CHAPTER III

.

ROMAN ARCHITECTURE.

CONTENTS.

Origin of style—The arch—Orders: Doric, Ionic, Corinthian, Composite— Temples—The Pantheon—Roman temples at Athens—at Baalbec.

CHRONOLOGICAL MEMORANDA.

DATES.

Foundation of Rome B.C. 753

Tarquinius Priscus—Cloaca Maxima, foundation of Temple of 616 Jupiter Capitolinus.

Temple of Jupiter Capitolinus dedicated 507

Scipio—tomb at Literium 184

Augustus—temples at Rome 31

Marcellus—theatre at Rome—died 23

Agrippa—portico of Pantheon—died 13

Nero—burning and rebuilding of Rome—died A.D. 68

Vespasian—Flavian amphitheatre built 70

Titus—arch in Forum 79

Destruction of Pompeii 79

Trajan—Ulpian Basilica and Pillar of Victory 98

Hadrian builds temple at Rome, Temple of Jupiter Olympius 117 at Athens, &c.

Septimius Severus—arch at Rome 194

Caracalla—baths 211

Diocletian—palace at Spalato 284

Maxentius—Basilica at Rome 306

Constantine—transfer of Empire to Constantinople 328

The earliest inhabitants of Rome were an Aryan or, as they used to be called, Indo-Germanic race, who established themselves in a country previously occupied by Pelasgians. Their principal neighbour on one side was Etruria, a Pelasgian nation. On the other hand was Magna Græcia, which had been colonised in very early ages by Hellenic settlers of kindred origin. It was therefore impossible that the architecture of the Romans should not be in fact a mixture of the styles of these two people. As a transition order, it was only a mechanical juxtaposition of both styles, the real fusion taking place many long centuries afterwards. Throughout the Roman period the two styles remain distinct, and there is no great difficulty in referring almost every feature in Roman architecture to its origin.

From the Greeks were borrowed the rectangular peristylar temple, with its columns and horizontal architraves, though they seldom if ever used it in its perfect purity, the cella of the Greek temples not being sufficiently large for their purposes. The principal Etruscan temples, as we have already shown, were square in plan, and the inner half occupied by one or more cells, to the sides and back of which the portico never extended. The Roman rectangular temple is a mixture of these two: it is generally, like the Greek examples, longer than its breadth, but the colonnade never seems to have entirely surrounded the building. Sometimes it extends to the two sides as well as the front, but more generally the cella occupies the whole of the inner part though frequently ornamented by a false peristyle of three-quarter columns attached to its walls.

Besides this, the Romans borrowed from the Etruscans or Greeks a circular form of temple. As applied by the Romans it was generally encircled by a peristyle of columns, though it is not clear that the Etruscans so used it; this may therefore be an improvement adopted from the Greeks on an Etruscan form. In early times these circular temples were dedicated to Vesta, Cybele, or some god or goddess either unknown or not generally worshipped by the Aryan races; but in later times this distinction was lost sight of.

A more important characteristic which the Romans borrowed from the Etruscans was the circular arch. It was known, it is true, to the Egyptians, Assyrians, and Greeks; yet none of these people, perhaps excepting the Assyrians, seem to have used it as a feature in their ornamental architecture; but the Etruscans appear to have had a peculiar predilection for it, and from them the Romans adopted it boldly, and introduced it into almost all their buildings. It was not at first used in temples of Grecian form, nor even in their peristylar circular ones. In the civil buildings of the Romans it was a universal feature, but was generally placed in juxtaposition with the Grecian orders. In the Colosseum, for instance, the whole construction is arched; but a useless network of ill-designed and ill-arranged Grecian columns, with their entablatures, is spread over the whole. This is a curious instance of the mixture of the two styles, and as such is very characteristic of Roman art; but in an artistic point of view the place of these columns would have been far better supplied by buttresses or panels, or some expedient more correctly constructive.

After having thoroughly familiarised themselves with the forms of the arch as an architectural feature, the Romans made a bold stride in advance by applying it as a vault both to the circular and rectangular forms of buildings. The most perfect examples of this are the rotunda of the Pantheon and the basilica of Maxentius, commonly called the Temple of Peace, strangely like each other in conception, though apparently so distant in date. In these buildings the Roman architects so completely emancipated themselves from the trammels of former styles as almost to entitle them to claim the invention of a new order of architecture. It would have required some more practice to invent details appropriate to the purpose; still these two buildings are to this hour unsurpassed for boldness of conception and just appreciation of the manner in which the new method ought to be applied. This is almost universally acknowledged so far as the interior of the Pantheon is concerned. In simple grandeur it is as yet unequalled; its faults being principally those of detail. It is not so easy, however, to form an opinion of the Temple of Peace in its present ruined state; but in so far as we can judge from what yet remains of it, in boldness and majesty of conception it must have been quite equal to the other example, though it must have required far more familiarity with the style adopted to manage its design as appropriately as the simpler dome of the Pantheon.

