¹ Paus. iv 2.3.

² Hdt. vi 101.

³ Cf. Bérard, C., ‘Architecture érétrienne et mythologie delphique’, AK xiv (1971) 59–73, pl. 40Google Scholar.

⁴ Auberson, P., Eretria i (Bern 1968) 9, 24Google Scholar.

⁵ E.g. Kourouniotis, K., ‘Ἀνασκαϕαὶ ἐν Ἐρετρίᾳ’, Praktika 1900, 53–6Google Scholar: end of sixth cent.; Langlotz, E., Zur Zeitbestimmung der strengrotfigurigen Vasenmalerei und der gleichzeitigen Plastik (Leipzig 1920) 78, 117: 500–490Google Scholar; Beazley, J. D. and Ashmole, B., Greek Sculpture and Painting (Cambridge 1932) fig. 52Google Scholar: late sixth cent.; Lippold, G., Die griechischen Plastik (Munich 1950) 72 f.Google Scholar: c. 520–10; von Bothmer, D., Amazons in Greek Art (Oxford 1957) 126: c. 520–10Google Scholar; Lullies, R. and Hirmer, M., Griechische Plastik² (Munich 1965) pls 66–9: c. 510Google Scholar; Herdejürgen, H., Untersuchungen zur thronenden Göttin aus Tarent in Berlin (Waldsassen 1968) 162 f.Google Scholar: 500–490; Themelis, P., ‘Ἐρετριακὰ’, ArchEph 1969, 164–6: shortly before 510Google Scholar; Richter, G. M. A., Sculpture and Sculptors of the Greeks⁴ (New Haven 1970) 251: 510–500Google Scholar; Kleine, J., Untersuchungen der attischen Kunst von Peisistratos bis Themistokles, Ist. Mitt. Beih. vii (1973) 96Google Scholar: late sixth cent.; Delivorrias, A., Attische Giebelskulpturen und Akrotere des fünften Jahrhunderts (Tübingen 1974) 179–80: after 500–490Google Scholar; Ridgway, B. S., The Archaic Style in Greek Sculpture (Princeton 1977) 212Google Scholar: last quarter sixth cent.; Robertson, M., A History of Greek Art (Cambridge 1976) pl. 50bGoogle Scholar: end sixth cent.; Boardman, J., Greek Sculpture, the Archaic Period, a Handbook (London 1978) fig. 205.1–3: about 510Google Scholar; Gauer, W., ‘Das Athener Schatzhaus und die marathonischen Akrothinia in Delphi’, Forschungen und Funde, Festschrift Bernhard Neutsch (Innsbruck 1980) 131 f.Google Scholar: after 500/499–490.

⁶ Auberson, P. and Schefold, K., Führer durch Eretria (Bern 1972) 115Google Scholar; Cf. Robertson (n. 5) 164. On this reading of the iconography, we must presumably infer that the original architects of the temple (‘c. 530–20’) had envisaged a different decorative scheme from the one partially extant.

⁷ Kourouniotis (n. 5) loc. cit.

⁸ Auberson (n. 4) 24.

⁹ Auberson (n. 4) 10.

¹⁰ Compare, for example, J. J. Coulton's recent study in which he places the Eretria Temple in the same group as the Temple of Zeus at Olympia: ‘Doric capitals: a proportional analysis’, BSA lxxiv (1979) 102–4Google Scholar. The comparanda Auberson himself employed in his analysis ([n. 4] 19) were buildings dated according to a conservatively high view of the conventional chronology.

¹¹ IG xii.9 202.12–14: ἐν] τῶι ἰερῶ[ι][τοῦ ᾿Απόλλωνος τοῦ] Δαφνη[φό] [που] ; 204.8–9: πρὸ τοῦ Ναο]ῦ τοῦ ᾿Απόλ [λωνος τοῦ Δαφνηφ]όρου; similar expressions are to be found in 208.23–4; 210.28–9; 212.23; 215.12–13; 216.14–15; 220.20–1; 225.7–8; 229.4–5; 230.5.

¹² Livy xxxii 16; Paus. vii 8.1.

¹³ Paus. vii 8.2.

¹⁴ Livy xxxii 21.7.

¹⁵ Konstantinu, J., ‘Aus dem Eretriagiebel’, AthMitt lxix/lxx (1954–1955) 41Google Scholar; von Bothmer (n. 5) 126.

¹⁶ Cf. Lullies (n. 5) 48; Boardman (n. 5) 156.

¹⁷ As Vessberg, O. notes (Studien zur Kunstgeschichte der römischen Republik [Lund/Leipzig 1941] 29Google Scholar, citing Livy xxxiv 52.4–5), this triumph is the first occasion on which Greek works of marble are explicitly reported to have been displayed in Rome.

¹⁸ Auberson–Schefold (n. 6) 121.

¹⁹ Cf. Livy xxxii 16.10.

²⁰ ADelt, Chron. xvii (1961/2) pl. 165b; Cf. Auberson–Schefold (n. 6) 42; ‘wunderbarer Erhaltung’, ibid. 30.

²¹ E.g. Lullies (n. 5) pls 66–9.

²² It is instructive to compare a description of the North Porch of the Erechtheum in 1837: ‘This side of the temple, being so well sheltered from the sea-breeze, has preserved its sculptured ornaments as fresh and sharp as if lately finished; and the columns of this portico being fluted with capitals elaborately worked and well-sheltered, have retained remains of colour …’. Letter of Braceridge, C. H., quoted in Paton, J. M. (ed.), The Erechtheum (Cambridge, Mass. 1927) 229Google Scholar. J. J. Coulton has, however, pointed out to us that the weathering in question at Eretria is more likely to have occurred following the destruction of the temple when the sculpture was lying on the ground than while it remained in the protected position of the gable.

