¹ The Turkish government representative was Ilhan Güceren of Isparta Museum, who contributed enormously to the success of the project. I am grateful also to Ümit and Gül Işın (Antalya) for their help, especially concerning Pednelissos, and to local officials in Korkuteli and Bucak for invaluable practical assistance. The Monuments and Museums Department of the Ministry of Culture swiftly and expeditiously issued a permit when it appeared that my application had gone astray. The cost of the survey was covered by a grant from the British Academy.

² Ptolemy V. 5. 1 lists Κρητόπολις at the head of the list of towns of Kabalia, at 61° 15′ and and 37° 30′. Early conjectures about its location are listed in the edition of C. Müller, vol. II (1900), 862. See the map produced by Bean, G. E., AS X (1960), 54 Fig. 2aGoogle Scholar.

³ Diodorus XVIII.41.1: (Eumenes) κατελάβετο χωρίον ὄχυρον ὅ προσηγορεύετο Νῶρα Strabo XII.2.6, 537: τὰ Νῶρα ὅ νῦν καλεῖται Νηροασσός, ἐν ᾧ Εὐμένης πολιορκούμενος ἀντέσχε πόλυν χρόνον.

⁴ The identification was firmly proposed by Ruge, W., RE XVII. 1 (1936), 923–4Google Scholar and endorsed by Magie, D., Roman Rule in Asia Minor II, 1096Google Scholar n. 6. See now Hild, F. and Restle, M., Tabula Imperii Byzantini II: Kappadokien (1981), 245–6Google Scholar with a site plan and description which admirably conforms with Diodorus' description (which they do not cite): “This fortress was very small with a circuit of not more than two stades, but of wonderful strength, for its buildings had been constructed close together on the top of a lofty crag, and it has been marvellously fortified, partly by nature, partly by the work of men's hands” (Loeb. trans.).

⁵ Diodorus XVIII.44.2: ὀξεῖαν δὲ καὶ παντελῶς ἐπι τετασμένην τὴν πορείαν ποιησάμενος ἐν ἡμέραις ἑπτὰ καὶ ταῖ ςἴσαις νυξὶ διήνυσε σταδίους δισχιλίους καὶ πεντακοσίουςεἰς τὴν ὀνομαζομένην Κρητῶν πόλιν.

⁶ The battle is also recalled by Polyaenus, , Strategemata IV.6. 7Google Scholar. The relationship of this description to Diodorus' account is discussed by Engel, R., Historia 21 (1972), 501–7Google Scholar.

⁷ Engels, D. W., Alexander the Great and the Logistics of the Macedonian Army (1978), 157–8Google Scholar discusses the bematists and the accuracy of the measurements relating to Alexander's own campaigns. On p. 155 he demonstrates the absurdity of the notion that “an army of 47,000 combatants marching with elephants can achieve a march rate of 40 miles per day for seven consecutive days”.

⁸ Polybius V.57.5–8. In 220 Achaios had declared himself king at Laodicea at Garsyeris' instigation and led an expedition against Antiochos II, who was then in Syria. His troops mutinied as they approached Lycaonia, and Achaios appeased them with booty plundered from Pisidian territory before he returned to base. No doubt his route to the east had been via Apamea, Apollonia and Pisidian Antioch (all Seleucid foundations) in the direction of Konya and the Lycaonian plain. The plundered territory in question would doubtless have been the northern part of Pisidia adjoining this route through Phrygia Paroreius. According to an inscription of 268/7 B.C. Achaios' likely grandfather, also called Achaios, owned a large property, including two villages, in the immediate neighbourhood of Laodicea (Wörrle, M., Chiron 5 (1975), 59–87Google Scholar, esp. 77). The foundation of Laodicea itself, almost certainly by Antiochos, II (RE XII, 727Google Scholar s.v. Laodikeia 5) may have provoked Achaios' own bid for power in 220, and this was launched from an area where his family exercised strong local influence.

⁹ Hall, A. S., AS XXXVI (1986), 137–57Google Scholar at 146. The most valuable ancient definition of the extent of the Milyas is that of Strabo XIII.4.17, 631, who says that it was the mountainous territory which stretched from the defile by Termessos and the passage through it which led to Isinda as far as Sagalassos and the territory of Apamea.

