Hebrew Sela, in Greek Petra. In the Old Testament we find it recorded of king Amaziah, that "he slew of Edom in the valley of salt, ten thousand, and took Sela by war, and called the name of it Joktheel unto this day." The prophet Isaiah also exhorts Moab to "send the lamb to the ruler of the land from Sela to (through) the wilderness unto the mount of the daughter of Zion;" alluding apparently to the tribute in sheep formerly paid to Israel. At this time, therefore, Sela would seem to have been in the possession of the Moabites; or at least they pastured their flocks as far south as to that region, much in the manner of the adjacent tribes at the present day.'These are the only certain notices of this city found in Scripture; and the last of them cannot be later than about 700 B. C."

About four centuries afterwards, as we have seen, the city was already known to the Greeks as Petra; it had passed into the hands of the Nabatheans, and had become a place of trade. The two expeditions sent against it by Antigonus before 301 B. C. have been already alluded to. In the first, Athenæus took the city by surprise, while the men were absent at a neighbouring mart or fair; and carried off a large booty of silver and merchandise. But the Nabatheans quickly pursued him, to the number of eight thousand men; and falling upon his camp by night, destroyed the greater part of his army. Of the second expedition, under the command of Demetrius, the Nabatheans had previous intelligence; and prepared themselves for an attack, by driving their flocks into the deserts, and placing their wealth under the protection of a strong garrison in Petra; to which, according to Diodorus, there was but a single approach, and that made by hand. In this way they succeeded in baffling the whole design of Demetrius.

Strabo, speaking of the Nabatheans during the reign of Augustus, describes the capital as follows: "The metropolis of the Nabatheans is Petra, so called; for it lies in a place in other respects plain and level, but shut in by rocks round about, precipitous indeed on the outside, but within having copious fountains for a supply of water and the irrigation of gardens. Beyond the enclosure, the region is mostly a desert, and espe

cially towards Judea." At this time the city had become the great place of transit for the products of the east, and was often resorted to by foreigners. The philosopher Athenadorus, Strabo's friend, spent some time in Petra; and related with admiration, that he found many Romans and other strangers residing there; that these often had lawsuits with one another and with the inhabitants; while the latter lived in peace among themselves, under excellent laws."

5

Similar, but more definite, is the testimony of Pliny in the first century: "The Nabatheans inhabit the city called Petra, in a valley less than two (Roman) miles in amplitude, surrounded by inaccessible mountains, with a stream flowing through it." About the same period, Petra is often mentioned by Josephus as the capital of Arabia Petræa, in all his notices of that kingdom and its connection with Jewish affairs. With that kingdom, it passed under the immediate sway of the Romans, during the reign of Trajan. His successor Adrian appears to have granted privileges to Petra, which led the inhabitants to give his name to the city upon coins. Several of these are still extant. In the fourth century, Petra is several times mentioned by Eusebius and Jerome; and in the Greek ecclesiastical Notitia of the fifth and sixth centuries, it appears as the metropolitan see of the Third Palestine. Of its bishops, Germanus was present at the council of Seleucia in A. D. 359; and Theodorus at that of Jerusalem in A. D. 536.

But from that time onwards, Petra suddenly vanishes from the pages of history. In the two Latin Notitiæ, referring in part to the centuries after the Muhammedan conquest and before the crusades, the name of Petra is no longer found, and the metropolitan see had been transferred to Rabbah. Whether Petra perished through the ruthless rage of the fanatic conquerors, or whether it had already been destroyed in some incursion of the hordes of the desert, is utterly unknown. The silence of

1 Strabo 16. 4. 21, Μητρόπολις δὲ τῶν Ναβαταίων ἐστὶν ἡ Πέτρα καλουμένη· κεῖται γὰρ ἐπὶ χωρίου τἄλλα ὁμαλοῦ καὶ ἐπιπέδου, κύκλῳ δὲ πέτρᾳ φρουρουμένου, τὰ μὲν ἐκτὸς κρημνοῦ ἀποτόμου, τὰ δ ̓ ἐντὸς πηγὰς ἀφθόνους ἔχοντος εἴς τε ὑδρείαν καὶ κηπείαν. Εξω δὲ τοῦ περιβόλου χώρα ἔρημος ἡ πλείστη, καὶ μάλιστα ἡ πρὸς Ιουδαία. * See above, p. 161. Strabo ibid.

✦ Plin. H. N. 6. 28 v. 32, "Deinde Nabathæi oppidum incolunt Petram nomine in convalle, paulo minus II mill. passuum amplitudinis, circumdatum montibus inaccessis amne interfluente."

See the references to Josephus above, p. 159. n. 4.

