[653] Coniungo is generally constr. with dat., or abl. and “cum:” but Cicero has ‘coniunctus’ with abl. alone, 1 De Or. 5. 17, “brevitas respondendi et lacessendi, subtili Venustate atque urbanitate coniuncta:” ib. 1. 57. 243, “dicendi vis egregia, summa festivitate et Venustate coniuncta,” and in other places. ‘Crepido’ a rising bank or elevation, as in Livy 27. 18 (Forc.), “eam quoque planitiem altera crepido haud facilior in adscensum (‘faciliori adscensu’ Madv.) ambibat:” comp. Stat. 1 Silv. 3. 43. ‘Crepido saxi’ probably = a projecting rock. “Fluctus ab undisoni ceu forte crepidine saxi” Val. Fl. 4. 44. Comp. κρηπὶς βωμῶν ῀ βωμός Soph. Trach. 993, κρηπῖδ᾽ Οὐλύμπου ῀ Ὄλυμπον Eur. Troades 215.
[654] Ponte v. 288 note. The ladder seems to be distinct from the bridge: comp. v. 657 “Exsuperatque moras et pontis transilit altos,” where the same is implied by the use of separate verbs.
[655] ‘Osirius’ Rom. ‘Qua advectus’ (= “advectus est”) like “quae passus” v. 162 above. It is hardly necessary (with Serv.) to identify Osinius with Massicus, leader of the Clusians, v. 166. It is quite in Virg.'s manner to mention out of its proper place the name of a subordinate chief which he had previously omitted. Comp. 9. 325 foll., where he mentions for the first time Rhamnes, a chief in the alliance of Turnus, whose name might certainly have been expected in the catalogue.
[656, 657] Med. has ‘huic’ for ‘huc,’ and ‘Aeneae trepida’ for ‘trepida Aeneae.’ With ‘huc . . . in latebras’ comp. G. 4. 423, “Hic iuvenem in latebris . . . Collocat.”
[658] Moras the ladders in his way: see on v. 654.
[659] Attigerit Pal. (apparently corrected from ‘attigerat’), and so originally Gud. ‘Rupit’ Rom. for ‘rumpit.’ The sequence ‘attigerat’ . . . ‘rumpit’ is the same as in 2. 705, “Dixerat ille; et iam per moenia clarior ignis Auditur.” Comp. 1. 586.
[660] Revoluta per aequora, on the retiring wave: so rightly Heyne and Wagn. in his small edition: though in his larger one he says “Aequora revoluta acceperim de itinere per aequora revoluto.”
[661] Brunck placed this and the following verse after v. 664, and Wagn. and the generality of subsequent editors have followed him. Two Paris MSS. are said to confirm the conjecture (see however on v. 705 below). In some respects the change would be an improvement, as v. 665 is more naturally contrasted with 661, 662 than with vv. 663, 664: but it might be urged on the other hand that ‘illum’ followed by ‘Turnum’ would be rather awkward, and the present order is at any rate unobjectionable. Heyne may be right in supposing that this and the following line were written by Virg. in the margin: but the supposition is not necessary. The early critics however appear to have felt a difficulty in the passage, as there was an old reading ‘Ille autem Aenean’ (still found in Rom., Pal. originally, and Gud. partially) which Urbanus ap. Serv. explained of Turnus venting his fury on sailors supposed to be in the ship. Serv. saw that, connected with the next line, this reading would give no sense. “In proelia poscere Turnum” 8. 614. A colon should be put after ‘poscit:’ not a full stop, with Wagn. and Forb.
[662] “Demittere morti” 5. 691, “neci” 2. 85. On the question whether “mors” is personified, see on G. 3. 480.
[664] “Nocti se inmiscuit atrae” 4. 570.
[665] Interea medio Rom. and one of Ribbeck's cursives. The abl. of extent ‘medio aequore’ is common in Virg.