¹⁰ Polybius V.72.4: It would be logical to look for Saporda along the route of the modern road up from the Pamphylian plain which runs west of Mercimek Dağ through the Çübuk Boğazı.

¹¹ AS X (1960), 51 no. 100Google Scholar.

¹² See Levick, B., Roman Colonies in Southern Asia Minor (1967), 146Google Scholar; her comments on the Latin inscriptions erected for members of the family of the Paccii are to be supplemented by the publication of another text by David French, in Mitchell, S. (ed.), Armies and Frontiers in Roman and Byzantine Anatolia (1981), 48–51Google Scholar. In fact, as Levick points out, the standard of Latin (to judge by the inscribed sample) was in sharp decline by the later second century A.D., and the preponderance of gravestones is in Greek.

¹³ See Denniston, J. D., The Greek Particles² (1954), 341Google Scholar, noting especially his section 4 on the “progressive” use of ἀλλὰ μήν.

¹⁴ See Levick, B., RE suppl. XI (1968), 859–71Google Scholar s.v. Comama for full documentation. Like Cremna, it appears to have minted as an independent community before the foundation of the Roman colony.

¹⁵ See J., and Robert, L., Bull. ép. 1978, 501 pp. 489–91Google Scholar; cf. Mitchell, S., in Schwertheim, E. (ed.), Forschungen in Pisidien, Asia-Minor-Studien 6 (1992), 6Google Scholar n. 17. The site is much more extensive than the published descriptions imply, but badly damaged by illegal excavation.

¹⁶ von Aulock, H., Münzen und Städte Pisidiens I (1977), 41 and 114–17Google Scholar nos. 1121–47. Kayhan Dörtlük informs me that many fine specimens of the mint, including unpublished types, are in the museum collections of Antalya and Burdur.

¹⁷ See Ruge, , RE XVIII.3 (1949), 586–7Google Scholar s.v. Panemoteichos, collecting and discussing the evidence.

¹⁸ Synekdemos 681 lists Pisidian communities in the following order: Kolbasa, Kremna, Panemouteichos, Ariassos, Maximianupolis, Ktema Maximianupolis, regio Salamara, demos Obrama, Kodrula, demos Sia, demos Sabaôn, Pednelissos, Selge.

¹⁹ De Thematibus I. xiv, 37 (ed. Pertusi, )Google Scholar: (the boundary) συνάπτει δὲ πρὸς τὴν Τλῶ καὶ Οἰνίανδα, εἶτα παρέρχεται Φίλητα καὶ αὐτὴν Ποδαλείαν; διεκτρέχει δὲ τὸκαλούμενον Ἀνεμότειχος καὶ συνάπτει πρὸς τὴν πόλιν Σαγαλασσόν It is remarkable that such obscure places were identified as marking a Theme boundary. Perhaps Panemoteichos and Podalia had been chosen to house some of the Thracian military settlements, which gave the Thrakesian theme its name; cf. Ostorgorsky, G., History of the Byzantine State² (1968), 95–100Google Scholar. His map, p. 96, and the more detailed one in the Tübinger Atlas des Vorderen Orients. B VII 18: Kleinasien. Die Erweiterung des byzantinischen Reiches im 10. und 11. Jahrhundert mark the boundary in question too far to the west.

²⁰ Bean, G. E., Anzeiger der Akademie in Wien 105 (1968), 157–63Google Scholar; Journeys in Northern Lycia 1965–67, Denkschr. Ak. Wien 104 (1971), 28–32Google Scholar; Jameson, S., RE suppl. XIV, 412–13Google Scholar, overlooking the reference to Constantine Porphyrogenitos.

²¹ Ruge, , RE XVIII. 3, 586Google Scholar, citing Schwartz, , Acta Conciliorum Oecumenicorum II. v. 58Google Scholar.

²² Ramsay, W. M., AJA 4 (1888), 266Google Scholar; compare his conjecture in BCH 7 (1883), 267Google Scholar that Cretopolis should be identified with the site at the north end of the Döşeme Boğazı. Radet, G., Revue archéologique 1888 II, 187–8Google Scholar. Radet's description is far from lucid and probably impenetrable to anyone who does not know the topography, but it emerges that he is distinguishing the Döşeme from the Çubuk Boğazı. He envisaged that Panemoteichos was at the top of the former, in a comparable position to Ariassos at the top of the Çubuk pass. See futher Ruge, , RE XVIII, 588Google Scholar.