²³ Hdt. vi 101.

²⁴ Hdt. v 101.

²⁵ Tod, M. N. in CAH v (1927) 17Google Scholar; Cf. the view of the numismatic evidence in Price, M. J. and Waggoner, N., Archaic Greek Silver Coinage: the ‘Asyut’ Hoard (London 1975) 56Google Scholar.

²⁶ Hdt. vi 101, 119; Plato, Epigr. 11, 12Google Scholar Page (‘Hellenistic literary exercises’: Page, D. L., Further Greek Epigrams [Cambridge 1981] 171–3)Google Scholar.

²⁷ If any credence can be placed in the dream of Apollonius of Tyana described by Philostratus—and it would be difficult to imagine a historical source with less to recommend it: see Penella, R. J., Athenaeum iii (1974) 295–300Google Scholar and Bowie, E. L., ANRW ii 16.2 (1978) 1652–99Google Scholar—the Eretrians in 490 were captured like fishes in nets; Philostratus goes on to say that 780 Eretrians were captured ‘not all fighting men, for the number included some women and old people, and I dare say children as well: for the main part of the Eretrians fled up to Caphareus and the highest mountains in the island’. (Philostr., VA i 25–6, tr. J. S. Phillimore, )Google Scholar.

²⁸ Hdt. viii 1, 46. These seven galleys in fact represent a larger contingent than the five the Eretrians contributed to the Ionian Revolt (Hdt. v 99).

²⁹ Hdt. ix 28, 31.

³⁰ ML no. 27. Dr Lewis has drawn our attention to the fact that the Eretrians are not included in Pausanias' record of the cities ‘who fought at Plataea against Mardonius and the Persians’ inscribed on the panhellenic dedication at Olympia (v 23.1). One might speculate that the omission of Eretria, Siphnos, Leucas and Thespiae was an oversight on Pausanias' part were it not for the fact that he explicitly states that the Plataeans were listed on the inscription ‘alone of the Boeotians’ thereby excluding the Thespians. Since, however, Pausanias does mention ‘Styreans from Euboea … and finally men of Chalcis on the Euripus’, perhaps the exclusion of the Eretrians, Styra's allies, was politically motivated. It is also possible that the Eretrians who survived took refuge in Styra until they were in a position to repair their ruined city following the defeat of Xerxes.

³¹ ATL i 270–1; Cf. iii 57 and 99 n. 1.

³² Mallwitz, A., Olympia und seine Bauten (Munich 1972) 94Google Scholar; Cf. Jeffery, L. H., Local Scripts of Archaic Greece (Oxford 1961) 88 no. 19Google Scholar, followed by Auberson–Scheold (n. 6) 30 f. Others, however, have seen this bull as a dedication from the late sixth century (e.g. Eckstein, F., Ἀναθήματα: Studien zu den Weihgeschenken strengen Stils im Heiligtum von Olympia [Berlin 1969] 50–3, 118–19, figs 12–13)Google Scholar, but see Gauer, W., Weihgeschenke aus den Perserkriegen, Ist. Mitt. Beih. ii (1968) 107 n. 506, 108Google Scholar.

³³ Ridgway (n. 5) 316 n. 15, where, it is encouraging to note, she considers the possibility that the Athena is part of a later [i.e. post–490] repair.

³⁴ Ibid.

³⁵ As W. Gauer has already observed (see n. 5).

³⁶ Hdt. v 97.3.

³⁷ Harrison, E. B., ‘Motifs of the city-siege on the shield of Athena Parthenos’, AJA lxxxv (1981) 295Google Scholar.

³⁸ Aesch., Eum. 688–93Google Scholar, Cf. Hdt. ix 27; Lysias ii 4; Isoc. iv 68.

³⁹ ADelt, Chron. xviii (1963) pl. 327Google Scholar; Cf. Goldberg, M. Y., ‘Archaic Greek akroteria’, AJA lxxxvi (1982) 213Google Scholar.

⁴⁰ ADelt, Chron. xvii (1961–1962) pl. 165 cGoogle Scholar.

⁴¹ The view espoused by Myres, J. L., Herodotus, Father of History (Oxford 1953) 183Google Scholar, though without supporting evidence. Eretria's participation is not mentioned by Hdt. ad v 77, nor in ML no. 15.

⁴² Hdt. v 102.

⁴³ Hdt. vi 99, viii 66; Meiggs, R., The Athenian Empire (Oxford 1972) 69 f.Google Scholar; Cf. Burn, A. R., Persia and the Greeks (London 1962) 237Google Scholar.

⁴⁴ Wallace, W. P., ‘The denies of Eretria’, Hesp. xvi (1947) 130–3Google Scholar.

⁴⁵ We hope to show elsewhere how the legend of Theseus was manipulated by Cimon along with his political patrons and allies in order to promote Athens' new aspirations for hegemony by land and sea. Cf. Barron, J. P., ‘Bakchylides, Theseus and a woolly cloak’, BICS xxvii (1980) 1–8Google Scholar.

⁴⁶ Robertson (n. 5) 164.

⁴⁷ Lyc. in Leocr. 81; Diod. xi 29.3.

⁴⁸ Siewert, P., Der Eid von Plataiai (Munich 1972)Google Scholar; Shear, T. L. Jr, Studies in the Early Projects of the Periklean Building Program (Diss. Princeton 1966) 16–65Google Scholar.

⁴⁹ But see Walsh, J., ‘The authenticity and date of the Peace of Callias and the Congress Decree’, Chiron xi (1981) 53 f., n. 58Google Scholar.