These two buildings may be considered as exemplifying the extent to which the Romans had progressed in the invention of a new style of architecture and the state in which they left it to their successors. It may however be worth while pointing out how, in transplanting Roman architecture to their new capital on the shores of the Bosphorus, the semi-Oriental nation seized on its own circular form, and, modifying and moulding it to its purpose, wrought out the Byzantine style; in which the dome is the great feature, almost to the total exclusion of the rectangular form with its intersecting vaults. On the other hand, the rectangular form was appropriated by the nations of the West with an equally distinct rejection of the circular and domical forms, except in those cases in which we find an Eastern people still incorporated with them. Thus in Italy both styles continued long in use, the one in baptisteries, the other in churches, but always kept distinct, as in Rome. In France they were so completely fused into each other that it requires considerable knowledge of architectural analysis to separate them again into their component parts. In England we rejected the circular form altogether, and so they did eventually in Germany, except when under French influence. Each race reclaimed its own among the spoils of Rome, and used it with the improvements it had acquired during its employment in the Imperial city.

ORDERS.

The first thing that strikes the student in attempting to classify the numerous examples of Roman architecture is the immense variety of purposes to which it is applied, as compared with previous styles. In Egypt architecture was applied only to temples, palaces and tombs. In Greece it was almost wholly confined to temples and theatres; and in Etruria to tombs. It is in Rome that we first feel that we have not to deal with either a Theocracy or a kingdom, but with a great people, who for the first time in the world’s history rendered architecture subservient to the myriad wants of the many-headed monster. It thus happens that in the Roman cities, in addition to temples we find basilicas, theatres and amphitheatres, baths, palaces, tombs, arches of triumph and pillars of victory, gates, bridges, and aqueducts, all equally objects of architectural skill. The best of these, in fact, are those which from previous neglect in other countries are here stamped with originality. These would have been noble works indeed had it not been that the Romans unsuccessfully applied to them those orders and details of architecture which were intended only to be applied to temples by other nations. In the time of Constantine these orders had nearly died out, and were only subordinately used for decorative purposes. In a little while they would have died out altogether, and the Roman would have become a new and complete style; but, as before remarked, this did not take place, and the most ancient orders therefore still remain an essential part of Roman art. We find the old orders predominating in the age of Augustus, and see them gradually die out as we approach that of Constantine.

DORIC.

Adopting the usual classification, the first of the Roman orders is the Doric, which, like everything else in this style, takes a place about half-way between the Tuscan wooden posts and the nobly simple order of the Greeks. It no doubt was a great improvement on the former, but for monumental purposes infinitely inferior to the latter. It was, however, more manageable; and for forums or courtyards, or as a three-quarter column between arcades, it was better adapted than the severer Greek style, which, when so employed, not only loses almost all its beauty, but becomes more unmeaning than the Roman. This fact was apparently recognised; for there is not, so far as is known, a single Doric temple throughout the Roman world. It would in consequence be most unfair to institute a comparison between a mere utilitarian prop used only in civil buildings and an order which the most refined artists in the world spent all their ingenuity in rendering the most perfect, because it was devoted to the highest religious purposes.

[Illustration: 179. Doric Order.]

The addition of an independent base made the order much more generally useful, and its adoption brought it much more into harmony with the other two existing orders, which would appear to have been the principal object of its introduction. The keynote of Roman architecture was the Corinthian order; and as, from the necessities of their tall, many-storeyed buildings, the Romans were forced to use the three orders together, often one over the other, it was indispensable that the three should be reduced to something like harmony. This was accordingly done, but at the expense of the Doric order, which, except when thus used in combination, must be confessed to have very little claim to our admiration.

IONIC.

The Romans were much more unfortunate in their modifications of the Ionic order than in those which they introduced into the Doric. They never seem to have either liked or understood it, nor to have employed it except as a _mezzo termine_ between the other two. In its own native East this order had originally only been used in porticoes between piers or _antæ_, where of course only one face was shown, and there were no angles to be turned. When the Greeks adopted it they used it in temples of Doric form, and in consequence were obliged to introduce a capital at each angle, with two voluted faces in juxtaposition at right angles to one another. In some instances—internally at least—as at Bassæ (Woodcut No. 142) they used a capital with four faces. The Romans, impatient of control, eagerly seized on this modification, but never quite got over the extreme difficulty of its employment. With them the angular volutes became mere horns, and even in the best examples the capital wants harmony and meaning.

