Serv. mentions three readings, ‘Diomeden,’ ‘Diomede,’ and ‘Diomedem.’ The first, which he rightly rejects as unmetrical, is found in Med., Pal., Rom., and Gud. The second, which he prefers, is supported more or less by two of Ribbeck's cursives, and is the original reading of Canon. Serv. however appears to be speaking simply on critical grounds, not on grounds of authority; and it is an obvious answer that such a representative of the Greek acc. is quite unexampled, at least in Virg. Macrob., Sat. 5. 17, defends ‘Diomede,’ also apparently on critical grounds, supposing the choice to lie between it and ‘Diomeden.’ ‘Diomedem’ is found in a few MSS., and is really supported by ‘Diomeden,’ the forms being constantly confused in MSS. It is approved by Lachm. on Lucr. 1.739. ‘Castra’ is applied to Diomede's settlement, to remind us of the encampment at Troy.
 ‘Contigimus’ apparently in entreaty, as no alliance was formed. Virg. probably thought of Aesch. Ag. 907, τὸν σὸν πόδ᾽, ὦναξ, Ἰλίου πορθήτορα, and perhaps also of Il. 24. 478, κύσε χεῖρας Δεινάς, ἀνδροφόνους, αἵ οἱ πολέας κτάνον υἷας. ‘Ilia tellus’ 9. 285, here put for “Ilium” to give the notion of utter destruction, like αὐτόχθον᾽ ὃν πατρῷον ἔθρισεν δόμον in Aesch. Ag. 536.
 Victor: he joined with Daunus against the Messapians, and received in recompense a part of their territory; a statement for which Heyne refers to Antoninus Liberalis, Metamorph. 37. ‘Iapygis,’ used improperly for “Apuli,” as Mount Garganus was not in the Iapygian part of Apulia. ‘Arvis,’ which Heins. restored for ‘agris,’ is found only in one of Ribbeck's cursives in an erasure.248]. Repeated from 1. 520.
 Virg. wavers, as we have seen, between two views of the past of Italy, a legendary and a semi-historical one: here he adopts the former, as if the Italian nations still lived in the halo of the golden age and knew nothing of war. ‘Antiqui Ausonii’ is part of the same feeling. “Saturnia regna” E. 4. 6.
 Quietos: comp. 7. 46, 623, 693, and contrast 8. 55.
 Cerda comp. Od. 3. 86, ἄλλους μὲν γὰρ πάντας, ὅσοι Τρωσὶν πολέμιζον, Πευθόμεθ᾽, ᾗχι ἕκαστος ἀπώλετο λυγρῷ ὀλέθρῳ. See also ib. 102 foll. “‘Violavimus’ quasi sacros,” Serv.: comp. vv. 277, 592., 2. 189 &c. So Hom.'s Ἴλιος ἱρή. Rom. has ‘populavimus.’
 Comp. 1. 100. ‘Premit,’ the reading before Heins., is found in none of Ribbeck's MSS.
 Vel Priamo miseranda manus is the counterpart of Aeneas' language 2. 6 foll. Serv. quotes Pacuvius inc. fr. 28, “Si Priamus adesset, et ipse eius commiseresceret.” There is another reading ‘domus,’ perhaps supported by Pal., which has ‘damus’ in an erasure. ‘Scit’ i.q. “testis est:” comp. G. 3. 474. ‘Triste Minervae sidus’ of the storm sent by Pallas on the return of the Greeks, Od. 3. 132 foll. ‘Sidus’ because of the connexion of storms with the appearance of certain stars, a reminiscence of the Georgics. See on 12. 451, “abrupto sidere.”
 The story was that Nauplius hung out false lights on Caphereus, the promontory of Euboea, in vengeance for the death of Palamedes. Sophocles wrote a tragedy on the subject. ‘Ultor’ refers not merely to the vengeance of Nauplius, but to the calamity regarded as a punishment for the destruction of Troy.
