Gud. has a variant ‘inmensa,’ doubtless from 6. 823, and others, including one of Ribbeck's cursives, give ‘infanda.’ Virg. probably thought of Il. 5. 676, τῷ ῥα κατὰ πληθὺν Λυκίων τράπε θυμὸν Ἀθήνη.
 For ‘adversos’ we might have expected ‘aversos:’ but the MSS. seem to have no variety. ‘Adversos’ is doubtless to be taken loosely, those on the opposite side, whether fighting or flying. ‘In adversos’ occurs again 10. 412., 11. 389.
 Excipit of catching in pursuit 11. 684, probably with a notion of doing it unexpectedly to the victim. This seems to suit the context better than Heyne's “excipit in se irruentem.” ‘Hinc,’ from the bodies of Phalaris and Gyges: comp. Od. 22. 271, τοὶ δ᾽ ἄρ᾽ ἐπήϊξαν, νεκύων δ᾽ ἐξ ἔγχἐ ἕλοντο. So Il. 13. 260 foll., δούρατα . . . τὰ κταμένων ἀποαίνυμαι. See 10. 342. “Raptas fugientibus ingerit hastas” 12. 330.
 ‘In tergum’ a sort of epexegesis of ‘fugientibus.’ Rom. and originally Pal. have ‘tergus,’ which Heins. restored and Heyne and Ribbeck retain, and so Charisius read p. 54. “Animum virisque addidit” above, v. 717. “Faces animumque ministrat” 5. 640.
 Comitem to those already killed. ‘Confixa Phegea parma’ like “succiso poplite Gygen” v. 762. Pal. corrected, Med. second reading, and Gud. originally have ‘confixum.’ Phegeus was probably flying with his shield slung behind him, 11. 619.
 They were standing on the rampart and attacking those without, not knowing that Turnus was within.
 Verbally from Il. 5. 678, where the persons are Lycians slain by Ulysses. Ov. also reproduces the line M. 13. 258, where Ulysses boasts of his exploits.
 Ab aggere with ‘dexter.’ Turnus is standing with his right hand to the rampart, and he takes a sweep with his sword in that direction. “Sublata dextra connixa” 5. 642. Med. originally and one of Ribbeck's cursives have ‘dextra’ here.
 Ribbeck reads ‘desectum’ from the original text of Gud., which would be neater, especially as ‘iacuit’ follows: but the authority is hardly sufficient. In Rom. the second, third, and fourth letters of the word seem to be obliterated. ‘Deiectum’ too is confirmed by Il. 20. 482, ὁ δὲ φασγάνῳ αὐχένα θείνας Τῆλ᾽ αὐτῇ πήληκι κάρη βάλε: comp. ib. 14. 497. In Sil. 13. 246, which is an imitation of the present passage, Heins. conj. “desectum.”
 Ungere or ‘unguere’ is the reading of all Ribbeck's MSS., though in Pal. the first letter is in an erasure. The inferior MSS. present considerable variety, whence Bentley wished to read “tingere.” For the anointing of arrows with poison comp. Od. 1. 261 foll., where it is mentioned as a thing of doubtful morality. It does not appear in the Iliad. ‘Manu’ is pleonastic. “Calamos armare veneno” 10. 140.
 Cordi above v. 615. ‘Numeros intendere nervis’ is one of Virg.'s usual efforts after variety. ‘Intendere nervos’ would be the common expression for stringing a lyre: so he chooses to represent the notes as strung on the chords.
 Equos probably of horses as used in war. Race-horses are among the stock subjects of lyric poetry (Hor. 4 Od. 2. 18, A. P. 84); but we must not gratuitously charge Virg. with an anachronism. ‘Arma virum’ 1. 119. The juxtaposition of the words is doubtless meant to remind us of Virg.'s own poem.
[778-818] ‘Mnestheus and Serestus rally the Trojans, who press upon Turnus. At last he leaps into the Tiber, swims to shore, and rejoins his army.’