¹ Karttunen, K., ‘A miraculous fountain in India’, Arctos xix (1985) 55–65, at 58Google Scholar, draws attention to this predilection of Ctesias. For bibliography on Ctesias' Indica see Bigwood, J.M., ‘Ctesias' Indica and Photius’, Phoenix xliii (1989) 302–16CrossRefGoogle Scholar, at 302 and Bigwood, , ‘Ctesias' parrot’, CQ xliii (1993) 321–7.CrossRefGoogle Scholar

² Lindegger, P., Griechische und römische Quellen zum Peripheren Tibet ii (Zürich 1982) 104.Google Scholar The comments of Karttunen, , India in early Greek literature (Helsinki 1989) 8–9 n. 18 are very brief.Google Scholar

³ E.g. by Lindegger (n. 2) 104, who also suggests influence by Herodotus. A recent article by Romm, J., ‘Belief and other worlds: Ktesias and the founding of the “Indian wonders”’ in Mindscapes: the geographies of imagined worlds, ed. Slusser, G.E. and Rabkin, E.G. (Carbondale IL 1989) 121–35Google Scholar, treats the work as in large measure a product of the author's fantasy.

⁴ The story is not mentioned by Brown, T.S., ‘Suggestions for a vita of Ctesias of Cnidus’, Historia xxvii (1978) 1–19Google Scholar, by Eck, B., ‘Sur la vie de Ctésias’, REG ciii (1990) 409–434CrossRefGoogle Scholar, or by Auberger, J., Ctésias: Histoires de l'orient (Paris 1991) 4–10Google Scholar, in her comments on Ctesias' life.

⁵ For Photius' emphasis on marvels and other aspects of his summary see Bigwood, ‘Ctesias' Indica’ (n. 1).

⁶ Hägg's, T. review of R. Henry, Photius, Bibliothèque, GGA ccxxviii (1976) 32–60Google Scholar, gives a fuller list (46 and 56) of the MSS variants in this passage, none of which affect my argument. He does not comment on the second in line 1, which also appears in Henry's, R., Clésias, La Perse, l'Inde: Les sommaires de Photius (Bruxelles 1947)Google Scholar, though not in his edition of the Bibliotheca.

⁷ Cf. throughout the fragments of this work expressions such as ‘they say’, or occasionally ‘the Indians say’; and cf. ‘the Bactrians say’ in F 45h line 11 (Ael. NA iv 27).

⁸ On this manuscript see Diller, A., ‘Some false fragments’, Classical studies presented to B.E. Perry (Chicago 1969) 27–30.Google ScholarPhilostratus, VA iii 45Google Scholar refers to the gold-producing well as a tall tale. However, his ‘stone which behaves like a magnet’ must be an allusion to Ctesias' pantarbe stone (described in iii 46; cf. 688 F 45.6), not to iron, as is supposed by C. Müller, ‘Ctesiae Cnidii fragmenta’ (appendix to Didot Herodotus, [Paris 1844]) 89 and by McCrindle, J.W., Ancient India as described by Ktesias the Knidian (Calcutta 1882) 9.Google Scholar

⁹ I discuss what he says about them in ‘Aristotle and the elephant again’, AJP cxiv (1993) 537–55. His description of the parrot (F 45.8) is surely also from personal observation; see Bigwood, ‘Ctesias' parrot’ (n. 1).

¹⁰ It becomes commonly used between 800 and 500 BC, according to B., and Allchin, R., The rise of civilization in India and Pakistan (Cambridge 1982) 345.Google Scholar (See 309 for the contro versy about the date of its introduction.)

¹¹ E.g. Ctesias F 45.46 (cf. F 45r=Ael. NA v 3) and F 45.49 (cf. F 45s=Antig. Hist. Mir. 150); Arrian Ind. 16.11 (=Nearchus 133 F 11); Diod. ii 36.2 (=Megasthenes 715 F 4; cf. Diod. ii 16.4). Cf. also in Curtius ix 8.1 the 100 talents of white iron (whatever is meant by this) brought, among other gifts, to Alexander by the Malli and Sudracae (Oxydracae) in the lower Punjab.

