¹ I am glad to be able to acknowledge here my profound gratitude to those who aided me in Jordan and Israel: His Excellency Salah Abu Zeid, Minister of Information in the Hashemite Kingdom of Jordan; Mr. David Harris, of the N.R.A. Soils Division in Amman; Prof. Fawzi el-Fakharani, formerly visiting professor at the University of Jordan and now Chairman of the Archaeology Department in the University of Libya at Benghazi; Prof. Zvi Yavetz of the University of Tel-Aviv; Dr. Avraham Negev, of the Hebrew University in Jerusalem; and Lt.-Col. Mordechai Gichon, of the University of Tel-Aviv. I have also to thank for help and criticism Professors T. D. Barnes, C. P. Jones, John Strugnell and F. V. Winnett. The present report is concerned chiefly with work in or on Arabia since the excavation of Jerash, completed in 1934 and published in 1938 (C. H. Kraeling, Gerasa: City of the Decapolis). The following abbreviations should be noted: AASOR = Annual of the American Schools of Oriental Research; ADAJ = Annual of the Department of Antiquities of Jordan; CIS = Corpus Inscriptionum Semiticarum; IEJ = Israel Exploration Journal; PEQ = Palestine Exploration Quarterly; RB = Revue Biblique; SDB = Supplément au Dictionnaire de la Bible; ZDPV = Zeitschrift des Deutschen Palästina-Vereins. Page references are to the initial page of a discussion.

² Mention should be made, honoris causa, of the serviceable and up-to-date guidebook The Antiquities of Jordan, by Harding, G. Lankaster, in the second (revised) edition of 1967Google Scholar.

³ SDB 886; cf. Starcky's earlier ‘The Nabataeans: a Historical Sketch’, The Biblical Archaeologist 18 (1955), 84Google Scholar. The SDB contribution is now the fundamental study of Nabataean civilization. N. Glueck, Deities and Dolphins (1965), is of much more limited scope, preponderantly concerned with art, and, in particular, at Khirbet Tannur. On this book, cf. Starcky, J., ‘Le temple nabatéen de Khirbet Tannur. À propos d'un livre récent’, RB 75 (1968), 206Google Scholar. Note also M. Lindner, Die Könige von Petra (1968), with fine colour photographs.

⁴ Diod. 19, 94–100.

⁵ SDB 903.

⁶ ibid.

⁷ F. V. Winnett and W. L. Reed, Ancient Records from North Arabia (1970), 99.

⁸ Negev, A., ‘Avdat, A Caravan Halt in the Negev’, Archaeology 14 (1961), 123Google Scholar. cf. PEQ 98 (1966), 95Google Scholar; PEQ 101 (1969), 5Google Scholar. Also A. Negev, Cities of the Desert (1966), 12. The Nabataeans were originally nomadic, as Diodorus' report (n. 4) shows (from a considerably earlier source); by Strabo's day they had become sedentary (p. 779). Tetradrachms with Aramaic lettering of a date before 310 have recently been discovered in southern Palestine, but there is no reason to think them Nabataean: Starcky, J., IX^e Congrès international d'archéologie classique: Rapports et Communications, Damascus, 1969, 23Google Scholar = Die Nabatäer (Catalogue of Munich Stadtmuseum exhibition, 1970), 81Google Scholar.

⁹ Journal of Biblical Literature 74 (1955), 160, n. 25Google Scholar.

¹⁰ PSI IV, 406.

¹¹ Glueck, N., Explorations in Eastern Palestine: I AASOR 14 (1934)Google Scholar; II, AASOR 15 (1935)Google Scholar; III, AASOR 18–19 (1939)Google Scholar; IV, AASOR 25–28 (1951)Google Scholar. Cf. also The Other Side of the Jordan (1940).

¹² Bāyir (on a desert route to ‘Ammān): Glueck, N., AASOR 14 (1934), 73Google Scholar; The Other Side of the Jordan (1940), 41; AASOR 25–28 (1951), 47Google Scholar. Al-Jawf, : AASOR 25–28 (1951), 16, 36, 44Google Scholar; Savignac, and Starcky, , RB 64 (1957), 196Google Scholar; Winnett and Reed, Ancient Records from North Arabia (1970), 15 and, in the same volume but by Milik and Starcky, 144. Sakāka: ibid. 7, 144.

¹³ Ithrā: Winnett and Reed, op. cit. 60, 160. On the Wādi Sirḥān: Glueck, The Other Side of the Jordan (1940), 40; AASOR 25–28 (1951), 34Google Scholar; Stein, A., Geographical Journal 95 (1940), 434Google Scholar.

¹⁴ Glueck, N., AASOR 25–28 (1951), 39Google Scholar.

