, there is a slight discrepancy between the scales. I believe the plan to be the more correct of the two, though I have no means of being quite certain on the point.

[314]. ‘Dispareri d’Architettura.’

[315]. Within the last few years a façade has been added to Sta. Croce, but about which the less said the better. It is wretched in design.

Transcriber’s Notes

This book often uses inconsistent spelling, particularly with respect to accents. These were left as printed unless the author showed a clear preference for one form.

Some presumed printer’s errors have been corrected, including normalizing punctuation. Page number references in the Table of Contents were corrected where errors were found. Further corrections are listed below with the original text (top) and the corrected text (bottom).

every pains has been taken

every pain has been taken p. [xxii]

progres

progress p. [48]

cotemporary

contemporary p. [50]

formula

formulæ p. [77]

Sedinag

Sedinga Illustration [27.]

longed ceased

long ceased p. [219]

Nor is is

Nor is it p. [247]

ines

lines p. [372]

Roumeia

Roumeïa p. [372]

Nimes

Nîmes p. [385]

Vogüe

Vogüé p. [423]

neo-Byzantine

Neo-Byzantine p. [455]

iconicon

icon p. [460]

orginally

originally p. [538]

turned the

turned to the p. [558]

100 ft. to

100 ft. to 1 in. Illustration [451.]

467. Illustration [467] (missing number added)

next

next to p. [596]