^{page 29 note 1} Theben, (Freiburg i. B. 1890). See Frazer, , Paus. v. pp. 31 ff.Google Scholar; Hitzig, und Blümner, , Paus. iii. pp. 411 ff.Google Scholar; Tucker, , ed. Seven ag. Thebes, pp. xi. ff.Google Scholar; Baedeker, etc. For earlier theories see Ulrichs, H. N., Reisen u. Forschungen, ii. pp. 1 ff.Google Scholar; Bursian, , Geographie, i. pp. 226 ff.Google Scholar; Forchhammer, , Topographia Thebarum (Kiel, 1854, Progr.).Google Scholar

^{page 29 note 2} Πρακτικά, 1892, pp. 41–6: 1893, pp. 18 ff.; Soteriádes, , Παρνασσός, 1900, pp. 140–170.Google Scholar

^{page 29 note 3} Hermes, xxvi. pp. 191 ff., esp. 194.

^{page 29 note 4} v. p. 33. The chief objections to Fabricius' view are that the sherds he saw did not lie along the line of his supposed western wall, but scattered over a wide area, that they were not rooftiles, and that he has to suppose that the walls of Thebes were not of stone but of mud-brick, which is improbable in itself, for Thebes is not far from stone quarries (see esp. Ulrichs, op. cit. p. 23, n. 4), and impossible in view of the legend of Amphion: Wilamowitz, p. 197, and see Eur. Phoen. 115–116, 797. So much for his west wall; his eastern wall just disappears on Soteriádes' map. For the fragments of stone foundations seen by him and Kalopais, see the illuminating remarks of Soteriádes, p. 157.

It may be added that Fabricius' view gave Thebes a circuit of from 4½ to 5 miles, which he compared with the statement of Dionysios Kalliphon (ll. 93–5, Geogr. Gr. Min. i. p. 241) that the circumference was 43 stadia; while the extension eastwards supposed by Soteriádes implies a circuit more nearly approaching the 70 stadia given by Herakleides (ib. i. p. 102). Thus does literature prove a useful handmaid to archaeology. Dionysios (‘der arme Schächer,’ as Wilamowitz, p. 208, calls him) is generally believed rather than Herakleides: the latter's statement is either an exaggeration (due to Theban ‘Grosssprecherei’ according to Fabricius; as though this writer were a flatterer of Thebes), or a mistake of the MSS. The MSS. of Dionysios would be more trustworthy as being in verse.

^{page 30 note 1} Their highest points (to the South) are 90 and 80 m. respectively above the plain to the North; that of the Kadmeia 63 m. only: Fabricius, p. 11.

^{page 31 note 1} iii. 17. 1.

^{page 31 note 2} Cf. Fraser, iii. p. 344; Hitzig and Blümner, i. p. 800.

^{page 31 note 3} ἡ κάτω πόλις ix. 7. 6, 5. 2; ii. 6. 4: ἡ ὑπὸ τῇ Καδμειᾳ ii. 5. 2.

^{page 32 note 1} Op. cit. p. 11. Pindar (Fr. 196, ed. Bergk) also speaks of the acropolis as ‘the great rock,’ λιπαρᾶν Θηβᾶν μέγαν σκόπελον and cf., ὄχθος ἀκρότατος of the oracle, ap. Schol. Eur. Phoen. 638.

^{page 32 note 2} Geogr. Gr. Min. i. p. 102 (F.H.G. ii. p. 258).

^{page 32 note 3} καὶ ίπποτρόφος δὲ ἀγαθή, κάθυδρος πᾶσα, χλωρά τε καὶ γεώλοφος, κηπεύματα ἔχουσα πλεῖστα τῶν ἐν ῾Ελλάδι πόλεων κ.τ.λ. Mr. Fraser (v. p. 27) translates γεώλοφος ‘with deep soil’; and in his introduction (i. p. xlv.) he omits it altogether. The writer's use of the word is strange in any case, however, for he calls Chalkis hilly: ἡ δὲ τῶν Χαλκιδέων πόλις ἐστὶ μὲν σταδίων ό . . . γεώλοφος δὲ πᾶσα καὶ σύσκιος, κ.τ.λ. Chalkis comprises one or two hills, but they are all very low (except Karababá, which is the other side of the Euripos, but probably belonged to Chalkis (§ 27).

^{page 32 note 4} Op. cit. p. 6.

^{page 32 note 5} Op. cit. p. 14. Soteriádes (pp. 169–170) only quotes Herakleides at the end of his article, letting him do his work of destruction.

^{page 33 note 1} This of a traveller who apparently spent a winter and a summer in Thebes, and knew the inhabitants well.

^{page 33 note 2} Cf. Fraser, , Introd. i. p. xliii. n. 1Google Scholar: ‘the prosperous condition of Thebes which the writer depicts came to an end after Sulla's rigorous treatment of the city in 86 B.C. (Paus. ix. 7. 5 ff.).’