¹⁵ Most recently, Deities and Dolphins (1965), 6.

¹⁶ On Medā'in Ṣalih, Doughty is very much worth reading still. Subsequent and scholarly, especially Jaussen, A. and Savignac, R., Mission archéologique en Arabie I (1909), 107, 301Google Scholar and II (1914), 78; Winnett and Reed, op. cit. 42. On Midian (al-Bad') A. Musil, The Northern Hegaz (1926), 109; H. St. J. Philby, The Land of Midian (1957), 257; Parr, P. J., RB 76 (1969), 392Google Scholar. Aramaic inscriptions from the oasis of Thaj, near Bahrein, are probably of Mesopotamian, not Nabataean origin: Starcky, op. cit. (n. 8), 23–4 (French) = 81 (German).

¹⁷ Newell, Late Seleucid Mints (1939), 92.

¹⁸ BM Coins Rom. Rep. 1, 483.

¹⁹ On that expedition: J. Pirenne, ‘L'expédition d'Aelius Gallus en Arabie du sud’, in Le Royaume Sud-Arabe de Qatabān et sa Datation (1961), not known to Jameson, S., ‘Chronology of the Campaigns of Aelius Gallus and C. Petronius’, JRS 58 (1968), 71Google Scholar. See also A. Dihle, ‘Der Zug des Aelius Gallus’, in Umstrittene Daten (1964), 80.

²⁰ See below, n. 39.

²¹ IEJ 13 (1963), 113Google Scholar.

²² Assuming the correction of ἀναβαταίως to Ναβαταίων ἀνάβασις; therefore, εἰς Πέτπαν πρὸς Μαλίχαν βασιλέα Ναβαταίων ἀνάβασις.

²³ R. Dussaud, La Pénétration des Arabes en Syrie avant l'Islam (1955), 211; J. Pirenne, op. cit. 167; Altheim, F. (with Stiehl, R.), Die Araber in der alten Welt I (1964), 40, 100, 106, 134Google Scholar.

²⁴ Starcky, J., RB 71 (1954), 161Google Scholar. A. Dihle's criticism of the hypothesis of a third Malichus and related matters (Umstrittene Daten [1964], 13) was attacked in turn by Altheim, ‘Zu einem Buch A. Dihle's’, op. cit. IV (1967), 492. Dihle's good argument would have been even stronger had he recognized that Malichus' name did not actually stand on the papyrus.

²⁵ cf. Starcky in SDB 918. The lot of other En-Geddi papyri: IEJ 12 (1962), 238Google Scholar.

²⁶ Bowersock, G. W., Zeitschr. f. Pap. u. Epig. 5 (1970), 44Google Scholar. And below, 231.

²⁷ cf. the discussion by A. Dihle, op. cit. 29. (On Annius Plocamus, see 27, n. 24.)

²⁸ ‘The Chronology of the Middle Nabataean Period’, PEQ 101 (1969), 5Google Scholar.

²⁹ ‘The Date of the Petra-Gaza Road’, PEQ 98 (1966), 89Google Scholar.

³⁰ Moa has been identified with Bīr Madkuūr (Alt, ZDPV [1935], 24) and with Moyet ‘Awād (Abel, , Géographie de la Palestine II [1938], 181)Google Scholar. Moahile has been placed at Qaṣr Maḥalle (Abel, op. cit., II. 182). This road also served for merchants going to Rhinocoloura (el-‘Arīsh), to which Strabo says the Nabataeans conveyed their goods from Petra (p. 781). After ‘Avdat the way would probably be by Nessana. On Nabataean commercial links with Egypt, note the traffic in bitumen from the Dead Sea for use in embalming: Hammond, P., ‘The Nabataean Bitumen Industry at the Dead Sea’, Biblical Archaeologist 22 (1959), 40Google Scholar. On Nabataean Nessana, see Colt, H. D., Excavations at Nessana I (1962)Google Scholar.

³¹ PEQ 101 (1969), 12Google Scholar.

³² Archaeology 14 (1961), 125Google Scholar.

³³ IEJ 13 (1963), 121Google Scholar; IEJ 17 (1967), 46Google Scholar. Also A. Negev, Cities of the Desert (1966), 23.

³⁴ The graffiti: Winnett, F. V., ‘Atiqot 2 (1959), 146Google Scholar; A. Jamme, op. cit. 150. Negev's interpretation of the graffiti: IEJ 13 (1963), 122Google Scholar; PEQ 101 (1969), 11Google Scholar. This interpretation was followed by N. Glueck in his Deities and Dolphins (1965), 527.

³⁵ Starcky, J. and Strugnell, J., ‘Deux nouvelles inscriptions nabatéennes’, RB 63 (1966), 236Google Scholar.

