CHAPTER IV

.

LOMBARD AND ROUND-ARCHED GOTHIC.

CONTENTS.

Chapel at Friuli—Churches at Piacenza and Novara—St. Michele, Pavia—St. Ambrogio, Milan—Cathedral, Piacenza—Churches at Verona—Churches at Toscanella—Circular Churches—Towers.

When, in the early centuries of the Christian era, the great mass of Gothic barbarism moved up the Valley of the Danube towards the west, one great division followed that river to its source, and thence penetrated into and settled in the Valley of the Rhine. Though sufficiently numerous to be able almost wholly to obliterate all traces of former civilisation, they had virtually no style of their own, and it seems probable that the edifices left by the Romans sufficed for the early wants of the people.

The other great division of the horde turned to the Sömmering Alps and, penetrating into Italy by way of Udine and Conegliano, settled in the Valley of the Po. They may have been as numerous as the others; but Italy in those days was far more densely peopled than Germany, and the inhabitants were consequently able to resist obliteration far more successfully than on the north of the Alps, and even where the new element prevailed most strongly its influence was far less felt than in the more sparsely-peopled Rhenish provinces. This was generally more apparent along the coast than in the interior. Venice did not exist, and Ravenna, though overwhelmed, became the great centre of Romano-Byzantine art. Pisa and Lucca resisted throughout. Florence was divided. The Barbarian influence was strongly felt at Siena, more feebly at Orvieto; but there it was stopped by the influence of Rome, which throughout the Middle Ages remained nearly uncontaminated.

Notwithstanding the almost insuperable barrier of the Alps which stretched between them and the different influences to which they were subjected, the connection between the northern and southern hordes remained intimate during the whole of the Middle Ages. Milan was as much German as Italian; and, indeed, except from a slightly superior degree of elegance in the southern examples, it is sometimes extremely difficult to distinguish between the designs of Lombard and of Rhenish churches. As the Middle Ages wore on, however, the breach between the two styles widened; and there is no difficulty, in the later pointed schools, in seeing how Italy was gradually working itself free from German influence, till at last they became distinct and antagonistic nationalities, practising two styles of art, which had very little in common the one with the other.

Whoever the Barbarians were who in the 5th and 6th centuries swarmed into Italy—Austro-Goths, Visi-Goths, or Lombards—they certainly did not belong to any of the great building races of the world. Few people ever had better opportunities than they of employing their easily-acquired plunder in architectural magnificence, if they had any taste that way; but, though we hear everywhere of the foundation of churches and the endowment of ecclesiastical establishments during the Carlovingian period, not one important edifice of that age has come down to our time. The monumental history of the early Romanesque style is as essentially a blank in Italy as it is in Saxon England. One or two circular buildings remain tolerably entire; some small chapels let us into the secrets of the style, but not one important edifice of any sort attests the splendour of the Lombard kingdom of Northern Italy. Aryans they must have been, and it was not till the beginning of the 11th century, when their blood was thoroughly mixed with that of the indigenous inhabitants and a complete fusion of races had taken place, that we find buildings of a monumental character erected, which have come down to the present day.

[Illustration: 439. Chapel at Friuli. (From Gailhabaud.)]

Among the smaller monuments of the age none has been preserved more complete and less altered than the little chapel at Friuli; which, though extremely small (only 18 ft. by 30 inside the walls), is interesting, as retaining all its decorations almost exactly as they were left by Gertrude, duchess of Friuli, who erected it in the 8th century. It shows considerable elegance in its details, and the sculpture is far better than it afterwards became, though perhaps its most remarkable peculiarity is the intersecting vault that covers it— _pulchre testudinatum_, as the old chronicle terms it. This is one proof among many, how early that feature was introduced which afterwards became the formative principle of the whole Gothic style, and was as essentially its characteristic as the pillars and entablatures of the five orders were the characteristics of the classical styles of Greece and Rome. As before remarked, it is this necessity for a stone roof that was the problem to be solved by the architects, and to accomplish which the style took almost all those forms which are so much admired in it.

From this example of the Carlovingian era we are obliged to pass to the 11th and 12th centuries, the first great building age of the Lombards. It is true that there is scarcely a single important church in Pavia, in Verona, or indeed in any of the cities of Lombardy, the original foundation of which cannot be traced back to a much earlier period. Before the canons of architectural criticism were properly understood, antiquaries were inclined to believe that in the buildings now existing they saw the identical edifices erected during the period of the Lombard sway. Either, however, in consequence of the rude construction of the earlier buildings, or because they were too small or too poor for the increased population and wealth of the cities at a later period, every one of the original churches has disappeared and been replaced by a larger and better-constructed edifice, adorned with all the improvements which the experience of centuries had introduced into the construction of religious edifices.