^{page 33 note 3} When the ‘two rivers’ of Thebes are mentioned the Dirke and Ismenos seem always to be meant. See Aesch. Sept. 273, Eur. Bacch. 5, and the Schol. to Eur. Phoen. 818–825; and cf. Eur. Suppl. 621. If Fabricius were right, it would mean that Herakleides does not mention the Ismenos among the rivers ‘that water the plain.’ Note, too, that Pausanias never mentions the smaller stream (cf. Forchhammer, p. 9).

^{page 34 note 1} Cf. Strabo, ix. 1. 16, p. 396: τὸ δ᾿ ἄστυ αὐτὸ πέτρα ἐστὶν ἐν πεδίῳ περιοικουμένη κύκλῳ

^{page 34 note 2} Varro is the only ancient writer who says Thebes was hilly: de Re Rust. iii. 1. 6; so Bursian, i. p. 225. But Varro is not comparable as an authority on such a point with the Greek travellers.

^{page 34 note 3} Op. cit. 1. p. 102. Καὶ γὰρ ποταμοὶ ῥέουσι δι ᾿αὐτῆς δύο τὸ ὑποκείμενον τῇ πόλει πεδίον πᾶν ἀρδεύοντες Φέρεται δὲ καὶ ὰπὸ τῆς Καδμείας ὔδωρ ἀφανὲς διὰ σωλήνων ἀγόμενον ὑπὸ Κάδμου τὸ παλαιόν, ὡς λέγουσι, κατεσκευασμένων

^{page 34 note 4} Op. cit. p. 4. See also Unger, , Thebana Paradoxa, p. 111.Google Scholar

^{page 34 note 5} Op. cit. p. 14.

^{page 35 note 1} Xen. Hell. v. 4. 8; Arrian, i. 8. 6; Plut. de gen. Socr. 4, p. 577 B; cf. Ulrichs, op. cit. p. 17; Fabricius, pp. 19, 31.

^{page 35 note 2} Cf. Eubulos, ap. Athen. x. 11, p. 417. κοπρῶν᾿ ἔχει ᾿πὶ ταῖς θύραις ἔκαστος to complete the picture. The former statement of Herakleides about the climate of Thebes I can well believe: it is at the present day very much fresher and greener than most towns in Greece. To the latter I can testify from personal experience, ranging from November to March. Cf. Wilamowitz, op. cit. p. 207, who alone of modern scholars has suggested an extension of the town northwards.

^{page 35 note 3} Neither Fabricius nor Soteriádes suggest the possibility of a change.

^{page 36 note 1} Anabasis, i. 7 ff.

^{page 36 note 2} Ibid. i. 8; and cf. the opening words of the preface.

^{page 36 note 3} Ibid. i. 7. 7.

^{page 36 note 4} Paus. ix. 23. 1.

^{page 36 note 5} Ibid. ix. 18. 1. Note that a hippodrome, a stadion, and the theatre were all near this gate, the first two πρὸ τῶν πυλῶν the latter κατὰ τὰς πύλας (Paus. ix. 23. l) The stadion was a γῆς χῶμα like that at Epidauros and Olympia, that is, the sides were formed by artificial embankments: it was not formed by the side or sides of a hill as at Delphi and to some extent at Athens; though in the hilly country at Thebes one might have expected it: this perhaps is additional evidence that the gate was down in the flat plain.

^{page 36 note 6} The ῾Αμφεῖον is near the στοά (i.e. the ὰγορά) in Xen. Hell. v. 4. 8; cf. 2. 29; Paus. ix. 17. 1–2 (and Fraser, ad loc.); Soph. O.T. 161; Plut. Vit. Arist. 20, fin.

^{page 36 note 7} Op. cit. p. 8.

^{page 36 note 8} Sev. ag. Thebes, 1. 377. If DrVerrall, (ed. Seven ag. Thebes, 1887, Introd. p. xix.)Google Scholar was right in supposing the order of the gates as enumerated by Aeschylus to be the topographical order too, then the Proitidian Gate will be on the South-East, for it is next to the Elektran on the South (v. infra). And the Neistan (leading to the West) will be S.W. of the town, not N.W., where Fabricius places it. If Fabricius' views on the topography of Thebes are correct, the Neistan Gate could not possibly be to the South-West. Note that an ᾿Αγοραῖος ῾Ερμῆς and the shrine of Artemis Soteira (which was in the agora, Soph. O.T. 161) was near the gate Proitides (Paus. ix. 17. 1, 2).

