appear of equal force. There is no doubt, however, that the asyndeton, occurring, as it does, in an address signally free from excitement and passion, is exceedingly harsh; nor can it be justified by such passages as Elektr. 719, ἤφριζον, εἰσέβαλλον ἱππικαὶ πνοαί, and still less by v. 115 below, xew xugi, Qridov undir. The phraseology is drawn from the language of the hunting-field, and contains a reference to the practice of driving the wild animal after which the chase was instituted into a circumscribed inclosure, where it was subsequently entangled in nets and slain. See St. John's "Manners and Customs of Ancient Greece," Vol. I. pp. 222 sqq. We find the same metaphor in Elektr. 1477, iv μécois àgnvστάτοις πέπτωκα; Ibid. 829, χρυσοδέτοις ἕρκεσι γυναικῶν of the necklace of Eriphyle ; Eur. Elektr. 165, ὀλόμενος δολίοις βρόχων ἕρκεσιν ; Æsch. Agam. 1620, ¡dívta Toûtov tñs díxns iv igxov. On the use of gxos in the sense of a net, a snare, see Hdt. 7. 85; Ar. Avv. 528; Pind. Nem. 3. 89; Pyth. 2. 147. HESYCHIUS: ἕρκεσι· δικτύοις. Σοφοκλῆς ̓Αθάμαντι δευτ

τέρω.

óvou, the MSS. Edd., read póvov, On the perpetual

61. Kärur' irudń. Krüger, ad Dion. Hist. p. 376, remarks, “ ἔπειτα id non cacophonon visum est Græcis." The same collocation is found, not only in the poets, as Eur. Sisyph. I. 9, Ar. Vesp. 322, Rann. 923, Eccl. 273, but in prose-writers also, as Dem. c. Near. 1375. 13, Dio Cass. 38. 32, 40. 64, Thuk. 5. 65, 8. 67, all which passages, with many others, are cited in Lobeck's note. In place of La., Lb., and one or two others, with Aldus and the old to which, as Schäfer justly observes, rouds is opposed. interchange of these nouns in ancient manuscripts, see Jacobs, Anth. Gr. I. 2. p. 227; Porson to Eur. Or. 1559. There can be no question that Tovov, as a word of larger import, is both more poetical and better suited to the verb λωφάω (fr. λόφος), which signifes to liberate the necks of cattle from the yoke, and hence generally to relieve from labor. HESYCHIUS: λwφῆσαι· ἀπὸ τοῦ τραχήλου τὸ ἄχθος ἀποθέσθαι, παῦσαι, λῆξαι, ἀναπαῦσαι, ἡσυχάσαι. PHOTIUS: κυρίως δὲ εἴρηται λωφῆσαι τὸ τὸ βάρος ἀπὸ τοῦ τραχήλου ἀποθέσθαι· λόφος γὰρ ὁ τράχηλος. It is constructed with the genitive in Æsch. Prom. 376, 655, Plat. Phædr. p. 251. C, Legg. II. p. 934. B (see Kühn. Gr. 514, ed. Jelf); with ảó and the genitive in Thuk. 6, 12; and in all the passages in which it is found, with the single exception of Esch. Prom. 27, is used intransitively in the meaning above mentioned.

63. is doμovs noμílεraι, he brings with him to his own dwelling. So Hdt. 6. 118, Θηβαῖοι ἐκομίσαντο (τὸ ἄγαλμα) ἐπὶ Δήλιον ; Ar. Vesp. 833, ἐγὼ

δραμὼν αὐτὸς κομιοῦμαι δρύφακτον ἔνδοθεν ; Eur. I. Τ. 774, κόμισαί μ ̓ ἐς "Αργος. In the preceding verse the words δεσμοῖσι συνδήσας are to be referred simply to τοὺς ζῶντας βοῶν, he brings to his own dwelling the still surviving oxen, when he had further bound them together with thongs. 65. Καὶ νῦν κατ' οἴκους. With the apparent redundancy of this expression after ές δόμους in v. 63, Lobeck aptly compares Trach. 689, κατ' οἶκον ἐν δόμοις. Hermann writes συνδέτους with the MSS. Aug. Β. C. Yet the same συμπλοκή occurs in several compounds with δυς, as δυσξύνεEur. Phan. 1510; δυσξύνθετος, Plut. Mor. p. 975. F; and τῆς ξυμ μάχου, below, ν. 90.