⁵⁰ Delivorrias (n. 5) has suggested that the absence of sculpture from the east pediment of the Eretria temple may be evidence for the building having been incomplete in 490.

⁵¹ Harrison, E. B., Archaic and Archaistic Sculpture, Agora xi (Princeton 1965) 5Google Scholar. Cf. Ridgway (n. 5) 8, and see too Francis, E. D. and Vickers, M., Burl. Mag. cxxiv (1982) 41–2Google Scholar.

⁵² Löwy, E. (Der Beginn der rotfigurigen Vasenmalerei, SBWien ccxvii. 2 [1938] 16–17Google Scholar) gives the arguments; for older discussion, see Hitzig-Blümner ad Paus. x 11.2 (in vol. iii.2, 692 ff.). For a recent bibliography of the Siphnian Treasury, see Büsing-Kolbe, A., ‘Frühe griechische Türen’, JdI xciii (1978) 86 n. 98Google Scholar.

⁵³ On the distribution of ὅτε in Hdt., see Brackett, H. D. (‘Temporal clauses in Herodotus’, Proc. Amer. Acad. Arts & Sciences xii.8 [1905] 211 f.Google Scholar) who, however, includes this passage as the sole exception to the pattern he has otherwise reconstructed. Since Brackett himself recognised that ‘by far the most frequent tense in clauses of pure antecedence is the aorist’ (Brackett 211; Cf. ὅτε + aorist at Hdt. iv 78, no; v 30; vi 69), he presumably judged ἐποιεῦντο to be ‘antecedent’ on other than grammatical grounds. A. D. Godley (ad loc.) thus translates the clause: ‘when they were making …’ (Loeb [London 1921]); Cf. Legrand, Ph.: ‘à l'époque où ils faisaient …’ (Budé [Paris 1967[)Google Scholar, and Feix, J.: ‘als sie das Schatzhaus bauten …’ (Heimeran [Munich 1963])Google Scholar. Had Hdt. intended to report antecedence we assume that he would have written the aorist ἐποιήσαντο or used a different conjunction (e.g., ἐπεί). ὅτε could also be causal in this context: ‘because they were having the treasury built …’, but we see no clear basis for preferring cause over time (Cf. Powell, J. E., A Lexicon to Herodotus [Cambridge 1938] 273Google Scholar, s.v. ὅτε 3). This difference is in any case trivial once it has been recognised that the tense of ἐποιεῦντο denotes contemporaneity. (Beside ὅτε + impf. of contemporaneous action, note its comparable use with περ to mean ‘at the very time when’, e.g. Hdt. v 99.1; vi 106.1; Cf. Powell, loc. cit., s.v. ὅτεπερ).

⁵⁴ Note the contrast between ἐχρέωντο (impf.) of the Siphnians' consultation and ἔχρησε (aor.) of the oracle's response, implying that Delphi may itself, for whatever reason, have temporised before finally replying to the islanders' insistent enquiries. We thank Mr J. G. Griffith for drawing our attention to this detail.

⁵⁵ Herodotus clearly thinks (iii 57.2) that the Treasury was built from the tithe, whereas Pausanias perhaps suggests that the tithe was only levied once the Treasury had been built. This raises the disputed question of the degree of Pausanias' dependence on Herodotus. T. R. Glover, for example, thought that Pausanias had ‘Herodotus at his finger ends’ (in ‘Prince of Digressors’, Springs of Hellas [London 1945] 162)Google Scholar. Heer, J. (La personnalité de Pausanias [Paris 1979] 97 ff.Google Scholar) is also of this view, whereas G. Daux asserts Pausanias' complete independence from Herodotus (Pausanias à Delphes [Paris 1936] 182Google Scholar). J. J. Coulton, however, suggests that Pausanias' double phrase may imply that the Treasury was built with early instalments of the tithe which then continued to be paid even when the building was complete.

⁵⁶ Cf. Löwy (n. 52) 26–7.

⁵⁷ E.g. Cimon's ostracism after Ithome, the fate of the leaders of Elis and Mantinea following the battle of Plataea, the aristocratic reaction in Argos during the early 460s, and the successive changes of government at Athens following the occupation of Decelea.

⁵⁸ Ridgway (n. 5) 9.

⁵⁹ Cf. Boardman (n. 5) 158.

⁶⁰ Frazer ad loc.; Cf. Heer (n. 55) 98: ‘cet événement a done eu lieu plus tard que de son vivant’.

⁶¹ Bury, J. B. and Meiggs, R., A History of Greece to the Death of Alexander the Great⁴ (London 1975) 135Google Scholar.

⁶² Wagner, G. A. and Weisgerber, G., ‘The ancient silver mine at Ayos Sostis on Siphnos (Greece)’, Archaeophysika x (1979) 222Google Scholar. Other useful recent geological studies include: Gale, N. H., ‘Some aspects of lead and silver mining in the Aegean’, Misc. Graeca ii (1979) 9–60Google Scholar; Wagner, G. A., Gropengiesser, H. and Gale, N. H., ‘Early Bronze Age lead-silver mining and metallurgy in the Aegean: the ancient workings on Siphnos’, in Craddock, P. T. (ed.), Scientific Studies in Early Mining and Extractive Metallurgy, Brit. Mus. Occasional Paper xx (1980) 63–80, pls 1–8Google Scholar; Gale, N. H. and Stos-Gale, Z., ‘Lead and silver in the ancient Aegean’, Scientific American cciv (1981) 176–92CrossRefGoogle Scholar; id., ‘Cycladic lead and silver metallurgy’, BSA lxxvi (1981) 169–244.

⁶³ Price–Waggoner (n. 25) 69 ff.