Judging from the rudeness of the earliest churches which we know to have been erected in the 11th century, it is evident that the progress made, up to that period, was by no means equal to what was accomplished during the next two centuries.

[Illustration: 440. Plan of San Antonio, Piacenza. (From Osten.[293]) Scale 100 ft. to 1 in.[294]]

This will appear from the plan and section of St. Antonio at Piacenza (Woodcuts Nos. 440 and 440a), built in the first years of the 11th century, and dedicated in 1014 by Bishop Siegfried.

[Illustration: 440a. Section of Church of San Antonio at Piacenza. (From Osten.)]

[Illustration: 441. Plan and section of Baptistry at Asti. (From Osten.) Scale 50 ft. to 1 in.]

Its arrangement is somewhat peculiar; the transepts are near the west end, and the octagonal tower rising from the intersection is supported on eight pillars, the square being completed by four polygonal piers. The principal point, however, to observe is, how completely the style has emancipated itself from all Roman tradition. A new style has grown up as essentially different from the early Christian as the style of Cologne or of York Cathedral. The architect is once more at liberty to work out his own designs without reference to anything beyond the exigencies of the edifices themselves. The plan, indeed, is still a reminiscence of the Basilica; but so are all the plans of Mediæval cathedrals, and we may trace back the forms of the pillars, the piers, and the arches they support, to the preceding style. All these were derived from Roman art, but the originals are forgotten, and the new style is wholly independent of the old one. The whole of the church too is roofed with intersecting vaults, which have become an integral part of the design, giving it an essentially different character. On the outside buttresses are introduced—timidly, it is true, but so frequently, as to make it evident that already there existed no insuperable objection to increase either their number or depth, as soon as additional abutment was required for wider arches.

The windows, as in all Italian churches, are small, for the Italians never patronised the art of painting on glass, always preferring frescoes or paintings on opaque grounds. In their bright climate, very small openings alone were requisite to admit a sufficiency of light without disturbing that shadowy effect which is so favourable to architectural grandeur.

Being a parochial church, this building had no baptistery attached to it; but there is one at Asti (Woodcut No. 441) so similar in style and age, that its plan and section, if examined with those of San Antonio, will give a very complete idea of Lombard architecture in the beginning of the 11th century, when it had completely shaken off the Roman influence, but had not yet begun to combine the newly-invented forms with that grace and beauty which mark its more finished examples. One peculiarity of this building is the gloom that reigns within, there being absolutely no windows in the dome, and those in the aisles are so small, that even in Italy the interior must always have been in comparative darkness.

[Illustration: 442. Plan of the Cathedral at Novara. Scale 100 ft. to 1 in.]

The cathedral of Novara, which in its present state is one of the most important buildings of the 11th century in the North of Italy, shows the style still further advanced. The coupling and grouping of piers are here fully understood, and the divisions of the chapels which form the outer aisle are, in fact, concealed buttresses. The Italians were never able to divest themselves of their partiality for flat walls, and never liked the bold external projections so universally admired on the other side of the Alps. They therefore gladly had recourse to this expedient to conceal them; and when this was not available they used metallic ties to resist the thrust of the arches—an expedient which is found even in this example. As will be seen from the annexed plan, the atrium connecting the basilica with the baptistery is retained, which seems to have been an arrangement almost universal in those early times. The half section, half elevation of the front (Woodcut No. 443) shows very distinctly how far the invention of the new style had then gone; for except some Corinthian pillars, borrowed from an older edifice, no trace of debased-Roman architecture is to be found in it. The design of the façade explains what it was that suggested to the Pisan architects the form to which they adapted their Romanesque details. In both styles the arcade was the original model of the whole system of ornamentation. In this case it is used first as a discharging arch, then as a mere repetition of a useful member, and lastly without pillars, as a mere ornamental string-course, which afterwards became the most favourite ornament, not only in Italy, but throughout all Germany.

[Illustration: 443. Elevation and Section of the Façade of the Cathedral at Novara. (From Osten.)]

Interesting as such an example is to the architectural antiquary who is tracing back and trying to understand the forms of a new style, it would be difficult to conceive anything much uglier and less artistic than such a façade as this of Novara or that of San Antonio, last quoted. Their sole merit is their history and their expression of rude energy, so characteristic of the people who erected them.

[Illustration: 444. Section of San Michele, Pavia. (From Agincourt.) No scale.]