[Illustration: 180. Ionic Order.]

When used as a three-quarter column these alterations were not required, and then the order resembled more its original form; but even in this state it was never equal to the Greek examples, and gradually deteriorated to the corrupt application of it in the Temple of Concord in the Forum, which is the most degenerate example of the order now to be found in Roman remains.

CORINTHIAN.

The fate of this order in the hands of the Romans was different from that of the other two. The Doric and Ionic orders had reached their acme of perfection in the hands of the Grecian artists, and seem to have become incapable of further improvement. The Corinthian, on the contrary, was a recent conception; and although nothing can surpass the elegance and grace with which the Greeks adorned it, the new capital never acquired with them that fulness and strength so requisite to render it an appropriate architectural ornament. These were added to it by the Romans, or rather perhaps by Grecian artists acting under their direction, who thus, as shown in Woodcut No. 181, produced an order which for richness combined with proportion and architectural fitness has hardly been surpassed. The base is elegant and appropriate; the shaft is of the most pleasing proportion, and the fluting gives it just the requisite degree of richness and no more; while the capital, though bordering on over-ornamentation, is so well arranged as to appear just suited to the work it has to do. The acanthus-leaves, it is true, approach the very verge of that degree of direct imitation of nature which, though allowable in architectural ornaments, is seldom advisable; they are, however, disposed so formally, and there still remains so much that is conventional in them, that, though perhaps not justly open to criticism on this account, they are nevertheless a very extreme example.

[Illustration: 181. Corinthian Order. From the Temple of Jupiter Stator.]

The entablature is not so admirable as the column. The architrave is too richly carved. It is evident, however, that this arose from the artist having copied in carving what the Greeks had only painted, and thereby produced a complexity far from pleasing.

The frieze, as we now find it, is perfectly plain; but this undoubtedly was not the case when originally erected. It either must have been painted (in which case the whole order of course was also painted), or ornamented with scrolls or figures in bronze, which may probably have been gilt.

The cornice is perhaps open to the same criticism as the architrave, of being over-rich, though this evidently arose from the same cause, viz., reproducing in carving what was originally only painted; which to our Northern eyes at least appears more appropriate for internal than for external decoration, though, under the purer skies where it was introduced and used, this remark may be hardly applicable.

The order of the portico of the Pantheon is, according to our notions, a nobler specimen of what an external pillar should be than that of the Temple of Jupiter Stator. The shafts are of one block, unfluted; the capital plainer; and the whole entablature, though as correctly proportional, is far less ornamented and more suited to the greater simplicity of the whole.

The order of the Temple of Antoninus and Faustina is another example intermediate between these two. The columns are in this instance very similar to those of the Pantheon, and the architrave is plain. The frieze, however, is ornamented with more taste than any other in Rome, and is a very pleasing example of those conventional representations of plants and animals which are so well suited to architectural purposes— more like Nature than those of the Greeks, but still avoiding direct imitation sufficiently to escape the affectation of pretending to appear what it is not and cannot be.

The Maison Carrée at Nîmes presents an example of a frieze ornamented with exquisite taste, while at Baalbec, and in some other examples, we have them so over-ornamented that the effect is far more offensive, from utter want of repose, than the frieze in the Temple of Jupiter Stator ever could be from its baldness.

Besides these there are at least fifty varieties of Corinthian capitals to be found, either in Rome or in various parts of the Roman Empire, all executed within the three centuries during which Rome continued to be the imperial city. Some of them are remarkable for that elegant simplicity which so evidently betrays the hand of a Grecian artist, while others again show a lavish exuberance of ornament which is but too characteristic of Roman art in general. Many, however, contain the germs of something better than was accomplished in that age; and a collection of them would afford more useful suggestions for designing capitals than have yet been available to modern artists.

COMPOSITE ORDER.

Among their various attempts to improve the order which has just been described, the Romans hit upon one which is extremely characteristic of their whole style of art. This is known by the distinguishing name of the Composite order, though virtually more like the typical examples of the Corinthian order than many of those classed under the latter denomination.

