Here as elsewhere the invocation indicates that the poet is awaking a louder strain. As Germ. remarks, the hint is from Il. 16. 112, ἕσπετε νῦν μοι, Μοῦσαι Ὀλύμπια δώματ᾽ ἔχουσαι, Ὅππως δὴ πρῶτον πῦρ ἔμπεσε νηυσὶν Ἀχαιῶν.
 The grounds for believing the event are old (‘fides’ as in Ov. 1. ex Pont. 5. 32); as we should say, the evidence is lost in the past, but the fame is perpetual. “Prisca fides” in a different sense 6. 878.
 Petenti 4. 127.
 Domito Olympo refers to services rendered by Cybele to Jupiter, enabling him to become master of heaven, either, as Serv. thinks, in saving him from his father, who sought to devour him, or, as Heyne suggests, in helping him against the Titans, or both. Heyne's own interpretation, understanding ‘domito Olympo’ ‘in that thou art the master of heaven, and as such able to do all I wish,’ would be flat. He objects that the help given by Cybele was of too old a date to be appealed to at the time of the taking of Troy. But the whole history of the gods as gods belongs to a ‘divine foretime,’ and the events affecting them after the heroic age has begun are comparatively few, so that they naturally live as it were upon the past, and refer to things which happened long ago as if they were still fresh.
 Ribbeck asterizes this line, supposing that Virg. intended it as an alternative to vv. 86, 87. Heyne had asterized vv. 86, 87 on similar grounds, thinking the mention of a grove in the citadel absurd, and inconsistent with the building of the fleet on Ida. Wagn. defends all three, making v. 85 an independent sentence: ‘I have a pine-forest; in this stood a grove of pitch-trees and maples, which I allowed Aeneas to use,’ ‘arce summa’ being understood with Serv. of Gargarus, the summit of Ida. As in 6. 743, 744, the truth seems to lie between the two views. We could not get rid of any part of the passage without sacrificing something: on the other hand we cannot say that in its present state it is altogether coherent. Virg. would doubtless have altered it had he lived to complete his poem: but we cannot point out the precise change which he would have made. Meantime Wagn. appears right in his view of the grammatical structure of the whole, breaking it up into two sentences, as there would be awkwardness in constructing ‘pinea silva’ in apposition to ‘lucus,’ or in making one the predicate, the other the subject. It is better, at the risk of a little harshness, to understand ‘est’ with ‘mihi’ than to make ‘dilecta’ the verb, with Ruhkopf. “Multos servata per annos” 7. 60.
 In arce summa would most naturally refer to the Trojan acropolis: comp. 1. 441, “lucus in urbe fuit media,” and the story of the bay-tree 7. 61, “inventam primas cum conderet arces,” as also the story of the olive in the acropolis of Athens. Where the passage is assumed to be unfinished, we cannot argue from the context: but it would be undoubtedly possible to understand ‘arce’ of the mountain, and v. 92 may be pleaded for this. It is a question of probabilities, and one that from the nature of the case must remain to some extent open. Virg. may have intended to make Aeneas get his timber from a sacred grove in the citadel, which might possibly have been conceived of as remaining unburnt, like the Athenian olive, after the sack of the city: but this is mere conjecture without data. ‘Quo’ refers to ‘lucus.’
 Urguet was the reading before Heins., but it has only the support of two MSS. of no authority. Heyne recalled it on internal grounds, regarding ‘anxius angit’ as “inepta allitteratio.” Wagn. well replies that the alliteration is intentional, expressing rhetorically the intensity of the anxiety, for which he comp. Cic. De Orat. 1. 1, “maxumas moles molestiarum,” and supported by Lucr. 3.993, “exest anxius angor,” while the archaic turn of the expression suits the speech of a primeval goddess. He also quotes Cic. Tusc. 4. 12 to show that ‘anxius angit’ is not a mere tautology: “Differt anxietas ab angore: neque enim omnes anxii qui anguntur aliquando, nec qui anxii semper anguntur.”
