Temples

Under this caption I have grouped all the ruins of buildings in Bosra which are probably to be classed as religious. They were, for the most part, columnar structures, and are three in number. The first I have called the Temple. It occupies the crest of the high ground in the eastern part of the city, which we may call the "acropolis", and consists of a section of heavy wall, a standing column, and many fragments of details; but these various features are so far separated from each other, and the site is so encumbered with the dwellings of the present inhabitants of the town, that it is impossible to discover if they did not originally belong to more than one structure. The second building of this class I have called the Nymphaeum. It is now represented by the four beautiful Corinthian columns that still stand at the intersection of the two principal streets. To the third I have given the name of Kalybe because of its resemblance to a class of open temples found in the Haurân, and given this name by M. de Vogue who discovered the term in a Greek inscription upon one of them. The first of these buildings is presented here with measurements of all the parts that I was able to find; the second and third are not fully, or satisfactorily, published herewith ; for only the ground plans and one or two of the details are represented by measured drawings, and I feel that some explanation of these omissions is due to the readers of this Part. It seemed inadvisable to duplicate the work of the more recent competent scholars at a site like Bosra where there was so much to be done, and where there were so many buildings that had never been published, or even noticed. Professor Briinnow had been in Bosra in 1898, and had sent me proofs of his drawings of the Theatre, the West Gate, and the Central Arch, which showed that these buildings were to be well published in his Provincia Arabia. Professor Puchstein had been here in 1902, only two years before us: his report1 stated that two clays had been spent in work upon the Theatre, the two Nymphaea, and the "Great Triumphal Arch". A few days before leaving Jerusalem on this expedition, I had received a letter from Professor Puchstein explaining that the two Nymphaea of the report had reference to the buildings which I have called the Nymphaeum and the Kalybe respectively, and assuring me that I need not take time for the study of these building's, since he had all the data for a complete publication of both, having had permission from the Turkish Government for excavations necessary to the taking of measurements. I at once determined to depend upon the work of these scholars in case I decided to publish a corpus of the monuments of Bosra, and devoted my time to the study of buildings that had not been satisfactorily published, or not published at all, taking only such measurements of the ground-plans of these buildings that had been studied, as would be required by Mr. Norris for the plan of the city, and sketching and measuring a few details for my own pleasure. Professor Brunnow's work has been extensively used in my presentation of two of the monuments described above, and I shall depend upon it wholly for all the plans of the Theatre which I shall present; but the untimely death of Professor Puchstein has greatly delayed the publication of the material which he collected in Syria, and which, but for his death, would have been published before now in his own masterly style. I have been unable by correspondence to secure any information from his notes, and I am therefore obliged, since I wish to include all the buildings of Bosra in this publication, to present these two buildings, by means of such data as I have at hand, <that is, by photographs, by measured ground-plans which were made for the survey, and by measurements and sketches of detail which I happened to make in leisure moments.

TEMPLE. Among the crowded group of modern dwellings on the high ground in the southeast quarter of the town are many remnants of an ancient building, or buildings, that crowned this most imposing of the sites in the city. The most conspicuous of these remains are two columns, one of which is complete, and which are marked as "Corinthian Columns" on our map of the city, standing almost due east of the East Arch, and north of Church No. 2. It was found impossible to trace the outlines of the ground-plan of any ancient structure among the crude hovels and cramped courtyards and stables which fill this quarter of the modern village. A piece of heavy wall of highly finished masonry
with two storeys of niches in it (A in Plan, 111. 219), another bit of wall, not unlike a

