Plan of the City. The map of Bosra published herewith is based upon the first accurate survey ever made of the ancient city. Rey's map is by far the best that has been published hitherto; but is wanting in certain details. Other maps have been published; but most of them have been compiled from notes and rough sketches made by various travellers, and are both inaccurate and misleading. One such map which often has been reproduced, indicates a city of symmetrical plan, with walls forming a great rectangle, with gates regularly placed, with streets evenly subdividing the great rectangle, and with public buildings more or less regularly disposed with reference to the streets; quite as if the Romans had laid out a new and complete scheme upon the site of the more ancient Bosra. Our survey demonstrates that such a plan departs far from the truth. The outline of the city is most irregular, the streets follow no symmetrical scheme, and the buildings are placed in hap-hazard fashion, indicating that the builders of the city of the Roman period found themselves confronted by the existing plan of an ancient Oriental city, and limited in conforming it to a symmetrical scheme by the position of buildings that could not be destroyed. In presenting the details of our survey it will be observed that no attempt has been made to give the plan of the city in other than Roman, Christian and Mediaeval times. The plan includes remains of certain buildings that may have been in existence before the Roman occupation of the city ; but does not show the maze of narrow and tortuous streets of the present Arab village. Black ink has been used to indicate the buildings of the city that are earlier than the seventh century, and red ink to distinguish the constructions of the earlier Mohammedan period. In the plan of the older city solid black indicates that the walls or columns are standing to a height of at least two metres. Shading shows that foundations are traceable, grey dotted surfaces represent the tops of walls that are sunken, like those of the reservoirs, and small circles not blackened are used to mark the rows of conjectured columns. The walls follow no regular lines nor fixed directions for any considerable distance; that on the west of the city sweeps in a series of obtuse angles, which almost compose a curve, toward the northeast, and, quite imperceptibly, becomes the north wall. The north and east walls take no consistent course, and the south wall is entirely wanting, as the ruins stand today, except at the southwest angle of the city where a long section of pre-Roman wall may or may not give the course of the wall of the later period. It is not possible to discover, without excavations, if the theatre and the great reservoir were within the enclosure of the city ; but it is probable that the Hippodrome was not included within the walls. The existence of two principal gates — the North gate and the West Gate - is demonstrable from the ruins. Three minor gates are to be found in the northwest quarter of the city. No signs of other gates were discovered, though they undoubtedly existed. The ancient streets, that is the more important of them, which were provided with colonnades, are to be traced by the ruins of their columns; here by broken shafts that lie in a more or less direct line, there by a few bases or stumps of columns that are still in situ, and, occasionally, by complete columns that have been incorporated with the walls of late buildings, or by pieces of ancient paving that have been uncovered to form the pavement of modern courtyards. It is only by marking such remains which are often invisible from a distance, and by-finding out, by means of a transit instrument, that they lie in a straight line, that the courses of the ancient streets can be determined. The less important streets could not be traced without excavations. There can be little doubt that the colonnaded streets given on our map existed as they are shown. The two principal avenues which roughly bisect the city at right angles were not carried through the city from wall to wall, so far as we could discouver. The Ions? avenue which begins at the West Gate passed through the heart of the city, leading toward a slight elevation in the east quarter of the city, and terminated at an arch less than three quarters of the way to the east wall. This elevation, upon which there are abundant remains of a great temple, may have constituted a sort of akropolis. The avenue which led in from the North Gate seems to have terminated at its juncture with the other avenue. Two more streets lying east and west, and one more lying north and south are to be traced in ruins of colonnades, as I have described above; but there were in all probability other colonnaded streets which would be revealed by systematic excavations, or even by further search prosecuted from house to house along the intricate lanes and by-ways of modern Bosra. We may assume from the evidence offered in the ruins, that the colonnaded streets of Bostra were flanked by rows of buildings which stood behind the columns on both sides of the streets. These buildings, which may have been houses, or shops, or both, are shown at several points on the plan where the evidence of their existence is conclusive; but I believe that similar structures might have been drawn on the plan along the sides of all the streets without departing from well supported conjecture. The buildings which date unquestionably from the Roman, or an earlier, period are not all found to be symmetrically placed with regard to the streets any more than the Christian buildings are. The Theatre, the Market, and the Central Baths, do conform to the direction of the streets, but the Palace is set without reference either to streets or to other buildings. The Cathedral, and the entire group of ecclesiastical buildings near it, were oriented in such a way as to throw them out of all symmetry with the Roman streets; and, though I believe that the colonnades remained in place, and in use, during the. fine Christian period of the sixth century, it is difficult to imagine how these buildings were architecturally accommodated to the monumental plan of the city already adopted. Other churches were placed with reference to the direction of streets, and are consequently oriented at a different angle from the Cathedral. One might assume that the mosques were erected while the colonnaded streets were still intact, if it were not for the fact that several of them have details of the despoiled colonnades built into their walls, and that one employs as interior supports columns from these streets ; for three of them at least are placed near the colonnaded streets and at right angles to them. It may be that the paved streets were still free and in use during the early Mohammedan period, even after the colonnades which flanked them had been wholly or in part removed.