The greatest defect of the Corinthian capital is the weakness of the small volutes supporting the angles of the abacus. A true artist would have remedied this by adding to their strength and carrying up the fulness of the capital to the top. The Romans removed the whole of the upper part and substituted an Ionic capital instead. Their only original idea, if it may be so called, in art was that of putting two dissimilar things together to make one which should combine the beauties of both, though as a rule the one generally serves to destroy the other. In the Composite capital they never could hide the junction; and consequently, though rich, and in some respects an improvement on the order out of which it grew, this capital never came into general use, and has seldom found favour except amongst the blindest admirers of all that the Romans did.

[Illustration: 182. Composite Order.]

[Illustration: 183. Corinthian Base, found in Church of St. Praxede in Rome.]

In the latter days of the Empire the Romans attempted another innovation which promised far better success, and with very little more elaboration would have been a great gain to the principles of architectural design. This was the introduction of the Persian or Assyrian base, modified to suit the details of the Corinthian or Composite orders. If they had always used this instead of the square pedestals on which they mounted their columns, and had attenuated the pillars slightly when used with arcades, they would have avoided many of the errors they fell into. This application, however, came too late to be generally used; and the forms already introduced continued to prevail. At the same time it is evident that a Persepolitan base for an Ionic and even for a Corinthian column would be amongst the greatest improvements that could now be introduced, especially for internal architecture.

COMPOSITE ARCADES.

The true Roman order, however, was not any of these columnar ordinances we have been enumerating, but an arrangement of two pillars placed at a distance from one another nearly equal to their own height, and having a very long entablature, which in consequence required to be supported in the centre by an arch springing from piers. This, as will be seen from the annexed woodcut, was in fact merely a screen of Grecian architecture placed in front of a construction of Etruscan design. Though not without a certain richness of effect, still, as used by the Romans, these two systems remain too distinctly dissimilar for the result to be pleasing, and their use necessitated certain supplemental arrangements by no means agreeable. In the first place, the columns had to be mounted on pedestals, or otherwise an entablature proportional to their size would have been too heavy and too important for a thing so useless and so avowedly a mere ornament. A projecting keystone was also introduced into the arch. This was unobjectionable in itself, but when projecting so far as to do the duty of an intermediate capital, it overpowered the arch without being equal to the work required of it.

[Illustration: 184. Doric Arcade.]

The Romans used these arcades with all the 3 orders, frequently one over the other, and tried various expedients to harmonise the construction with the ornamentation, but without much effect. They seem always to have felt the discordance as a blemish, and at last got rid of it, but whether they did so in the best way is not quite clear. The most obvious mode of effecting this would no doubt have been by omitting the pillars altogether, bending the architrave, as is usually done, round the arch, and then inserting the frieze and cornices into the wall, using them as a string-course. A slight degree of practice would soon have enabled them—by panelling the pier, cutting off its angles, or some such expedient—to have obtained the degree of lightness or of ornament they required, and so really to have invented a new order.

This, however, was not the course that the Romans pursued. What they did was to remove the pier altogether, and to substitute for it the pillar taken down from its pedestal. This of course was not effected at once, but was the result of many trials and expedients. One of the earliest of these is observed in the Ionic Temple of Concord before alluded to, in which a concealed arch is thrown from the head of each pillar, but above the entablature, so as to take the whole weight of the superstructure from off the cornice between the pillars. When once this was done it was perceived that so deep an entablature was no longer required, and that it might be either wholly omitted, as was sometimes done in the centre intercolumniation, or very much reduced. There is an old temple at Talavera in Spain, which is a good example of the former expedient; and the Roman gateway at Damascus is a remarkable instance of the latter. There the architrave, frieze and cornice are carried across in the form of an arch from pier to pier, thus constituting a new feature in architectural design.

[Illustration: 185. View in Courtyard of Palace at Spalato]

In Diocletian’s reign we find all these changes already introduced into domestic architecture, as shown in Woodcut No. 185, representing the great court of his palace at Spalato, where, at one end, the entablature is bent into the form of an arch over the central intercolumniation, while on each side of the court the arches spring directly from the capitals of the columns.

Had the Romans at this period been more desirous to improve their external architecture, there is little doubt that they would have adopted the expedient of omitting the entire entablature: but at this time almost all their efforts were devoted to internal improvement, and not unfrequently at the expense of the exterior. Indeed the whole history of Roman art, from the time of Augustus to that of Constantine, is a transition from the external architecture of the Greeks to the internal embellishment of the Christians. At first we see the cells of the temple gradually enlarged at the expense of the peristyle, and finally, in some instances, entirely overpowering them. Their basilicas and halls become more important than their porticoes, and the exterior is in almost every instance sacrificed to internal arrangements. For an interior, an arch resting on a circular column is obviously far more appropriate than one resting on a pier. Externally, on the contrary, the square pier is most suitable, because a pillar cannot support a wall of sufficient thickness. This defect was not remedied until the Gothic architects devised the plan of coupling two or more pillars together; but this point had not been reached at the time when with the fall of Rome all progress in art was effectually checked for a time.