τος,

66. Δείξω δὲ καὶ σοί. SCHOL. : πιθανὴ ἡ παρείσοδος τοῦ Αἴαντος· οὕτω γὰρ μεῖζον γίνεται τὸ πάθος τῆς τραγῳδίας, τῶν θεατῶν νῦν μὲν παραφρο νοῦντα, ὀλίγῳ δ ̓ ὕστερον ἔμφρονα θεωμένων· καὶ ἵνα ἰδὼν ὁ Ὀδυσσεὺς ἐξείπη τοῖς ἄλλοις Ἕλλησιν· οὕτω δὲ καὶ ἡ εὔνοια τῆς ̓Αθηνᾶς ἐνδείκνυται εἰς Οδυσσέα.

68. Θαρσῶν δὲ μίμνε μηδὲ . . . . ἄνδρ'. The sense is, μηδὲ συμφορὰν ἡγοῦ τὸν ἄνδρα γενήσεσθαι. So Ar. Eccl. 512, μὴ ξυμφορὰ γενήσεται τὸ πρᾶγμα. Lobeck and Schäfer consider μηδὲ συμφορὰν δίχου as inserted διὰ μέσου between μίμνε and its accusative, τὸν ἄνδρα. On the other hand, Erfurdt and Hermann deny that μíuvs can be so constructed. It is most probable, however, that the accusative belongs equally to both imperatives, as in Hom. Π. 13. 476, ὡς μένεν Ἰδομενεὺς δουρικλυτὸς, οὐδ ̓ ὑπεχώρει, Αἰνείαν ἐπιόντα βοηθόον, where ὑποχωρεῖν has the same construction as in Thuk. 2. 88; Euthyd. p. 133; Luc. Tox. 36. See note to v. 451 below.

69. ἀποστρόφους. By prolepsis for ὥστε ἀποστρόφους εἶναι, the sense being ὄμματα Αἴαντος ἀποστρέψω καὶ ἀπείρξω ἀπὸ τοῦ εἰσιδεῖν τὴν πρόσοψιν. Cf. Herm. ad Vig. p. 897; Seidl. ad Eur. Elektr. 442; Reisig, Comm. Cr. ad Ed. Kol. 1227; Stallb. Plat. Prot. 327. C; Valckn. Diatrib. 205 ; Kühn. Gr. 440. 2, ed. Jelf. So (Ed. Kol. 1200, τῶν σῶν ἀδέρκτων ὀμμάτων τητώμενος ; Virg. En. 1, age diversos. The reading πρόσοψιν, for which the simple ὄψιν is more common, is confirmed by Eur. Phan. 1353, εἰσορῶ πρόσοψιν ἀγγέλου.

Athene now addresses herself

71. Οὗτος. Kühn. Gr. 476, ed. Jelf. to Aias. The MSS. La. Lb. Harl. αἰχμαλώτιδας as a proparoxytone, and this adjective is frequently so written by the old copyists, as at Eur. Hek. 1096, and several other places. They were doubtless misled by the analogy of termination in such words as ἡπειρῶτις, στρατιῶτις, etc. The expression δεσμοῖς ἀπευθύνοντα does not occur elsewhere, and has been variously explained. SCHOL. ἀπευθύνοντα· τιμωρίαν ἀπαιτοῦντα· ἐπι·