¹² Ed. L. Casson (Princeton 1989) 114. Indian iron is also listed as a dutiable import under M. Aurelius and Commodus (Dig. xxxix 4.16.7).

¹³ On how far Photius' vocabulary is that of the original see Bigwood, ‘Ctesias' Indica’ (n. 1) 306–8. In Herodotus, xiphos is applied to the Spartan short sword (vii 224.1), as well as to the akinakes or ‘Median’ dagger (vii 54.2; cf. iii 64.3 and iii 78.5). On the Greek sword in general see Anderson, J.K., Military theory and practice in the age of Xenophon (Berkeley and Los Angeles 1970) 37–8Google Scholar; cf. Anderson, , ‘Hoplite weapons and offensive arms’, in Hoplites: the classical Greek battle experience, ed. Hanson, V.D. (London and New York 1991) 15–37 at 25.CrossRefGoogle Scholar It is not of course certain that Ctesias talks of Indian swords, rather than swords made in Persia of Indian iron. For bulk metal as a gift see Curtius ix 8.1 (cf. n. 11).

¹⁴ For the Indian evidence see von Hinüber, O., Arrian ed. and tr. Wirth, G. and von Hinüber, O. (Zürich 1985)Google Scholar on Arr. Ind. 16.6, and Pant, G.N., Indian arms and armour ii (New Delhi 1980) 6 ff.Google Scholar For the identification of the Indian peoples on the Iranian monuments see Roaf, M., ‘The subject peoples on the base of the statue of Darius’, CDAFI iv (1974) 73–160Google Scholar, especially 144–7. Delegates in the processions of ‘tribute-bearers’ on the Persepolis reliefs are mostly unarmed and their gifts (which include weapons) are not necessarily representative of the area from which they come; see Sancisi-Weerdenburg, H., ‘Gifts in the Persian empire’ in Le tribut dans l'empire perse, ed. Briant, P. and Herrenschmidt, C. (Paris 1989) 129–46 at 136–7.Google Scholar (The weapons brought as gifts by the Indians and Gandarans do not in fact include swords.) The throne-bearers on the royal tombs (see Schmidt, E., Persepolis viii [Chicago 1970] 108 ff.Google Scholar) are with one exception armed, in most cases with a sword. It either hangs from a belt at the waist (or is tucked into the waist-band) or, as in the case of the delegates from the Indic provinces, is suspended from the shoulder. Vogelsang, W.J., The rise and organisation of the Achaemenid empire (Leiden 1992) 140Google Scholar compares the long sword of Arr. Ind. 16.9. However, ten throne-bearers in all are given this type of sword and one suspects some stylisation (cf. Schmidt 116).

¹⁵ For the problems of Herodotus' army list see the useful comments of Lewis, D.M., ‘Persians in Herodotus’, in The Greek historians: literature and history. Papers presented to A. E. Raubitschek (Stanford 1985) 101–17.Google Scholar According to Ctesias, the dog-headed Indian tribe acquires xiphe, as well as other weapons, by barter and as gifts from the Indian king (F 45.41).

¹⁶ Four items are specifically mentioned in the fragments of Ctesias as gifts of the Indian to the Persian king. Two, i.e. animals (F 45dβ=Ael. NA iv 21) and woven materials (F 45γ=Ael. NA iv 46) are among the more common kinds of gifts depicted on the Persepolis reliefs. For the third, a fragrant oil in an alabaster container (F 45.47), compare Cambyses' gift of an alabaster container of myrrh (Hdt. iii 20). For the fourth item, a drug or poison (F 45m=Ael. NA iv 41), cf. the pharmaka of Polycleitus 128 F 3a (Strabo xv 3.21), and also the aphrodisiac sent by the Indian king Sandrocottus to Seleucus (Phylarchus 81 F 35). H. Sancisi-Weerdenburg (n. 14) 129–46 discusses the types of gifts brought by the peoples of the empire and also those bestowed by the king.