^{page 37 note 1} ὡς μήτε ἔξωθεν τινὰ τοῖς ἐγκατειλημμένοις δύνασθαι ἐπωφελεῖν μήτε αὐτοὺς ἐκθέοντας βλάπτειν τι σφᾶς ὅποτε τοῖς ἔξω πολεμίοις προσφύγοιντο. This implies, I think, that it was necessary for the Thebans to get between the Kadmeia and the besiegers, and hence that, except on one side (the North), the lower town did not touch the Kadmeia—did not, as supposed, surround it on three sides. Cf. the parallel passage in Diodoros (xvii. 8. 4): τὴν Καδμείαν τάφροις βαθείαις καὶ σραυρώ- μασι πυκνοῖς περιέλαβον, ὥστε μήτε βοήθειαν αὐτοῖς δύνασθαι, μήτ᾿ ἀγορὰν εἰσπέμψαι. But Diodoros is perhaps not much help: see below.

^{page 37 note 2} Paus. ix. 8. 7. Cf. Eur. Bacch. 780 ff.; Suppl. 651 ff.

^{page 37 note 3} Fabricius and Soteriádes on their maps give the line of the ancient road to Athens, and the position of the Elektran gate, below the Kadmeia to the East.

^{page 37 note 4} To do this it was necessary to capture and cross the two χάρακες οτ προφυλακαί that the Thebans had formed: they then pursued them within these till they reached the Herakleion. This indicates, too, that the κοίλη ὁδός is the bed of the stream just below the Kadmeia to the East, and that the defences of the Thebans were at its Southern end and along the hill East of it (as well, presumably, as West of the Kadmeia). On Fabricius' plan there would be no room for a pursuit between the outworks and the wall of the town. (Soteriádes, however, loc. cit. p. 163, thinks it unlikely that the Thebans would have built outworks on all three sides of the Kadmeia; and hence that the Eastern and Western walls of the citadel were within the city circuit.)

^{page 38 note 1} See Fabricius, pp. 18, 31.

^{page 38 note 2} This would suit very well, too, with the position of the ῾Ηράκλειον to which the κοίλη ὁδός led, if the shrine mentioned by Arrian is the same as that mentioned by Pausanias as near the Elektran Gate. ‘Left of the gate called Elektran is the house of Amphitryon . . . ᾿Ενταῦθα ῾Ηράκλειόν ἐστι ᾿ (ix. 11. 4). Ulrichs (p. 7) and Fabricius (p. 22) suppose it to have occupied the site of the present chapel of St. Nicholas, South of the Kadmeia. It was clearly outside the city. Arrian implies this, and Pausanias seems to do so too; for the house of Amphitryon, next door to the shrine, was ‘left of the gates’ just as the Ismenian hill was ‘to the right of the gates’; and τοῦ ῾Ηρακλείου γυμνάσιον ἔχεται καὶ στάδιον (ix. 11. 7), and the latter at least must have been outside. Pherekydes, fr. 39 (ap. Anton. Lib. 33. 3), supports this view of the site of the Elektran Gate: after the defeat of Eurystheus, Hyllos and the other Herakleidae returned to Thebes: ᾤκουν δὲ παρὰ τᾶς ῾Ηλέκτρας πύλας, ὅθιπερ καὶ ὁ ῾Ηρακλῆς ἐν τῇ ἀγορᾷ. For the Agora was probably situated in the hollow just East of the hill called Ampheion (see Fabricius' map reproduced on Pl. XIX. A.). Herakles was born near this gate according to Schol. T. on Iliad, xix. 99. (In the Teubner ed. of Antoninos, Wachsmuth's conjecture ἐν ορώτῃ ὤρᾳ Γογ ἐν τῇ ἀγορᾁ is adopted.)

If this gymnasium and stadion near the Herakleion is the same as the gymnasium and stadion of Iolaos, near the Proitidian Gate (see above, p. 36) which seems possible, though Pausanias would be guilty of a confusion if this were so, then the Proitidian Gate would certainly be next the Elektran, where Dr. Verrall placed it.