³⁶ Wright, G. R. H., ‘Structure of the Qaṣr Bint Far'ūn: A Preliminary Review’, PEQ 93 (1961), 8Google Scholar; cf. Parr, P. J., Ex Oriente Lux 19 (1965–1966), 555Google Scholar; Syria 45 (1968), IGoogle ScholarPubMed = ADAJ 11–12 (1967–1968), 5Google Scholar. Starcky, however, had dissented from the old communis opinio: SDB 977.

³⁷ Parr, , Ex Oriente Lux 19 (1965–1966), 556Google Scholar; Syria 45 (1968), 20Google Scholar = ADAJ 11–12 (1967–1968), 17Google Scholar.

³⁸ On the arched gate, Wright, G. R. H., ‘Structure et date de l'arc monumental de Pétra’, RB 73 (1966), 404Google Scholar; on the Khazneh, Wright, , ‘The Khazneh at Petra: A Review’, ADAJ 6–7 (1962), 24Google Scholar. See also Parr, P. J., ‘The Beginnings of Hellenization at Petra’, VIII^e Congrès international d'archéologie classique, Paris (1965), 527Google Scholar.

³⁹ Hammond, P., ‘The excavation of the Main Theater at Petra’, ADAJ 8–9 (1964), 81Google Scholar. See J. Starcky's perceptive comments on the flowering of Syro-Palestinian civilization in the age of Aretas IV: ‘La Civilization nabatéenne: état des questions’, IX^e Congrès international d'archéologie classique: Rapports et Communications, Damascus (1969), 22Google Scholar. Likewise (in German), Starcky, in Die Nabatäer, Catalogue of the Munich Stadtmuseum exhibition (1970), 81Google Scholar. On page 25 (French) and page 82 (German) Starcky compares with the Petra monuments the temple of Ba'āl Shamīn at Sia', the Herodian temple at Jerusalem, the temple of Juppiter at Damascus, the temple of Qōs at Tannūr, the temple of Bēl at Palmyra, the temple of Juppiter at Ba'ālbek, and (assigned to Rabbel II) the temple of Allāt on Jebel Ramm.

⁴⁰ Parr, P. J., ‘La date du barrage du Siq à Pétra’, RB 74 (1967), 45Google Scholar, or perhaps Roman, cf. p. 49.

⁴¹ Starcky, J., ‘Nouvelle épitaphe nabatéenne donnant le nom sémitique de Pétra’, RB 72 (1965), 95Google Scholar. Also in ADAJ 10 (1965), 44Google Scholar. The RB publication is more complete.

⁴² Strabo, p. 784.

⁴³ Ch. Clermont-Ganneau, , Études d'archéologie orientale I (1895), 146Google Scholar.

⁴⁴ Note κάπαρ on a boundary-stone: ‘Atiqot 2 (1959), 152Google Scholar.

⁴⁵ Wright, G. R. H., ‘Strabo on Funerary Customs at Petra’, PEQ 101 (1969), 113Google Scholar.

⁴⁶ In SDB 903 and 906.

⁴⁷ ‘Die Zeit des Schriftstellers Uranius’, Arch. f. Rel. III (1908), 239Google Scholar; cf. SDB 906. See also J. Pirenne, La Royaume Sud-arabe de Qatabān et sa datation (1961), 128.

⁴⁸ Steph. Byz. s. v. Μωθώ.

⁴⁹ Obodas the god: e.g. CIS II, 354 (Petra). The ‘Avdat temple: Archaeology 14 (1961), 125Google Scholar; IEJ 17 (1967), 55 [Zeus Obodas]Google Scholar.

⁵⁰ See above, n. 19. Note Strabo, p. 819: εἰ δὲ μὴ ὁ Συλλαῖος αὐτὸν (i.e. Γάλλον) προυδίδου, κἄν κατεστρέψατο τὴν εὐδαίμονα πᾶσαν. Cf. p. 780 : (Augustus) προσοικειοῦσθαι δὴ διενοήθη τούτους ἤ καταστρέφεσθαι.

⁵¹ Strabo, p. 780; Jos, , AJ 16, 273Google Scholar. On his way to Rome to see Augustus Syllaeus left behind at least two dedications, both bilingual (Nabataean and Greek): at Miletus (Cantineau, , Le Nabatéen II [1932], 46Google Scholar) and at Delos (unpublished, but see SDB 913).

⁵² ILS 140, ll. 9–10.

⁵³ Dio 55, 10a, 5, which is explicit. Dio's notice that Gaius was consul in Syria (ἔν τε τῇ Συρίᾳ ὄντα καὶ ὑπατεύοντα) does not fit with the statement of the cenotaph; but Dio is simply reporting the news that Phraataces heard, presumably that Gaius had arrived in Syria and become consul. He cannot be assumed to have passed his entire consulate there: the other evidence is decisive.