large parotid (B), and the two columns (C), are shown in the accompanying plan, together with the apse and side chambers of a Christian
Church (D), into the walls of which several details of the more ancient building were incorporated, these are the only visible remains of this ancient edifice that can be given on a plan. The other remains are fragments of architectural ornament.In the ground-plan fill. 219) I have attempted to show these features in their mutual relations; but I am unable to suggest the form of the building to which they belonged. There is no doubt that the heavy wall (A) is the southwest angle of a building; but the two columns, which appear to have been within the building, are set on a diagonal opposite to the interior angle, and, being about 14 m. distant from it, seem too far removed to have constituted interior supports on the plan of an octagon within a square outer wall. The modern dwelling between (A) and (B) makes it impossible to discover any connexion between these two walls. Two superposed niches in the wall (A) are given in a drawing beside the plan, and the profile of (B) is also represented. The column (C) is presented in a photograph fill. 220). Only the upper storey of the niches is above the present ground level, the niche of the lower storev shown herewith had been disclosed recently by the natives while excavating for building stones. The two niches are not shown here in their actual relation, the upper niche being the one next to the angle, and the lower being the second from it. The niche above this particular lower niche is less well preserved, but was substantially like it. Both are interesting ; the Doric half columns of the lower niche are unusual in this particular usage, and the absence of a Doric entablature is significant. The erect and the inverted conches are not unusual in Syria ; the little cross at the centre of the upper conch was added in early Christian days. The two columns, one of which is represented in a photograph (111. 220) are elevated upon pedestals.

Their shafts are nearly eight diameters high. The capitals are of good design; but were not executed with the exquisite technique that is to be observed in the capitals of the Nymphaeum; for the spaces underneath the angle volutes are solid instead of being cut through as they are in the capitals of the other building, the abacus is flat, and its profile is poor; but the acanthus leaves and the bud are well executed. There is a column almost precisely similar to these standing about 28 m. to the northeast.
In the courtyard of the same house with the two columns are fragments of a fine large portal in the same style as the portals of the two temples at Atil, and the outermost gate of the temenos at Si', all of which belong to the period of the earlier Antonine emperors1. The fragments show sections, or courses, of panels separated by quarter-columns the grooves of which are filled with a cable. The panels and quarter-columns were set upon a plain orthostate, like a dado, as in the example at Sic quoted above. The panel carving is rich and beautiful, being in flowing designs naturalistically treated. The outer panels are the wider, and have a rinceau of acanthus with starlike and lily-like flowers of large and graceful form (111. 221). The inner panels are carved with the grapevine (111. 222). Among other architectural fragments found here is a piece of ornament showing the Greek fret, or meander, with a pair of grotesque heads in relief in the lacuna (111. 223). These heads must be taken either as decorative grotesques, or as earlier than the other fragments, perhaps as of Nabataean workmanship ; for the heads wrought into the meander carving of the gate at Si', which belong unquestionably to the same period as the panels described above — the period of the Antonines — are distinguished for their grace and beauty. The photographs of the details herewith presented (111. 221, 222, 223) were taken from casts now in the Princeton collection. The originals were not so placed that they could easily be photographed. Although none of these details, i. e., columns, niches, and fragments of ornament, is of a scale so large as to prove the former existence of a large temple

NYMPHAEUM. The ruins of this building (111. 224) and of the one directly to the east of it (111. 225), with their standing columns, are among the most imposing of the monuments of Bosra, and are certainly the subjects most frequently chosen for photographs by .visitors to the ancient city. The four columns of the Nymphaeum, which are beautiful in themselves, and are quite perfectly preserved, are practically all that remains of the building of which they formed the facade. They stand on a line diagonal to the two main streets of the town which intersect at this point (see Plan in 111. 226)