βαλόντα μετ ̓ εὐθύνης. Billerbeck renders the participle by castigantem, and the whole phrase, captivos manibus laqueo ligatis verberantem. Passow translates xigas droμoïs år. “die Hände starr in Fesseln schlagen," and Wunder, regere (i. e. adstringere) manus vinculis, an operation which we learn from v. 62 to have been already over. Neue seems nearer the truth in interpreting retorquentem, comparing (Ed. R. 1154. The language of Athene, together with the employment of the participle present, appears to point to some occupation in which Aias was engaged at the moment of her address, and the action ascribed to him in v. 108, öglòv ävw xiovi dńcas, may perhaps suggest that the participle should here be rendered by exporrigentem, guiding the hands of the captives upwards, i. e. tying them to the pillar to which he is subsequently represented as having bound them. 73. Αἴαντα φωνῶ. "So frequently in Attic poetry, = καλῶ σε, ὦ Αίαν. So below, v. 789, καλῶ θ ̓ ἅμα πομπαῖον Ερμήν, κ. τ. λ., te invoco, Mercuri; v. 793 sq., naλã d'.... · σεμνὰς Ερινυς ; whilst at 801, ἔτ ̓, ὦ ταχεῖαι .'Egvúss. Hence it is not surprising to find the vocative and nomina. tive frequently associated in the same address, as at v. 819, ŵ xλrwai ̓Αθῆναι καὶ τὸ σύντροφον γένος, which is χαίρετε, ὦ κλειναὶ ̓Αθῆναι καὶ χαιρέτω τὸ σύντροφον γένος. Cf. Philokt. 530, 867, 986." WUNDER. 75. undè derλíav ages. The MSS. г. La. Lb. ägns. See Dawes, Misc. Cr. p. 221 ; Trach. 1183, οὐ θᾶσσον οἴσεις μηδ ̓ ἀπιστήσεις ἐμοί; Eur. Hipp. 498, οὐχὶ συγκλείσεις καὶ μὴ μεθήσεις ; Plat. Symp. p. 175. Α, οὔκουν καλεῖς αὐτὸν καὶ μὴ ἀφήσεις; The future indicative seems always preferred in such formulæ, to express an energetic command interrogatively; and Elmsley (Addend. ad Eur. Bacch. v. 344) is therefore incorrect in stating, that, if the penult of the aorist tenses of age were not always long, the subjunctive aorist would deserve the preference in our passage. "When the command is negative, où μn is used; and when a positive and negative command stand together, où is used with the former, and μ added to the latter." KÜHNER. Since, however, où μn is especially used in forbidding, it is better to consider, with Neue, that où runs through the sentence, and is common to both clauses; on which principle, unde, in the latter, becomes nai où μn. See Matth. Gr. Gr. § 498. c, § 517, Obs. 2. Lastly, diav alguy is concipere timiditatem. So v. 129, below, yxov algs, where see note; Ed. R. 914, buμòv age. In this signification algola is elsewhere found. See Ed. R. 635, 1225; Antig. 907. In the expression devòv žαíge μivos, v. 1010 below, the verb seems to be employed in its own stricter meaning.

76. Μὴ πρὸς θεῶν. SCHOL. : παραιτεῖται Οδυσσεύς, οὐχ ὡς κωμῳδοῦντος

τοῦ ποιητοῦ δειλίαν τοῦ ἥρωος, οὕτω γὰρ ἀφαιρεθείη τῆς τραγωδίας τὸ ἀξί ωμα· ἀλλὰ τὸ εὐλαβὲς ἐνδείκνυται· ἔμφρονος γὰρ ἦν τὸ τῷ μεμηνότι παρα χωρεῖν· ὅρα γὰρ, οἷον ἦν προσελθόντα ἐκεῖνον θεάσασθαι τὸν νεκρόν (Brunck, ἐχθρόν). Οι ἔξω κάλει αὐτόν may be supplied to μή, which must on no account be referred to the words immediately preceding, as Hermann teaches. On the construction of the participle μivwv, see Matth. Gr. Gr. 297. The full expression would be: àgzsírw roúrov iv xλioiy péveiv, satis esto, ut in tentorio maneat.

77. Τί μὴ γένηται ;

SCHOL. : μὴ τί γένηται φοβῇ; οὐχ ὁ αὐτὸς ἦν καὶ πρότερον ἄνθρωπος, ἀλλ ̓ οὐδὲν ὑμᾶς εἰργάσατο κακὸν, ἡμῶν ἐπιμελουμένων ; “So rí μù yśvnrai, Eur. Suppl. 544; ví μù wońow, Soph. Elektr. 1276; which last, although of different signification, agrees in form with the Latin quidni faciam? Latin writers at one time place the final conjunction before the interrogative pronoun: ut quo nos reciperemus? Liv. 44. 39. 5; ut qualiter sentiremus, Plin. H. N. 13. 13; at another time after it : quid uti facerent? Cic. p. Sext. 39. In Greek authors the latter is always postpositum : (Ed. Kol. 1725, ὡς τί ῥέξομεν; Elektr. 398, ὅπως τί δράσης ; Eur. Ion. 527, s dù rí Qiúyus ; In all these passages the conjunction s is indisputably final; but since the verb dependent upon it is omitted, its real power is so obscured that it would seem altogether redundant. It serves, however, to connect the interrogation which follows with what goes before. Compare ¿s rí dý, quid ita, Eur. I. A. 1342, ås ægòs rí,