⁵⁴ Pliny, , NH 6, 141Google Scholar; 12,56; 32,10; cf. FGrH in A 17s, F 1–3.

⁵⁵ Pliny, , NH 6, 160Google Scholar. S. Jameson apparently forgot about Gaius' presence in this area when she wrote that after Aelius Gallus ‘the region attracted no further interest until the time of Nero’ (JRS 58 [1968], 79Google Scholar).

⁵⁶ ibid. As K. Wellesley has proved, there is no possibility that Gaius attacked Aden (as some inferred from the Periplus of the Red Sea 26): ‘The Fable of a Roman Attack on Aden’, Par. del Pass. 9 (1954), 401Google Scholar. J. Innes Miller, The Spice Trade of the Roman Empire (1969), 254 even says that Gaius sailed around the Arabian peninsula.

⁵⁷ Strabo, p. 780 : οὐδὲ κατὰ γῆν σφόδρα πολεμισταί εἰσιν; p. 781 (lack of interest in warfare) κοινὸν δὲ τοῦτο πᾶσι τοῖς Ἀράβων βασιλεῦσιν; Jos., , AJ 14, 31Google Scholar: οὐκ εὖ πρὸς πόλεμον διακείμενοι.

⁵⁸ See the researches of M. Gichon, notably ‘The Origin of the Limes Palestinae and the Major Phases in its Development’, Studien zu den Militärgrenzen Roms, Beiheft 19 d. Bonner Jahrb. (1967), 175. Gichon argues against the view of Avi-Yonah that there was no Palestine limes until the age of Diocletian, cf. also IEJ 17 (1967), 27Google Scholar; Provincialia: Festschr. Laur-Belart (1968), 317.

⁵⁹ IEJ 17 (1967), 47Google Scholar, citing inscriptions; cf. N. Glueck, Rivers in the Desert (1959), 201; A. Negev, Cities of the Desert (1966), 20. See also the Nabataean irrigation arrangements (water-rights) recorded in the En-Geddi papyri: IEJ 12 (1962), 243Google Scholar; Ex Oriente Lux 17 (1963), 232Google Scholar. For a professional reconstruction of an ancient desert farming-system, see M. Evenari and others, The Negev: The Challenge of the Desert (1971). cf. also the agricultural regime at Byzantine Nessana: Meyerson, P. in Excavations at Nessana I (1962), 211Google Scholar.

⁶⁰ He appears in the En-Geddi papyri: IEJ 12 (1962), 239Google Scholar. Cited also in Ex Oriente Lux 17 (1963), 230Google Scholar.

⁶¹ cf. n. 21 above.

⁶² Negev, A., PEQ 101 (1969), 14Google Scholar.

⁶³ N. Glueck, Deities and Dolphins (1965), 3, 45.

⁶⁴ Amm. Marc. 14, 8, 3.

⁶⁵ ibid.

⁶⁶ Semitic Inscriptions, Part IV of the Publications of an American Archaeological Expedition to Syria in 1899–1900 (1904), 143, no. 45. Professor F. V. Winnett has generously shown me an unpublished Safaitic text containing the same three words. It was found in 1958–9 by Winnett and Harding north of the H-4 pumping station and will be published by them as W-H no. 2113.

⁶⁷ E. Littmann, Thamūd und Safā (1940), 122, no. 14; R. Dussaud, La Pénétration des Arabes en Syrie avant l'Islam (1955), 139. Professor Winnett has also shown me another of his unpublished Safaitic texts (see note 66) and allowed me to make it public. It will be numbered W-H no. 1734 = 2815 and mentions SNT MRDT NBṬ ‘L 'L RM, ‘the year of the Nabataean revolt against the people of Rome.’ Unfortunately, the R of RM is not a certain reading, and a secure date cannot be inferred from the script.

⁶⁸ cf. Préaux, C., Phoibos 5 (1950–1951), 132Google Scholar.

⁶⁹ On milestones of C. Claudius Severus: PIR ² C 1023.

⁷⁰ Described with quotations in IEJ 12 (1962), 235Google Scholar and Ex Oriente Lux 17 (1963), 227Google Scholar. Four of the least interesting of the papyri have been published with Hebrew commentary by Polotsky, H. J. in Eretz-Israel 8 (1967), 46Google Scholar. These are documents nos. 12, 27, 28, 29. Note also that the third period of building at Khirbet Tannūr belongs to the time of the early province: N. Glueck, Deities and Dolphins (1965), 138.

⁷¹ Doubted by Brünnow and Domaszewski, vol. III, 268 and by A. Kammerer, Pétra et la Nabatène (1929), 286.