⁶⁴ Gale, N. H., Gentner, W. and Wagner, G. A., ‘Mineralogical and geographical silver sources of archaic Greek coinage’, Metallurgy in Numismatics, Royal Numismatic Soc. Special Publication xii (1980) 36–43Google Scholar.

⁶⁵ Although on present evidence fifth-century coin production at Siphnos appears to have been fairly limited (Kraay, C. M., Archaic and Classical Greek Coinage [London 1978] 45)Google Scholar, we cannot be certain as to the extent to which this may have been due to the decree of Clearchus (ATL ii 61–8, D 14; iii 278, 281), which forced the use of Attic coinage on most of the allied states, probably in 449/8 (but see Mattingly, H., Historia x [1961] 148 ffGoogle Scholar. and JHS ci [1981] 86Google Scholar for a later date). Such coins as do exist, however, bear heads of Apollo ‘the patron deity of the Delian League of which Siphnos was a member’ (Kraay 47).

⁶⁶ Meiggs (n. 43) 61 citing Beloch, Gr. Gesch.² ii.2356–71Google Scholar.

⁶⁷ The fourth-century dedication on the Athenian Acropolis by the Siphnians of a golden στέϕανος weighing 66 drachmae, 5 obols is, however, scarcely an indication of poverty (IG ii² 1425.125).

⁶⁸ Davies, J. K., Athenian Propertied Families (Oxford 1971) 590–2, C 12Google Scholar. Dr Norman Ashton draws our attention to the fact that these Siphnians are the only non-Athenians mentioned in the surviving records of the mining leases and lessees of the latter half of the fourth century BC (cf. Hesp. xix [1950] 189 ff.Google Scholar).

⁶⁹ Anth. Pal. ix 421Google Scholar (Antipater of Thessalonica); Cf. [Dem.] xiii 34 (date uncertain).

⁷⁰ Diod. xxxi 45.

⁷¹ Cf. n. 66.

⁷² Cf. Hdt. iii 57.

⁷³ Ibid.

⁷⁴ On the analogy of Siphnos, suggests Labarbe, J. in La loi navale de Thémistocle (Paris 1957) 39CrossRefGoogle Scholar; cf. Macan ad Hdt. vii 144. Another similarity between the circumstances at Siphnos in the 520s and Athens in the 480s is found in the advice the Delphic oracle gave each state prior to their respective invasions. In A. H. Sayce's translation, Siphnos received the following warning:

When the town-hall in Siphnos is white,

And white-browed the market where judgements are said,

A wise man is needed to guard

'Gainst an ambush of wood and a herald in red.

Siphnos, however, unlike Athens, did not heed this Delphic counsel ‘'gainst an ambush of wood’, and paid a heavy price for its self-assurance. The actual price is worth considering since it suggests a reason for Delphi's attitude towards Siphnos in the first place. At first the Samians demanded ten talents of silver, but the Siphnians refused. The Samians then tried a new approach, laying waste the Siphnian countryside, thereby drawing the Siphnians out of their city. The Siphnians not only got the worse of the battle that ensued, but found their retreat cut off. Unable to return to their city, the Siphnians finally came to terms ten times more disadvantageous than those they were originally offered. Although Herodotus does not explicitly say so, we might infer that the ten talents represent Delphi's tithe and the hundred talents the islanders’ annual revenue.

⁷⁵ Cf. Aesch., Pers. 240Google Scholar, and Price–Waggoner (n. 23) 139 n. 246.

⁷⁶ Burn (n. 43) 442.

⁷⁷ Hdt. viii 46; Cf. 66.

⁷⁸ Paros had medised in 490. A Parian trireme had accompanied the Persian fleet to Marathon (Hdt. vi 133; did it carry, as Raubitschek, A. E. has suggested [in Charites: Festschrift Ernst Langlotz (Bonn 1957) 239Google Scholar], the block of Parian marble the Persians brought with them with which they intended to celebrate a victory and from which the statue of Nemesis at Rhamnus was subsequently made [Paus. 1 33.2–3]?). Paros, moreover, is included ad Aesch., Pers. 884Google Scholar in a list of Greek islands which, as Aeschylus' Persian chorus clearly implies, had belonged to Xerxes' realm before his defeat at Salamis. The Parians had learnt a lesson from their medism in 490 (Cf. Hdt. vi 133), but only in part for, like the Corcyreans, they temporised before Salamis (Hdt. viii 67, Cf. vii 168), and this temporising was probably the cause of their failure to appear either on the Serpent Column or in the Olympia dedication (Paus. v 23). The fact that the Parians stayed behind in Cythnos (itself on the patriotic side [Macan ad Hdt. viii 67] and on the Delphic and Olympian lists) implies malingering rather than Parian support for the Greek cause. Paros' relatively high tribute quotas (e.g. Meiggs [n. 4.3] 558–9) might indeed be considered punitive unless Meiggs is correct in his view that ‘the comparatively high tribute of Paros should be attributed to her marble quarries’ (ibid. 61). (A similar observation could, of course, be made with regard to the Siphnian quota.)

⁷⁹ ML 59–60.

⁸⁰ Rawlinson, G. (ed.), History of Herodotus iv (London 1862) 393Google Scholar.

⁸¹ ML 60.

⁸² ATL iii 57, 265–74, esp. 267–8, and, in the judgement of the authors of ATL, Siphnos also perhaps contributed ships. See too n. 66.

⁸³ ATL iii 197–9.

⁸⁴ Francis, E. D. and Vickers, M., ‘Leagros kalos’, PCPS ccvii (1981) 97–136Google Scholar.