^{page 38 page 3} Diodoros' account of the capture of Thebes does not help us much (xvii. 8–14): it differs largely from Arrian's account, and always for the worse: he never seems to speak with anything like the same authority. He is much vaguer in telling the story, and the details that he does give concern the turning of the water of the Dirke into blood and not the military operations of Alexander and the Thebans. As far as it goes (if it has any topographical value at all) it supports the view that the town lay in the plain; for the fighting takes place outside the city, and cavalry and the Macedonian phalanx, both unsuited for operations on hilly ground, play a prominent part; §11. 2. ἱππεῖς ἐντὸς τοῦ χαρακώματος (round the city wall), αὐτοὶ μὰχην πρὸ τῆς πόλεως ἡτοιμάζ- οντο . . .μέγας ἀγὼν συνίστατο . . . διὰ τὸ βάρος τῆς φάλαγγος. Finally (§12. 3) Alexander sends Perdikkas in by τινὰ πυλίδα inadvertently left open … 4. εὐθὺς ἀνεχώρησαν (sc. οἱ Θηβαῖοι) ἐντὸς τῶν τειχῶν . . . 5. οἱ . . . ἱππεῖς . . . συνέτρεχον εὶς τὴν πόλιν . . . 6. οἱ δὲ τὴν Καδμείαν φρουρο῀ντες ἐκχυθέντεδ ἐκ τῆς ὰκροπόλεως ἀπήντων τοῖς Θηβαίοις καὶ τεταραγμένοις ἐπιπεσόντες πολύν ἐποίουν φάνον. No other military details does he give. Wilamowitz (op. cit. p. 202, Anm. 1) considers the whole to be a poetized account, a κατεψευσμένη ἱστορία modelled on Euripides' Supplices; while the description in Pseudo-Kallisthenes (i. 46) is still more highly poetized. Plut. Vit. Alex. 11 tells us nothing about the topography, but gives us the number of casualties (6,000 killed, 30,000 taken prisoners; with which Diodoros § 14, agrees). Justin (xi. 2–4, the epitome of Pompeius Trogus) gives us no information either.

In general, note Eur. Suppl. 618, τὰ καλλίπυργα πεδία πῶς ἱκοίμεθ᾿ ἄν and observe that in the Phoenissae all the enemy's forces can be seen at the same time from one spot on the Kadmeia (ll. 1356–8; cf. 101–2). This would be easy enough if they surrounded a town down in the plain; so that it is not necessary to adopt Dr. Verrall's hypothesis of a town confined to the Kadmeia (see below, p. 48, n. 3).

^{page 39 note 1} See above, p. 34, n. 3.

^{page 39 note 2} Op. cit. p. 6.

^{page 39 note 3} Ps. Plut. de fluv. ii. (δ Ἰισμηνός), 1 (init.).

^{page 39 note 4} Var. Hist. xii. 57. ἡ δὲ περὶ τὸν ᾿Ισμήνιον καὶ αὐτὰ τὰ τείχη ῥέουδα κρήνη καλουμένη Δίρκη καθαρῷ καὶ ἡδεῖ ῥέουσα ὕδατι παρὰ πάντα τὸν πρόσθεν χρόνον ἄφνω καὶ παρ᾿ ἐλπίδα αἴματος ἀνεπλήσθν

^{page 40 note 1} Paus. ix. 10. 2. Pausanias says nothing as to whether the river Ismenos flowed through or outside the walls, merely adding here that it flowed by the hill Ismenion.

^{page 40 note 2} ix. 2. 24, pp. 408–9: ὁ ῾Ετεωνὸς δὲ Σκάρφη μετωνομάσθη, καὶ αὕτη δὲ τῆς Παρασωπίας. ὁ γὰρ Ασωπὸς καὶ δ ᾿Ισμηνὸς διὰ τοῦ πεδίου ῥέουσι τοῦ πρὸ τῶν Θηβῶν ἔστι δὲ καὶ ἡ Δίρκη κρήνη καὶ Πότνιαι . . .

^{page 40 note 3} Fraser, V. pp. 134–5.

^{page 40 note 4} vv. 823 ff., see below, p. 44, n. 3.

^{page 40 note 5} ix. 25. 1–2.

^{page 41 note 1} Paus. ix. 25. 5, 26. 5.

^{page 41 note 2} ix. 25. 3.

^{page 41 note 3} Op. cit. pp. 25–6.

^{page 41 note 4} Ulrichs (op. cit. p. 8) supposed that Pindar lived in Kynoskephalai where he was born, and that Kynoskephalai was a suburb of Thebes immediately West of Dirke; but this Fabricius (loc. cit.) showed to be unfounded. See the references in Frazer (v. p. 135) and Hitzig (iii. pp. 466 and 468).

^{page 42 note 1} Or he may be supposed to be approaching the city from the North (ch. 24–5), mentioning first what was to be seen near the gate, then entering as far as Pindar's house, then returning to the road westwards.

^{page 42 note 2} ix. 8. 4: Θηβαίοις δὲ ἐν τῷ περιβύλῳ τοῦ ἀρχαίου τείχους ἐπτὰ ἀριθμὸν ἧσαν πύλαι, μένουσι δὲ καὶ ἐς ἡμᾶς ἔτι (Wilamowitz, loc. cit., claims that there never were seven gates; but at any rate their supposed sites would be shown to Pausanias, and that is sufficient for following his topography.)

^{page 42 note 3} In which case Arrian's ‘Gate leading to Eleutherai and Attica’ might not be the same as the Elektran Gate, but a new one in the new town-wall. But there would not be much difference in the position of gates on the South side of the town. Keramópoullos (᾿Εφ.᾿Αρχ 1909, pp. 110–111, n. 1) recognizes that the ‘old wall’ of Pausanias is not the same as the later circuit wall, but thinks it was the acropolis wall, and that all the seven gates were in the Kadmeia.