[666] Ignarus rerum, ignorant of the truth: “haud nescia rerum” in a somewhat different sense 12. 227. ‘Ingratus salutis’ is a poetical constr. Cicero does not apparently even use ‘gratia’ or ‘gratus’ with the gen.; Livy has “ob eius gratiam meriti” 23. 15, and “haud minorem eius rei . . . gratiam futuram” 26. 50: a constr. of which ‘ingratus’ with gen. is an extension. Virg. may have been led to it by wishing to balance ‘ignarus rerum:’ he has elsewhere similar licences from a similar cause: e. g. “fortunatusque laborum Egregiusque animi,” &c., 11. 415, “nec veterum memini laetorve malorum,” ib. 280.
[667] “Tendoque supinas Ad caelum cum voce manus” 3. 176: see note on 2. 378. ‘Duplicis’ note on 1. 93. ‘Tollit’ one of Ribbeck's cursives and some inferior copies.
[668] Tanto for ‘tanton'’ Pal. and Gud. So in 3. 319 (note) ‘Pyrrhi’ is read by some copies for ‘Pyrrhin'.’ ‘Tanton'’ here is confirmed by Serv., Diomedes 430, and Priscian 1032, 1224. With the thought comp. Il. 11. 404, Ὦ μοι ἐγώ, τί πάθω; μέγα μὲν κακὸν αἴ κε φέβωμαι: and Hor. 3 Od. 27. 37 foll.
[670] “Unde quo veni?” Hor. 1. c. ‘Quove reducit’ Med. with ‘quemve’ added in the margin. With ‘quem’ (=“qualem”） Serv. well comp. “Hunc ego te ‘Euryale’ adspicio” 9. 481, where “hunc”=“talem.” Comp. for the thought Catull. 62 (64). 177 “Nam quo me referam, quali spe perdita nitar?” ‘Reducet’ one of Ribbeck's cursives, and so Heyne. ‘Reducit’ seems best taken with Wagn. “reducit ex acie,” which will account for the present, the retreat being already in progress. If it were to be understood with Peerlkamp of return to the field after flight, we should rather have had the future.
[671] Videbo can I bear to behold? Comp. the questions in 4. 534 foll., and “nec generum Aenean captiva videbo” 12. 63.
[672] Quid as we should say “What is to be said of?” so τί in Greek: e. g. Plato Gorg. p. 502 A, τί δὲ ὁ πατὴρ αὐτοῦ Μέλης; ἦ πρὸς τὸ βέλτιστον βλέπων ἐδόκει σοι κιθαρωδεῖν; (Jelf § 872. 2. g.) See on 7. 365, G. 3. 258. With ‘me meaque arma secuti’ comp. 3. 156, “nos te, Dardania incensa, tuaque arma secuti.”
[673] Quosque Med. a m. p. and Rom.: ‘Quove’ Med. a m. s., Pal., Gud., and another of Ribbeck's cursives. “‘Quosve’ . . . Asper ‘quosne’ legit, et adnotavit ‘ne’ pro ‘ve’” Serv. ‘Quosne,’ which is found in Leid. and one of Ribbeck's cursives, was rightly adopted by Heins. and the edd. after him: comp. Catull. 62 (64). 180, “an patris auxilium sperem, quemne ipsa reliqui?” a passage obviously in Virg.'s mind. See also Hor. 1. S. 10. 21 (Wagn.), and Hand, Tursellinus 4, p. 77. So “quiane” 4. 538 . ‘Relinqui’ Pal. corrected for ‘reliqui.’
[674] Pallentes Med. a m. p. For the transition from relative to demonstrative, involving a change of case, comp. Soph. Aj. 457, “Καὶ νῦν τί χρὴ δρᾶν, ὅστις ἐμφανῶς θεοῖς Ἐχθαίρομαι, μισεῖ δέ μ᾽ Ἑλλήνων στρατός;” Apoll. R. 3. 836, “Λήθῃ ἀχέων τά οἱ ἐν ποσὶν ἦε Θεσπέσἰ, ἄλλα τ᾽ ἔμελλεν ἀεξήσεσθαι ὀπίσσω”. ‘Cadentem’ Gud. originally.