⁸⁵ Vitr. i 5. Carya civitas Peloponnensis cum Persis hostibus contra Graeciam consensit, postea Graeci per victoriam gloriose bello liberati communi consilio Caryatibus bellum indixerunt. Itaque oppido capto viris interfectis civitate desacrata matronas eorum in servitutem abduxerunt, nec sunt passi stolas neque ornatus matronales deponere, uti non una triumpho ducerentur sed aeterno servitutis exemplo gravi contumelia pressae poenas pendere viderentur pro civitate. Ideo qui tunc architecti fuerunt aedificiis publicis designaverunt earum imagines oneri ferundo conlocatas, ut etiam posteris nota poena peccati Caryatium memoriae traderetur.

⁸⁶ Frazer, J. G., Pausanias iii (London 1913) 320ad iii 10.7 (citing Preller, below, n. 90)Google Scholar.

⁸⁷ Attributed by Pliny to Diogenes of Athens, NH xxxvi 37.

⁸⁸ Lessing, G. E., ‘Karyatiden’, in Petersen, J. and von Olshausen, W. (eds), Lessings Werke 17. Schriften zur antiken Kunstgeschichte (Berlin, etc. n.d.) 385–6Google Scholar.

⁸⁹ Blomfield, C.J., ‘Some remarks on the Caryatides of ancient architecture’, Museum Criticum vii (1826) 401Google Scholar.

⁹⁰ Preller, L., ‘De caussa nominis Caryatidum’, Ann. dell'Inst. arch. xv (1843) 396–406, esp. 404Google Scholar ( = Ausgewählte Aufsätze [Berlin 1864] 136–44Google Scholar). Vitruvius' testimony is now usually dismissed as a late fabrication: e.g. for Borchhardt, J., Die Bauskulptur des Heroons von Limyra: Das Grabmal des lykischen Königs Perikles, Ist. Forsch. xxxii (1976) 44, n. 100Google Scholar, it ‘beweist nur dass zu seiner Zeit die Bedeutung der Karyatiden nicht mehr verstanden wurde’; Cf. Drerup, H., ‘Zur Bezeichnung “Karyatide”’, MWPr 1975–1976, 11Google Scholar; Lauter, H., ‘Die Koren des Erechtheion’, Ant. Plastik xvi (1976) 14 ff., esp. 15, n. 47Google Scholar.

⁹¹ Walpole, R., Memoirs Relating to European and Asiatic Turkey² (London 1818) 602CrossRefGoogle Scholar.

⁹² Hdt. viii 73.

⁹³ Paus. viii 13.6, 14.1; Arcadian Caryae is not even mentioned by Hejnic, J., Pausanias the Perieget and the Archaic History of Arcadia (Prague 1961)Google Scholar.

⁹⁴ Paus. iii 10.7; iv 16.9.

⁹⁵ Xen., Hell. vi 5.25Google Scholar; vii 1.28. Cf. Meineke, A., Anakcta alexandrina (Berlin 1843) 362Google Scholar; Lauter (n. 90) n. 47.

⁹⁶ Cf. the ‘useful concept’ described by A. R. Burn (n. 43) 545, citing Pritchett, W. K., AJA lxi (1957) 20 n. 68Google Scholar, viz. that the usual Spartan conception of warfare was ‘agonal, i.e. they fought for admitted victory, as in the games, and not for the total destruction of an enemy’.

⁹⁷ Xen., Hell. vii 1.28Google Scholar.

⁹⁸ Cf. Serv. ad Verg., Ecl. viii 30Google Scholar; Stat., Theb. iv 225Google Scholar; Burkert, W., ‘Demaratos, Astrabakos, und Herakles: Königsmythos und Politik zur Zeit der Perserkriege (Herodot 6.67–69)’, MusHelv xxii (1965) 172Google Scholar (for further refs see Frazer, , Pausanias iii 320Google Scholar).

⁹⁹ Lewis, D. M., Sparta and Persia (Leiden 1977) 147 n. 80Google Scholar.

¹⁰⁰ Huxley, G. L., ‘The medism of Caryae’, GRBS viii (1967) 30Google Scholar, citing Brunt, P. A., Historia ii (1953/1954) 137 and 149 on Hdt. vii 132Google Scholar.

¹⁰¹ Huxley (n. 100) 31; Cf. Hdt. vii 235.

¹⁰² See Burn (n. 43) 505.

¹⁰³ Cf. Thuc. v 55.3; Cartledge, P. A., Sparta and Lakonia: a Regional History, 1300·363 BC (London 1979) 205CrossRefGoogle Scholar.

¹⁰⁴ Burn (n. 43) 506.

¹⁰⁵ Plommer, W. H., ‘Vitruvius and the origin of Caryatids’, JHS xcix (1979) 97, 101Google Scholar. Plommer correctly notes that the pertinent architectural members of the Erechtheum were described as korai (cf. IG i² 372.86, = IG i³ 474.86), but this does not of itself refute Vitruvius' claim that such statues could also be known as Caryatids: the figures in the Erechtheum frieze are, as Wolters, P. saw (Zeits.für bildenden Kunst vi [1895] 37Google Scholar; Cf. Drerup [n. 90] 12), only described in the most general way: ‘the man near the altar, the woman with the child, etc’ The inscriptions in question were legal documents, not religious or artistic treatises.

¹⁰⁶ Vitr. i 5.8–11.

¹⁰⁷ Cf. Plommer (n. 105) 99; Athen. 241d (pace Lautcr [n. 90] 14). We suggest that the reference of the term Caryatid could have been generalised to include an architectural form of the kind to which Athenaeus alludes.