²³ Rott, H., Kleinasiatische Denkmäler (1908), 28Google Scholar; Moretti, G., Annuario della scuola archeologica di Atene 6–7 (1923–1924), 555–6Google Scholar. The confusion of these early descriptions, and the unavailability of good maps led even Ruge into error, RE XVIII. 2, 1584–5Google Scholar s.v. Osia, where he uncharacteristically confuses the site described by Rott (which was resurveyed in November–December 1993 by David French) with that of (O)Sia, which was discovered by Bérard, V., BCH 16 (1892), 435Google Scholar with inscriptions nos. 66–9, in the hills to the north-east near Karaot (see further Bean, , AS X (1960), 74Google Scholar and Mitchell, , Asia-Minor-Studien 6 (1992), 17–19Google Scholar).

²⁴ Honigmann, E., Le Synekdèmos d' Hiérokles, 30, 681.3Google Scholar; rejected by Ruge, , RE XVIII. 3, 588Google Scholar.

²⁵ Hyia: Bean, G. E., AS X (1960), 81Google Scholar no. 134; Apollo sanctuary: Robert, L., BCH 107 (1983), 576–7CrossRefGoogle Scholar.

²⁶ Apart from the coins, which all preserve the spelling Πανεμοτ(ε)ιχ(ε)ιτῶν for the ethnic, there is considerable confusion about the name in the ecclesiastical sources, and forms beginning Panemu- or Πανεμον)- tend to predominate (see Ruge, , RE XVIII. 586–7Google Scholar); but note Panemotichi in Acta Conc. Oecum. II. v. 58Google Scholar and Ἀνεμότειχος in the De Thematibus.

²⁷ Compare AS XXXV (1985), 52Google Scholar no. 1 (Tymbriada), with note.

²⁸ Mitchell, S., Anatolia. Land, Men, and Gods in Asia Minor I (1993), 217–25Google Scholar.

²⁹ See especially Robert, L., in de Gagnières, J., Laodicée du Lycos (1969), 317–19Google Scholar; in Pisidia, cf. Nollé, J., I. Selge (IGSK 37), 70Google Scholar no. 2 and note.

³⁰ Magie, D., Roman Rule in Asia Minor II, 1516–17Google Scholar n. 48.

³¹ Schallmayer, E. et al. , Der römische Weihbezirk von Osterbürken I. Corpus der griechischen und lateinischen Beneficiarier-Inschriften des römischen Reiches (1990), nos. 697, 698, 708 and 726Google Scholar for examples of the designation οὐέτρανος ἀπὸ βενεϕικιαρίου in Greek inscriptions dating from the second and third centuries; the last, from Bostra in Arabia, was also a local councillor. There are many more examples in Latin inscriptions of veterani ex beneficiariis.

³² See Heberdey, R., Termessische Studien, Denk. Ak. Wien 69 (1929)Google Scholar; Devijver, H., “The Inscriptions of the Neon-Library of Roman Sagalassos”, in Waelkens, M. (ed.), Sagalassos II (1994), 107–23Google Scholar.

³³ See Halfmann, H., “Die Senatoren aus den kleinasiatischen Provinzen”, in Epigrafia e ordine senatorio (1982), 641–3Google Scholar. All the senators which he identifies from Pisidian cities belong in the third century. For the conjecture that L. Rutilius Rufus, cos. suff. in A.D. 125 came from Cremna, see S. Mitchell, in Cremna in Pisidia. An Ancient City in War and Peace, forthcoming.

³⁴ These texts will be edited in a forthcoming corpus of the inscriptions of Cremna by G. H. R. Horsley and S. Mitchell, as nos. 26, 27 and 44.

³⁵ Senators: Marcellus, Aufidius Coresnius, IGR III 357Google Scholar; cf. 367 = ILS 8835; M. Ulpius Calippianus (perhaps related to the M. Ulpii at Cremna), Eck, W., RE suppl. XIV, 936Google Scholar; Proculus, Cl. Dometillianus, IGR III 356Google Scholar. For Terentius Marcianus, see Mitchell, S., in French, D. H. and Lightfoot, C. S. (eds.), The Eastern Frontier of the Roman Empire I (1989), 322–3Google Scholar.