 ‘Protei columnas’ on the analogy of “Herculis columnae,” for the extremity of Egypt. For Proteus see Od. 4. 351 foll., and for the rationalized form of the story, which made him the king of Egypt, Hdt. 2. 112 foll., and the Helena of Euripides. ‘Adusque’ Hor. 1 S. 1. 97., 5. 96. Comp. “abusque” 7. 289.
 Ribbeck, following a hint of Wagn.'s, puts this and the following line after v. 268, that ‘invidisse’ may be governed by ‘referam.’ For the construction of ‘invidisse’ see below. That Virg. intended the present order is shown not only by the consent of the MSS., but by the turn of the language. He dismisses Neoptolemus, Idomeneus, and the Locrians briefly: ‘Why talk of them? Agamemnon himself’ &c. ‘Regna,’ the fortunes of his kingdom, which, as we have seen 3. 333, was divided after his death. ‘Versos’ may be either i. q. “eversos,” as he was driven from his home (3. 121), or “mutatos,” as “vertere solum” is used of exile.
 The Opuntian Locrians followed Ajax son of Oileus (2. 527 foll.); and postHomeric legend seems to have fixed them, or some of them, on the coast of Africa after their leader's death. Serv. says that part of Ajax's followers settled in Italy in the Bruttian territory (see 3. 399), part in Africa, in the Pentapolis or some other place: but his notices are confused and perhaps interpolated, and he identifies the two parties respectively with the Locri Epizephyrii and Locri Ozolae. Perhaps the fact that the death of Ajax is related by Proteus in Od. 4. 499 foll. may have influenced the legend about the locality.
 Prima intra limina is either a vivid way of saying that he was slain immediately upon his return, or refers to some variety of the story different from those given respectively by Hom. and Aesch. Ribbeck reads ‘inter’ for ‘intra,’ from Pal., Gud. originally, and another cursive.
 Oppetiit: 1. 96 note. ‘Subsidere’ is used with acc. of lying in wait for a person or thing Lucan 5. 227 (quoted by Serv.), Sil. 13. 221, being apparently a technical term in hunting (comp. Forc. s.v. ‘Subsessor’). This seems also to be the sense of the word in Manil. 5. 303 (speaking of Philoctetes), “Maior et armatis hostis subsederat exsul,” where “hostis” (which Wakef. wished to alter into “hosti”） may be acc. pl. And this might very well be its sense here, ‘devictam Asiam’ being understood as a bold expression for “victorem Asiae,” something like “Alpes apertas” 10. 13, which however is not an exact parallel. But it seems more probable that Virg. refers to the Greek ἔφεδροι, and means that Aegisthus took up the combat with Agamemnon when Asia was worsted. Mr. Price, formerly of St. John's Coll. Camb., acutely objects that Aegisthus would rather be called ἔφεδρος to Agamemnon than to Asia, as Philoctetes in the passage just cited is to the enemy: but if we suppose Virg. to have coined the expression, he might exercise a certain liberty in the matter, and the mention of Asia was quite necessary if the image was to be used at all: he might recollect too that ἔφεδρος is used with a gen. in the sense of successor to a person. Mr. Price's own interpretation is that Aegisthus lay at the bottom of Asia, like lees in a cask, as a further difficulty (he comp. 5. 498, Lucr. 5.497); and this view, with a little modification, would be quite reconcilable with that advocated above, and with the passage from Manilius. There is no other instance of ‘subsidere’ with an acc. in this sense: but it is only a variety of the other, both meaning to watch for; and the use of “subsidia” for reserves (comp. Varro L. L. 5. 90) would naturally lead Virg. to the employment of the verb. Cerda says that ἔφεδρος is actually used of an adulterer: but he gives no instances, nor do Lidd. and Scott notice the use. Val. Max. twice uses the words “subsessor alieni matrimonii” (2. 6. 5., 7. 2, ext. 1), apparently referring to treacherous lying in wait, and Arnob. 4. 23., 5, 20 employs “subsessor” as a synonyme for “adulter.” The difficulty of the expression has led to two remarkable varieties of reading, ‘devicta Asia,’ mentioned with preference by Serv. and found in three of Ribbeck's cursives (in two of them, including Gud., from a correction), and ‘possedit,’ Rom., Med. second reading, and a variant in Gud., perhaps from a gloss of Serv., who gives as possible explanations of ‘subsedit’ “post possedit” and “dolo possedit.” Forb. thinks ‘devictam Asiam’ is the captured wealth of Asia, which would be less forcible than either of the interpretations given above. It is in keeping with Virg.'s love of variety that he should mention Clytaemnestra alone in the preceding line, Aegisthus in this.