¹⁰⁸ Homolle, T., ‘L'origine des Caryatides’, RA v (1917) 18–67Google Scholar; Marcadé, J., ‘Les bras des danseurs’, Mélanges Helléniques offerts à Georges Daux (Paris 1974) 239–54Google Scholar; Ridgway (n. 5) 204–5 and n. 24.

¹⁰⁹ Cf. n. 98.

¹¹⁰ Note Vitruvius' interest in the effects of the Persian Wars on the history of architecture. His account of the Stoa Persica at Sparta, which follows on immediately after that of the Caryatids, is well known: i 4; Cf. Paus. iii 11.3 and Coulton, J.J., The Architectural Development of the Greek Stoa (Oxford 1976) 39Google Scholar. We believe him to be recording a genuine tradition when he describes the re-use of ‘the yards and masts of the ships captured from the Persians’ to roof the Odeion by the Theatre of Dionysus at Athens: v 91 (though he was mistaken in attributing the building to Themistocles rather than Pericles; Cf. Broneer, O., ‘The tent of Xerxes and the Greek theater’, U. Cal. Publ. in Class. Arch. i [1944] 305–12Google Scholar; von Gall, H., ‘Das persische Königszelt und die Hallenarchitektur in Iran und Griechenland’, Festschrift für Frank Brommer [Mainz 1977] 119–32Google Scholar; id., ‘Das Zelt des Xerxes und seine Rolle als persischer Raumtyp in Griechenland’, Gymnasium lxxxvi [1979] 444–62, pls 13–14; Francis, E. D., ‘Greeks and Persians: the art of hazard and triumph’, in Schmandt-Besserat, D. [ed.], Ancient Persia: the Art of an Empire, Invited Lectures on the Middle East, U. Texas at Austin iv [Malibu 1980] 82–3Google Scholar. It would be interesting to determine if the architecture of the front of the stage building in the theatre at Eretria [which ‘ähnelt der eines persischen Palastes’, Auberson–Schefold (n. 6) 47–9] belongs in this tradition). Note too, Vitruvius' ‘not only detailed but accurate information’ regarding the Telesterion at Eleusis: vii praef. 16–17; Shear (n. 48) 175–8.

¹¹¹ FdD iv. 2, pl. 16; Richter, G. M. A., BCH lxxxii (1958) 92 ffCrossRefGoogle Scholar.

¹¹² Cf. Ridgway (n. 5) 101 n. 19: ‘The Knidian Karyatids are usually dated before 540 …’. See too Büsing-Kolbe (n. 52) 88 n. 104.

¹¹³ Jeffery, L. H., Archaic Greece, the City-States c. 700–500 BC (London 1976) 200Google Scholar.

¹¹⁴ ATL iii 213; Cf. Meiggs (n. 43) 55–6.

¹¹⁵ Aesch., Pers. 891Google Scholar.

¹¹⁶ Hdt. i 174.

¹¹⁷ Jeffery (n. 113) 199.

¹¹⁸ Hdt. vii 93, with Macan's n. ad loc.

¹¹⁹ Paus. x 11.4. Pausanias' doubt necessarily renders his testimony inconclusive: ‘effet de style chez Pausanias, lecture erronée, ou témoignage fidèle en présence d'un texte peu explicite?’ enquires Salviat, F. (‘La dédicace du trésor de Cnide’, Études Delphiques, BCH Suppl. iv [1977] 23–36)Google Scholar. The treasury inscription is too fragmentary to be decisive. The concluding words are usually restored to suggest an epinician dedication, but δεκάτ[αν ἀπὸ τῶμ πολεμί]ων (so Homolle, Th., BCH xx [1896] 591Google Scholar; FdD iii.i 289; Salviat 33) would probably imply a victory against Greeks (cf. πολεμίων) for which we see no appropriate context [leg. δεκάτ[αν ἀπὸ τῶμ Μηδικ]ῶν vel. sim.?). Now the Cnidians had participated in the commercial success of the eastern Greeks at Naucratis through their membership of the Hellenion (Hdt. ii 178). They also seem to have enjoyed special privileges in the Naucratite precinct of Miletus (Jeffery [n. 113] 352). G. Roux accordingly interpreted Pausanias' ἐπίδειξις εὐδαιμονίας in the sense of ‘economic prosperity’ and proposed δεκάτ[αν ὁ δᾶμος ὁ Κνιδί]ων (Énigmes à Delphes [Paris 1963] 68Google Scholar; Cf. Salviat 35). Salviat, however, restores the inscription as part of the architrave of the facade so that the lacunae become even longer than before and the area of uncertainty enlarged. He judiciously leaves the lacuna open.

¹²⁰ D.S. xi 36.5 f.; Diodorus is probably relying upon Ephorus, Cf. Burn (n. 43) 550.

¹²¹ Hdt. ix 106.

¹²² Cf. Paus. v 6 (on the ‘Chest of Cypsclus’ at Olympia).

¹²³ For examples of Cnidian script, see e.g. Jeffery (n. 32) pl. 68 and ‘Table of Letters’.

¹²⁴ We might compare the Mantineans and the main force of the Eleans who, arriving late for the Battle of Plataea, ‘professed their regret and offered (or afterwards pretended to have offered) to pursue Artabazus and his corps, a ludicrous proposal which Pausanias of course vetoed. When they got home, they banished their generals. Probably the delay was not involuntary but political and indicates that the party in power was not whole-hearted for the cause of Hellas’ (CAH v 340, Cf. Hdt. ix 77; see also Andrewes, A., Phoenix vi [1952] 1–5, esp. 2CrossRefGoogle Scholar). The Siphnians, without a warfleet, were unable to make a similar gesture by offering, for example, to pursue the Persians to Mycale, and so elected to redeem their promissory note in the hope of avoiding the disfavour of the Delphic authorities.