¹⁷ See the table in Walser, G., Die Völkerschaften auf den Reliefs von Persepolis (Berlin 1966) 103.Google Scholar

¹⁸ I follow here the identification of the various delegations given by Walser (n. 17). On the differences between the two kinds of dagger see Calmeyer, P., ‘Greek historiography and Achaemenid reliefs’. Ach. Hist. ii ed. Sancisi-Weerdenburg, H. and Kuhrt, A. (Leiden 1987) 11–26 at 13.Google Scholar

¹⁹ van Buitenen, J.A.B., The Mahābhārata ii (Chicago 1975).Google Scholar The epic is believed to have been gradually shaped over the period c. 400 BC to c. 400 AD, its more or less final form being reached in the Gupta period (4th-6th centuries AD); see Karttunen (n. 2) 147.

²⁰ For the golden akinakes as a gift of honour from the king see Xen. Anab. i 2.27. On redistribution of gifts see Sancisi-Weerdenburg (n. 14) 137 ff.

²¹ Sancisi-Weerdenburg, H., ‘Decadence in the empire or decadence in the sources?’ Ach. Hist. i ed. Sancisi-Weerdenburg, H. (Leiden 1987) 33–45 at 36Google Scholar, stresses our lack of knowledge

²² See Griffiths, A., ‘Democedes of Croton: a Greek doctor at the court of Darius’, Ach. Hist. ii ed. Sancisi-Weerdenburg, H. and Kuhrt, A. (Leiden 1987) 37–51.Google Scholar Eck (n. 4) 413 accepts the large house and the honour of sharing the king's table (Hdt. iii 132).

²³ On the inaccuracies see Bigwood, , ‘The ancient accounts of the battle of Cunaxa’, AJP civ (1983) 340–57 at 344–8.Google Scholar

²⁴ On the sources see also Westlake, H.D., ‘Diodorus and the expedition of Cyrus’, Phoenix xli (1987) 241–54.CrossRefGoogle Scholar Xen. (Anab. i 8.24 ff.), who was not near the king and is not necessarily correct, gives Ctesias' version some support, but disagrees over the gravity of the wound. His Artaxerxes does not withdraw from the battle. On Deinon's account (690 F 17 = Plut. Art. 10.1–3), where Artaxerxes apparently does not receive a wound and which seems to reflect Artaxerxes' propaganda, see Stevenson, R.B., ‘Lies and invention in Deinon's Persica’, Ach. Hist. ii ed. Sancisi-Weerdenburg, H. and Kuhrt, A. (Leiden 1987) 27–35 at 30–31.Google Scholar Diodorus' account of Artaxerxes' wound (xiv 23.6) is perhaps influenced by Ctesias.

²⁵ This is rightly emphasised in H. Sancisi-Weerdenburg's discussion of Parysatis (n. 21) especially 40–44, and also ‘Exit Atossa: images of women in Greek historiography on Persia’, in Images of women in antiquity ², ed. Cameron, A. and Kuhrt, A. (London 1993) 20–33 at 31–33Google Scholar, although she is perhaps too sceptical. On Parysatis see also Lewis, D.M., Sparta and Persia (Leiden 1977) 21–22.Google Scholar

²⁶ See Stolper, M.W., Entrepreneurs and empire (Istanbul 1985) 63–64Google Scholar, and for further discussion, Briant, P., ‘Dons de terres et de villes: l'Asie mineure dans le contexte achéménide’, REA lxxxvii (1985) 53–72 at 59–60CrossRefGoogle Scholar, and Cardascia, G., ‘La ceinture de Parysatis’, Marchands, diplomates et empereurs: Études … offertes à Paul Garelli, ed. Charpin, D. and Joannes, F. (Paris 1991) 363–9.Google Scholar

²⁷ It is well-known that Ctesias acted as an intermediary for Artaxerxes in the last years of his stay in Persia (F 30 = Phot., and F 32 = Plut. Art. 21–4), though he no doubt greatly exaggerated his own importance.

²⁸ See Bigwood (n. 23) 345 and 356.

²⁹ Megasthenes (715 F 32 Strabo xv 1.55) describes the king hunting with an escort of armed women; in the Arthaśāstra attributed to Kautilya (i 21.1; tr. R.P. Kangle, Bombay 1963) women armed with bows guard the king's bed-chamber. (On the date of this work, some of the information in which may go back to c. 300 BC, see Karttunen [n. 2] 146–7.)

