For the story of Polyxena see the Hecuba of Euripides.
 Troiae sub moenibus altis is used loosely in any case, as Polyxena's death happened after the sack of Troy. Euripides makes Polyxena's death take place in the Thracian Chersonese; Virg. followed a different story, placing the tomb of Achilles on the Sigean promontory.
 The captives were divided by lot, Eur. Tro. 240 foll., where however it is said of Andromache, v. 274, καὶ τήνδ᾽ Ἀχιλλέως ἔλαβε παῖς ἐξαίρετον. Andromache's feeling is like that of Creusa 2. 785 foll. For the indicative where we might have expected the subjunctive see on G. 2. 460.
 Diversa per aequora vectae 1. 376, a comparison of which will show that ‘patria incensa’ here may be the local abl., though the abl. abs. seems more natural. ‘Patria’ of the city Troy, as in 5. 624.
 Servitio enixae defines ‘tulimus:’ Andromache was the slave of her master's passion, and had offspring by him. The name of Andromache's son was Molossus, who is one of the dramatis personae in Eur. Andr. It is strange that Jahn should have attempted to fix another sense on ‘enixae,’ “semper enitentes ut servitio exiremus,” though he appears right in saying that this absolute use of ‘eniti’ for bringing forth is not common. It would be possible to explain ‘fastum’ and ‘iuvenem superbum’ of the insolence of Pyrrhus in leaving Andromache after making her his concubine; but ‘servitio’ is in favour of the other interpretation, which is also perhaps more in keeping with ancient feeling.
Serv. may be right in supposing
that the epithets contain a taunt: “cum
ingenti felle: ac si diceret, infelices maritis,
semper ut Paridi, ut Deiphobo.” For the
story comp. Eur. Andromache, parts of
which Virg. seems to have imitated here,
e. g. vv. 24 foll.—
“κἀγὼ δόμοις τοῖσδ᾽ ἄρσεν᾽ ἐντίκτω κόρον,
πλαθεῖσ᾽ Ἀχιλλέως παιδί, δεσπότῃ δ᾽
. . . . . . . .
ἐπεὶ δὲ τὴν Λάκαιναν Ἑρμιόνην γαμεῖ τοὐμὸν παρώσας δεσπότης δοῦλον λέχος.
 Wagn. appears rightly to explain ‘que’ as coupling two clauses which are co-ordinate in sense, though not grammatically, ‘transmisit me famulo’ (to be the wife of his captive) ‘famulamque’ (and to be a captive wife). Of the parallel instances which he quotes from Virg. the most apposite seem to be “Ipse gravis graviterque ad terram pondere vasto” 5. 447, “Extremus galeaque ima subsedit Acestes” ib. 498, “Obvius adversoque occurrit” 10. 734. See also on 2. 87, G. 2. 428. The old reading ‘Me famulam famuloque,’ supported by a few inferior MSS., looks like a correction.
 ‘Flammatus’ fragm. Vat., Pal., Gud. a m. p., ‘inflammatus’ Med. External evidence seems in favour of the simple word, which I have preferred with Ribbeck. On internal grounds there seems little to say: the dictionaries would lead us to suppose that the simple word was more common in poetry, the compound in prose. See on 4. 54. Henry remarks that Orestes is represented as impelled to kill Pyrrhus by two causes—he is not himself, and he has sustained a personal injury. With the sense so produced he compares Ausonius' epitaph on Pyrrhus (Epitaph. Heroum 9), “Inpius ante aras quem fraude peremit Orestes, Quid mirum? caesa iam genetrice furens.”
Excipit E. 3. 18. With the
language of this line Henry comp. 1. 349,
“ante aras . . . Clam ferro incautum superat,”
observing that ‘patrias ad aras’ explains
‘incautum,’ Pyrrhus being attacked
at home in his own penetralia. There is
however a doubt about the reference of
‘patrias ad aras’ here, which is generally
explained of an altar raised in honour of
Achilles at Delphi (a fact apparently resting
only on a doubtful statement of Serv.),
Delphi being the place where Pyrrhus was
said to have met his death (Dict. B.
‘Neoptolemus’). Virg.'s brevity will
hardly allow us to decide definitely. In
any case, as the language shows (comp. 2.
663), we are meant to think of Pyrrhus'
death at the altar as a retribution for his
murder of Priam. Heyne refers to Tryphiodorus
“σχέτλιος, ἦ μὲν ἔμελλε καὶ αὐτῷ πότμος
ἔσσεσθαι παρὰ βωμὸν ἀλαθέος Ἀπόλλωνος
ὕστερον, ὁππότε μιν ζαθέου δηλήμονα νηοῦ
Δελφὸς ἀνὴρ ἐλάσας ἱερῇ κατέπεφνε μαχαίρῃ.
 Different stories were told of the connexion of Pyrrhus with Epirus: see Dict. B. ‘Neoptolemus.’ The names Neoptolemus and Pyrrhus were both given to kings of Epirus in historical times. The sense of ‘reddita’ is disputed. Forb. explains it, not improbably, with reference to Helenus' position at Troy, to which he was now in some sort restored. But it may be suggested that the word may have the sense of giving in succession, giving to one person after another, though I know of no other instance which would support it. Mr. Long suggests that ‘re’ here simply expresses a result from an antecedent. ‘Cessit’ above v. 297.
 The name ‘Chaones’ was one of greater antiquity than that which is here ascribed to it, the Chaonians being connected by tradition with the Pelasgians (Dict. Geogr. ‘Epeirus’). Various stories of this Chaon are mentioned by Serv., the general result being that he was Helenus' brother or comrade, who was either killed by him accidentally or died for him voluntarily.