 Invidisse deos can only be constructed as an exclamation, unless we consider with Wagn. and others that vv. 266—268 are out of their places, having been either added as an after-thought by Virg. or deranged by his transcribers. To suppose that Virg. could have intended to interrupt the construction after v. 265 and return to it again here is quite out of the question. There is no inappropriateness, as Wagn. thinks, in the exclamation, which is a sort of reflection on the divine φθόνος, ‘surely the gods need not have grudged me this.’ See Madv. § 399. The feeling is the same in 4. 550., 5. 82, though the expression is different. ‘Aris’ is read by all Ribbeck's MSS. but two cursives, one of which has the word altered into ‘agris’ by a late correction, while the other (the Codex Minoraugiensis) has ‘oris,’ found also in the ‘Oblongus Pierii.’ In spite of its external authority, ‘aris’ seems exceedingly improbable, as there seems no reason why the altars should be mentioned, and the word may have been introduced from a recollection of 3. 332. But it is not easy to say what word should take its place. ‘Agris’ would seem most probable in itself, as ‘oris’ looks too much like an introduction from v. 281 below: but perhaps it is best to let external authority decide in favour of the latter, which was the old reading before Heins. ‘Argis,’ the reading of some of Pierius' copies, would have great plausibility (comp. 2. 95): but though it would be natural in the mouth of Diomede under other circumstances (see on v. 246), it does not agree with ‘Calydona.’ No MS. appears to give ‘arvis,’ which is another possible variety. In the parallel passage Ov. M. 14. 476 the MSS. are divided between ‘agris,’ ‘Argis,’ and ‘arvis.’ In Catull. 62 (64). 132 the best MSS. give ‘patriis—aris,’ so that possibly it and the present passage may confirm each other. For ‘ut’ two MSS. have ‘ne,’ as in v. 43 above: there however the previous accusative makes some difference, though probably either construction would be possible here, just as in Greek the same verb will sometimes take either the simple inf. or the inf. with μή, the one being really acc. of object, the other cogn. acc. So here ‘ut viderem’ is equivalent to an acc. or an inf. “Redditus his terris” 6. 18.
 Coniugium for “coniugem” 2. 579. Calydon was the seat of Diomede's family: but his father Tydeus migrated to Argos. He was variously connected with Calydon according to different versions of his story: Virg. seems to have followed that which made him go to Aetolia to assist his grandfather Oeneus after the expedition of the Epigoni. Virg. may have meant to translate Καλυδῶνος ἐραννῆς Il. 9. 531, 577.
 Horribili visu i. q. “horribile” or “horribilia visu,” probably constructed with ‘sequuntur,’ though it might go with ‘portenta.’ ‘Nunc etiam sequuntur,’ follow me even into my exile, the birds in question belonging to the islands called “insulae Diomedeae,” off the coast of Apulia. Heyne, in an Excursus, treats of these birds, which are called “artenas” by moderns, and have been identified with puffins. Serv. says that Virg. has departed from the true account, which represented them as transformed in consequence of their grief for the death of Diomede. They are mentioned by Lycophron v. 597, and described by Ov. M. 14. 457 foll., Pliny 10. 44. Ov. speaks of their form as “ut non cycnorum, sic albis proxima cycnis.” They were said to be friendly to Greek sailors, unfriendly to Italians.