TEMPLES.

There is perhaps nothing that strikes the inquirer into the architectural history of Rome more than the extreme insignificance of her temples, as compared with the other buildings of the imperial city and with some contemporary temples found in the provinces. The only temple which remains at all worthy of such a capital is the Pantheon. All others are now mere fragments, from which we can with difficulty restore even the plans of the buildings, far less judge of their effect. We have now no means of forming an opinion of the great national temple of the Capitoline Jove, no trace of it, nor any intelligible description, having been preserved to the present time. Its having been of Etruscan origin, and retaining its original form to the latest day, would lead us to suppose that the temple itself was small, and that its magnificence, if any, was confined to the enclosure and to the substructure, which may have been immense.

Of the Augustan age we have nothing but the remains of three temples, each consisting of only three columns; and the excavations that have been made around them have not sufficed to make even their plans tolerably clear.

The most remarkable was that of Jupiter Stator in the Forum, the beautiful details of which have been already alluded to and described. This temple was octastyle in front. It was raised on a stylobate 22 ft. in height, the extreme width of which was 98 ft., and this corresponds as closely as possible with 100 Roman ft. The angular columns were 85 ft. from centre to centre. The height of the pillars was 48 ft., and that of the entablature 12 ft. 6 in.[165] It is probable that the whole height to the apex of the pediment was nearly equal to the extreme width, and that it was designed to be so.

The pillars certainly extended on both flanks, and the temple is generally restored as peristylar, but apparently without any authority. From the analogy of the other temples it seems more probable that there were not more than eight or ten pillars on each side, and that the apse of the cella formed the termination opposite the portico.

The temple nearest to this in situation and style is that of Jupiter Tonans.[166] The order in this instance is of slightly inferior dimensions to that of the temple just described, and of very inferior execution. The temple, too, was very much smaller, having only six columns in front, and from its situation it could not well have had more than that number on the flanks, so that its extreme dimensions were probably about 70 ft. by 85.

The third is the Temple of Mars Ultor, of which a plan is annexed; for though now as completely decayed as the other two, in the time of Ant. Sabacco and Palladio there seem to have been sufficient remains to justify an attempt at restoration. As will be seen, it is nearly square in plan (112 ft. by 120). The cella is here a much more important part than is usual in Greek temples, and terminates in an apse, which afterwards became characteristic of all places of worship. Behind the cella, and on each side, was a lofty screen of walls and arches, part of which still remain, and form quite a new adjunct, unlike anything hitherto met with attached to any temple now known.

[Illustration: 186. Temple of Mars Ultor. (From Cresy’s ‘Rome.’) Scale 100 ft. to 1 in.]

The next class of temples, called pseudo-peripteral (or those in which the cella occupies the whole of the after part), are generally more modern, certainly more completely Roman, than these last. One of the best specimens at Rome is the Temple of Antoninus and Faustina, a small building measuring 72 ft. by 120. There is also a very elegant little Ionic temple of this class called that of Fortuna Virilis; while the Ionic Temple of Concord, built by Vespasian, and above alluded to, appears also to have been of this class. So was the temple in the forum at Pompeii; but the finest specimen now remaining to us is the so-called Maison Carrée at Nîmes, which is indeed one of the most elegant temples of the Roman world, owing probably a great deal of its beauty to the taste of the Grecian colonists long settled in its neighbourhood. It is hexastyle, with 11 columns in the flanks, 3 of which stand free, and belong to the portico; the remaining 8 are attached to the walls of the cella. The temple is small, only 45 ft. by 85; but such is the beauty of its proportions and the elegance of its details that it strikes every beholder with admiration.

[Illustration: 187. Plan of Maison Carrée at Nîmes. Scale 100 ft. to 1 in.]

The date of this temple has not been satisfactorily ascertained. From the nail-holes of the inscription on the frieze it has been attempted to make out the names of Caius and Lucius Cæsar, and there is nothing in the style of its architecture to contradict this hypothesis. Even if the buildings in the capital were such as to render this date ambiguous, it would scarcely be safe to apply any argument derived from them to a provincial example erected in the midst of a Grecian colony. But for their evidence we might almost be inclined to fancy its style represented the age of Trajan.

