“Devenere locos” 1. 365. Homer's
Elysium is not part of the infernal
regions, but a separate region (Od. 4.
563 foll.), which later legends (Hesiod,
Works 170 foll., Pind. Ol. 2. 61 foll., fr. 95,
Bergk, ed. 1) developed into the ‘Islands
of the Blest.’ Virg. has not copied Hom.,
whose description speaks of a place where
there are no storms but always cooling
zephyrs: from the second of Pindar's
elaborate pictures he has taken the pursuits
of the heroes. Perhaps the nearest
parallel to his language is Aristoph. Frogs
154 foll., where Heracles says to Dionysus
that after passing the place of punishment
he will come to a region described as follows:
“ἐντεῦθεν αὐλῶν τίς σε περίεισιν πνοή,
ὄψει τε φῶς κάλλιστον, ὥσπερ ἐνθάδε,
καὶ μυρ᾽ῥινῶνας, καὶ θιάσους εὐδαίμονας
ἀνδρῶν γυναικῶν, καὶ κρότον χειρῶν πολύν.
 Fortunatae Insulae is the Latin equivalent of μακάρων νῆσοι. Plaut. however gives it “fortunatorum insulae:” see Forc. ‘fortunatus.’ With the transference of the epithet we may comp. “lugentes campi” above v. 441. “Apparet divum numen sedesque quietae” Lucr. 3.18, a passage which Virg. may have been thinking of, as the next verse seems to show.
 Virg. copies Hom.'s description of Olympus, Od. 6. 44, ἀλλὰ μάλ᾽ αἴθρη Πέπταται ἀνέφελος, λευκὴ δ᾽ ἐπιδέδρομεν αἴγλη, the preceding lines of which resemble those on Elysium referred to on v. 638, and so would naturally be associated with them by a reader of the Odyssey. It is this passage which Lucr. imitates 3. 18 foll.; and Virg. would seem here again to be somewhat indebted to his language, “semperque innubilus aether Integit, et large diffuso lumine ridet.” ‘Vestit’ is from another passage in Lucr. 2.148, “Convestire sua perfundens omnia luce,” of the sun, and perhaps from Cic. Arat. 60, “Quem cum perpetuo vestivit lumine Titan.” ‘Largior vestit’ is meant to express μάλα πέπταται, the transparency of the ether being conceived of as a superabundant fluid which permeates every part of the region. ‘Lumine purpureo’ as plainly is meant to render λευκὴ αἴγλη, ‘purpureus’ having its Roman sense of dazzling. See on E. 5. 38, G. 4. 373. Wordsworth, imitating this passage, talks in his Laodamia of “fields invested with purpureal gleams,” as Gray, imitating “lumen iuventae purpureum” 1. 591, talks of “purple light of love.” The expression in each case would convey a false notion to a reader unacquainted with Latin, being only defensible if understood not as a translation of Virg.'s epithet, but as a quotation of it. ‘Largior’ is a predicate, and so is coupled with ‘lumine purpureo,’ both qualifying ‘vestit:’ see on 5. 498. For ‘campos’ fragm. Vat. a m. p., Rom., and Gud. a m. s. have ‘campus,’ Pal. a m. p. ‘campis.’
 They have a sun and stars of their own, distinct from those in the upper world. Pind. fr. l. c. apparently says that the sun visits them when it leaves us: elsewhere however he gives them a sun which shines night and day alike (Ol. l. c.). Wakef. ingeniously but erroneously makes ‘sidera’ nom., ‘sua’ and ‘suum’ being used reciprocally; ‘there is a new sun, with stars of its own, new stars, with a sun of their own.’ ‘Norunt’ = ‘notos habent,’ like ‘novere’ G. 4. 155, though there perhaps rather more is intended: see note.
 From Pind. fr. l. c. καὶ τοὶ μὲν ἱππείοις γυμνασίοις, τοὶ δὲ φορμίγγεσι τέρπονται. ‘Palaestra’ may be either the place or the exercise: but the former seems more likely, both on account of ‘in,’ and because ‘graminea palaestra’ would be a little harsh, though not unexampled, for ‘palaestra in gramine.’ So perhaps “agresti palaestrae” G. 2. 642 may mean the place.