 Et epexegetical. ‘Amissi’ probably means not simply lost, but separated from me: see on 2. 148. Ov. l. c. makes them transformed for a fresh insult offered to Venus during their wanderings. Serv. mentions a variant ‘admissis,’ which Pierius says he found in some old MSS. It was adopted by Heins., who interpreted it “concitatis,” like “admissis equis:” but it would seem more natural to understand it as “indutis” or “sumptis.” Pal. originally had ‘amissis,’ and another has ‘admissi.’ “Altum nidis petiere relictis” G. 2. 210.
 Caelestia corpora may refer to Mars as well as Venus (comp. Il. 5. 882): but as we do not hear that the former outrage was visited on Diomede, it is more probable that the plural is to be understood generally. Comp. Il. 5. 407., 6. 129, the former of which passages Virg. had in his mind, as appears from his use of ‘demens.’
 Laetor is coupled closely with ‘memini,’ ‘ve’ after the negative having the same power as ‘que,’ as it was not that he did not remember the war, but that he took no pleasure in the remembrance. This doubtless helps the construction, though Virg. probably thought also of such expressions as εὐδαιμονίζειν τινὰ τύχης (πώματος ἥσθη, Soph. Phil. 715). See on v. 73 above. With the sentiment generally comp. 2. 3, and contrast 1. 203.
 Two of Ribbeck's cursives have ‘portastis,’ which was read by the older editions. Comp. 4. 598 note. One MS., the Parrhasian, which is frequently interpolated, has ‘desertis,’ doubtless, as Heyne remarks, for ‘defertis.’
 Vertite ad seems to combine the notion of προστρέπεσθαι, gifts offered in supplication (comp. Tac. A. 3. 5, “versi ad caelum ac deos”), with that of changing the direction. Serv. distinguishes ‘stetimus tela aspera contra’ from ‘contulimus manus’ as distant from hand-to-hand fighting, which is possible (see on v. 284), but not necessary.
 The meaning of ‘in clipeum adsurgat’ is not clear. The choice seems to lie between rising against the shield of the enemy, so as to level the spear over it (Cerda comp. Il. 12. 404: comp. also Il. 23. 820), and rising upon one's own shield, lifting it up in the rush of a hand-to-hand encounter, for which Wagn. comp. 12. 711 foll., 723 foll. The latter is supported by “consurgit in ensem” 9. 749., 12. 729, “adsurgentis dextra” 10. 797. Stat. Ach. 1. 485 (quoted by Gossrau on 9. 749) has “Odrysiam Gradivus in hastam Surgeret,” evidently of rising to throw or charge with the lance, which may perhaps show that he understood Virg. here in this way. In Il. 5. 297 foll. Aeneas leaps down from his car with shield and lance to protect Pandarus against Diomede. ‘Turbine’ of the wind of the lance, as in 1. 45 of the wind of the thunderbolt.
 Imitated, as Cerda remarks, from Il. 2. 371 foll., where Agamemnon says that if he had ten such counsellors as Nestor, Troy would soon fall. ‘Praeterea’ = “praeter Aeneam.” “Tibi uni concedam, praeterea nemini” Cic. Fam. 4. 3. It is hardly worth while considering whether Virg. meant two exclusive of Aeneas and Hector, or two exclusive of Aeneas. He doubtless expressed himself loosely, the mention of Hector v. 289 being an after-thought. “Idaeos campos” 7. 222.
 So 2. 193, “Ultro Asiam magno Pelopea ad moenia bello Venturam, et nostros ea fata manere nepotes.” ‘Inachius,’ an epithet of Argos (7. 372), is apparently extended to the Grecian cities generally, unless we prefer saying that Virg. is thinking of Argos and Mycenae, or that the pl. is used loosely.