The church of San Michele at Pavia, which took its present form either at the end of the 11th or beginning of the 12th century, is one of the most interesting of this age, and presents in itself all the characteristics of a perfect round-arched Gothic church. Indeed there is hardly any feature worth mentioning which was invented after this date except the pointed arch—a very doubtful improvement—and window tracery, which the Italians never cordially adopted or understood. The section (Woodcut No. 444) shows the general arrangement of San Michele in its present state. The researches of M. de Dartein,[295] however, have shown that, when first built, the nave was covered over with two square quadripartite vaults, as might in fact have been divined from the difference in size[296] of the centre and two other piers. The existing oblong vaulted compartments date from the 15th century, when secondary shafts were carried up above the ground storey shafts of piers 1 and 3. The section, however, shows that well-marked vaulting shafts spring from floor to roof, that the pier arches in the wall are probably distinct and well understood, and that the angles of these piers are softened and ornamented by shafts and other subordinate members. Altogether, it is evident that that subdivision of labour (if the expression may be used) which was so characteristic of the true Gothic style had here been perfectly understood, every part having its own function and telling its own story. To complete the style only required a little experience to decide on the best and most agreeable proportions in size and solidity. In a century from the date of this church the required progress had been made; a century later it had been carried too far, and the artistic value of the style was lost in mere masonic excellence. San Michele and the other churches of its age fail principally from over-heaviness of parts and a certain clumsiness of construction, which, though not without its value as an expression of power, wants the refinement necessary for a true work of art. Externally, one of the most pleasing features of this church is the apse with its circular gallery. In Italian churches the gallery is usually a simple range of similar arcades; here, however, it is broken into three great divisions by coupled shafts springing from the ground, and these again subdivided by single shafts running in like manner through the whole height of the apse. The gallery thus not only becomes a part of the whole design, instead of looking like a possible afterthought, but an agreeable variety is also given, which adds not a little to the pleasing effect of the building.

[Illustration: 445. View of the Apse of San Michele, Pavia. (From Du Somerard, ‘Les Arts au Moyen-Age.’)]

There are at least two other churches in Pavia, which, though altered in many parts, retain their apsidal arrangements tolerably perfect. One of these, that of San Teodoro (1150), may be somewhat later than the San Michele, and has its gallery divided into triplets of arcades by bold flat buttresses springing from the ground. In the other, San Pietro in Cielo d’Oro, dating from 1132, the arcade is omitted round the apse, though introduced in the central dome. It has besides two subordinate apses of graceful design, but inferior to the other examples.

Though Milan must have been rich in churches of this age, the only one now remaining tolerably entire is San Ambrogio, which is so interesting as almost to make amends for its singularity. Historical evidence shows that a church existed here from a very early age. It was rebuilt in the 9th century by Bishop Angelbert, aided by the munificence of Louis the Pious, and an atrium was added by Bishop Anspertus; but except the apse and “the canons’” tower, nothing remains of even that church, all the rest having been rebuilt in the 11th or 12th century. During the late restoration the bases of some of the columns of the 9th-century church were discovered, and one of them is now visible in the pulpit enclosure.

The disposition of the building will be understood from the annexed plan, which shows both the atrium and the church. The former is virtually the nave; in other words, had the church been erected on the colder and stormier side of the Alps, a clerestory would have been added to the atrium, and it would have been roofed over; and then the plan would have been nearly identical with that of a Northern cathedral.

[Illustration: 446. Plan of San Ambrogio, Milan. (From Ferrario.) Scale 100 ft. to 1 in.]

The third (sexpartite) bay was revaulted in the 14th century with two oblong quadripartite vaults, but these are now replaced by sexpartite vaulting. The dome is probably an addition of the end of the 12th century, and it is raised over what would otherwise have been the fourth bay of the church. As it is, the atrium (Woodcut No. 446) is a highly pleasing adjunct to the façade, removing the church back from the noisy world outside, and by its quiet seclusion tending to produce that devotional feeling so suitable to the entrance of a place of worship. The façade of the building itself, though, like the atrium, only in brick, is one of the best designs of its age; the upper loggia, or open gallery, of five bold but unequal arches, producing more shadow than the façade at Pisa, without the multitude of small parts there crowded together, and with far more architectural propriety and grace. As seen from the atrium, with its two towers, one on either flank, it forms a composition scarcely surpassed by any other in this style.

As now restored, the simplicity and line effect of the vaulted interior is remarkable, and it is also a museum of ecclesiological antiquities of the best class. The silver altar of Angilbertus (A.D. 835) is unrivalled either for richness or beauty of design by anything of the kind known to exist elsewhere, and the _baldacchino_ that surmounts it is also of singular beauty: so are some of its old tombs, of the earliest Christian workmanship. Its mosaics, its pulpit, and the bronze doors, not to mention the brazen serpent—said to be the very one erected by Moses in the wilderness—and innumerable other relics, make this church one of the most interesting of Italy, if not indeed of all Europe.