 There is an unmetrical reading ‘Pergamiamque’ in Med. (supported by a correction in Pal.), which Serv. mentions as the usual one in his time, though he condemns it. With ‘iugis addidit arcem’ comp. 6. 774.
 Andromache means, as Heyne remarks, to ask Aeneas how he has come to Epirus—by stress of weather, or by destiny, or divine intervention, ‘qui’ having virtually the force of ‘quomodo’ (E. 1. 54 note). The alternatives are scarcely meant to exclude each other, being rather different ways of stating the same thing. With ‘quae fata dedere’ comp. 4. 225, “fatisque datas non respicit urbes.”
 Ignarum is explained by ‘deus,’ a divine intervention having brought Aeneas to a country which he did not know to be a friendly one. ‘Deus adpulit oris’ is repeated below v. 715. Before Heins. ‘quis te’ was read for ‘quisnam.’
 Quid puer Ascanius? see on G. 3. 258. ‘Superat’ as in E. 9. 27. ‘Vescitur aura’ 1. 546. Here some MSS. give ‘auras,’ which is supported by a reading in fragm. Vat., and approved by Jahn; but the construction is not found elsewhere in Virg.
 A solitary instance in Virg. of a hemistich where the sense is incomplete. The copyists of the inferior MSS. have attempted to supply the deficiency in different ways—“peperit fumante Creusa” “obsessa est enixa Creusa,” “natum fumante reliqui.” Later critics, as Heyne Gossrau, and Ladewing, have fancied that the passage has been interpolated. Wagn. and Forb. complain that, as the text stands, Andromache makes no mention of Creusa, whom she could not know to be lost, and accordingly adopt, as does Ribbeck, ‘quae’ for ‘quem’ from the ‘Menagianus alter,’ separating ‘et vescitur aura’ from ‘superatne.’ (Ribbeck cites Med. for ‘quae:’ but a friend who consulted the MS. for me assures me that it distinctly reads ‘quem.’) They account for ‘amissae’ v. 341 by supposing that Aeneas gives some sign which shows that his wife is no more— an expedient which would scarcely be natural in an ancient drama, but is ridiculous in an epic. (Ribbeck supposes a lacuna.) The words of the next line clearly show that Andromache—how, we know not, but may imagine for ourselves— was aware of Creusa's fate. They are not such as would occur to her on the moment of hearing a piece of news like this: they are precisely what might be spoken under other circumstances by a mother possessed with the image of her own lost boy, and wondering whether the separation had really entailed a breach in their love of each other. On the whole, there seems no good reason to doubt that we have the passage as Virg. left it. If we cannot complete the hemistich satisfactorily, we may console ourselves with thinking that be could not either.
 Repeated 12. 440. ‘Avunculus,’ because Creusa, according to one account, was Priam's daughter. Serv. mentions a criticism of his day, “quidam ‘avunculus’ humiliter in heroico carmine dictum accipiunt.”
[344-355] ‘As she was speaking, Helenus appears. He welcomes me to his city, built after the model of old Troy, and entertains my companions.’
 Pal. and others give ‘Helenus multis’ for ‘multis Helenus:’ but Wagn. rightly observes that the present line does not stand on the same footing as v. 295 above, ‘Priamides’ being here joined with ‘heros.’
 Pal. and others have ‘lacrimans,’ a reading as old as Serv., which, as Pierius says, might be explained “interfundit verba singula multum lacrimans:” but the received reading is clearly right, though the adverbial accusative ‘multum’ is unusual where another accusative is expressed. Serv. observes judiciously, “Bene verba Heleno post Andromacham non dedit, ne frigeret.” This reticence indeed is one of Virg.'s most noble characteristics, though it must be admitted that, in his anxiety not to weary the reader, he sometimes fails to inform him sufficiently.
 Simulata magnis, as Cic. Att. 9. 8 talks of “Minervam simulatam Mentori.” Donatus remarks, “In omnibus istis versibus, ne minuat magnitudinem Troiae patriae, ostendit hanc non tantam esse, sed similem: nam dixit ‘parvam Troiam,’ et ‘simulata Pergama, arentemque rivum Xanthi.’”
 Germanus attempts to show that the features of this river are intended not to be contrasted but to be paralleled with those of the real Xanthus, which was itself comparatively a small stream, as Horace Epod. 13. 13 talks of “frigida parvi Scamandri flumina.” But this is obviously contrary to Virg.'s meaning, which is evidently to contrast Hom.'s ποταμὸς δινήεις with its miniature, which “derives its course from thrifty urns and an unfruitful source.”
 Aeneas embraces the gate in token of recognition, as the women in 2. 490 embrace the doors in token of farewell.
 Accipere of entertaining guests, as in Ter. Eun. 5. 8. 52, “Accipit hominem nemo melius.” The “porticus” seem to have surrounded the “aula,” which appears to be used by Virg. in the case of a palace as equivalent to “atrium.” See Lersch, § 72. “Atria” and “porticus” are connected 2. 528., 12. 474 foll. The banquet probably extended to both, as all Aeneas' companions appear to have been entertained. Symmons finds a difficulty here: but the circumstance doubtless did not count for much with Virg., who is apt to exaggerate in such matters (comp. 1. 634, 635 &c.), and merely wished to convey a notion of Helenus' hospitality.
 ‘Aulai’ is one of the archaisms which Virg. admits into the Aeneid: see on 1. 254. ‘In medio,’ the reading of some MSS., is supported by citations in several grammarians: but the preposition is omitted by Med., Pal., and others.
[356-373] ‘Wishing to sail, I consult Helenus about my voyage, telling him that every divine intimation, save that of Celaeno, has been in favour of the journey to Italy, and asking him what I have to be on my guard against.’