and were so placed as to cut off the sharp angle which would have been produced if the colonnades of the two streets had met; for the streets do not meet at an exact right angle; but at an angle rather more acute. The north colonnade of the east-and-west street, and the west colonnade of the north-and-south street, terminated against the two end columns of this facade. The central intercolumniation is wider than the other two which are equal. This is all that can be seen from the streets of the modern village; for whatever else remains of the building is hidden behind high walls, as may be seen in the photograph (111. 224). Both Laborde1 and Rey2, who saw the building many years ago, refer to an apsidal construction behind the columns; the former gives approximate measurements of this construction in his text, the latter gives more exact figures, and shows a plan on very small scale in his map of the city8. When Puchstein was here he excavated to find this curved wall, and we found his excavation almost completely filled up when we entered the enclosure behind the wall. I took measurements of as much of the wall as I could find, which would be sufficient for our Plan of Bosra, and from these I have drawn the accompanying plan (111. 226). Puchstein called the building a Nymphaeum and I have followed him. The plan is sufficiently like that of the Nymphaeum of Gerasa to make this identification practically certain. Knowing that Puchstein had measured all the details of the columns, and had excavated to find the bottom of one pedestal, I did not undertake the trouble and expense of erecting a scaffolding about one of the columns, or of reopening his excavation which had filled up. I trust that his results may soon be given to the world; for the details here are of more than ordinary interest. I contented myself with measuring some of the details of the base of one of the columns which were easily reached. For a restoration I have assumed the height of the pedestals, and, knowing the diameter of a column (1.20 m.), I have erected the four columns by finding their approximate height in diameters from photographs. The capital was drawn from a sketch and from photographs. The present level of the soil is shown by a broken line. The visible parts of the building are shaded to represent basalt. The rear wall was restored from a similar plan in the central portion of the Nymphaeum at 'Amman, where the wall is standing. An apsis must be covered by a half dome; at cAmman the half dome springs from the level
of the tops of the columns. This disposition of the half dome suggests an arched entablature even if the wide middle intercolumniation did not demand it. I have chosen to strike the semicircle of the intrados of the arched architrave from the level of the tops of the columns rather than from the level of the top of the architrave, because the remnant of the arcuated architrave of the Central Arch here in Bosra, and other examples of arched entablatures in the Haurân have this form. The members of the entablature are drawn roughly from other buildings in the neighbourhood, the roof is wholly conjectural. It is , very probable that the walls adjoining the apsis had two or three storeys of niches in them, like the walls beside the great niche at 'Amman, and it is not impossible that the apsis itself was also provided with superposed niches, like the one at Djerash, but I found no proof or suggestion of these details.
The upper part of the pedestal and the base of a column are. given herewith in a measured drawing (111. 226). The octagonal form of the pedestal and of the plinth below the Attic base is interesting, and was probably employed in the present case to accommodate the end columns of the street colonnades which approached them at an angle. I have made the adjoining bases of the columns of the colonnades to correspond with them. The four capitals, which are of unusual beauty, present charming variations in minor details, as may be seen by a careful observation of the photograph. The proportions of the capital given in my drawing may not be absolutely accurate,, as they were not measured; but certain observations may be made which are, in the main, correct. The capitals are unusually tall, the part between the acanthus leaves and the abacus being almost as high as the taller leaves. The acanthus leaves have a prominent V section, the cauliculi, or stalks, are tall and flowing. The angle volutes and the intermediate spirals are quite free from the bell, as in the early Greek capitals of this order from Epidauros. Each volute and spiral was composed of a slender moulded fillet, and was cut entirely free as a detached member, in the most delicate manner possible, springing out of the stalk and touching only the angle of the abacus in the case of the volutes, and only the lip of the bell in the case of the spirals. The flower in the middle of the abacus has several varieties, and the sides of the abacus are slightly moulded. The execution could not have been more perfect in the finest quality of marble. The contrast between these capitals and that shown in 111. 220 is most marked.
KALYBE : The term which I have applied to this almost unique building which stands directly across the street from the Nymphaeum, is one that requires an explanation. I used it to designate a large structure which is connected with the Palace at Shehba (Philippopolis) in the Djebel Hauran, and which is the only building I know that is in any way analagous to this one. The word was first employed by M. de Vogue who took it directly from the Greek y.a/.vfir}, which he found in an inscription at Umm iz-Zetun where it was used to describe a sort of shrine, or open-air temple, and was extended by him to cover other somewhat similar buildings in the Haurân. These buildings consist of an open, arched, sanctuary flanked by walls with niches in them. That they were religious in character is proven by the adjective itoa which appears before the word xaXv& ; in two inscriptions.