[Illustration: 447. Atrium of San Ambrogio, Milan. (From Ferrario.[297])]

Generally speaking, the most beautiful part of a Lombard church is its eastern end. The apse with its gallery, the transepts, and above all the dome that almost invariably surmounts their intersection with the choir, constitute a group which always has a pleasing effect, and is very often highly artistic and beautiful. The sides of the nave, too, are often well designed and appropriate; but, with scarcely a single exception, the west end, or entrance front, is comparatively mean. The building seems to be cut off at a certain length without any appropriate finish, or anything to balance the bold projections towards the east. The French cathedrals, on the contrary, while they entirely escape this defect by means of their bold western towers, are generally deficient in the eastern parts, and almost always lack the central dome or tower. The English Gothic architects alone understood the proper combination of the three parts. The Italians, when they introduced a tower, almost always used it as a detached object, and not as a part of the design of the church. In consequence of this the façades of their churches are frequently the least happy parts of the composition, notwithstanding the pains and amount of ornament lavished upon them.

[Illustration: 448. Façade of the Cathedral at Piacenza. (From Chapuy, ‘Moyen-Âge Monumental.’)]

The elevation of the cathedral at Piacenza is a fair illustration of the general mode of treating the western front of the building, not only in the 11th and 12th centuries, but afterwards, when a church had a façade at all—for the Italians seem to have been seldom able to satisfy themselves with this part of their designs, and a great many of their most important churches have, in consequence, not even now been completed in this respect.

Instead of recessing their doors, as was the practice on this side of the Alps, the Italians added projecting porches, often of considerable depth, and supported by two or more slight columns, generally resting on the backs of symbolical animals. No part of these porches, as an architectural arrangement, can be deemed worthy of any commendation; for, in the first place, a column planted on an animal’s back is an anomaly and an absurdity, and the extreme tenuity of the pillars, as compared with the mass they support, is so glaring that even its universality fails in reconciling the eye to the disproportion. In the present instance the porch is two storeys in height, the upper being a niche for sculpture. Its almost exact resemblance to the entrance porch below is therefore a defect. Above there is generally a gallery, sometimes only in the centre; sometimes, as in this instance, at the sides, though often carried quite across; and in the centre above this there is almost invariably a circular window, the tracery of which is frequently not only elaborately but beautifully ornamented with foliage and various sculptural devices.

Above this there is generally one of those open galleries mentioned before, following the slope of the roof, though frequently, as in this instance, this is replaced by a mere belt of semicircular arches, suggesting an arcade, but in reality only an ornament.

VERONA.

Almost every important city in Lombardy shows local peculiarities in its style, arising from some distinction of race or tradition. The greater number of these must necessarily be passed over in a work like the present, but some are so marked as to demand particular mention. Among these that of Verona seems the most marked and interesting. This Roman city became the favourite capital of Theodoric the Goth—Dietrich of Berne, as the old Germans called him—and was by him adorned with many noble buildings which have either perished or been overlooked. There is a passage in the writings of his friend Cassiodorus which has hitherto been a stumbling-block to commentators, but seems to find an explanation in the buildings here, and to point to the origin of a mode of decoration worth remarking upon. In talking of the architecture of his day he speaks of “the reed-like tenuity of the columns making it appear as if lofty masses of building were supported on upright spears, which in regard to substance look like hollow tubes.”[298] It might be supposed that this referred exclusively to the metal architecture of the use of which we find traces in the paintings at Pompeii and elsewhere.[299] But the context hardly bears this out, and he is probably alluding to a stone or marble architecture, which in the decline of true art had aspired to a certain extent to imitate the lightness which the metallic form had rendered a favourite.

To return to Verona:—The apse of the cathedral seems to have belonged to an older edifice than that to which it is now attached, as was often the case, that being the most solid as well as the most sacred part of the building. As seen in the woodcut (No. 449) it is ornamented with pilasters, classical in design, but more attenuated than any found elsewhere; so that I cannot but believe that this is either one of the identical buildings to which Cassiodorus refers, or at least an early copy from one of them.

[Illustration: 449. Apse of the Cathedral, Verona. (From Hope’s ‘History of Architecture.’)]

At a far later age, in the 12th century, the beautiful church of San Zenone shows traces of the same style of decoration (Woodcut No. 450), pilasters being used here almost as slight as those at the cathedral, but so elegant and so gracefully applied as to form one of the most beautiful decorations of the style. Once introduced, it was of course repeated in other buildings, though seldom carried to so great an extent or employed so gracefully as in this instance. Indeed, whether taken internally or externally, San Zenone may be regarded as one of the most pleasing and perfect examples of the style to be found in the North of Italy.