¹²⁵ The historical context of the ‘ex-Cnidian Caryatid’ is too fragmentary to allow us to judge its iconographic intent, but there is evidence outside Delphi (but still Apolline) which we must take into account. Pausanias' description of the Throne of Apollo at Laconian Amyclae (iii 18.8–19.5) has been thought to include references to architectural members of Caryatid type (᾿Ανέχουσιν ἔμπροσθεν αὐτόν, κατὰ ταὑτα δὲ καὶ ὄπισθεν, Χάριτές τε δύο καὶ ῾Ωραι δύο; Cf. Buschor, E. and von Massow, W., AthMitt lii [1927] 79–80Google Scholar), so that Caryatids resembling those of the Siphnian Treasury have frequently been incorporated in restorations (Buschor–v. Massow 79–80; Fiechter, E., JdI xxxiii [1918] 166 ff.Google Scholar; Martin, R., ‘Bathyklès de Magnésie et le “trône” d'Apollon à Amyklae’, RA 1976, 205–18Google Scholar). It is therefore important for us to consider both the occasion of this dedication, and the identity of its ‘Caryatids’. Scholars (e.g. Dinsmoor, W. B., The Architecture of Ancient Greece [London 1950] 142 n. 1)Google Scholar have dated the Throne to about the same period as the Siphnian Treasury and the two monuments share comparable extant architectural features (e.g. Fiechter pls 13–14). If the Siphnian Treasury dates from the 470s, then could a similar date be appropriate for the Throne at Amyclae? Furthermore, our information about its architect, Bathycles of Magnesia (=Overbeck nos 360–1) is chronologically inconclusive. We do not know when he lived, but he was almost certainly not (pace Levi ad Paus. iii 18.9) ‘an early or mid-sixth century sculptor’ and a contemporary of Croesus; Paus. iii 10.8 precludes such an assumption since Pausanias states that the image of Apollo for which Croesus sent gold was a crude piece of work not made by Bathycles (ἔργον δὲ οὐ Βαθυκλέους ἐστὶν ἀλλὰ ἀρχαῖον καὶ οὐ σὺν τέχνῃ πεποιημένον). Against a view that the Throne was refurbished in the 470s it might, however, be argued that Amyclae lacked the political clout to win such attention, for ‘in contrast with the Spartan cult of Orthia, Amyclae was something of an outsider in Spartan politics’ (Cartledge [n. 103] 107–8). On the other hand, Amyclae is prominent in two crucial events involving Spartan conduct during the Persian Wars. In 490 the Spartans were reluctant to depart for Attica because they were celebrating the Carneian festival at Amyclae (Hdt. vi 106). Eleven years later in 479, however, the Amyclaean Hyacinthia was disrupted when the Spartan army left Laconia before the conclusion of the festival (Hdt. ix 7–11). This action was a breach of custom, contrast Aristomenes' conduct in the Second Messenian War in bringing about a truce so that his army could return to celebrate the Hyacinthia (Paus. iv 19.3) or the behaviour of the hoplites from Amyclae who in 390 were allowed to go home from the invasion of Corinth, even though they were on active service (Xen., Hell. iv 5.11Google Scholar). For a general discussion, see Pritchett, W. K., Ancient Greek Military Practices i ( = The Greek State at War i) (Berkeley/L.A. 1971) 125Google Scholar. A post-war Throne of Apollo could represent a thank-offering to the god (on the part of a state unlikely, for example, to build a Treasury at Delphi) and a recompense for having disrupted Apollo's festival in the year of Plataea. Pausanias' ‘Charites and Horae’ (appropriate enough in an epinician context) may in fact have been ‘Vitruvian’ Caryatids if J. Borchhardt is correct in his view that by the Roman period ‘die Bedeutung der Karyatiden nicht mehr verstanden wurde’.

¹²⁶ M. Y. Goldberg (n. 39) 199, 212–13, has assembled a phalanx of Nike acroteria of ‘525’; it would be good to know what victories they reflect.

¹²⁷ E.g. Hdt. i 4; the Eion poem (Jacoby, F., Hesp. xiv [1945] 186–7, 203Google Scholar); Paus. iii 9.3 (Agesilaus of Sparta sacrificing at Aulis before attacking Persia).

¹²⁸ Boardman, J., ‘The Parthenon frieze—another view’, Fests. F. Brommer (Mainz 1977) 39–49, pl. 16Google Scholar.

¹²⁹ Clairmont, C., JHS lxxix (1959) 211–12Google Scholar.

¹³⁰ Hdt. i 4: on this passage, however, see White, M. E., ‘Herodotus' starting-point’, Phoenix xxiii (1969) 39–48CrossRefGoogle Scholar.

¹³¹ Watrous, L. V., ‘The sculptural program of the Siphnian Treasury at Delphi’, AJA lxxxvi (1982) 169–71Google Scholar, but we do not agree with his specific interpretation, for example, of their helmet crests.

¹³² This image has inherently chauvinistic implications at Delphi and may have served as a particular warning to Heraclid states. For example, ‘not many years before Xerxes' invasion of Greece’, the Phocians made a dedication at Delphi to commemorate a victory over the Thessalians (Hdt. viii 27; according to Westlake, H. D., ‘The medism of Thessaly’, JHS lvi [1936] 15Google Scholar, the war with Phocis ‘can scarcely have been concluded before 500’). With a tithe of the booty won in battle, the Phocians had set up groups of large statues (μεγάλοι ἀνδρίαντες) at Abae and around the tripod in front of the Temple of Apollo at Delphi (Hdt. loc. cit.). The Delphian statues were still there in Pausanias’ day: ‘Heracles and Apollo are holding out the tripod and are set to fight over it, with Athena restraining Heracles' anger and Leto and Artemis restraining Apollo’ (Paus. x 13.4). For recent discussions of the Struggle for the Tripod, see Watrous (n. 131) 167 nn. 67 ff. The best account remains that of Passow, F., ‘Herakles der Dreyfussräuber auf Denkmaalen alter Kunst’, in Böttiger, C. A. (ed.), Archäologie und Kunst (Breslau 1828) 125–64Google Scholar.