³⁰ Discussed by Sancisi-Weerdenburg, H., ‘A typically Persian gift (Hdt. ix 109)’, Historia xxxvii (1988) 372–4.Google Scholar

³¹ E.g. Tomyris in Hdt. i 205–214, Atossa daughter of Ariaspes in Hellanicus 4 F 178a, Zarinaea and Sparethra in Ctesias (F 5 = Diod. ii 34.3–5, F 8a and F 8b, and F 9.3), to name a few. For archaeological evidence suggesting that some Scythian and Sauromatian women took part in military ventures, see Melyukova, A.I., Cambridge history of early inner Asia, ed. Sinor, D. (Cambridge 1990) 106 and 111–12Google Scholar, and Rolle, R., The world of the Scythians, Eng. tr. (London 1989) 86–91.Google Scholar

³² Photius presumably means a sword of this iron is planted in the ground, is translated by McCrindle (n. 8) 9 as ‘thunderstorm’, and by Lassen, C., Indische Alterthumskunde ² ii (Leipzig 1874) 564Google Scholar as ‘Blitzstrahl’; cf. LSJ s.v. ‘hurricane…attended by lightning’. Whatever its meaning here, at F 45.18 it clearly means ‘tornado’ or ‘whirlwind’. Melville, C., ‘Meteorological hazards and disasters in Iran: a preliminary survey to 1950’, Iran xx (1984) 113–150CrossRefGoogle Scholar collects data from the seventh century AD onwards on the destructiveness of rain, hail, thunderstorms, dust-storms and the like in Iran.

³³ For the first view, see Baehr as reported by Müller (n. 8) 89; for the second, see Lassen (n. 32) 564.

³⁴ See SirFrazer, J.G., The golden bough ³ i (London 1911) 244–331.Google Scholar

³⁵ Herodotus adds that they also sacrificed to the Greek local deities, Thetis and the Nereids. Additional classical references to Magi as manipulators of the weather are given by Fiedler, W., Antiker Wetterzauber (Stuttgart 1931) 19–21.Google Scholar For similar attempts on the part of Indian brahmans see Fiedler 17 and 45.

³⁶ Discusseci by Calmeyer, P., ‘Der “Apollon” des Dareios’, AMI xxii (1989) 125–30Google Scholar and by Nagel, W. and Jacobs, B., ‘Königsgötter und Sonnengottheit bei altiranischen Dynastien’, IA xxiv (1989) 337–89.Google Scholar

³⁷ See SirFrazer, J.G., The golden bough ³ iii (London 1911) 232–6Google Scholar, citing Indian beliefs among others, and Mundle, I.RAC vi (1964) s.v. ‘Erz’ 479–490.Google Scholar For the sword in magic see Mouterde, R., ‘Le glaive de Dardanos: objets et inscriptions magiques de Syria’, Mélanges de l'université Saint-Joseph xv.3 (Beirut 1930) 61Google Scholar n.2 and Mundle op. cit. 489. Cf. also the apotropaic seven swords of adamant which King Ganges fixes in the earth (Philostratus VA iii 21), although it is not certain that there is anything Indian or oriental about this story. According to Anderson, G., Philostratus (Beckenham, Kent 1986) 211Google Scholar the context is Pythagorean. For the mixture of Indian and other ideas in this section of the work see Anderson 199–220.

³⁸ According to the Denkard vii 5.9 (Pahlavi Texts v p. 76, tr. W.E. West), a compilation of the 9th or 10th centuries AD, Zoroaster taught rites to dispel hail and similar evils. For the beliefs attributed to the Magi and to Zoroaster by the Graeco-Roman world see especially Beck, R., ‘Thus spake not Zarathuštra: Zoroastrian pseudepigrapha of the Graeco-Roman world’, in Boyce, M. and Grenet, F., A history of Zoroastrianism iii (Leiden 1991) 491–565.Google Scholar Earlier views are found in Bidez, J. and Cumont, F., Les mages hellénisés, 2 vols. (Paris 1938).Google Scholar

³⁹ For gold cf. also his tale of the griffins in F 45.26 (Phot.) and F 45h (Ael. NA iv 27). For gold production in general in ancient India see O. von Hinüber (n. 14) 1123–24 on Arr. Ind. 15.4. The gold of Dardistan in the north is discussed by Bernard, P., ‘Les Indiens de la liste des tributs d'Hérodote’, Stud. Iran. xvi (1987) 177–91CrossRefGoogle Scholar, and somewhat differently by W. J. Vogelsang (n. 14) 204–6.