The cathedral at Modena is another good example, though not possessing any features of much novelty or deserving special mention. That of Parma is also important, though hardly so pleasing. Indeed, scarcely any city in the Valley of the Po is without some more or less-perfect churches of this date, none showing any important peculiarities that have not been exemplified above, unless perhaps it is the apse of the church of San Donato on the Murano near Venice, which is decorated with a richness of marble decoration to which the purer Gothic style never attained, and which entitles this church to rank rather with the Byzantine than with the Lombard buildings of which we are treating, or a style so curiously exceptional as to make it one of the most interesting churches, historically, to be found in the North of Italy.

[Illustration: 450. Façade of San Zenone, Verona. (From Chapuy.)]

Recent discoveries in Syria[300] have proved almost beyond a doubt that the carved slabs with which it is adorned externally were borrowed from some desecrated building on the coast of Syria—destroyed probably by the Moslems—and brought to Venice, probably at the time when the church acquired the remains of San Donato, in the beginning of the 12th century. Whether brought then or at an earlier period, they belong to the age of Justinian, certainly came from the East, and, mixed up with Italian details of the period, make up an exterior as picturesque as it is interesting to the student of the history of art in those days.

It is extremely difficult to draw a line between the pointed and round-arched Gothic styles in Italy. The former was so evidently a foreign importation, so unwillingly received and so little understood, that it made its way but slowly. Even, for instance, in the church at Vercelli, which is usually quoted as the earliest example of the pointed style in Italy (built 1219-1222), there is not a pointed arch nor a trace of one on the exterior. All the windows and openings are round-headed, and, except the pier-arches and vaults, nothing pointed appears anywhere. Even at a later date than this the round arch, especially as a decorative form, is frequently placed above the pointed one, and always used in preference to it. Instead, therefore, of attempting to draw a line where none exists in reality, it will be better now to pass on from this part of the subject, and to take up the older style at a point from which we can best trace the formation of the new. The latter does not essentially differ from the former, except in the introduction of the French form of the pointed arch and its accompaniments. It remains only to say a few words on the peculiarities which the round form of churches took in the hands of the early Lombard architects, as well as on the campanile, which forms so striking a feature in the cities of Northern Italy.

TOSCANELLA.

On the boundary-line which separates the Guelfic from the Ghibelline influence, there exist at Toscanella, near Viterbo, two churches of great beauty of detail; but which, as might almost be predicated from their situation, defy any attempt at classification. They are not Gothic, for they have no vaults, nor does their style suggest any vaulting contrivances. They are scarcely debased Roman, for the tracery of their circular windows, their many-shafted doors, and generally their details are such as to indicate a Northern rather than a Roman affinity; and the Byzantine sculpture which is found in the pulpit was probably taken from an earlier church—though an Italian Byzantine influence can be traced in much of its decoration. Under these circumstances, it is better to treat them as exceptional, than to attempt to give them a name which might mislead without conveying any correct information.

The elder of these two churches, Sta. Maria, was erected in the beginning of the 13th century (1206?), but is so unlike most buildings of that age, that it is usually ascribed to the 6th or 7th. On a close examination, however, all its details are found to be full of advanced Romanesque forms. The pillars are rude Corinthian, with a Lombardic abacus. They are widely spaced, having no vault to support; and the mouldings of the arches are what we should call “Transitional Early English.”

[Illustration: 451. Plan of Sta. Maria, Toscanella. (From Gailhabaud.) Scale 100 ft. to 1 in.]

Externally the façade is too plain to be quite pleasing, but this arises from its depending originally on painting for its decoration—some traces of which still remain, but the greater part has perished. Its three doorways are richly and beautifully ornamented with shafts and sculptured foliage, quite equal in detail to anything of the class to be found in Italy, and its great circular window would not be thought out of place at Chartres or Lincoln.

[Illustration: 452. View of the Interior of Sta. Maria, Toscanella. (From Gailhabaud.)]

The church of St. Pietro is probably a century later than that of Sta. Maria, and its façade is richer and more elegant—a difference arising more from those details being in this instance carved which in the earlier church were painted. The design, however, deserves attention for its historical, perhaps, even more than its artistic claims; for it was this class of façade that Palladio and the architects of the cinque-cento period seized upon, and, applying pilasters and pediments of classical type, converted it into the fashionable churches which are to be found in every part of Europe.[301]

[Illustration: 453. Elevation of the Exterior of Sta. Maria, Toscanella. (From Gailhabaud.) No scale.]

