210 Thucydides (i. 20) denies that such a division or regiment existed in the Spartan army.

211 Even in the Spartan army indiscipline was apt to make its appearance without the existence of such a substantial motive as in the present instance. Cf. the insubordination of the Spartan officers at the battle of Mantinea in 418 B.C. (Thuc. v. 72).

212 That they never reached the rocky ὑπωρέη is plain from the incidents of the battle that followed.

213 Of A 4 and A 5.

214 The ὑπωρέη of Herodotus.

215 Cf. the tale H. ix. 58.

216 Δρόμῳ διαβάντας τὸν Ἀσωπὸν (H. ix. 59).

217 It will be seen, when the details of the Athenian retreat come to be examined, how noticeably this detail accords with the account which Herodotus gives of that retreat.

218 Some modern commentators have regarded this detail mentioned by Herodotus as a convincing proof of the Athenian bias in his narrative. To me it seems eminently natural, after the experience of the previous days, that Pausanias or any other commander should have summoned help under the circumstances. I shall, moreover, have occasion to show that the Athenians did undoubtedly diverge from their march to the Island in the direction in which the Spartan battle with the Persians took place.

219 It is clear from Herodotus’ subsequent account of the proceedings of the Greek centre that this battle took place out of sight of that part of the army which had retired to Platæa.

220 H. ix. 62: Ἤδη ἐγίνετο ἡ μάχη ἰσχυρὴ παρ’ αὐτὸ τὸ Δημήτριον.

221 This incidental detail mentioned by Herodotus peculiarly supports the view that the temple must have stood on the site of the church of St. Demetrion.

222 This is clearly shown in Herodotus’ narrative. He distinctly speaks of the Athenians as having at the beginning of the movement “turned down towards the plain” (H. ix. 56, κάτω τραφθέντες ἐς τὸ πεδίον); and in a still more remarkable passage he says that, when Mardonius led his Persians across the Asopos in pursuit of the Greeks, “he did not see the Athenians, who had turned down towards the plain, by reason of the (intervening) hills” (H. ix. 59). The hills mentioned are evidently the northern extension of the Asopos ridge.

223 A 1 in the map.

224 Ridge 5.

225 Thus far διὰ τῆς ὑπωρέης (H. ix. 69).

226 I.e. ridges 3 and 2; cf. H. ix. 69, διὰ ... τῶν κολωνῶν.

227 Ridge 5.

228 I confess I cannot understand the argument of those who regard Herodotus’ account of Platæa as being tainted throughout with a lying Athenian tradition. In so far as the narrative provides evidence of its source or sources, there is at least as much matter in it which may be attributed to Spartan as to Athenian origin.

229 The Asopos ridge, the Long ridge, and the Plateau.

230 The treatment meted out to the Æginetans in the narrative of Platæa, as contrasted with the account which Herodotus gives of their conduct at Salamis, points to the very various character of the sources from which he drew his history. This part of the Platæan narrative is undoubtedly drawn from a tradition highly coloured by the relations which existed between Athens and Ægina twenty years after Platæa was fought.

231 Xen. Anab. iii. 2, 27. The striking words are μὴ τὰ ζεύγη ἡμῶν στρατηγῇ.

232 I have had occasion to speak of the Thermopylæ narrative under various aspects in relation to the sources from which it is derived.

To prevent any misconception, I should like to sum up briefly my conclusions.

(1) The whole “motivation” of the story is derived from a version of official origin at Sparta.

(2) The incidents of the actual fighting may be derived partly from a Spartan source, probably of an unofficial character. The description of some of them, however, rests on information picked up by Herodotus at Thermopylæ itself from natives of the region.

233 For these two points cf. H. vii. 152.