^{page 43 note 1} Aesch. Seven ag. Thebes, 358–660; Eur. Phoenissae, 1120–54.

^{page 43 note 2} Apollodor. iii. 77 (ed. Wagner): ᾿Αμφιαράῳ δὲ φεύγοντι παρὰ ποταμὸν ᾿Ισμηνόν, πριν ὑπὸ Περικλυμένου τὰ νῶτα τρωθῇ, Ζεὺς κεραυνὸν βαλὼν τὴν γὴν δείστησεν So too Pindar, (Nem. ix. 22 f.)Google Scholar—

᾿Ισμηνοῦ δ᾿ ἐπ᾿ ὄχθαισι γλυκὺν νόστον ἐρεισάμενοι κ.τ.λ.

^{page 43 note 3} Seven ag. Thebes, 377–9.

Τυδεὺς μὲν ἤδη πρὸς πύλαισι Προιτίσιν βρέμει, πόρον δ᾿ ᾿Ισμηνὸν οὐκ ὲᾷ περᾶν ὁ μάντις οὐ γὰρ σφάγια γίγνεται καλά

In Apollodoros (iii. 68) it is Amphiaraos who is stationed at the Proitidian gate, and so near the Ismenos.

^{page 43 note 4} ll. 383–4.

^{page 44 note 1} See above, p. 36, and n. 8.

^{page 44 note 2} It is perhaps supported by Pausanias (ix. 8. 3), who says that the place where Amphiaraos was swallowed up by the earth was to the right of the road from Potniai to Thebes, that is, to the South-East of the town.

^{page 44 note 3} 818 ff.

῾Αρμονίας δέ ποτ᾿ εἰς ὑμεναίους ἤλυθον οὑρανίδαι, φόρμιγγί τε τείχεα Θήβας τᾶς ᾿Αμφιοϝίας τε λύρας ὔπο πύργος ἀνέσταν διδύμων ποταμῶν πόρον ἀμφὶ μέσον Δίρκας, χλοεροτρόφον ἄ πεδίον

^{page 44 note 4} Op. cit. p. 8.

^{page 44 note 5} Op. cit. pp. 25–6.

^{page 44 note 6} Op. cit. iii. p. 466.

^{page 44 note 7} Dindorf, , Scholia in Euripidis tragoedias, iii. p. 231Google Scholar: λείπει τὸ ὄς ὄς ὁ πόρος κατὰ τὸ πεδίον τῆς Δίρκης ἔμπροσθεν τοῦ ᾿Ισμηνοῦ καταδεύει. οὖτοι γάρ εἰσιν οἰ δύο ποταμοί, ἠ τε Δίρκη καὶ ὁ ῾Ισμηνός. See Wilamowitz, op. cit. p. 200, who says the whole passage as it stands is unintelligible, and would read (after Schenkl): πόρον ἀμφὶ μέσον, Δίρκα χλοεροτρόφον ᾆ πεδίον κ.τ.λ.. But the corruption must be older than the Scholiast, who has apparently read ἆμ πεδίον. Wilamowitz also takes the ‘two rivers’ to be Dirke and Ismenos.

^{page 45 note 1} ἀνέστη (i.e. τῆς πόλεως τὰ τείχη) δὲ περὶ τὸ μέσον τῶν δύο ποταμῶν τῆς Δίρκης, ἤγουν οὔς ἡ Δίρκη ποιεῖ σχιζομένη, ἤτις καταρρέει εἰς τὸ πεδίον ἔμπροσθεν τοῦ ᾿Ισμηνοῦ ποταμοῦ (ibid. p. 229, 1. 28), and on v. 825: ἄλλως. περὶ τὸν μέσον τῶν διδύμων ποταμῶν τῆς Δίρκης, ἤτις σχιζομένη δύο ἀποτελεῖ ποταμούς, οὔς καὶ Διδύμους ἐκάλεσαν. ἐπεὶ ἔξ ἐνὸς οἱ δύο ἀποτελοῦνται (ibid. p. 231). Others seem to combine the two meanings, as (on v. 825): ἤτις Δίρκη εἰς δύο σκιζομένη τὸ ἔμπροσθεν τοῦ ᾿Ισμηνοῦ πεδίον καταρδεύει. οὖτοι γάρ εἰσιν οἱ δύο ποταμοὶ ἡ τε Δίρκη καὶ ὁ ᾿Ισμηνός (ibid., and cf. the confusing n. to 1. 818, ibid. p. 229, 1. 8). The latter of these two interpretations, that supposing that the Dirke divided into two, certainly appears to fit the Greek of Euripides better. And there are other instances where the Greeks supposed that a river might divide into two near its source and become two rivers: the Danube was thought to have a long branch flowing southward into the Adriatic, as well as that flowing into the Euxine (Bunbury, , Hist. Gk Geography, i. pp. 384, 388Google Scholar; cf. ii. pp. 25–6, 357, 398), and Aristotle thought that the Tanais was a branch of the Araxes (that is, probably, the Iaxartes), which flowed into the Palus Maeotis (the Sea of Azov), while the main river flowed into the Caspian (ibid. pp. 399–400, 433–4; cf. the ‘island’ formed by the Oeroe near Plataia, Hdt. ix. 51). But to believe this of great and very distant rivers is one thing; to believe it of a small stream so near as Thebes is very different. No one who had been on the spot, or who was in contact with people who had been on the spot, could have had so strange a notion about the Dirke.