The difficulty which the Italians never entirely conquered, was how to amalgamate the sloping lines of the roofs of the aisles with the horizontal lines of the rest of the façade. The gallery over the central doorway enabled them very nearly to accomplish it in these Toscanella churches, and if the same string-courses had been carried all across, the whole might have been harmonised; but it was just missed, and, what is strange, more so in the second than in the first example.

CIRCULAR CHURCHES.

In the earliest times of Christian architecture, as we have already seen, the circular form of church was nearly as frequent as that derived from the Roman basilica. In process of time the latter was found to be much better adapted to the extended requirements of Christianity. Hence in the 11th and 12th centuries, when so many of the early churches were rebuilt and enlarged, most of the old circular buildings disappeared. Enough, however, remain to enable us to trace, though imperfectly, what their arrangements were.

Among those which have been illustrated, perhaps the most interesting is that known as the church of San Stefano at Bologna, or rather the circular centre of that congeries of seven churches usually known by that name.

It is one of those numerous churches of which it is impossible to predicate whether it was originally a baptismal or a sepulchral edifice. In old times it bore both names, and may have had both destinations, but latterly, at all events, the question has been settled by the compromise usually adopted in such cases, of dedicating it to the first martyr, to whom a sepulchral form of building is especially appropriate.

[Illustration: 454. Plan of the Duomo, Brescia. (From Hübsch.) Scale 100 ft. to 1 in.]

[Illustration: 455. Elevation of Duomo at Brescia. (From Hübsch.) Scale 50 ft. to 1 in.]

Notwithstanding a considerable amount of ancient remains mixed up in the details, no part of the present church seems older than the Carlovingian era; while, on the other hand, its extreme irregularity and clumsiness of construction point to a period before the 11th century. Its general form is that of an extremely irregular octagon, about 60 ft. in diameter, in the centre of which stands a circlet of columns, some coupled, some single, supporting a semicircular dome. The circumscribing aisle is covered with the usual intersecting ribbed vault of the 10th century, but the whole is so rude as scarcely to deserve mention except for its antiquity.

The Duomo Vecchio of Brescia is ascribed to the 8th or 9th century, but this date according to Cattaneo[302] can only be ascribed to an earlier basilica church, the crypt of which still exists on the east side of the Duomo. As will be seen from the plan, it is a large church, 125 ft. across over all, and is covered by a dome 65 ft. in diameter internally supported by eight piers of plain design. The mode in which light is introduced into the central compartment illustrates the various tentative expedients by which the architects in that age attempted to accomplish their object. First, there is a range of small windows in the dome below the springing of the dome. In the dome itself there are four circular sides, and, as if the architect felt that he was doing something unusual and inartistic, he managed externally to confuse these with the rudiments of the roof gallery.

[Illustration: 456. Section of Duomo at Brescia. (From Hübsch.) Scale 50 ft. to 1 in.]

[Illustration: 457. San Tomaso in Limine. Scale 100 ft. to 1 in.]

[Illustration: 458. San Tomaso. (From Isabelle, ‘Édifices Circulaires.’) Scale 50 ft. to 1 in.]

It is not clear whether originally it had or was intended to have an apse between its two round towers—where the foundations of some buildings can still be traced; but these may be the remains of the early church referred to.

Turning from these, we find the round-arched Gothic style completely developed in the church of San Tomaso in Limine, near Bergamo. From the annexed plan it will be seen that the circular part is the nave or entrance, as in Germany and England, in contradistinction to the French mode of arrangement, where the circular part is always the sanctum, the rectangular the nave or less holy place.

The general plan of this example is circular. It is not more than 30 ft. across internally. In the centre stand eight pillars, supporting a vaulted gallery, which forms a triforium or upper storey, and, with the dome and its little cupola, raise the whole height to about 50 ft. A small choir with a semicircular niche projects eastward.

The dimensions of the building are so small that it hardly deserves notice, except as a perfect example of the style of the 11th or 12th century in Lombardy, and for a certain propriety and elegance of design, in which it is not surpassed, internally at least, by any building of its age. It is to be regretted that the idea was never carried out (at any rate no example remains) on such a scale as to enable us to judge of the effect of such a domical arrangement as is here attempted. The great defect of all one-storeyed domes is their lowness, both internally and more especially externally. This method of building a dome in two storeys would seem calculated to obviate the objection; but though common in small sepulchral chambers, it has never been tried on a scale sufficiently large to enable us to judge of its real effect. After this period the circular shape was so completely superseded by the rectangular, that no further improvement took place in it.

TOWERS.