The temple of Diana in the same city is another edifice of singular beauty of detail, and interesting from the peculiarity of its plan. Exclusive of the portico it is nearly square, 70 ft. by 65, and consists of a cella which is covered with a stone ribbed vault, the thrust of which is counteracted by smaller vaults thrown across two side passages or aisles which are, however, not thrown open to the cella. The columns in the cella are detached from the wall, which is singularly interesting as the origin of much which we find afterwards in Gothic work. (A somewhat similar arrangement is found in the small temple at Baalbec (Woodcut No. 197) where, however, the peristyle occupies the position and serves the same purpose as the aisles at Nîmes, viz., to resist the thrust of the vault over the cella.)

[Illustration: 188. Plan of Temple of Diana at Nîmes. Scale 100 ft. to 1 in.]

Throughout this building the details of the architecture are unsurpassed for variety and elegance by anything found in the metropolis, and are applied here with a freedom and elegance bespeaking the presence of a Grecian mind even in this remote corner of the Empire. Another interesting feature is the porch. This was supported by four slender columns of singularly elegant design, but placed so widely apart that they could not have carried a stone entablature. It is difficult to guess what could have been the form of the wooden ones; but a mortice which still exists in the walls of the temple shows that it must have been eight or ten feet deep, and therefore probably of Etruscan form (Woodcut No. 167); though it may have assumed a circular arched form between the pillars.[167]

[Illustration: 189. View of the Interior of the Temple of Diana at Nîmes. (From Laborde.)]

Another peculiarity is, that the light was introduced over the portico by a great semicircular window, as is done in the Buddhist caves in India; which, so far as I know, is the most perfect mode of lighting the interior of a temple which has yet been discovered.

Not far from the Colosseum, in the direction of the Forum, are still to be seen the remains of a great double temple built by the Emperor Hadrian, and dedicated to Venus and Rome, and consisting of the ruins of its two cells, each about 70 ft. square, covered with tunnel-vaults, and placed back to back, so that their apses touch one another. These stand on a platform 480 ft. long by 330 wide; and it is generally supposed that on the edge of this once stood 56 great columns, 65 ft. in height, thus moulding the whole into one great peripteral temple. Some fragments of such pillars are said to be found in the neighbourhood, but not one is now erect,—not even a base is in its place,—nor can any of its columns be traced to any other buildings. This part, therefore, of the arrangement is very problematical, and I should be rather inclined to restore it, as Palladio and the older architects have done, with a corridor of ten small columns in front of each of the cells. If we could assume the plan of this temple to have been really peripteral, as supposed, it must have been a building worthy of the imperial city and of the magnificence of the emperor to whom its erection is ascribed.

More perfect and more interesting than any of these is the Pantheon, which is undoubtedly one of the finest temples of the ancient world. Externally its effect is very much destroyed by its two parts, the circular and the rectangular, being so dissimilar in style and so incongruously joined together. The portico especially, in itself the finest which Rome exhibits, is very much injured by being prefixed to a mass which overpowers it and does not harmonise with any of its lines. The pitch, too, of its pediment is perhaps somewhat too high, but, notwithstanding all this, its sixteen columns, the shaft of each composed of a single block, and the simple grandeur of the details, render it perhaps the most satisfactory example of its class.

[Illustration: 190. Plan of Pantheon at Rome. Scale 100 ft. to 1 in.]

The pillars are arranged in the Etruscan fashion, as they were originally disposed in front of three-celled temples. As they now stand, however, they are added unsymmetrically to a rotunda, and in so clumsy a fashion that the two are certainly not part of the same design and do not belong to the same age. Either it was that the portico was added to the pre-existing rotunda, or that the rotunda is long subsequent to the portico. Unfortunately the two inscriptions on the portico hardly help to a solution of the difficulty. The principal one states that it was built by M. Agrippa, but the “it” may refer to the rotunda only, and may have been put there by those who in the time of Aurelius[168] repaired the temple which had “fallen into decay from age.” This hardly could, under any circumstances, be predicated of the rotunda, which shows no sign of decay during the last seventeen centuries of ill-treatment and neglect, and may last for as many more without injury to its stability, but might be said of a portico which, if of wood, as Etruscan porticoes usually were, may easily in 200 years have required repairs and rebuilding. From a more careful examination on the spot, I am convinced that the portico was added at some subsequent period to the rotunda. If by Agrippa, then the dome must belong to Republican times; if by Severus it may have been, as is generally supposed, the hall of the Baths of Agrippa.[169] Altogether I know of no building whose date and arrangements are so singular and so exceptional as this. Though it is, and always must have been, one of the most prominent buildings in Rome, and most important from its size and design, I know of no other building in Rome whose date or original destination it is so difficult to determine.