^{page 45 note 2} Two other lines in the same play (101–2) perhaps support this view, though they are not conclusive; the retainer says to Antigone:

σκόπει δὲ πεδία καὶ πὰρ᾿ ᾿Ισμηνοῦ ῥοάς Δίρκης τε νᾶμα, πολεμίων στράτευμ᾿ ὄσον

^{page 45 note 3} Phoen. 730.

^{page 45 note 4} ll. 131–2. No other details of Tydeus' station are given here. At ll. 1119–20 he is at the gate Homoloides.

^{page 46 note 1} Ll. 1009–12:

ἀλλ᾿ εῖμι, καὶ στὰς ἐξ ἐπάλξεων ἄκρων σφάξας ἐμαυτὸν σηκὸν ἐς μελαμβαθῆ δράκοντος, ἔνθ᾿ ὁ μάντις ἐξηγήσατο, ἐλευθερώσω γαῖαν

Cf. ll. 932 ff.

^{page 46 note 2} Published by Mahaffy, J. P. in Hermathena, viii. 1893, pp. 38–51Google Scholar; Antiope C, right col. ll. 30 ff. Hermes as deus ex machina is telling Lykos to throw Dirke's bones into Ares' spring (and then tells Amphion and Zethos to build the walls of Thebes)—

ὡς ἄν τὸ Δίρκης ὄνομ᾿ ἐπώνυμον λάβῃ κρήνης ἀπορροὺς ὄς δίεισιν ἄστεος πεδία τὰ Θήβης ὄδασιν ἐξάρδων ἀεί. ὑμεῖς δ᾿ ἐπειδὰν ὄσιος ῃ Κάδμου πόλις χωρεῖτε [παῖδε]ς ἄστυ δ᾿ ᾿Ισμηνὸν πὰρα ἔπτάστομον πύλαισιν ἐξαρτύετε.

Wilamowitz at the end of his article in Hermes (pp. 241–2) quotes this fragment in an Appendix to prove his point that Ares' spring and Dirke are one and the same, but fails to notice that it also disproves his contention that the Dirke flowed outside the walls, for which he relies on the passage in the Phoenissae quoted above.

^{page 46 note 3} It may be that the inconsistency of the statements with regard to the Ismenos (especially the Antiope-fragment, which seems to distinguish the positions of the Ismenos and the Dirke) can be explained by the extension of the city boundaries by the Lacedaemonians after the battle of Tanagra in 457 (Diod. xi. 81): before this the Ismenos was outside the walls, afterwards inside; and confusion would not be unnatural. But I do not think it probable (see above, p. 42).

^{page 47 note 1} Fabricius tries to reconcile the evidence of the Tragedians with his theories; Soteriádes, more logically, denies their trustworthiness on the matter—a very possible view—but then his plan is even more inconsistent with Arrian, Herakleides, and Pausanias than with the fifth-century evidence. (The description of the battle before Thebes in Euripides Supplices is not sufficiently clear to us to afford any valuable evidence, and the text may have been disturbed.)

^{page 47 note 2} Strabo, ix. 2. 42, p. 416.

^{page 47 note 3} Θηβαίοις δὲ ἐπὶ μὲν Κασσάνδρου πᾶς ὁ ἁρχαῖος περίβολος ἀνῳκίσθη (ix. 7. 4); neither Diodoros (xix. 53. 2, 54. 1) nor Plutarch (de republ. ger. xvii. 9) hints at any change.

^{page 47 note 4} τῆς μὲν τῶν Θηβαίων πόλεως μείζονα τὸν περίβολον κατεσκεύασαν (Diod. xi. 81. 3).

^{page 48 note 1} ix. 5. 6.