There is perhaps no question of early Christian archæology involved in so much obscurity as that of the introduction and early use of towers. The great monumental pillars of the Romans—as, for instance, those of Trajan and Antoninus—were practically towers; and latterly their tombs began to assume an aspiring character like that at St. Remi (Woodcut No. 231), or those at Palmyra and elsewhere in the East, which show a marked tendency in that direction. But none of these can be looked upon as an undoubted prototype of the towers attached to the churches of the Christians.

At Ravenna, as early as the age of Justinian, we find a circular tower attached to St. Apollinare in Classe (Woodcut No. 412), and in the other churches of that place they seem even then to have been considered necessary adjuncts.[303] At the same time it is by no means clear that they were erected as bell-towers; indeed the evidence is tolerably clear that bells were not used in Christian churches till the time of Pope Adrian I., some two centuries later. What, then, were they? There is, I think, no trace of their being sepulchral monuments, or that they were designed or used as tombs; and unless they were, like the _sthambas_ of the Buddhists, pillars of victory, or towers erected to mark sacred or remarkable spots, it is difficult to say what they were, or where we are to look for an analogy.

Be this as it may, the oldest circular towers with which we are acquainted are those of Ravenna; while the last of the series is the famous leaning one at Pisa, commenced in the year 1174. The gradations between these two extremes must have been the same that marked the changes in the architecture of the churches to which they are attached; but the links are more completely wanting in the case of the towers than in that of the churches.[304]

[Illustration: 459. Tower of Sta. Maria in Cosmedin. (From Gutensohn and Knapp.)]

The tower of St. Apollinare in Classe, above referred to, the most perfect of those of Ravenna, is a simple brick tower (see Woodcut No. 412), nine storeys in height, the lower windows being narrow single openings; above there are two, and the three upper storeys are adorned with four windows of three lights each.

In Rome, as far as we know, the first tower attached to a church was that said to have been built by Pope Adrian I. in front of the atrium of St. Peter’s; but there are no examples now existing in Rome which can be said to be earlier than the 11th century, and that date applies only to the lower portion of them. In the 12th and 13th centuries they became common, and we find them attached to the churches of S. Lorenzo without the walls, S. Croce in Gerusalemme, SS. Giovanni e Paolo, S. Giorgio in Velabro (13th century), and others. All these are square in plan and extremely similar in design, no improvement and scarcely any change having taken place between the first and the last, as if the form were an old and established one when we find it first adopted. That attached to Sta. Maria in Cosmedin (Woodcut No. 459) is perhaps one of the best and most complete. Its dimensions are small, its breadth being little more than 15 ft., and its height only 110; but notwithstanding this there is great dignity in the design, and, in a city where buildings are not generally tall, its height is sufficient to give it prominence without overpowering other objects,—a characteristic which renders these Roman towers not only beautiful structures in themselves, but appropriate ornaments to the buildings to which they are attached.

The chief interest of these towers is derived from the numerous progeny to which they gave birth: for though there is scarcely an instance of a square Romanesque tower beyond the walls of Rome during the period in which this style flourished, the form was seized upon with avidity by the Gothic architects in all the countries of Europe; and whether as a detached campanile (as in Italy), or as an integral part of the building (as we soon find it employed on this side of the Alps), it forms the most prominent, and perhaps also the most beautiful, feature in the aspiring architecture of the Middle Ages.

There is certainly no architectural feature which the Gothic architects can so justly call their own as the towers and spires which in the Middle Ages were so favourite, so indispensable a part of their churches and other edifices, becoming in fact as necessary parts of the external design as the vaults were of the internal decoration of the building.

It is true, as before remarked, that we neither know where they were first invented, nor even where they were first applied to Christian churches—those of Rome and Ravenna being evidently not the earliest examples; nor have they any features which betray their origin—at least none have yet been pointed out, though it is not impossible that a closer examination would bring some such to light. They certainly are as little classical, in form or details, as anything that can well be conceived; and belong to an undefined Romanesque style.

Those of which we have already spoken are all church-towers—_campaniles_ or bell-towers attached to churches. But this exclusive distinction by no means applies to the Gothic towers. The tower of St. Mark at Venice, for instance, and the Toraccio at Cremona, are evidently civic monuments, like the belfries of the Low Countries—symbols of communal power wholly distinct from the church, their proximity to which seems only to arise from the fact of all the principal buildings being grouped together. This is certainly the case with a large class of very ugly buildings in Italy, such as those attached to the town-halls of Florence and Siena, or the famous Asinelli and Garisenda towers at Bologna. They are merely tall square brick towers, with a machicolated balcony at the top, but possessing no more architectural design than the chimney of a cotton factory. Originally, when lower, they may have been towers of defence, but afterwards became mere symbols of power.