The building at Shehba is quite well preserved, as may be seen by the plan given in A.A.E.S. II, Fig. 133; here the central sanctuary, which is square in the example

at Umm iz-Zetun, has the form of a half-domed exedra. The whole structure is more elaborate; the flanking walls with niches in them describe exterior and interior angles, and are then brought forward like two antae. The edifice stands so much higher than the level of the aqueduct of the city that it can hardly have been a Nymphaeum, as it has been called by more than one explorer, and it seems to require a term of its own to designate it.
The remains of the building in Bosra (111. 225) are not sufficient to warrant an unqualified statement that it was an edifice of the same type as that at Shehba. They consist of two very tall columns standing 24.60 m. apart on centres, and an anta wall joined to one of the columns by an entablature, and having three storeys of niches in its outer face (See Plan, 111. 226). From the inner face of the anta wall, and at a right angle with it, extends a fragment of a wall, 1.90 m. thick, with remains of a niche facing the column. The entablature which connects the column with the anta has the three usual members, an architrave, frieze and cornice, on both sides and across the front, showing that the column was not joined to other columns, but stood, like its fellow 25 m. away, as a purely decorative feature. This much of the plan does not suggest a temple of any well known type, but the anta wall with its niches, and the certainty that there was a second anta opposite the other column, do suggest the anta walls of the building at Shehba which are less than three metres farther apart, and the transverse wall, also with a niche, suggest the same feature in the example at Shehba. So that one has only to reconstruct a plan, as I have done, providing a large central exedra,

arched openings into side chambers, and wall surfaces to contain niches, and he will have a building similar in almost all details to the building at Shehba which I have called a Kalybe. The substitution of an anta and a free standing column for a solid anta wall is a minor difference. It should be understood that I found no tangible evidence in foundation walls for this restoration, that Puchstein's researches may have disclosed material for a very different restoration, and that this plan must be taken as a tentative one.
The two columns stand on a line with the east colonnade of the north-and-south street, and the south wall of the building was planned in a most interesting way to accommodate it to the obtuse angle at the intersection with the east-and-west street. The pilaster of the anta is on axis with the column, at right angles to the front line of the building; but the wall behind the anta is extended at an obtuse angle to this axis, and grows thicker toward the east in order to maintain a right angle for the chamber within. This wall terminates in a three-quarter column which takes up anew the line of the street colonnade. This angle, shown at (C) on the plan, has misled man)' observers to believe that the street turned slightly at this point; the plan shows that this is not necessary. The column at the southeast angle which shows but three quarters of its circumference, and belongs to the main street colonnade, has a console below which is an inscription1 recording the setting up of a "torch holder". It is interesting to think of these brackets as carrying street lamps as well as statues.

For the same reason that I omitted to make measured drawings of the details of the Nymphaeum, I omitted to draw to scale the superstructure of this edifice. The building has attracted the most unfavourable comment from almost all the travellers who have written about Bosra, chiefly because of the unusual and decidedly uncanonical

height of the order. The side wall adjoining the anta has three, and probably had four, storeys of niches (111. 225). The pilaster of the anta extends to the full height of this wall; but the column was elevated upon a pedestal; still the height of the anta was such that the shaft of the column was made more than 13 diameters high, and gave the whole column a height of about 16 diameters which is of course absurd. But the ornamental details of this order do not deserve the reputation they have been given as examples of bad taste. The bases are of white marble and have a delicate profile. The capital of the column and the cap of the pilaster were well designed and beautifully wrought in the free and open manner which characterizes the capitals of the Nymphaeum. The entablature is rather over-elaborately decorated, in the style of the later Antonines, and the proportions are not perfect; but the work was executed with rare skill in the hardest of materials. The three bands of the architrave are separated by bead-and-reel mouldings, and this member is finished off with a richly carved cymatium having a bead-and-reel below, an ovolo carved with widely spaced egg-and-dart in the middle, and a cavetto above. The pulvinated frieze is adorned with a deeply cut and flowing rinceau of acanthus. The bed mouldings which are carved upon a separate course of stone, consist of an egg-and-dart, a dentil band, and a narrow cyma reversa. The modillions of the cornice are low, but project well to carry an overhanging: corona ornamented with vertical grooves and a sima carved with anthemions and acanthus leaves. It will be noticed in the photograph that the frieze was pierced perpendicularly with a large hole which was probably also cut through the cornice and the architrave. It was perhaps intended to carry a mast of wood like those with which the cornice of the Coliseum in Rome was provided.