⁴⁰ Karttunen (n. 2) 8–9 n. 18, is tempted to connect it with Nuristani traditions about magic lakes containing precious items. For these see Jones, Schuyler, ‘Silver, gold and iron. Concerning Katara, urei, and the magic lakes of Nuristan’, KUML Årbog for Jysk arkeologisk selskab 1973–1974, 251–61Google Scholar (English version) at 255–8 and Tucci, G., ‘On Swāt. The Dards and connected problems’, East and West xxvii (1977) 9–103Google Scholar at 28–29. One tradition goes back to the sixth century AD at least.

⁴¹ Cf. the well in F 45.49 and also the Ethiopian pool (Fl=Diod. ii 14.4). In both cases we have the label as well as measurements.

⁴² See Oppenheim, A. L., ‘A fiscal practice of the ancient Near East’, JNES vi (1947) 116–20Google Scholar, and Asheri, D., Erodoto, le storie, libro iii (Milan 1990) 322–23 on Hdt. iii 96.Google Scholar

⁴³ Cf. P. Bernard (n. 39) 180–81. For evidence of such practices from the Lydian period at Sardis see Hanfmann, G.M.A., Sardis from prehistoric to Roman times (Cambridge, MA 1983) 34–41.Google Scholar

⁴⁴ See especially Jacoby, F., RE xi (1922)Google Scholar s.v. ‘Ktesias’ 2059 and, for this story, Lindegger (n. 2) 104.

⁴⁵ Although it was not necessarily Indian practice. Ctesias' griffin story implies that some Indians knew how to refine gold; see F 45h lines 31–2 (Ael. NA iv 27). However, Strabo xv 1.30, perhaps following Onesicritus (134 F 21), and speaking of the territory of Sopeithes, comments on the primitive technology of the Indians (of the area?), while Megasthenes (715 F 23b=Strabo xv 1.44) says something similar of the Derdae (Dards). For the evidence relating to early gold-mining see especially Allchin, F.R., ‘Upon the antiquity and methods of gold-mining in ancient India’, JESHO v (1962) 195–211.Google Scholar

⁴⁶ On the cult of Artemis Karyatis see Wide, S., Lakonische Kulte (Stuttgart 1893) 102–3Google Scholar and 108 and Calame, C., Les choeurs de jeunes filles en Grèce archaïque i (Rome 1977) especially 264–76.Google ScholarSchmidt, E., Geschichte der Karyatide, (Würzburg 1982) 14–32Google Scholar reviews some of the literary evidence for the term ‘Karyatides’ and also comments on the problem of the relationship of the dancing girls to the architectural support figures.

⁴⁷ Statius Theb. iv 225 refers to their singing.

⁴⁸ On these identifications see Fuchs, W., Die Vorbilder der neuattischen Reliefs, JDAI Erg. H. 20 (1959) 91–92Google Scholar and Schmidt (n. 46) 23. Fuchs gives some well-known examples of kalathiskos dancing girls and bibliographical information in n. 53 (pp. 91–92). Further examples and bibliography in Cook, A.B., Zeus iii.2 (Cambridge 1940) 990–1012.Google Scholar

⁴⁹ For maenads see Boardman, J., Greek gems and finger rings (London 1970) 216.Google Scholar Boardman no. 167 depicts a ‘mantle-dancer’ no. 409 (Arcadian) depicts a dancer with tambourine. Dancers also appear in nos. 566 and 1029.

⁵⁰ See Boardman (n. 49) for the first (no. 718) and for the third (233, pl. 822). For the second (from S. Russia) see Cook (n. 48) 1009 and Greifenhagen, A., Antike Kunstwerke (Berlin 1960) 30CrossRefGoogle Scholar, pl. 93,2. Cook (1011) also refers to similar figures on gems of Roman imperial date.