⁷² cf. recently Negev, A., ‘New Dated Nabataean Graffiti from the Sinai’, IEJ 17 (1967), 252Google Scholar.

⁷³ Rothenberg, B., PEQ 102 (1970), 18Google Scholar.

⁷⁴ Bietenhard, H., ‘Die Dekapolis von Pompeius bis Traian’, ZDPV 79 (1963), 44Google Scholar.

⁷⁵ Répertoire d'fipig. Sém. II, 1128 (cf. II, 1175). See also the graffiti of soldiers in Syria 22 (1941), 219Google Scholar.

⁷⁶ Archaeology 22 (1969), 139 and 325Google Scholar.

⁷⁷ A. Musil, The Northern Hegaz (1926), 185.

⁷⁸ op. cit., 258.

⁷⁹ H. St. J. Philby, The Land of Midian (1957), 146, cf. 154. On the bilingual text see also A. Grohmann. Arabien (1963), 73. This inscription is due to receive a definitive publication from Milik. For a provisional text, see Teixidor, J., Syria 47 (1970), 378Google Scholar; cf. Altheim-Stiehl, op. cit. (n. 23), V/2 (1969), 24 and figs. 2–5.

⁸⁰ Syria 34 (1957), 259Google Scholar.

⁸¹ Parr, P. J., ‘Exploration archéologique du Hedjaz et de Madian’, RB 76 (1969), 390Google Scholar. A full report of this exploration is appearing in the Bulletin of the London Institute of Archaeology: for Part I, see Bulletin (1970), 193.

⁸² IEJ 12 (1962), 258Google Scholar; cf. Bowersock, G. W., Zeitschr.f. Pap. u. Epig. 5 (1970), 39Google Scholar.

⁸³ Yadin, Y., IEJ 12 (1962), 257, n. 52Google Scholar; with more detail, Ex Oriente Lux 17 (1963), 234Google Scholar; cf. G. W. Bowersock, op. cit., 44. It will be seen that I now maintain a still more cautious attitude to Yadin's proposals than in my earlier study.

⁸⁴ The new name is widely attested on the coins and inscriptions of the city. For Bostra as caput viae in the north, see Mittmann, S., ZDPV 80 (1964), 113 [German]Google Scholar = ADAJ 11 (1966), 65 [English]Google Scholar, cf. also Beyer, G., ZDPV 58 (1935), 143Google Scholar.

⁸⁵ Michigan Papyri, vol. 8 (1951), no. 466, l. 48Google Scholar.

⁸⁶ Milik, J. T., Syria 35 (1958), 244Google Scholar.

⁸⁷ Harding, G. Lankaster, PEQ 90 (1958), 14Google Scholar; Kirkbride, D., ADAJ 4–5 (1960), 120Google Scholar; Bennett, C. M., Archaeology 15 (1962), 243Google Scholar; J. Starcky, SDB 948.

⁸⁸ One should not forget Strabo, p. 779: μητρόπολις δὲ τῶν Ναβαταίων ἐστιν ἡ Πέτρα καλουμένη, cited by Starcky, and Bennett, in Syria 45 (1968), 53, n. 1Google Scholar = ADAJ 12–13 (1967–1968), 39, n. 40Google Scholar à propos an inscription from Petra mentioning Ἁδριανὴ Πέτρα μητρόπολις.

⁸⁹ G. W. Bowersock, op. cit., 40.

⁹⁰ Michigan Papyri, vol. 8 (1951), nos. 465 and 466Google Scholar.

⁹¹ Préaux, C., ‘Une source nouvelle sur l'annexion de l'Arabie par Trajan’, Phoibos 5 (1950–1951), 123Google Scholar.

⁹² R. Dussaud, La Pénétration des Arabes en Syrie avant l'Islam (1955), 154; Syme, R., Historia 14 (1965), 353, n. 53Google Scholar; Petersen, L., Klio 48 (1967), 160 with n. 3Google Scholar.

⁹³ Préaux, op. cit. 127. The evidence is CIL III, 14147⁶ (with the correct date given there).

⁹⁴ cf. A. Stein, Die Reichsbeamten von Dazien (1944), 54; Smallwood, E. M., JRS 52 (1962), 131Google Scholar. Hence the problem raised in JRS 48 (1958), 4 vanishesGoogle Scholar. Oὑπατικός just means governor.

⁹⁵ III Cyrenaica in 119: BGU 1. 140. II Traiana in 127: CIL III, 42. The Michigan documents (note 90) imply that Apollinarius was recruited in Egypt. It might therefore be argued that he was serving with a formerly Egyptian force now in Arabia (i.e. all or part of III Cyrenaica). This, however, would mean either cutting the Egyptian garrison in half or leaving the new Arabian province without a legion; neither possibility seems likely. Further, Egyptians are found in other eastern legions than those in Egypt: note 22 of them in X Fretensis in 125/6 (PSI IX, 1026 c).