³⁶ IGR III 392Google Scholar (cf. Eck, RE suppl. XIV, 942)Google Scholar; Halfmann, loc. cit. (n. 32) suggests a relationship with Quirinia Patra, the well-connected wife of Bryonianus Lollianus of Side.

³⁷ Alföldy, G., ZPE 34 (1979), 279 fGoogle Scholar.

³⁸ Carrié, J.-L., ZPE 35 (1979), 213–24Google Scholar.

³⁹ Nollé, I., Chiron 16 (1986), 202Google Scholar; Mitchell, , Anatolia I, 238Google Scholar with refs.

⁴⁰ See Brandt, H., Gesellschaft and Wirtschaft Pamphyliens und Pisidiens im Altertum, Asia-Minor-Studien 7 (1992), 150–60, 192–7Google Scholar.

⁴¹ It should be noted, however, that this simple headgear is not an example of the elaborate “Büstenkronen”, decorated with busts of deified members of the imperial family, which are particularly characteristic headgear for the high priests of the imperial cult in Asia Minor; see Inan, J. and Alföldy-Rosenbaum, E., Römische und frühbyzantinische Porträtplastik aus der Türkei. Neue Funde (1977), 38–47Google Scholar, Wórrle, M., Stadt und Fest im kaiserzeitlichen Kleinasien (1988), 187–8Google Scholar, and Price, S. R. F., Rituals and Power. The Imperial Cult in Roman Asia Minor (1984), 170–1Google Scholar.

⁴² For comparative material and detailed discussion see Bergmann, Marianne, Studien zum römischen Porträt des 3. Jahrhunderts n. Chr. Antiquitas Reihe 3. Bd. 18 (1977)Google Scholar. For a selection of Asia Minor comparanda, see Inan, and Rosenbaum, , Roman and Early Byzantine Portrait Sculpture in Asia Minor (1966), nos. 85, 104, 251, 252, 274, 282Google Scholar and the material collected in their subsequent volume (see last note).

⁴³ There were four at Adada (Price, , Rituals and Power, 269)Google Scholar; both the main temples at Sagalassos were at least shared by the imperial cult (Price, , Rituals and Power, 270)Google Scholar. For imperial temples at Cremna, see S. Mitchell, Cremna in Pisidia, forthcoming. It may be no coincidence that the abundant archaeological and epigraphical remains of Termessus have yielded few clear traces of the imperial cult (see Price, 271), for the place was proud of its status as a free city. For bath houses as a hallmark of Roman cities in Asia Minor in the second and third centuries, see Mitchell, , Anatolia I, 216–17Google Scholar.

⁴⁴ For Ariassos, see Mitchell, S., AS XLI (1991), 159–60Google Scholar and Schulz, A., in Asia-Minor-Studien 6 (1992), 35–6Google Scholar.

⁴⁵ The most recent survey of the sites in the northern Zivint Ova is that of Bean, , AS X (1960), 51–74Google Scholar. For a comprehensive discussion of Isinda, see David French, in Studies in the History and Topography of Lycia and Pisidia. In Memoriam Alan S. Hall, British Institute of Archaeology at Ankara Monograph no. 19 (1994), 53–86.

⁴⁶ For the Tauros, , RE Va, 39Google Scholar s.v. Tauros 4; for the Kolobatos, , RE XI, 1107Google Scholar.

⁴⁷ French, D. H., in Studies in the History and Topography of Lycia and Pisidia, 90–105Google Scholar.

⁴⁸ Woodward, A. M. and Ormerod, H. S., BSA 16 (1909/1910) 90–105Google Scholar.

⁴⁹ Woodward, A. M., BSA 17 (1910/1911), 205Google Scholar no. 1, copied by “Mr. Nikolas Michael of Adalia”, who also provided the text of another inscription (206 no. 2), carved on a base set up for the fortune and victory of Marcus Aurelius Antoninus Severus Augustus” (i.e. Caracalla): (? three lines lost).

⁵⁰ CIL III 6885Google Scholar, cf. Ramsay, , JRS 6 (1916), 132Google Scholar. See further below n. 53.