A third class, and by far the most numerous, of these buildings are undoubtedly ecclesiastical erections; they are either actually attached to the churches, or so placed with regard to them as to leave no doubt on the matter. There is not, however, I believe, in all Italy a single example of a tower or towers forming, as on this side of the Alps, an integral part of the design.

Sometimes they stand detached, but more generally are connected with some angle of the building, the favourite position being the western angle of the southern transept. Occasionally we find one tower placed at the angle of the façade, but this is seldom the case when the tower and the church are of the same age. It is so in the cathedral at Lucca, and San Ambrogio at Milan; in the latter of which a second tower has been added more recently to balance the older one. It does also happen as in the instance of Novara, before quoted (Woodcut No. 443), that two towers are actually parts of the original design; this, however, is certainly the exception, not the rule.

In design the Italian campaniles differ very considerably from those on this side of the Alps. They never have projecting buttresses, nor assume that pyramidal form which is so essential and so beautiful a feature in the Northern examples. In plan the campanile is always square, and carried up without break or offset to two-thirds at least of its intended height. This, which is virtually the whole design (for the spire seems an idea borrowed from the North), is generally solid to a considerable height, or with only such openings as serve to admit light to the stairs or inclined planes. Above the solid part one round-headed window is introduced in each face, and in the next storey two; in the one above this three, then four, and lastly five, the lights being merely separated by slight shafts, so that the upper storey is virtually an open loggia (see Woodcut No. 498). There is no doubt great beauty and propriety of design in this arrangement; in point of taste it is unobjectionable, but it wants the vigour and variety of the Northern tower.

So far as we can judge from drawings and such ancient examples as remain, the original termination was a simple cone in the centre, with a smaller one at each of the angles.

At Verona an octagonal lantern is added, and at Modena and Cremona the octagon is crowned by a lofty spire, but these hardly come within the limits of the epoch of which we are now treating. So greatly did the Italians prefer the round arch, that even in their imitation of the Northern styles they used the pointed shape only when compelled—a circumstance which makes it extremely difficult, particularly in the towers, to draw the line between the two styles; for though pointed arches were no doubt introduced in the 13th and 14th centuries, the circular-headed shape continued to be employed from the age of the Romanesque to that of the Renaissance.

One of the oldest and certainly the most celebrated of the Gothic towers of Italy, is that of St. Mark’s at Venice, commenced in the year 902; it took the infant republic three centuries to raise it 180 ft., to the point at which the square basement terminates. On this there must originally have been an open loggia of some sort, no doubt with a conical roof. The present superstructure was added in the 16th century; but though the loggia is a very pleasing feature, it is overpowered by the solid mass that it surmounts, and by the extremely ugly square extinguisher that crowns the whole. Its locality and its associations have earned for it a great deal of undue laudation, but in point of design no campanile in Italy deserves it less. The base is a mere unornamented mass of brickwork, slightly fluted, and pierced unsymmetrically with small windows to light the inclined plane within. Its size, its height, and its apparent solidity are its only merits. These are no doubt important elements in that low class of architectural excellence of which the Egyptian pyramids are the type; but even in these elements this edifice must confess itself a pigmy, and inferior to even a second-class pyramid on the banks of the Nile, while it has none of the beauty of design and detail displayed by the Giralda of Seville, or even by other Italian towers in its own neighbourhood.

The campanile at Piacenza (Woodcut No. 448) is, perhaps, more like the original of St. Mark’s than any other, and certainly displays as little beauty as any building of this sort can possess.

That of San Zenone at Verona is far more pleasing. It is, indeed, as beautiful both in proportion and details as any of its age, while it exemplifies at once the beauties and the defects of the style. Among the first is an elegant simplicity that always is pleasing, but this is accompanied by a leanness and poverty of effect, when compared with Northern examples, which must rank in the latter category.

Mr. Jackson, in his work on Dalmatia and Istria, gives illustrations of several towers in those countries which, in beauty of design, excel many of the Italian examples. The Romanesque style would seem to have had a much longer duration on the east side of the Adriatic than in Italy. Thus the tower of Spalato, a lofty campanile of six storeys in height, commenced in the beginning of the 13th century and not terminated till 1416 (except the upper octagon and spire), is virtually in the same pure Romanesque style throughout. Mr. Jackson notes also the continued influence of Roman work of the 3rd century, by which it is surrounded, and that fragments of ancient material, columns and capitals, have been used up in its construction. The campaniles of Zara and in the island of Arbe are both fine examples of Romanesque design.

##