⁹⁶ cf. Bowersock, op. cit., 43.

⁹⁷ Negev, A., ‘Seal Impressions from Tomb 107 at Kurnub (Mampsis )’, IEJ 19 (1969), 89Google Scholar; also Negev in Die Nabatäer, Catalogue of the Munich Stadtmuseum exhibition (1970), 40.

⁹⁸ Negev, , IEJ 19 (1969), 90Google Scholar.

⁹⁹ IEJ 12 (1962), 260Google Scholar. It is worth remembering that there are no coins of Characmoba before Elagabalus.

¹⁰⁰ Ben-Dor, S., ‘Petra Colonia’, Berytus 9 (1948), 41Google Scholar.

¹⁰¹ Ben-Dor, ibid., 43.

¹⁰² The evidence is set forth in FGrH III A 281.

¹⁰³ Stein, A., ‘Kallinikos von Petrai’, Hermes 58 (1923), 448Google Scholar. Stein argued from Callinicus' πρὸς Κλεόπατραν περὶ τῶν κατ' Ἀλεξάνδρειαν ἱστοριῶν βιβλία δέκα. The only evidence that Zenobia took the name Cleopatr a is in the Historia Augusta, Aug. 27, 3; Prob. 9, 5 (Corp. Pap. Jud. III, 1449 is not secure).

¹⁰⁴ Suid. s. v. Гέσιος. cf. Bernays, J., ‘Ein nabatäischer Schriftsteller’, Ges. Abhand. II (1885), 291Google Scholar, an expanded version of Rh. Mus. 17 (1862), 304Google Scholar. Note also the brisk activity in Negev cities during the Byzantine era: N. Glueck, Rivers in the Desert (1959), 252; A. Negev, Cities of the Desert (1966), 33. On life in Byzantine Nessana, see Kraemer, C. J. Jr., Excavations at Nessana, vol. 3 (1958)CrossRefGoogle Scholar: non-literary papyri.

¹⁰⁵ Parr, P. J., PEQ 89 (1957), 15Google Scholar. Some of the fragments had been found previously by George and Agnes Horsfield during their pioneering investigations at Petra : cf. Quarterly of the Dept. of Antiquities in Palestine 7 (1938), 1Google Scholar; 8 (1938), 87; 9 (1942), 105.

¹⁰⁶ ibid. For the vigour of Petra, note the panegyriarchs delegated from Adraa to Petra: inscriptions in Brünnow-Domaszewski, vol. I, 220 (cf. the festival at Petra in the fourth century as reported by Epiphanius, Panarion II, 51, 22).

¹⁰⁷ On Adraa (and danger from desert Arabs as well as Palmyrenes) see Pflaum, H.-G., ‘La Fortification de la ville d'Adraha d'Arabie (259–60 à 274–5) d'après des inscriptions récemment découvertes’, Syria 29 (1952), 307Google Scholar. On Palmyrene destruction of a temple at Bostra see Seyrig, H., Syria 22 (1941), 46Google Scholar. cf. Malalas, Chron. 299 (Bonn) on Zenobia and Arabia.

¹⁰⁸ On second and early third century Gerasa: C. H. Kraeling, Gerasa: City of the Decapolis (1938), 52. There are many Hadrianic inscriptions from the year of Hadrian's visit: Welles apud Kraeling, op. cit. inscriptions no. 30, 58, 143–5. Colonia Antoniniana: inscriptions no. 179, 191—perhaps this honour was also due to Elagabalus, just as in the cases of Petra and Bostra.

¹⁰⁹ Syria 45 (1968), 56Google Scholar [French] = ADAJ 12–13 (1967–1968), 42 [English]Google Scholar. Refuting this, Bowersock, G. W., Zeitschr. f. Pap. u. Epig. 5 (1970), 41Google Scholar.

¹¹⁰ Bowersock, op. cit., 45.

¹¹¹ IEJ 12 (1962), 259Google Scholar.

¹¹² ibid.

¹¹³ ibid.

¹¹⁴ CIL III, 6028.

¹¹⁵ Above, n. 94.

¹¹⁶ Vol. I, 470.

¹¹⁷ The main highway goes on to ‘Ain Sadaqa. Thomsen, (ZDPV 40 [1917], 35)Google Scholar and Abel, following him (Géographie de la Palestine II [1938], 229Google Scholar), consider the line from Doshak to Petra a branch road which rejoins the putative main road at ‘Ain Ṣadaqa. This is difficult to credit in view of Petra's importance, not to mention that two of the stops on the alleged branch road (‘Ain Nejl and Petra) appear on the Peutinger Table on the main line south. In addition, Petra is the caput viae of the via nova Traiana in the south: cf. ZDPV 58 (1935), 129Google Scholar. Need one say more?