 Translated from Od. 8. 264, πέπληγον δὲ χορὸν θεῖον ποσίν, where however it would seem from a preceding line, v. 260, that χορόν is the place of dancing, not the dance, though the other construction, χορόν as a cogn. acc., would be sufficiently idiomatic. The sense of clapping the hands, though the most usual sense of ‘plaudo,’ does not seem to be the primary one, at least if we may argue from its derivation “plaustrum” (comp. “claudo,” “claustrum”), which may either be, as Scaliger ap. Forc. thinks, “a plaudendo terram,” or perhaps ‘a thing hammered,’ as κροτεῖν is used of applause, of hammering, and of the rattle of a car (ὄχεα κροτέοντες Il. 15. 453), ὄχεα κροτητά Soph. El. 714 being taken by some ‘hammered,’ by others ‘made to rattle.’ Κρότος ποδῶν is used of dancing Eur. Heracl. 583, Tro. 746, like ‘pedibus plaudunt’ here, and the parallel may be completed by comparing κροτητὰ μέλη Soph. Thamyris fr. 221, music struck out by the πλῆκτρον, with the similar action expressed by ‘plaudere’ in the Ciris v. 179, “Non Libyco molles plauduntur pectine telae.” ‘Carmina dicunt’ G. 1. 350, where it is mentioned in connexion with dancing.
 Orpheus was one of the mythical fathers of song, and his name was associated with revelations about the lower world, supposed to be preserved by secret societies (Dict. M. Orpheus), so that he is naturally made the harper who plays while the blessed spirits dance and sing. He is called ‘sacerdos,’ as in Hor. A. P. 391 he is called “sacer interpresque deorum.” The long robe was characteristic of musicians, as Cerda shows, comp. Prop. 3. 23. 16, “Pythius in longa carmina veste sonat” (of the statue of Apollo in the Palatine temple), and also Hor. A. P. 215, Ov. F. 6. 654, 688, where the long robes of the ‘tibicines’ are mentioned and accounted for. ‘Cum veste’ above v. 359. Elsewhere we have ‘in veste,’ as 12. 169, “puraque in veste sacerdos.”
 This line is unusually difficult, as owing to Virg.'s love of artificial expressions we cannot be sure whether he has used particular words in their natural or in their accommodated sense. ‘Obloqui’ is evidently intended i. q. ἀντιφωνεῖ, though its usual sense is not so much that of answering as of interruption or contradiction: see Forc. ‘Septem discrimina vocum’ is evidently intended to express the seven notes produced by the seven strings of the lyre, known as the Heptachord of Orpheus. But whether in this line Orpheus is intended to produce the notes with his voice in singing, or with the lyre as he plays, is not clear. ‘Obloquitur’ seems as if it may be intended to translate ἀντιφωνεῖ, in the sense of ‘accompanies,’ as though the verb in Greek does not appear to be technically applied to music, ἀντίφωνος is used in Aristot. Probl. § 19 &c. of notes one of which is the octave of the other. Virg. is not likely to have intended this restricted sense, but he may have taken the general notion of accompanying from the special one of accompanying in a higher or lower octave. There is a further difficulty about ‘numeris,’ which may be either abl. or dat, meaning in the former case ‘by means of numbers,’ the notes either of the voice or of the lyre, according as we understand ‘obloquitur,’ in the latter case ‘to the numbers,’ either the notes of the harp, the notes of his own singing, or the beat of the dancers. In my first edition I had understood the present line of singing, and the next of playing, making ‘numeris’ the dat. and explaining it of the dancers. Now I return to the more ordinary opinion, taking ‘numeris’ dat. of the dancing, but supposing that Orpheus only plays, and so regarding the next line as epexegetical.
 Fragm. Vat. has a curious variant, ‘dictis’ for ‘digitis.’ Markland ingeniously conj. ‘fidem:’ but ‘eadem’ is supported by the context, the meaning apparently being that Orpheus accompanies himself and the dancers on the harp. ‘Pectine:’ “Though the Romans adopted into their own language the Greek word ‘plectrum,’ they used the Latin ‘pecten’ to denote the same thing, not because the instrument used in striking the lyre was at all like a comb in shape and appearance, but because it was held in the right hand and inserted between the stamina of the lyre as the comb was between the stamina of the loom.” Dict. A. ‘Tela.’
 Comp. v. 580, “Hic genus antiquum Terrae, Titania pubes.” Cerda remarks that Virg. seems intentionally to begin the enumeration of the respective inmates of Tartarus and Elysium in the same way. ‘Pulcherrima proles’ Heyne thinks may be borrowed from the mention of Ganymede in a similar passage, Il. 20. 231 foll. Ἶλός τ᾽ Ἀσσάρακός τε καὶ ἀντίθεος Γανυμήδης, Ὃς δὴ κάλλιστος γένετο θνητῶν ἀνθρώπων. Heyne comp. Hesiod, Works 158 foll. δικαιότερον καὶ ἄρειον, Ἀνδρῶν ἡρώων θεῖον γένος, οἳ καλέονται Ἡμίθεοι προτέρῃ γενεῇ κατ᾽ ἀπείρονα γαῖαν.