[Illustration: 191. Half Elevation, half Section, of the Pantheon at Rome. Scale 50 ft. to 1 in.]

Internally perhaps the greatest defect of the building is a want of height in the perpendicular part, which the dome appears to overpower and crush. This mistake is aggravated by the lower part being cut up into two storeys, an attic being placed over the lower order. The former defect may have arisen from the architect wishing to keep the walls in some proportion to the portico. The latter is a peculiarity of the age in which I suppose this temple to have been remodelled, when two or more storeys seem to have become indispensable requisites of architectural design. We must ascribe also to the practice of the age the method of cutting through the entablature by the arches of the great niches, as shown in the sectional part of the last woodcut. It has already been pointed out that this was becoming a characteristic of the style at the time when the circular part of this temple was arranged as it at present appears.

Notwithstanding these defects and many others of detail that might be mentioned, there is a grandeur and a simplicity in the proportions of this great temple that render it still one of the very finest and most sublime interiors in the world, and the dimensions of its dome, 145 ft. 6 in. span by 147 in height, have not yet been surpassed by any subsequent erection. Though it is deprived of its bronze covering[170] and of the greater part of those ornaments on which it mainly depended for effect, and though these have been replaced by tawdry and incongruous modernisms, still nothing can destroy the effect of a design so vast and of a form so simply grand. It possesses moreover one other element of architectural sublimity in having a single window, and that placed high up in the building. I know of no other temples which possess this feature except the great rock-cut Buddhist basilicas of India. In them the light is introduced even more artistically than here; but, nevertheless, that one great eye opening upon heaven is by far the noblest conception for lighting a building to be found in Europe.

Besides this great rotunda there are two other circular temples in or near Rome. The one at Tivoli, shown in plan and elevation in the annexed woodcuts (Nos. 192 and 193), has long been known and admired; the other, near the mouth of the Cloaca Maxima, has a cell surrounded by twenty Corinthian columns of singularly slender proportions. Both these probably stand on Etruscan sites; they certainly are Etruscan in form, and are very likely sacred to Pelasgic deities, either Vesta or Cybele.

[Illustration: 192. Plan of Temple at Tivoli. Scale 100 ft. to 1 in.]

[Illustration: 193. Restored Elevation of Temple at Tivoli. Scale 50 ft. to 1 in.]

Both in dimensions and design they form a perfect contrast to the Pantheon, as might be expected from their both belonging to the Augustan age of art: consequently the cella is small, its interior is unornamented, and all the art and expense is lavished on the external features, especially on the peristyle; showing more strongly than even the rectangular temple the still remaining predominance of Grecian taste, which was gradually dying out during the whole period of the Empire.

It is to be regretted that the exact dates of both these temples are unknown, for, as that at Tivoli shows the stoutest example of a Corinthian column known and that in Rome the slenderest, it might lead to some important deductions if we could be certain which was the older of the two. It may be, however, that this difference of style has no connection with the relative age of the two buildings, but that it is merely an instance of the good taste of the age to which they belong. The Roman example, being placed in a low and flat situation, required all the height that could be given it; that at Tivoli, being placed on the edge of a rock, required as much solidity as the order would admit of to prevent its looking poor and insecure. A Gothic or a Greek architect would certainly have made this distinction.

One more step towards the modern style of round temples was taken before the fall of the Western Empire, in the temple which Diocletian built in his palace at Spalato. Internally the temple is circular, 28 ft. in diameter, and the height of the perpendicular part to the springing of the dome is about equal to its width. This is a much more pleasing proportion than we find in the Pantheon; perhaps the very best that has yet been employed. Externally the building is an octagon, surrounded by a low dwarf peristyle, very unlike that employed in the older examples. This angularity is certainly a great improvement, giving expression and character to the building, and affording flat faces for the entrances or porches; but the peristyle is too low, and mars the dignity of the whole.[171]

[Illustration: 194. Plan and Elevation of Temple in Diocletian’s Palace at Spalato. Scale for Plan 100 ft. to 1 in.; for Elevation 50 ft. to 1 in.]

To us its principal interest consists in its being so extremely similar to the Christian baptisteries which were erected in the following centuries, and which were copies, but very slightly altered, from buildings of this class.

ATHENS.