[675] One inferior copy rather ingeniously gives ‘quid ego.’ ‘Et quae’ Med. first reading for ‘aut quae:’ Heins. conj. ‘ecquae.’ ‘Dehiscet’ Rom., and so Pal. originally in 12. 883, where nearly the same words are put into the mouth of Juturna. ‘Ima’ goes closely with ‘dehiscat’ (see on 4. 24), ‘can open its foundations deep enough.’
[676] ‘O’ is omitted in Rom. and erased in Pal. “Vos O mihi Manes Este boni” 12. 646. Comp. Il. 6. 346 foll. (ὥς μ᾽ ὄφελε . . .) Οἴχεσθαι προφέρουσα κακὴ ἀνέμοιο θύελλα Εἰς ὄρος ἢ εἰς κῦμα πολυφλοίσβοιο θαλάσσης (Cerda).
[677] ‘Rupes’ are cliffs, ‘saxa’ smaller rocks and stones. With ‘volens adoro’ comp. “Iunoni cane vota libens” 3. 438 note.
[678] Syrtes Pal. corrected and Gud. ‘Syrtis’ sing. 4. 41 (Heyne). ‘Saxa’ in connexion with ‘vada,’ as in 3. 706 “vada dura lego saxis Lilybeia caecis.”
[679] Conscia that knows of my shame.
[680] Hunc Med. for ‘huc,’ as also originally ‘illum’ for ‘illuc.’ Wagn. thinks this accounts for a reading ‘animum’ found in some MSS., and partially supported by one of Ribbeck's cursives. “Atque animum nunc huc celerem, nunc dividit illuc” 4. 285.
[681] ‘Mucroni’ Pal. originally, Rom.: ‘Mucrone’ Med., Pal. corrected, Gud., and two other of Ribbeck's cursives: and so Serv. and Priscian (1199, 1205). ‘Sese mucrone induat’ = throw himself upon his sword. Virg. has used the abl. where a prose writer would have used the dat. or “in” with acc. In Caesar B. G. 7. 73, “sese ipsi acutissimis vallis induebant” (Pierius) the subst. is probably in the dat.: so ib. 83, Ov. 2 Am. 10. 31 (Forb.).
[682] ‘Inbuat’ Pal. ‘Durum’ Rom. for ‘crudum.’ “Crudum Transadigit costas et cratis pectoris ensem” 12. 507. ‘Crudum’ = hard, cruel: comp. ὠμόφρων σίδαρος Aesch. Theb. 728. So “cruda bella” Ov. Am. 3. 8. 58; “cruda tyrannide” Juv. 8. 223. ‘Exigere,’ to drive right through, v. 815 below.
[683] Iactet Rom. for ‘iaciat.’ With ‘iaciat fluctibus’ (= “in fluctus”) comp. “ponto iaciunt” 9. 712.
[684] “Reddite me Danais” 2. 669.
[685] Cerda well comp. Il. 8. 169 (of Diomed) Τρὶς μὲν μερμήριξε κατὰ φρένα καὶ κατὰ θυμόν, Τρὶς δ᾽ ἄρ᾽ ἀπ᾽ Ἰδαίων ὀρέων κτύπε μητίετα Ζεύς. Comp. also Il. 1. 188 foll., where Athene is sent by Hera to restrain Achilles from slaying Agamemnon. ‘Conatus viam’ seems to be on the analogy of “conari opus.”
[686] Some inferior copies have ‘animo,’ which is adopted by Wagn. on account of the apparent difficulty of the gen. ‘Animi’ all Ribbeck's MSS., and so Serv. and Arusianus p. 246 L. Heyne takes ‘animi miserata’ as = ‘pitying him for his spirit;’ Serv. more naturally as = ‘pitying him in her mind’ (θυμῷ φιλέουσά τε κηδομένη τε Il. 1. 196). See on 6. 332, “sortemque animi miseratus iniquam,” where ‘animi’ has been restored from Med. for ‘animo.’ “Fidens animi” 2. 61, “furens animi” 5. 202, “victus animi” G. 4. 491.
[687] Aestu secundo with the tide (or current?) to back him, Ov. M. 13. 629, 727 (Forb.).
[689-754] ‘The ἀριστεία of Mezentius, who kills among others Hebrus, Evanthes, Acron, and Orodes.’