 Imitated, as Cerda remarks, from Catull. 62 (64). 22, “O nimis optato saeclorum tempore nati Heroes, salvete, deum genus.” Forb. thinks ‘melioribus annis’ refers specially to the days of Troy's prosperity, but the general reference to a happier divine foretime is more probable.
 Comp. G. 3. 35, 36. Here Dardanus seems to be mentioned as a descendant of Teucer: but it is not easy to say which of the legends about them Virg. followed. See on 3. 107, 108, 168.
 Arma coupled with ‘currus’ as in v. 485, 1. 16, 17. ‘Virum’ seems to go with both, as “arma virum” are combined 1. 119., 9. 777: but it might be constructed with ‘inanis,’ like “caelestium inanes” Pers. 2. 61. ‘Inanis’ however seems to mean ghostly, as it is a constant epithet of the dead: see Forc. ‘Mirantur’ is found in fragm. Vat., Med., &c., and is perhaps right, as there is not much force in Pierius' objection that the Sibyl was not likely to wonder at what she had seen before. Pierius however says that he found ‘miratur’ in all his oldest MSS. but one: it is the reading of Pal. and Rom.; and the sing. may have been altered by some one who supposed ‘currus’ to be the nom., a very common source of error: see Wagn. Q. V. 8. 2. a. It is not clear whether the cars are represented as empty or as filled by their ghostly riders. ‘Stant terra defixae hastae’ is in favour of the former view, meaning apparently, like its prototype, Il. 3. 135, quoted on the next line, that the warriors are enjoying relaxation: on the other hand ‘quae gratia currum’ &c. seems to show that martial exercises are still going on. Perhaps we may say that the present line refers to one class who are exercising, the next to another, who have done their exercise and are now feeding or grooming their horses, the same distinction which seems to be drawn in the words ‘quae gratia,’ &c. and ‘quae cura,’ &c. But the difficulty may be best removed by regarding the sentence ‘quae gratia—repostos’ as a conclusion drawn from the picture actually presented: the warriors have their spears and horses near them, to use when they please.
 παρὰ δ᾽ ἔγχεα μακρὰ πέπηγεν Il. 3. 135, of the armies resting before the combat of Paris and Menelaus. For ‘terra’ fragm. Vat. and two of Ribbeck's cursives have ‘terrae,’ as in G. 2. 290: but it is not likely that Virg. should have preferred the jingle, while the clerical error is natural enough.
 Campum is restored by Wagn. from Med., Pal., fragm. Vat. &c. for ‘campos’ (Rom.). The MSS. vary between ‘currum’ and ‘curruum,’ inclining however to the contracted form, which is also supported by Priscian 7798 P, and by Serv. Priscian says that ‘curruum’ might be scanned as a hypermeter. Wagn. prefers accounting for it by crasis: but it is difficult to see why that should be resorted to when the next verse begins with a vowel, though it is of course admissible in such cases as v. 33 above. Meanwhile it seems safest to recall ‘currum.’ ‘Gratia’ with gen. of the thing or person wherein pleasure is felt, like χάρις, 7. 402.
 Nitentis perhaps with ‘pascere:’ comp. E. 6. 4. note. ‘Nitidi’ is used similarly of sleek horses 7. 275. For the care taken by the Homeric warriors of their horses comp. Il. 8. 185 foll. ‘Cura pascere:’ see on G. 1. 213.
 Sequitur, as we should say, follows them beyond the grave.
 Per herbam, feasting on the grass, like the Trojans 1. 214., 3. 221 foll., 7. 109. The object is to give a picture of natural golden-age simplicity, with which we may contrast the elaboration of art in the infernal banquet above, v. 603. Compare Milton's language about Mammon, Par. Lost, Book 1.
 Vescentis without a case, as in Livy 37. 20, “pars vescentes sub umbra,” comp. by Forb. The word has been objected to as homely, but ‘vescitur’ is used of a sacrificial feast 8. 182. The Paean at banquets is as old as Hom. Il. 1. 473, the words of which seem copied here, καλὸν ἀείδοντες Παιήονα. ‘Choro’ in a band or chorally. If the singers are the same as the banqueters, they can hardly be dancing.