 ‘Neu cursu’ Med., Rom., Gud. corrected, ‘ne cursu’ Pal., Gud. originally, and another of Ribbeck's cursives. There seems no internal reason for choosing between them, though Heyne and Ribbeck prefer ‘ne.’ ‘Neu’ is of course equivalent to “ut neque,” and as such may be used with the first as well as with a subsequent clause, as in v. 42 above. ‘Quassatae’ is constructed with ‘vincantur,’ but from its position it has the effect of a second verb. “Quassatam ventis classem” 1. 551. ‘Cursu’ of a voyage 6. 338. Virg. may be thinking of Hom.'s language about the ships of the Phaeacians Od. 9. 562, οὐδέ ποτέ σφιν Οὔτε τι πημανθῆναι ἔπι δέος οὔτ᾽ ἀπολέσθαι.
 Torquet of the revolutions of the heaven and heavenly bodies, which Jupiter is supposed to guide: comp. 4. 269, 482. “Sidera mundi” Lucr. 2.328 &c. Comp. generally Id. 5. 1209, “ne quae forte deum nobis inmensa potestas Sit, vario motu quae candida sidera verset.”
 Ordinarily the fates are said to call men, who have to follow their bidding: comp. 10. 472., 5. 709: here Cybele, in attempting to change destiny, is said to call it to leave its path. Comp. G. 2. 52, where ‘vocare’ is used of an attempt to cultivate trees. “‘Istis’ utrum precibus an navibus?” Serv. Most of the editors say the latter, Peerlkamp the former. With the former comp. “his monuit nos” Juv. 11. 114, with the latter “his moenia quaere” 2. 294.
 Inmortale fas seems to be i. q. “id quod fas est inmortalibus,” that which divine law allows to the immortals. So ‘mortalis’ is used rather widely as an epithet when the meaning is not that the thing is mortal but that it belongs to a mortal (comp. E. 8. 35, G. 3. 319). ‘Fas’ may be spoken of as binding the gods, as in 4. 113., 5. 800., 8. 397.
 Habeant is explained by ‘petis:’ ‘dost thou ask that they should have?’ ‘Certus:’ Jupiter puts a second objection, also in the form of an allegation of incompatibility. Dangers are uncertain ex vi termini: and if Aeneas, being a man, has to encounter them, that he should be assured against them is a contradiction. It may be said that this contradiction is incurred already, as Aeneas knows that he shall reach Italy: but whatever may be the worth of the knowledge, it does not prevent him from contemplating the possibility of drowning, 1. 94 foll. ‘Lustret,’ traverse, like “lustrandum navibus aequor” 3. 385.
 Immo: Jupiter as it were amends the proposition, so that it is not, as Hand thinks, i. q. “at.” ‘Defunctae’ 6. 83. ‘Finem’ is explained by the context, the end of the voyage. ‘Portus tenebunt’ 1. 400.
 Olim is rightly connected by Wagn. with what precedes, not with what follows. ‘Undis:’ Serv. mentions another reading ‘undas,’ which is the more usual construction in Virg., and might be supported by 5. 689: but it is found only in one or two inferior copies. Comp. 11. 702 note. One ship was lost in the storm off Africa (1. 584), four were burnt in Sicily (5. 699), so that Aeneas must have landed with fifteen, the original number having been twenty (1. 381). Two of these had gone with Aeneas to Pallanteum, 8. 79; thirteen consequently remained.
 Rom. has ‘aut’ for ‘et.’ ‘Pectore,’ as appearing with their breasts out of the water, “nutricum tenus exstantes e gurgite cano” Catull. 62 (64). 18. quoted by Gossrau. So 10. 212 of Triton, “spumea semifero sub pectore murmurat unda.”
 Virg. has mixed up the nod which pledges Jupiter (Il. 1. 525) and the oath by the Styx which binds the gods (see passages referred to on 6. 324). ‘Stygii per flumina fratris’ like “Corythi Tyrrhena ab sede” 7. 209, ‘Stygii’ really belonging to ‘flumina.’
 Torrens is applied to a violent river from the connexion of the notions of heat and vehement motion (comp. “aestus”). Here advantage is taken of the double meaning of the word to apply it to the infernal river, which is described in language taken partly from Acheron (6. 296), which is a violent muddy stream, partly from Phlegethon (6. 550), which is a river of fire. Comp. Plato Phaedo p. 111, where the mixture of fire and mud is illustrated from the eruptions of Aetna.