^{page 48 note 2} As he does in the case of the Arcadian Orchomenos, which was once on a hill, afterwards in the plain below: ῾Ορχομενίοις δὲ ἡ προτέρα πόλις ἐπὶ ὄρους ἦν ἄκρᾳ τῇ κορυφῇ καὶ ἀγορᾶς τε καὶ τειχῶν ἐρείπια λείπεται τὴν δὲ ἐφ᾿ ἡμῶν πόλιν ὑπὸ τὸν περίβολον οἰκοῦσι τοῦ ἀρχαίου τείχους (viii. 13. 2).

^{page 48 note 3} Cf. esp. Paus. ix. 5. 6; ii. 6. 4; Diod. xix. 53. 4. 5. See Unger, , Thebana Paradoxa (Halle, 1839), i. chaps. 1 and 2.Google Scholar

^{page 48 note 4} Od. xi. 263–5.

^{page 48 note 5} Paus. ix. 16. 7.

^{page 48 note 6} Paus. viii. 33. 2; ix. 7. 6; cf. Dio Chrys. Orat. vii. vol. i. p. 136, ed. Dindorf (p. 263, ed. Reiske); Strabo, ix. 2. 5, p. 403; 25, p. 410.

^{page 48 note 7} Ελλ. Χωρογραφία (3rd ed. Athens, 1901), i. p. 116.

^{page 48 note 8} DrVerrall, , however (ed. Sev. ag. Thebes, pp. xix. ff.)Google Scholar, held that in the Septem Aeschylus definitely thought of the city as occupying even less than the entire surface of the Kadmeia; and though what Aeschylus thought about the early extent of Thebes would not be very important or relevant, yet, as Dr. Verrall has shown that he is in all probability following a very old version of the story, it may be worth while to examine this contention. Dr. Verrall argued principally on the ground that on two points Aeschylus and Pausanias disagree, namely, as to the positions of the altar of ᾿Αθηνᾶ ᾿´Ογκα or or Ὄγγα, and of the grave of Amphion, both of which Aeschylus states to be outside the gates, while Pausanias (he claimed) says they were inside. With regard to Athena Onka, Aeschylus is quite clear: the shrine was near the Elektran Gate, outside the city (ll. 164–5, 486–7, 501–2). ‘But Pausanias says it was in the upper city.’ This is certainly what he implies; but it is to be noted that his language is vague: ἔδει δὲ ἄρα Κάδμον καὶ τὸν σὺν αὐτῳ̑ στρατὸν ἐνταῦθα οἰκῆσαι κατὰ τοῦ θεοῦ τὴν μαντείαν, ἔνθα ἡ βοῦς ἔμελλε καμοῦσα ὀκλάσειν. ἀποφαίνουσιν οὖν καὶ τοῦτο τὸ χωρίον. ἐνταῦθα ἔστι μὲν ὑπαίθρῳ βωμὸς καὶ ἄγαλμα[᾿Αθηνᾶς].ἀναθεῖναι δὲ αὐτὸ Κάδμον λέγουσι (ix. 12 2). He then proceeds to mention the house of Kadmos, the site of which was in the agora of Pausanias' day, on the Kadmeia itself (see now Keramópoullos, , Ἐφ. Ἀπχ. 1909, pp. 57–122).Google Scholar Now if Pausanias really means by this that the image of Athena stood on the acropolis, we can only say that he is wrong; for every other author who mentions it says or implies that it was outside the walls (see the references given by Frazer, v. pp. 48–9): there is even said to have been a village Onkai on the spot (Schol. ad Pind. Ol. ii. 48; Tzetzes, ad Lyk. 1225Google Scholar). But we need not take Pausanias too literally; for the legend of the cow grew up because of the shrine, not vice versa, and a little way off from the Kadmeia would have sufficed for the myth-makers. The cow might have ‘grown tired and sat down’ a little early, and not had sufficient strength left to climb up to the acropolis. It is to be noted that Pausanias is here giving mythological, not geographical, information. He has just mentioned the sacrifices to Apollo Spodios. This is important, owing to his habit of going out of the true topographical order to talk about any myth or religious custom that at the moment may be relevant.

If the order of the gates in Aeschylus is the topographical order, then this shrine of Athena was to the West or North-West of the town, not South-West of the Kadmeia, where Ulrichs (p. 15) and Fabricius (p. 28) place it.

The grave of Amphion was also outside the walls, near the Πύλαι Βορραῖαι the North Gate (Septem, 526–8; cf. Eur. Phoen. 145, Schol. ad loc., Suppl. 663). Pausanias mentions it (ix. 17. 4) immediately after his description of the agora, and evidently wishes to imply that it was within the gates. But though it was inside in his day, and outside in early days, this would not confine the early city to the Kadmeia only, but merely make it of less extent than the city (or rather its ruins) of Pausanias' own day: which no one would deny. Pausanias certainly implies that the tomb was not on the acropolis.

