Nabataean Remains in Bosra

Before taking up in order the description of other structures in Bosra which belong certainly to the period of Roman domination in Arabia, and which are in a better or a worse state of preservation, it may be well to give a description of some remains of ancient buildings which bear a different mark from the edifices of the Roman or Christian periods, and which seem to be allied with architectural monuments which are known to be older, and to have been the work of the Nabataeans. These fragments have a striking resemblance to details of architecture found at St' in the Djebel Hauran, which are known by inscriptions in Nabataean script to belong to the first century of this era. Some of them are closely allied with the architectural features of the rockhewn tombs of ancient Hegra, near Medâin Sâleh, which are also known to be of Nabataean origin and of early date, and all are suggestive of the style of architecture illustrated in some of the famous rock-hewn façades of Petra which, for lack of inscriptions, are to be classed with the other monuments by analogy.

It may be said at the outset that Nabataean inscriptions are found in almost ever}' quarter of the city; but that the remains of buildings and the fragments of architectural details that can be called Nabataean have all been found between the East Arch and the line of the east wall of the town. The most conspicuous monument of this class is a tall half column engaged with a broad pier, which stands about 25 m. to the eastward of the East Arch. The scale of this half column (111. 211) which is 1.32 m. in diameter, suggests that the building to which it belonged was one of fairly large dimensions; but all other signs of a building in the vici-nity have disappeared, or are completety hidden in the closely packed cluster of modern dwellings that now occupies the spot. The half column itself rises 4.50 m. above the flat earthen roof of one of these dwellings, and the photograph was taken from that roof. To this height we may add at least 3.50 m. for the height of the house, and not less than 2 m. more for the depth of the accumulated soil above the original level upon which the base of the column rests, which will give 10 m. or between eight and nine diameters, as the height of the column. The capital here is the detail which attracts attention (111. 212), and I

have prepared a measured drawing on large scale to illustrate it. The drawing and the photograph both show a set of mouldings, an ovolo and a cyma recta below two slanting fascias separated by a groove of triangular section —, which compose a sort of echinus, and a high and heavy abacus with concave sides and salient angles which are cut off square at the ends and bevelled inwards at a steep angle to meet curved surfaces which carry the concave lines of the top of the abacus to the convex line of the semicircular top of the echinus. A projecting boss has been broken from the middle of the top of the face of the abacus. It appears that the mouldings of the echinus were returned outward on either side, and prolonged across the top of, the broad pilaster with

which the half column is engaged, but it is not to be known how the pilaster was treated above this moulding. This peculiar abacus which is the dominating feature of the capital, taken together with precisely similar details in the rock-hewn façades of Hegra and Petra and in one of the temples at Sic, may be said to create a new architectural order. In the façades alluded to this abacus appears with and without the moulded echinus; in the earliest type it is found as the cap of a flat pilaster with only a simple moulding below it; but it occurs also in connexion with an abacus, substantially like this one, as the capital of a half column, in façades that are known to belong to the first century a. d. In Hegra a small member, not unlike a primitive volute or a minute leaf, often appears underneath the salient angles of the abacus1. The order in the tomb façades of that place is completed in several different ways; in the simpler tombs by a plain architrave, a torus, and açavetto cornice which is frankly Egyptian and plainly shows Ptolemaic influence, although in these same tombs, where an abacus of this type occurs as the cap of a pilaster flanking the doorway, the

order above it consists of an architrave moulded at the top, a plain frieze, and a cornice made up of several mouldings , all beneath a pediment with raking cornice that is quite Classical in effect .In other tombs which are still early a moulding is given to the top of the architrave, and a plain frieze is inserted between the architrave and the cavetto cornice3. In a third type, in which a moulded echinus like the Bosra example is inserted below the abacus, the architrave and frieze are surmounted by a cornice of Classical appearance, and the Egyptian cornice is elevated upon an Attic storey4, and again, usually in portals, capitals of this style carry an entablature with a complete Doric frieze of triglyphs and metopes5. The fact that this sort of capital is found only in places known to have been occupied by the Nabataeans, and usually in connexion with Nabataean inscriptions, gives a reason for referring this detail to a Nabataean origin, and for applying the name Nabataean to the Order, at least in its simpler and less Classical form. Due east of the Nabataean half column, and about 100 metres distant, stands a group of similar half columns, three of which are without capitals; but the fourth retains the lower part, that is the echinus, of a capital similar to that described above and shown in illustrations 211 and 212. The group is almost completely hidden by modern dwellings and stables, and a considerable part of each half column is deeply buried. Each half column is engaged with a pier 1.20 m. wide and 1 m. deep. They stand at the angles of a rectangle, 8.10 m. by 11 m., facing inwards (111. 213). The two on the south are directly in line with the Nabataean column described above. The outer faces of the piers have finished surfaces, but the inner faces of the north pair and the south pair have projecting pilasters which are now much lower than the columns, and appear to have carried arches. The half column which stands at the southeast angle of the group is the only one of the four of which the lower part can be seen, and this shows that the half column was set upon a rectangular pedestal, or perhaps it would be more accurate to say that the lower part of the half column takes the form of a pier; for there is no sign of a base above the pedestal the sides of which are tangent to the semicircle of the half column. The amount of data to be secured here is so small that it is well-nigh impossible to suggest a probable restoration even for one pair of the piers, and it is quite out of the question to hazard a conjecture as to the probable extent or purpose of the structure of which they formed a part. Two facts however may be significant: The southernmost pair of piers stand in line of columniation with the Nabataean half column, and the distance between the two pairs is about equal to the width of the East Arch. And, since the half columns of the two pairs face each other, they may be thought of as standing on opposite sides of a street, 11 m. wide, that led eastward from the East Arch at an obtuse angle with the main east-and-west street of"the city. In A-B of the drawing (111. 213) I have taken one pair of piers, and, assuming a depth of soil of about 2 m. , have shown two pedestals

rising 2 m. above the line of the soil, this being the height of the one visible pedestal. From the pedestals rise the shafts of the half columns, and from one to the other of the pilasters on the inner sides of the piers I have thrown an arch. The capitals have been restored from the other Nabataean half column on the basis of the single echinus which is still in place, and is shown in detail in the drawing, and over these I have placed a purely conjectural epistyle, such as one might find in examples at Petra. Nothing, however, short of the removal of the present houses and a thorough excavation of the site, can reveal the original arrangement of these very interesting architectural details which probably belong to the early period of Bostrian history.