¹¹⁸ Limes congresses have not shown a particular interest in the limes Arabiae, although work on the limes Palestinae is sometimes relevant.

¹¹⁹ Explorations of Eastern Palestine: see above, n. 11.

¹²⁰ Thomsen, P., ‘Die römischen Meilensteine der Provinzen Syria, Arabia, und Palestina’, ZDPV 40 (1917), 1Google Scholar.

¹²¹ A. Poidebard, La Trace de Rome dans le désert de Syrie (1934).

¹²² See plate XV, 1 and 2.

¹²³ Stein, A., ‘Surveys on the Roman Frontier in Iraq and Trans-jordan’, Geographical Journal 95 (1940), 428Google Scholar. The aerial photographs are mentioned on page 438.

¹²⁴ cf. Thomsen, op. cit. (n. 120), with excellent map. El Hadῑd is often identified with Adittha of the Notitia (Hatita in the Tab. Peut.). See also Butler's study, ‘Trajan's Road from Boṣrā to the Red Sea’, in the Princeton Arch. Exped. Publications III, A, 2 (1911).

¹²⁵ Mittmann, S., ‘Die römische Strasse von Gerasa nach Adraa’, ZDPV 80 (1964), 113Google Scholar [German] = ADAJ II (1966), 65 [English]Google Scholar.

¹²⁶ Hadrianic milestones: op. cit., 123 in ZDPV and 74 in ADAJ. Aelius Severianus: op. cit., 126 in ZDPV and 77 in ADAJ. Furnius Julianus: op. cit., 127 in ZDPV and 78 in ADAJ.

¹²⁷ AASOR 15 (1935), 71Google Scholar.

¹²⁸ Stein, op. cit. (n. 123), 437.

¹²⁹ Musil, The Northern Hegaz (1926), 53; Savignac, , RB 41 (1932), 595Google Scholar; Frank, , ZDPV 57 (1934), 235Google Scholar; Glueck, , AASOR 15 (1935), 65Google Scholar; Stein, op. cit. (n. 123), 437. Stein, on p. 436, mentions that his surveyor prepared a map of the ancient road on a scale of 4 miles to 1 inch: the map survives in the Stein Collection at the Royal Geographical Society.

¹³⁰ Apud Brünnow-Domaszewski, vol. I, 477.

¹³¹ e.g., Musil, op. cit. (n. 129), 59; Stein, op. cit. (n. 123), 437. The name of this place is vocalized in some works as Ahmeime, which has induced a false belief that the spot is the site of the ancient Ammatha.

¹³² cf. Savignac, op. cit. (n. 129); Stein, op. cit. (n. 123), 437; Glueck, , AASOR 18–19 (1939), 15Google Scholar.

¹³³ Glueck, , AASOR 15 (1935), 58Google Scholar (plan of the fort on 173). Cf. Jaussen (visit of 1902), quoted apud Brünnow-Domaszewski, vol. I, 473: ‘Au Naqb Eštar seulement, j'ai pu distinguer une voie antique à côté du chemin actuel’.

¹³⁴ Apud Brünnow-Domaszewski, vol. I, 478.

¹³⁵ Musil, op. cit. (n. 129), 58; Frank, op. cit. 235.

¹³⁶ AASOR 15 (1935), 65Google Scholar.

¹³⁷ Stein, op. cit. (n. 123), 437.

¹³⁸ AASOR 15 (1935), 68Google Scholar.

¹³⁹ ZDPV 57 (1934), 235Google Scholar; AASOR 15 (1935), 70Google Scholar.

¹⁴⁰ Stein, op. cit. (n. 123), 438.

¹⁴¹ Its name implies a connection with the Jebel Shabῑb in the south and the Qaṣr Shabῑb near Zarqᾱ. All these are probably traces of the medieval Shabῑb ibn Tubbai, ruler of the land from below Ma‘ān to Mt. Hermon (cf. Doughty apud Brünnow-Domaszewski, vol. II, 223).

¹⁴² Stein, op. cit. (n. 123), 437; Aharoni, , IEJ 13 (1963), 41Google Scholar.

¹⁴³ Near ‘Ain el Hawwᾱre: ZDPV 58 (1935), 24Google Scholar.

¹⁴⁴ Musil, op. cit. (n. 129), 59, n. 20.

¹⁴⁵ After Oboda a road will have diverged to Nessana, from there presumably on to Rhinocoloura.

¹⁴⁶ cf. IEJ 17 (1967), 55Google Scholar: the earliest sign of resettlement is a burial dated to 242.