d. R. 1174. d. Kol. 1182, ὡς πρὸς τί χρείας; ὡς τί χρήζων; ὡς τί dù biawy; and see Matth. ad Phan. 621, from whose reasoning I dissent. Similarly, rrí; Demosth., öri rí dý; Lucian. Enc. Dem. § 22, are referred to what has been said previously; just as when a person having denied that he is about to do this or that, another asks or rí (sc. ysyśνηται); ὡς τί (sc. γένηται); Andocides, in Or. III. p. 26. 26, appends this verb to the final particle, writing iva nuv ri yivnrai; although it is generally omitted: iva dù ri goσionxe, Ar. Nubb. 1192. See Herm. to Vig. p. 849; Matth. Gr. Gr. § 620; and this is also imitated by later Latin writers, ut quid autem coletur? Aug. Civ. IV. 18, whilst more ancient authors seem to have said ut quid? only. Indeed, the collocation vari by frequent usage coalesced so entirely into one word that the Etym. M. 471. 16 calls it an iripinua igwrnośws, and hence, whether the dependent verb is expressed or omitted, it is often written with but one accent, as in Ar. Eccl. 719, Plat. Apol. p. 26. c, Evang. Matth. ix. 4, in which passages the more recent editors have erased the hyphen. That this was, nevertheless, an exceedingly ancient mode of writing is testified

by Arcadius, p. 184, and by the Grammaticus Hermanni, p. 460, rò ivatí καὶ διατί ἐν τῇ συνηθείᾳ (συνεπεία) ὀξύνεται.” LOBECK.

πρόσθεν οὐκ ἀνὴρ ὁδ ̓ ἦν; "Nonne hic vir antea fuit tecumque versatus est sine ullo tuo periculo? cur ejus hominis præsentiam extimescis, quem toties antea vidisti imperterritus? The goddess marvels that Odysseus should now fear the appearance of Aias, whom he had so frequently seen before without any such emotion." LOBECK. 66 "avng hic non virum, sed hominem, i. e. mortalem, neque ultra mortale robur validum significat." HERMANN. The last critic objects to the explanation of Lobeck, num antea non fuit vir fortissimus? which is adopted by Wunder and Dindorf, that it can be admitted only upon the supposition that Aias had been previously the foe of Odysseus. Since this was not the case, and Odysseus could therefore have no reason to fear Aias, he holds that the explanation of the Scholiast must not be rejected. Both explanations are, however, frigid, and inappropriate to the rejoinder of Odysseus, Εχθρός γε τῷδε τἀνδρὶ καὶ τανῦν ἔτι, in which the words xai Tavuv T, referring clearly to goods , distinctly show that the whole verse is to be connected closely with the language just uttered by Athene. As, then, the thought nonne antea hic vir fuit is not absolute, it would seem that Odysseus in his reply interrupts the language of Athene, who was intending to pay a tribute to the prowess of Aias, and to remove the fears of Odysseus. If this view be correct, we ought to follow the example of those who place the sign of interrogation after T, and the sign of interruption after v.

78. τῷδε τανδρί. SCHOL. : δεικτικῶς ἀντὶ τοῦ ἐμοὶ, τῷ Οδυσσεῖ. Cf. Schäf. Meletem. Cr. p. 114; Seidl. Eur. I. T. 1402; Matth. Gr. Gr. 470. 9. See below, vv. 397, 421. Similarly, ride xugi. Cf. Porson to

Eur. Med. 389.

80. iv dóμois μévv. Such is the reading of Aldus and the majority of the manuscripts. In the Codd. Ien. Dresd. a. is dóuous, for which the MSS. La. Lb. exhibit is dóμous, the former, however, with iv os suprascriptum. SCHOL. ROM. is dóμovs · TOINTIxãs ȧvrì Toũ iv dóμois. Hermann has received this latter reading, believing that it could not have sprung from the copyists or interpreters, and defending it by Eur. I. T. 620, ἀλλ ̓ εἰς ἀνάγκην κείμεθ', ἣν φυλακτίον, where, however, the preposition is referred to πεπτώκαμεν, in place of which κείμεθα, or the result of falling, has been substituted.

81. Join μεμηνότα περιφανῶς, and comp. v. 66 above, περιφανῆ νόσον. The inquiry of Athene must not be understood as made with the intention of depreciating Odysseus in the estimation of the audience. It serves merely to