The plan (Ill. 289), as a temple plan, is most unusual, being like that of a Christian basilica short in proportion to its width and without longitudinal or other interior roof supports. It has three portals — a wide doorway between two narrow ones — in its facade, a large niche, like an apse, (Ill. 290) in the wall opposite the entrance, and side chambers in two storeys flanking the apse, all concealed by a flat rear wall. The floor of the apse was elevated, and the apse connects by a doorway with the upper side chamber on the west side. It is impossible, without excavations, to determine whether the porch of columns, which I have shown as conjectured in the plan, (Ill. 289) existed or not.
In the earlier publications cited above, the exterior order has been made equal in height to the interior order; but now, more accurate measurements of the heights of the exterior of the south wall show that the exterior pilasters were 7 m. in height, which would raise the tops of their caps 30 cm. above the top of the cornice of the interior order, and the exterior entablature, which is known by its fragments to have been almost exactly equal in scale to the interior entablature, when set upon its pilasters, raises the outer cornice 2 m. above the interior cornice, and makes its full height correspond to the height of the crown of the arcuated cornice above the apse (Ill. 292 Sect. A.-B.). This arrangement of the two orders at once raises the problem of the roofing of the building, and the question as to whether the outer entablature was crowned by pediments on the north and south. It is quite certain that the interior space between the portals and the apse was not roofed over·, for it is plain that there were no interior free-standing columns to carry a root. The entablature above the engaged columns of the side walls is continuous, as is shown by an excellent photograph of the west wall of the interior published by M. Kondakow,1 and there is no provision for connecting the engaged columns with free-standing columns within the chamber. I am fully persuaded that this part of the building was open to the sky; but that does not mean that there was not a gabled roof above the apse and a corresponding gable above the north wall, or a gable roof over a portico on the north, as I have indicated in my restored section (Ill. 289, Sect. C-D). The question of the presence or absence of a porch for the facade is another matter. I incline to the opinion that there was a porch of four columns, the inner pair centred upon the space between the portals. This spacing would require the arcuation of the entablature over the middle intercolumniation, which would correspond with the arched entablature within. The falling off of the ground in the spaces opposite the side portals suggested the arrangement of steps given in the plan (Ill. 289). There is ample material in the fragments lying all about the Tychaion to warrant the restoration which I have suggested; it is only a question whether some of them may not have come from another building; but, considering the similarity in technique and in scale between the fragments and the details of the building which are in place, this seems hardly likely. The restoration of the north wall was not a difficult problem. In my drawing (Ill. 292) I have called this the façade, which it would be if we could be certain that there was no portico outside of it. Much more of this wall is in situ than would appear in a casual glance at the wall as it is now, encumbered with the confusion of a late rebuilding. It will be seen, by reference to the photograph (Ill. 288), that the pilasters at the ends and the two side portals are preserved to a height of about 3 m.; the doorway on the right preserving its frieze and one console, that on the left, its frieze and parts of its cornice above the socket for a console. The details of these small portals are given on large scale in Plate XIX (A). Of the great portal only two courses of the right hand jamb are in place; but the lintel, a beautiful example of late Classic carving·, has been reset in the middle of the wall at a level much lower than the original. The probable height of the doorway, as shown in Ill. 292, was obtained from doorways in the region that are well preserved, and from the relation of the ornament above the lintel to the niches which flanked it. A piece of the frieze of the door-cap is to be seen in the modern wall at the left in Ill. 288, and one end of its cornice and one of its consoles were found in the crude wall in front of the building, shown in the same photograph. All these details are presented in a scale drawing in Plate XIX (B). The niches which must have been placed above the smaller doorways are represented by two sections of coupled colonettes and two coupled Corinthian capitals built into the north wall (Ill. 288), and by fragments of two conchas and the end of one pediment built into the other crude wall. They are not difficult to restore with every probability of correctness from these fragments, and are almost exactly similar to niches correspondingly placed in the front walls of the Basilica at Shakka and the little temple at Br6keh 3 in the Ledja, except that the order in these other examples was the Ionic. Other details appearing in the new wall are a corner pilaster cap, a piece of heavy cornice with modillions, pieces of richly carved architrave, which differ slightly in measurement from the interior architraves, and several bits of frieze wrought with the guilloche, some pieces flat and others curved for use in an arcuated entablature. Some of these fragments undoubtedly belonged to the interior entablature, others certainly did not, and were quite certainly parts of the exterior enrichment of the temple. Other interesting details of this facade of which we have record, but of which no remnant remains, are four bronze torch holders which were affixed to the wall at both ends of the lintels of the two small portals. There are parts of four inscriptions, dedicating the torch holderss, still in place in the wall, and over two of them are the sockets which received the supports of the torch holders, as shown in Ill. 292. The restoration of the façade, or north wall, should be compared with the drawing made by M. Rey of the front wall of the Pretorium at Mismiyeh now entirely destroyed. The beauty of the interior ornament is shown in the photographs (Ill. 290 and 291) and in Plate XIX. Very little restoration was necessary, as will be seen in Ill. 292 Section A-B. There are few monuments of antiquity in which the finer details have been so well preserved. The illustrations speak for themselves, and such parts as are in situ are indicated by stippling in Ill. 292. It is astonishing to find such refinement of technique in such a material as basalt, and at a date which we know to have been near the close of the second century after Christ. Certain points of interest are to be noted in departures from the canons of Classic architecture: for instance, the interior entablature is much too heavy for the columns which were already reduced in scale by being elevated upon pedestals, and the capitals which were given height to suit the entablature were greatly compressed all around to fit the shafts (Pl. XIX). A similar error is to be seen in the little pilasters which flank the upper openings beside the apse, where the caps are much too tall for the pilasters. The omission, above these pilasters, of a complete entablature, and the substitution of the beautiful volutes are other features which show that the work is not to be called Roman, and has an independent descent from the Hellenistic architecture of Asia. In many respects this architecture resembles that of the Classic Revival more closely than it does the contemporaneous architecture of Rome. Along the side walls, between the engaged columns of the interior, are projecting brackets to hold statues, and each bracket has its inscription.5 Another inscription,6 one upon the curved wall of the apse, records the gilding of the statue of Tyche and of the niche itself. The statue of the divinity was perhaps one of heroic size proportioned to the apse in which it stood. On the south side of the birkeh, that is the side opposite the temple, are the remains of a colonnade (Ill. 293). It would be impossible, without systematic excavations, to discover the original form of this structure or its extent; but, from the part which is standing, and from dispersed fragments, it would seem that it enclosed the pool of water. It is evident that the lower part of the fragment now in situ is deeply buried; for the Corinthian columns are very short in proportion to their diameter. Ihis fragment consists of a rectangular pier and two columns carrying an architrave with two bands and a highly finished cymatium; but its important feature is a remnant of an upper storey consisting of a pier, with a pilaster on its face, carrying the spring- stones of two small moulded arches. Its original form is easily restored. The frieze above the main architrave is a low member with a rightlined moulding at the top; and, in place of a cornice, there is a plain course of stone upon which the piers of the upper storey were set. It is practically certain that there was a pier above each column, and that these piers carried narrow arches, and that pilasters were run up between the arches to carry a continuous entablature above them. It is both interesting and important to find an arcade above a colonnade at so early a period; for the details of these columns and the mouldings of the arcade above them are not appreciably later than the Tychaion. These remains are worthy of thorough and systematic investigation by means of excavations, and their remoteness from dwellings or other modem buildings would render such an undertaking both simple and inexpensive.