 Dardanus, the Trojan, as in 4. 662, here for the Trojans generally (Heyne). There would however be some force in referring it to the founder of the Trojan race himself, and supposing him poetically to lead his descendants against Greece. ‘Versis fatis’ abl. abs., i. q. “versa fata.” Comp. “conversa numina” 5. 466.
 Quidquid cessatum est may be a clause standing for an acc. of duration, or we may say that Virg. intended to finish the sentence in some other way, so as to give the sense ‘was due to Hector and Aeneas.’ ‘Durae,’ difficult to take. Contrast Hor. 2 Od. 4. 10 foll., “ademptus Hector Tradidit fessis leviora tolli Pergama Graiis.” For ‘cessatum’ Med. (second reading) has ‘certatum,’ apparently a conj. of Apronianus, and certainly inferior.
 Manu abl. instr., as if it had been “manu retardata est.” That which Turnus 9. 155 naturally attributed to Hector alone, Diomed attributes to Hector and Aeneas jointly. Hom. classes them together more than once as the bravest of the Trojans, Il. 6. 77 foll., 17. 513.
 The elder Seneca (Suasor. 2, p. 23), after quoting a line from Abronius Silo, “belli mora concidit Hector,” goes on to say, “Notate prae ceteris quanto decentius Vergilius dixerit hoc, quod valde erat celebre Quidquid—haesit. Messala aiebat hic Vergilium debuisse desinere: quod sequitur et—annum explementum esse. Maecenas hoc etiam priori comparabat.” Heyne agrees with Messala: but Wagn. rightly makes allowance for Virg.'s love of variety. ‘Vestigia rettulit’ is to be interpreted by ‘haesit,’ not of actual retreat, but of slipping back on being pulled up, like “Obstipuit retroque pedem cum voce repressit” 2. 378. To explain it with Serv. from 2. 169, “retro sublapsa referri Spes Danaum,” would be to introduce a new metaphor.
 The first hint of Aeneas' reputation for piety seems to be in Il. 20. 298, where Poseidon says of him κεχαρισμένα δ᾽ αἰεὶ Δῶρα θεοῖσι δίδωσι τοὶ οὐρανὸν εὐρὺν ἔχουσιν. In Il. 24. 66 foll., however, Hector has, if anything, a preference given to him in this respect by Zeus himself: so that it may be doubted whether the Homeric Diomed would have made any such distinction as that here expressed. Aeneas' filial piety is not noticed in the Iliad, though it may have formed part of the cycle of tradition, as it was supposed to have been exhibited chiefly at the taking of Troy. See Introd. to Book 2. ‘Coire’ and ‘dextrae’ are both words equally applicable to peace and war, so that ‘in foedera’ has a sort of tacit opposition to “in proelia” or some such expression in Virg.'s mind. “In amicitiam coeant et foedera iungant” 7. 546.
 Rom. and some others have ‘optime regum,’ an obvious error, introduced from v. 353: it was however the old reading. In ‘regis’ following ‘rex’ there is the same sort of official accuracy as in 8. 17., 9. 369.
 ‘In hearing Diomed's reply, you have at the same time heard his opinion on the war.’ We should have expected ‘de magno’ (which Peerlkamp actually conj. for ‘sit magno’), but Virg. seems to have chosen the abl. of circumstance for variety's sake. ‘What advice he has to offer us under our heavy war.’ This seems better than to emphasize ‘magno,’ ‘seeing the war is so great,’ or to make it dat., ‘what the war thinks’ = ‘how the war is to be carried on,’ interpretations suggested by Forb.
[296-335] ‘After the ambassadors' speech, Latinus proposes to the assembly to assign a part of the kingdom to the Trojans, or to fit out a fleet for them to go elsewhere; the proposals being made by an embassy charged with presents to Aeneas.’