The difficulties in the way of Dr., Verrall's view are, firstly, that the Ismenos is evidently supposed to be quite near the gate Proitides, for the seer would not have opposed Tydeus' advance beyond it, if there was such a wide intervening space as exists between the Ismenos and the Kadmeía (ll. 377–9; above, p. 40); and secondly, that he had perforce, to suppose the Καδμεία πόλις was divided into an ἀκρόπολις and a lower area, all enclosed by the walls (p. xxii; and see his nn. on ll. 88–90, 204, 257). In this he had the support of Wilamowitz, who would ‘die Kadmeia auf eine Citadelle auf der Südkuppe beschränken’ (op. cit. p. 238). But no such division is possible; the Kadmeia is not two hills, but one hill, gently sloping without a break from South to North, till it reaches the steeper slopes at the Northern end. There is no ‘Südkuppe.’ Keramópoullos holds that in prehistoric times the Kadmeia had four separate peaks, and that the palace of ‘Kadmos’ was on the peak furthest North but one, the smooth surface which it now possesses being due to later accretions of soil (᾿Εφ.᾿Αρχ 1909, pp. 107–9). But none of these would have been sufficiently isolated to form an acropolis. (Nor would Aeschylus have known of this.) If the Kadmeia only, was occupied at the time of the Seven, then there was no lower town, only an acropolis. And this is, I think, improbable on other grounds.

^{page 49 note 1} Travels in Northern Greece, ii. p. 226. Leake says the walls were 28 ft. thick, built of large, roughly-hewn stones, like the Τιρύνθιον πλίνθευμα. These were at the North end of the acropolis, near the Frankish tower. But Keramópoullos, (Ἐφ. Ἀρχ. 1907, pp.205–8Google Scholar) asserts that Leake was mistaken and that they belonged to the early classical period. There are some remains of a later wall in the same place, built of isodomous masonry; and it may be remarked that the stones were quarried from the hills round Thebes.

^{page 50 note 1} Bursian, , Geographie, i. p. 225.Google Scholar

^{page 50 note 2} Murray, , Rise of the Greek Epic, p. 36.Google Scholar

^{page 50 note 3} xxvi. (1891), pp. 131–242: ‘Nur für den Angriff auf die sieben Thore hat es Bedeutung dass sieben Argeienhelder gezählt werden; nur für den Sturm der Sieben gegen Theben ist Theben die Stadt der sieben Thore.’

^{page 51 note 1} Pausanias says the πύλαι ῾´γψισται were near a sanctuary of Zeus ῾´γψιστος (ix. 8. 5); and Aristodemos (ap. Schol. ad Eur. Phoen. 419) says the πύλαι ῾Ομολωῖδες were so called διὰ τὸ πλήσιον εἶναι τοῦ ῾Ομολώου ἤρωος (or ὄρους see Wilamowitz, op. cit. pp. 215 f., and Steph. Byz. s.v. ῾Ομόλη and cf. Müller, C., F. H. G. iii. p. 309).Google Scholar Neither of these explanations tells us anything; but they may have been full of meaning to a contemporary.

^{page 51 note 2} loc. cit. p. 258, 12: [πόλις ] ἀρχαία μὲν οὖσα,καινῶς δ᾿ ἐρρυμοτομημένη διὰ τὸ τρὶς ἤδη, ως φασιν αἱ ἰστορίαι, κατεσκάφθαι διὰ τὸ βάρος καὶ τὴν ὑπερηφνίαν τῶν κατοικούντων

^{page 51 note 3} Diodoros, , frag. xxi. 14Google Scholar; Plut. Dem. 39–40.

^{page 51 note 4} Op. cit. v. pp. 26–7.

^{page 52 note 1} The entire description of the campaign is summed up in about four lines (frag. xxi, 14): ὄτι Δημήτριος ὁ βασιλεὺς τὸ δεύτερον ἀποστατησάιτων Θηβαίων πολιορκίᾳ τὰ τείχη καθελών, τὴν πόλιν κατὰ κράτος εἶλε, δέκα μόνους ἀνελὼν τοὺς τὴν ἀποστασίαν κατεργαζομένους. This was evidently not a repetition of the events of 335.

^{page 52 note 2} Herakleides, 14.

^{page 52 note 3} Zosimos, v. 5 (Frazer, loc. cit.). Wilamowitz attributes the very weak and vacillating policy of Thebes during the wars of 197 and 146 B.C. to its defenceless condition: Livy, xxxiii. 1–2; Plut. Tit. 6; and Paus. vii. 15. 10; Polyb. xxxix. 9; Livy, Epit. lii.

^{page 53 note 1} p. 224.

^{page 53 note 2} Ridder, De in Bulletin de Correspondance hellénique, xviii. (1892), pp. 271–310Google Scholar; Noack, , Athenische Mitteilungen, xix. (1894), pp. 154, 405–85.Google Scholar