⁵¹ Sagalassos, , Waelkens, M., AS 40 (1990), 190–3Google Scholar; Cremna, , Mitchell, S., AS XXXVI (1986), 44–5Google Scholar.

⁵² IGR III 385Google Scholar (from the architrave of a building, cf. Bean, , AS X (1960), 77Google Scholar n. 72) and 386; I recopied both texts in 1993.

⁵³ Voconius Saxa, who was honoured at Comama (see n. 50), Perge (Kaygusuz, I., Epigraphica Antolica 2 (1983), 37–9Google Scholar), and Phaselis, (SEG XXXI (1981) 1300Google Scholar), apparently played a key role in organizing assistance for the stricken cities of Lycia and Pamphylia, as emerges from several references in the inscription of Opramoas at Rhodiapolis, (TAM II, 3Google Scholar, col. XI. 42, XV. 55. XVII, 59; see Hanslik, R., RE suppl. IX, 1834 ffGoogle Scholar. no. 14).

⁵⁴ Typallia: Mitchell, S., Studies in the History and Topography of Lycia and Pisidia (1994), 92–3Google Scholar, suggesting the area east of Isparta; Palaiopolis: presumably in western Pisidia, see von Aulock, H., Münzen und Städte Pisidiens I (1977), 40–1Google Scholar. But the number of topographical puzzles in Pisidia has been sharply reduced recently. Keraitai, Kolbasa and Malos have all firmly fixed; see Studies in the History and Topography of Lycia and Pisidia, 92. A newly published inscription from Lycian Xanthos provides the text of a decree passed in honour of a Xanthian judge by the hitherto unknown city of the Angeireis. The editors, Bousquet, J. and Gauthier, Ph., REG 16 (1993), 12–23CrossRefGoogle Scholar, suggest that the city was Pisidian, and that its territory should be identified with the ager Agerensis of Cicero, , De Lege Agraria II, 19, 50Google Scholar, a region which had come into Roman hands as a result of the campaigns in the Taurus of Servilius Isauricus. However, it is not self-evident that the Angeireis and the ager Agerensis were identical. In the Proceedings of the Tenth International Congress of Classical Archaeology, Ankara-Izmir 1973 (1978), 311–8Google Scholar at 313 n. 9 I suggested that the reading of the Cicero passage agrum Agerensem (nominative: ager Agerensis) might have been corrupted by dittography, and that the original reading might have been the historically plausible agrum Isaurensem.

⁵⁵ See n. 18.

⁵⁶ Bean, , AS X (1960), 79–80Google Scholar.

⁵⁷ H. von Aulock, op. cit, 32 citing the finds of Bean, , AS 1960, 69Google Scholar n. 57 and of Woodward, and Ormerod, , BSA 16 (1909–1910), 132Google Scholar.

⁵⁸ Weiss, P., “Pisidisches aus Kodrula”. Studien zum antiken Kleinasien, Asia-Minor-Studien 3 (1991), 69–73Google Scholar.

⁵⁹ It may be legitimate to cite the “Phrygian” character of the finds recently recovered from the Bayındır tumulus near Elmalı as evidence for Phrygian influence in the culture of the whole Milyadic region; cf. JHS Archaeological Reports 1989/1990, 87–8Google Scholar.

⁶⁰ See H. von Aulock, op. cit., 32; the same point was made by Ramsay, W. M., REG 6 (1893), 256–7Google Scholar. For the distribution of inscriptions and reliefs depicting this triad of deities, see Robert, L., BCH 107 (1983), 553–79CrossRefGoogle Scholar. He observes that Pisidia and the Milyas are the centre of these cults (p. 567). The case is strong, but one must add the reservation that coin types relating to this cult occur in eastern Pisidia at Pednelissos (von Aulock, op. cit., 1189, 1204), and in the north at Conana (von Aulock, , Münzen und Städte Pisidiens II, 793–6Google Scholar), at Prostanna (ibid., 1802–6), and Sagalassos (Chapouthier, F.,Les Dioscures au service d' une déesse (1935), 44Google Scholar). See further, Robert, , BCH 107 (1983), 577–8CrossRefGoogle Scholar.

⁶¹ In the absence of further new evidence the best hope of pinning Kodrula down probably lies in closer study of its interesting coin-types, and their find spots. There may be clues to the latter in the museum registers of Burdur and Antalya.