Even assuming that Hadrian completed the great Temple of Venus at Rome in the manner generally supposed, it must have been very far surpassed by the great Temple of Jupiter Olympius at Athens, which, though probably not entirely erected, was certainly finished, by that Emperor. It was octastyle in front,[172] with a double range of 20 columns on each flank so that it could not well have had less than 106 columns, all about 58 ft. in height, and of the most elegant Corinthian order, presenting altogether a group of far greater magnificence than any other temple we are acquainted with of its class in the ancient world. Its lineal dimensions also, as may be seen from the plan (Woodcut No. 195), were only rivalled by the two great Sicilian temples at Selinus and Agrigentum (Woodcuts Nos. 151, 152). It was 135 ft. wide by 354 in length, or nearly the same dimensions as the great Hypostyle Hall at Karnac, from which, however, it differs most materially, that being a beautiful example of an interior, this depending for all its magnificence on the external arrangement of its columns. Mr. Penrose’s discoveries in 1884 show that there was an opisthodomus at the rear and a vestibule or court in front of the cella which may have been hypæthral so as to admit light into the interior. This arrangement became so common in the early Christian world that there must have been some precedent for it; which, in addition to other reasons,[173] strongly inclines me to believe that the arrangement shown in the plan is correct.

[Illustration: 195. Ruins of the Temple of Jupiter Olympius at Athens.]

[Illustration: 196. Plan of Temple of Jupiter Olympius at Athens.]

BAALBEC.

The temples of Palmyra and Kangovar have been already mentioned in speaking of that of Jerusalem, to which class they seem to belong in their general arrangements, though their details are borrowed from Roman architecture. This, however, is not the case with the temples at Baalbec, which taken together and with their accompaniments, form the most magnificent temple group now left to us of their class and age. The great temple, if completed (which, however, probably it never was), would have been about 160 ft. by 290, and therefore, as a Corinthian temple, only inferior to that of Jupiter Olympius at Athens. Only nine of its colossal columns are now standing, but the bases of most of the others are _in situ_. Scarcely less magnificent than the temple itself was the court in which it stood, above 380 ft. square, and surrounded on three sides by recessed porticoes of most exuberant richness, though in perhaps rather questionable taste. In front of this was a hexagonal court of very great beauty, with a noble portico of 12 Corinthian columns, with two square blocks of masonry at each end. The whole extent of the portico is 260 ft., and of its kind it is perhaps unrivalled, certainly among the buildings of so late a date as the period to which it belongs.

[Illustration: 197. Plan of Small Temple at Baalbec. Scale 100 ft. to 1 in.]

[Illustration: 198. Elevation of Small Temple at Baalbec. Scale 50 ft. to 1 in.]

The other, or smaller temple, stands close to the larger. Its dimensions, to the usual scale, are shown in the plan (Woodcut No. 197). It is larger than any of the Roman peripteral temples, being 117 ft. by 227 ft., or rather exceeding the dimensions of the Parthenon at Athens, and its portico is both wider and higher than that of the Pantheon at Rome. Had this portico been applied to that building, the slope of its pediment would have coincided exactly with that of the upper sloping cornice, and would have been the greatest possible improvement to that edifice. As it is, it certainly is the best proportioned and the most graceful Roman portico of the first class that remains to us in a state of sufficient completeness to allow us to judge of its effect.

The interior of the cella was richly ornamented with niches and pilasters, and covered with a ribbed and coffered vault, remarkable, like every part of this edifice, rather for the profusion than for the good taste of its ornaments.

One of the principal peculiarities of this group of buildings is the immense size of some of the stones used in the substructure of the great temple: three of these average about 63 ft. in length, 10 ft. 5 in. in breadth, and 13 ft. in height. A fourth, of similar dimensions, is lying in the quarry, which it is calculated must weigh alone more than 1100 tons in its rough state, or nearly as much as one of the tubes of the Britannia Bridge. It is not easy to see why such masses were employed. If they had been used as foundation stones their use would have been apparent, but they are placed over several courses of smaller stones, about half-way up the terrace wall, as mere binding stones, apparently for show. It is true that in many places in the Bible and in Josephus nothing is so much insisted upon as the immense size of the stones used in the building of the Temple and the walls of Jerusalem, the bulk of the materials used appearing to have been thought a matter of far more importance than the architecture. It probably was some such feeling as this which led to their employment here, though, had these huge stones been set upright, as the Egyptians would have placed them, we might more easily have understood why so great an expense should have been incurred on their account. As it is, there seems no reason for doubting their being of the same age as the temples they support, though their use is certainly exceptional in Roman temples of this class.

##