¹³³ Robertson, M., Between Archaeology and Art History (Oxford 1963) 19–20Google Scholar; Cf. id. (n. 5) 152; id., A Shorter History of Greek Art (Cambridge 1981) 41; Wegner, M., ‘Gleichzeitigesungleichartiges’, Fests. Hans Erich Stier (Münster 1972) 72–87Google Scholar.

¹³⁴ Robertson (n. 5) 156.

¹³⁵ Ibid. 155–6. Nevertheless, some scholars, for example J. Kleine, have tried to explain this stylistic discrepancy in chronological terms: ‘Danach wären Nord und Ostfries vor 525 begonnen, die (Katastrophe von Siphnos hätte eine Arbeitsunterbrechung bewirkt, Sud- und Westfries schliesslich wären um oder nach 520 gearbeitet worden’ (Kleine [n. 5] 32 n. 89). The assumption that a necessary correlation exists between apparent stylistic development and historical time derives in no small measure from the influence of H. Brunn on nineteenth-century attitudes to the study of Greek art. In this connection his observations regarding the Aeginetan pediments make for highly instructive reading: Über das Alter der aeginetischen Bildwerke (SBMünchen 1867.1 4) 405–28Google Scholar. We are most grateful to Prof. Philipp Fehl for bringing this reference to our attention. For evidence for the simultaneous production of works in several different styles at another period, see Vickers, M., ‘Fifth-century brickstamps from Thessaloniki’, BSA lxviii (1973) 292–4Google Scholar. For an excellent critique of the principles of method at issue in evaluating the relationship between style and history, see Stewart, A. F., Attika: Studies in Athenian Sculpture of the Hellenistic Age, Soc. Prom. Hell. Stud. Suppl. Paper xiv (London 1979) ch. 6Google Scholar, ‘Time and Style’, 133–54.

¹³⁶ Robertson 1981 (n. 133) 41.

¹³⁷ Cf. the illuminating remarks on the different ‘syntaxes’ employed by Mantegna, and Dürer, in Ivins, W. M., Prints and Visual Communication (New Haven 1953) 60–1Google Scholar.

¹³⁸ Robertson 1963 (n. 133) 19; Cf. Beazley, J. D., ‘Marpessa’, Charites: Fests. Ernst Langlotz (Bonn 1957) 136–9, esp. 138–9Google Scholar on ARV ² 362.21 and 280.18.

¹³⁹ Ridgway (n. 5) 270.

¹⁴⁰ Ibid., and 279; see also Jeffery (n. 113) 185.

¹⁴¹ Stewart, A. F., ‘Aristion of Paros’, AAA ix (1976) 257 ffGoogle Scholar.

¹⁴² Lullies (n. 5) 48.

¹⁴³ Stewart (n. 141).

¹⁴⁴ Robertson (n. 5) 150.

¹⁴⁵ Ashmole, B., ‘The relationship between coins and sculpture’, Transactions of the International Numismatic Congress (London 1938) 17–22Google Scholar; see further however Price–Waggoner (n. 25) 66–8.

¹⁴⁶ Kraay, C.M., Archaic and Classical Greek Coinage (London 1976) 62–3Google Scholar. A relatively low date for the origin of Early Owls has come in for criticism from those who still prefer to associate this coinage with Solon's currency reforms (notably Cahn, H. A., Kleine Schriften zur Münzkunde und Archäologie [Basel 1975] 81–97Google Scholar). Not only does ‘the evidence of finds appear to run counter to this conclusion’ (Kraay 56, and see next note), but the stylistic arguments made in support of such a view can also be accounted for (though not, of course, the association with Solon) if our dating of the Siphnian Treasury is accepted.

¹⁴⁷ Price–Waggoner (n. 25) 63. Their chronological conclusions have been challenged by Cahn, H. A. (‘Asiut: Kritische Betnerkungen zu einer Schatzfund-publikation’, Schweiz. Num. Rundschau lvi [1977] 279–87)Google Scholar who (281–2) regards the Persepolis foundation deposits, which include Greek coins, as providing a terminus ante quem of 513 BC. This view depends on Schmidt's, E. F. assumption (Persepolis i [Chicago 1953] 39–40, 63, 70Google Scholar; ii [1957] no, 113–14, pl. 84, nos 27–8) that the building of the Apadana was started before the campaign against the European Scythians because no reference is made to this people on the short list of countries under Persian rule on the foundation tablets. The fact that there is a notable lack of unanimity regarding the date of the Scythian expedition (Cf. Price–Waggoner 129; Root, M. C., The King and Kingship in Achaemenid Art [Leiden 1979] 75Google Scholar) is not directly relevant to the question. We concur with M. Roaf that the ‘omission [of the Scythians] is not conclusive’ and follow him in believing that the ‘Apadana reliefs and the East Door of the Central Building were probably designed and started a few years before Darius' death in 486’ (‘The subject peoples on the base of the statue of Darius’, Cahiers de la Délégation Française d'Archéologie en Iran iv [1974] 90–1Google Scholar). If such were indeed the case, then the Persepolis deposits would no longer constitute a major difficulty for the downdating of early Greek coinage.

¹⁴⁸ Price–Waggoner (n. 25) 56.