9

Mionnet Descr. de Médailles Antiques, Tom. V. p. 587. Eckhel Doctr. Nummor. Vet. II. p. 503. No less than eight coins of Petra are described, viz. three in honour of Adrian; one of Marcus Aurelius and Verus; two of Septimius Severus; and two of Geta. Most of them bear on the reverse the inscription: 'Αδριάνη Πέτρα Μητρόπολις. For this custom on the coins of cities, see above, p. 60. n. 3.

'Onomast. arts. Petra, Idumæa, Theman, etc. Reland Pal. p. 215, 217. Sea also in general the art. Petra in Reland p. 926 sq.

Reland Pal. pp. 933, 533. Le Quien Oriens Chr. III. p. 725.

Relund Pal. pp. 223, 226.

all Arabian writers as to the very existence of Petra, would seem to favour the latter supposition; for had the city still retained its importance, we could hardly expect that they should pass it over without some notice, in their accounts of the country and its conquest. As it is, this sudden and total disappearance of the very name and trace of a city so renowned, is one of the most singular circumstances of its history.'

2

The crusaders, as we have seen, found Petra at Kerak, just as they also found Beersheba at Beit Jibrîn; thus introducing a confusion as to Petra, which is not wholly removed even at the present day. It was not until the reports collected by Seetzen respecting the wonderful remains in Wady Musa, had been verified by the personal discovery and examination of them by Burckhardt, that the latter traveller first ventured to assume their identity with the site of the ancient capital of Arabia Petræa. This identity is now, I believe, admitted by all.-The arguments for the identity in question, are of a threefold nature, and all lie within a small compass.

First, the character of the site, as given by Strabo and Pliny in the passages above quoted; an area in a valley surrounded by precipitous rocks, with a stream running through it, and a single approach made by hand,' as mentioned by Diodorus ; all this corresponds entirely to Wady Musa as already described.

Again, the ancient specifications as to the distance of Petra from both the Dead Sea and the Elanitic gulf, all point to Wady Músa. Passing over the merely casual and indefinite estimates of Strabo and Pliny, we find in Diodorus Siculus, that Demetrius, on his return from Petra, marched three hundred stadia,

1 No Arabian writer mentions Petra; and the only ones who speak of Wady Mûsa, so far as yet known, are Kazwiny in the thirteenth century, and Ibn Iyâs (Ben Ayas) in the fifteenth. Both these authors merely relate a Muhammedan legend, according to which Moses died and was buried in this valley. See Kazwiny's Geogr. Lexicon, Athar el-Belad, Cod. Ms. Arab. Biblioth. Gothan. No. 234. fol. 80. Ibn Iyâs in his geogr. work, Neshek elEzhár, Cod. Gothan. No. 302. fol. 226. These notices from Mss. in the Library at Gotha, I owe to the kindness of Prof. Roediger of Halle.-For the forms Arce, Arcceme, Recem, etc. assigned by Josephus and others as the earliest name of Petra; as also for the er-Rakim of Arabian writers, see Note XXXVII, at the end of the volume. In like manner the place called by Arabian writers el-Hijr, (not el-Hajar a stone') has sometimes been mistaken for Petra; see the same Note.

2 See p. 167, above. Adrichomius p. 129. Raumer's Palist. p. 412 sq.

Burckhardt p. 431. The first published account of Burckhardt's visit to Wady Mûsa, seems to have been contained in a letter from him dated Cairo, Sept. 12, 1812, prefixed to his Travels in Nubia, Lond. 1819. But before this appeared, Ritter had already suggested the identity of Wady Mûsa and Petra, on the strength of Seetzen's reports in Zach's Monatl. Corr. XVII. p. 139. See Ritter's Erdkunde Th. II. p. 117. Berl. 1818.

See pp. 169, 170.

Strabo places it at three or four days' march from Jericho, 16. 4. 21. Pliny at 600 Roman miles from Gaza, and 135 from the Persian gulf; H. N. 6. 28 or 32. Here, as Cellarius suggests, the two numbers have probably been transposed; that is, it should read 135 miles from Gaza, and 600 from the Persian gulf. Cellar. Notit. Orb. II. p.

581.

and encamped near the Dead Sea. This distance is equal to about fifteen hours with camels; and if reckoned northwards from Wady Musa along the ancient road, extends to nearly opposite the south end of the sea. After all, this is doubtless also a mere estimate, and is if any thing too small; but at any rate, it could never apply to Kerak.-More exactly is the position of Petra laid down in the Peutinger Tables. The distance is there marked from Ailah along the ancient road to Petra, by the stations Ad Dianam, Præsidium, Hauara, and Zadagatta, at ninety-nine Roman miles in all, equivalent to about seventyeight and two thirds geographical miles. The actual direct distance between 'Akabah and Wady Musa, on a straight line, is about sixty-four geographical miles; and when we take into account the windings of the way and the steepness of the mountains, the comparison is here sufficiently exact.' On this route too, the name and site of Zadagatta (Zodocatha) still exist at Usdakah, about six hours south of Wady Musa. Further, the same Tables, although somewhat confused on the north of Petra, yet give the distance between it and Rabbah as at least over seventy-two Roman miles; which corresponds well with Wady Músa.