 The scent of the bay has been mentioned E. 2. 54 foll. ‘Lauri nemus,’ not unlike “picis lucos” G. 2. 438. ‘Superne’ is rightly understood by Henry and Ladewig to mean ‘in the upper world.’ The river is supposed to take its rise in the Elysian fields, just as in G. 4. 366 foll, we are told that Aristaeus saw the subterranean sources of all the rivers in the world, Eridanus included. The Eridanus was the subject of various myths, being placed in different parts of the globe, and turned into a constellation. The notion of its underground source doubtless comes from the fact, noticed by Heyne, that the Po, with which the Romans identified it, not far from its source, flows underground for two miles. ‘Plurimus’ &c., will then refer to its course through the upper world, not through the shades.
[660-678] ‘Seeing a crowd of worthies with Musaeus among them, the Sibyl inquires where Anchises is to be found. Musaeus replies that the blessed spirits have no certain habitation, but offers to guide them; and so they ascend a slope.’
 Manus—passi like “genus . . . pubes . . . deiecti” above vv. 580 foll. The latter part of the line is repeated 7. 182. Those who have been wounded are named rather than the slain, as all patriotic warriors are meant to be included.
 “Dum vita manebat” above v. 608. Nonius quotes the words with “maneret,” and Serv. explains “dum in communione vitae versarentur.” The strict use of ‘dum’ with the subj., for which see note on G. 4. 457, is not adhered to by postAugustan writers, and so is not to be looked for among the old grammarians.
 Pii = “casti.” A comment on the epithet as applied to poets is furnished by the well-known passage Hor. 2 Ep. 1. 126 foll. ‘Phoebo digna locuti’ is generally explained of their power of song, but it may also refer to their purity. Serv. thinks prophets are meant, and explains ‘Phoebo’ &c. of their truthfulness; but the presence of Musaeus shows that poets formed part of the fraternity, and Virg. would hardly have so little feeling for his order as to pass them over in silence.
 Vitam, not their life, but life generally, a usage common in Lucr. e. g. 6. 3, “Et recreaverunt vitam (Athenae).” The whole of the latter part of Lucr.'s 5th Book is in fact a commentary on this line. See also G. 1. 133. ‘Per artis’ i. q. ‘artibus,’ as “per artem” G. 1. 122 i. q. ‘arte.’
 A more general description of the benefactors of the human race. For ‘alios’ fragm. Vat. a m. pr., Pal., Rom., Med., Gud., and others give ‘aliquos,’ which is supported also by Serv.; by Donatus however is said to be for ‘alios,’ which is supported by corrections in fragm. Vat. and in two of Ribbeck's cursives, and is infinitely preferable. Wagn. accounts for the corruption by the proximity of ‘quique.’ ‘Memores’ of grateful recollection 4. 539. ‘Merendo’ by their services. Germ. comp. Prop. 5. 11. 101, “sim digna merendo.”
 Circumfusos, spread all about the place which Aeneas and the Sibyl were entering.
 Musaeus is the mythical father of poets, as Orpheus of singers. The tallness of his stature is described, rather unreasonably, in words copied from Hom.'s description of Ajax Il. 3. 227. ‘Medium turba hunc habet,’ a poetical variety for “hic turbae medius est.” Some of the early critics accused Virg. of jealousy in not rather naming Homer than Musaeus, as if a sense of obligation ought to have made him ready to encounter an anachronism. Silius has been able to repair the omission, introducing Homer into his Elysian fields, which by-the-bye he speaks of incongruously as “Stygia umbra,” in lines (13. 778 foll.) which have been excellently, though not quite accurately, translated by Chapman in the introductory verses to his Iliad.
 Amnis may be only a poetical plural: but Virg. apparently means Aeneas to have crossed three of the infernal rivers, though he only mentions the passage of one: see on v. 295. ‘Tranavimus’ of crossing in a boat, like “innare” v. 134. Ribbeck reads ‘transnavimus’ from Gud. a m. s., two other cursives, and the MSS. of Non. and Paulus. The original reading of Med. was ‘transnavibus.’
 Atque seems to mean immediately.
 ‘Riparum toros’ like “viridante toro herbae” 5. 388. ‘Recentia,’ an epithet transferred to meadows from the streams that freshen them: see on v. 635. For ‘rivis’ Rom. and another MS. give ‘silvis.’
 Sistam implies, what we should also infer from the context, that Musaeus leaves them when they have mounted the slope and see the way on the other side.
 Linquunt, Aeneas and the Sibyl: see on v. 676.
[679-702] ‘Anchises is in the valley beyond, surveying his future posterity, spirits that are hereafter to take flesh. He welcomes Aeneas with joy and surprise. Aeneas attempts to embrace him, but in vain.’