¹⁴⁷ On the camp at ‘Avdat: ibid., 47; also A. Negev, Cities of the Desert (1966), 18. Dr. Negev assigns considerable importance to the Nabataean army, but the issue is a delicate one: exclusively Nabataean sherds at the ‘Avdat camp are no proof that this was not a Roman camp. The role of the attested Nabataean 'SRTG' (στρατηγός)—cf., e.g., CIS II, 196—may not have been very militaristic. See n. 57 above on the unwarlike character of the Nabataeans. What did the RB MŠRYT' (‘chief of the camp[s]’) in CIS II, 196 do? The army supplied contingents (e.g. to Caesar [Bell. Alex. I] and Aelius Gallus [Strabo, p. 780]), but that does not mean it was more than a police force to keep the routes open.

¹⁴⁸ Ad Dianam is ‘Ain Ghadiᾱn, and Aridela is ᾱin Gharandel; cf. Alt, , ZDPV 58 (1935), 24Google Scholar. Toponymy is conclusive.

¹⁴⁹ See above, p. 225.

¹⁵⁰ Rothenberg, B., ‘An Archaeological Survey of South Sinai’, PEQ 102 (1970), 4Google Scholar. The account of the road begins on page 18.

¹⁵¹ ibid., 20.

¹⁵² cf. n. 72 above.

¹⁵³ See, for example, the inscriptions in R. Dussaud and F. Macler, Rapport sur une mission scientifique dans les régions désertiques de la Syrie moyenne, in Nouvelles Archives des Missions scientifiques et littéraires 10 (1903), 411Google Scholar.

¹⁵⁴ Stein, op. cit. (n. 123), 433.

¹⁵⁵ Stein's error is proven by the Arabic inscription dating the Qaṣr Burqu' to A.D. 700: Sauvaget, J., Journal Asiatique 231 (1939), 23Google Scholar and F. E. Day apud Henry Field, North Arabian Desert Archaeological Survey, 1925–50 (Papers of Peabody Museum, vol. 45, no. 2), 1960, 154. Glueck's observations: AASOR 25–28 (1951), 32Google Scholar. A Greek inscription was found at Burqu': S. Dow apud Field, op. cit. 161.

¹⁵⁶ Dussaud and Macler, op. cit. (n. 153), 670 no. 85 = IGR III, 1339.

¹⁵⁷ Stein, op. cit. (n. 123), 434. Note also a milestone reported to have been at Azraq in Antiquity 3 (1929), 89Google Scholar: Group Capt. Rees said he ‘rescued’ the stone.

¹⁶⁸ Constans, whose name would appear last in a listing of three Caesars, is obviously not included. I am particularly grateful to T. D. Barnes for his acute suggestions concerning this inscription.

¹⁵⁹ Stein, op. cit. (n. 123), 434.

¹⁶⁰ ibid.

¹⁶¹ cf. N. Glueck, The Other Side of the Jordan (1940), 41.

¹⁶² An inscription of the year 213, mentioning the governor Furnius Julianus, apud Brünnow-Domaszewski, vol. III, 291 (Princeton Arch. Exp. Syria III, A, 2, 21, no. 17). On the Arab desert buildings, see above all Sauvaget, J., ‘Remarques sur les monuments omeyyades,’ Journal Asiatique 231 (1939), 1Google Scholar.

¹⁶³ There may have been a short time when the south was detached from the north but was still distinct from Palestine; cf. below, and chiefly Brünnow-Domaszewski, vol. III, 277.

¹⁶⁴ Vol. III, 266. Against: M. Avi-Yonah, The Holy Land: A Historical Geography (1966), 170.

¹⁶⁵ For recently discovered Christian inscriptions at Bostra: Ann. Arch. Syrie 15 (1965), 73Google Scholar. Nabataean epigraphy: Rosenthal suggests from lettering that the Nessana inscriptions in Nabataean belong to 150–350 (apud Colt, H. D., Excavations at Nessana I [1962], 201)Google Scholar. This would make them among the latest Nabataean texts known. (A new Nabataean inscription which appears dated to A.D. 356 will be the latest so far discovered: Beiträge zur alten Geschichte: Festschrift für Altheim [1970], II, 87.Google Scholar) Note that in the fourth century Epiphanius says that certain cultic observances at Petra were conducted Ἀραβικῇ διαλέκτῳ and Ἀραβιστί (Panarion II, 51, 22). That may mean Arabic, not Nabataean.

¹⁶⁶ Alt, A., ‘Augusta Libanensis’, ZDPV 71 (1965), 173Google Scholar, against W. Seston, Dioclétien et la tétrarchie (1946), 373.

¹⁶⁷ ‘Atiqot 2 (1959), 153Google Scholar.