Lastly, Josephus, and also Eusebius and Jerome, testify expressly, that Mount Hor, where Aaron died, was in the vicinity of Petra. And to this day the mountain which both tradition and the circumstances of the case mark as the same, still rears its lonely head above the vale of Wady Musa.

These considerations appear to me to demonstrate the identity of Petra with Wady Musa; and also to show as conclu

1 Diod. Sic. 19. 98, #λholov tŷs 'Aopaλτίτιδος λίμνης.

The Audpa of Ptolemy, and the Havana of the Notitia Dignitatum. Ptol. 5. 17. Reland Pal. pp. 463, 230.

The following is the specification of the Tables: From Haila, xvi ad Dianam. XXI Præsidio. XXIII Hauara. xx Zadagatta. XVIII Petris. Summa XCIX.-The station 'ad Dianam' is common to both the routes from Ailah, to Jerusalem and to Petra. It must therefore have lain in the great valley; and the distance of 16 Roman miles from 'Akabah would bring it about opposite the Wady and fountain el-Hendis. It is marked as a small temple of Diana. This point must have been considerably north of the mouth of Wady el-Ithm. The Roman road to Petra appears therefore to have ascended the mountain north of that Wady; and the station Præsidium is probably to be sought in those mountains. It would consequently seem, that

el-Humeiyimch did not lie upon the great Roman road, of which Laborde found traces further north upon the mountain.

The geographical position of Petra is fixed on the accompanying map at 30° 25' N. Lat. and 35° 38′ 9′′ E. Long. from Greenwich. This latitude is the mean between that resulting from our routes and Laborde's. Moore and Beke give it at 30° 19', which appears to be even less correct than their observations at Jerusalem and Hebron; see above, p. 74, n. 2. Vol. I. p. 259, n. 1. The longitude is that found by a comparison of our routes with those of Laborde. See Kiepert's Memoir in the former edition of this work, III. App. p. 37.

sively, that it could not have been situated at Kerak, where the crusaders placed it.'

But how or when the name of Petra was dropped, or in what age that of Wady Musa was adopted, we have no means of ascertaining. The crusaders found the latter in current use, and speak here only of the "Vallis Moysi." They also speak of a building on the neighbouring mountain, consecrated to Aaron; but they appear to have discovered nowhere any trace of a Christian population."

Then came other centuries of oblivion; and the name of Wady Musa was not again heard of, until the reports of Seetzen in A. D. 1807. During his excursion from Hebron to the hill Madurah, his Arab guide of the Haweitât described the place, exclaiming: "Ah, how I weep, when I behold the ruins of Wady Musa!" The subsequent visits of Burckhardt, Irby and Mangles, Laborde, and others, have put the world in possession of most of the details; yet I apprehend that the historical and antiquarian interest of the place is by no means exhausted. The scholar who should go thither learned in the lore of Grecian and Egyptian arts and architecture, would be able, I doubt not, still to reap a rich harvest of new facts, illustrative of the taste, the antiquities, and the general history of this remarkable people.

Saturday, June 2d. As morning dawned, we rose from our couch of sand in the middle of the Arabah; and at 4 o'clock were again upon our way towards the fountain el-Weibeh. Looking back, we could see the Wady through which we had descended from the pass of Nemela, bearing S. 55° E. and marking the course we had travelled during the darkness. We were now more than half way across the 'Arabah; and continued to travel on nearly W. N. W. through a rolling gravelly desert, with rounded naked hills of considerable elevation. Our guides had usually in the 'Arabah kept one man ahead as a scout; and now, as we approached el-Weibeh, they took double precautions against any enemy; since this and other fountains in the valley, are the usual rendezvous of wandering parties.

For the question whether there was probably more than one Petra, see in Note XXXVII, at the end of the Volume.

2 See above, pp. 163, 165. 'Guibert speaks of this building as a church; 7. 36. p. 555. Another writer calls it an "Oratorium;" Gesta Dei p. 581. Fulcher incorrectly makes it a monastery dedicated to St. Aaron; c. 23, p. 405.

Not improbably there may have been here originally a Christian chapel, as on Jebel Mûsa and Mount St. Catherine; but there is no historical trace of any monastery on the mountain. See Note XXXVI, end of the volume.

Zach's Monatl. Corr. XVII. p. 136. See above, p. 165.